Merge pull request #2319 from bagerard/prepare_0_20_0

Bump version to 0.20.0 and update CHANGELOG accordingly

.gitignore

@@ -1,4 +1,25 @@
-*.pyc
-.*.swp
+!.gitignore
+*~
+*.py[co]
+.*.sw[po]
+.cache/
+.coverage
+.coveragerc
+.env
+.idea/
+.pytest_cache/
+.tox/
+.eggs/
+*.egg
 docs/.build
 docs/_build
+build/
+dist/
+mongoengine.egg-info/
+env/
+.settings
+.project
+.pydevproject
+htmlcov/
+venv
+venv3

.landscape.yml

@@ -0,0 +1,17 @@
+pylint:
+    disable:
+        # We use this a lot (e.g. via document._meta)
+        - protected-access
+
+    options:
+        additional-builtins:
+            # add long as valid built-ins.
+            - long
+
+pyflakes:
+    disable:
+        # undefined variables are already covered by pylint (and exclude long)
+        - F821
+
+ignore-paths:
+    - benchmark.py

.pre-commit-config.yaml

@@ -0,0 +1,12 @@
+fail_fast: false
+repos:
+    - repo: https://github.com/ambv/black
+      rev: 19.10b0
+      hooks:
+        - id: black
+    - repo: https://gitlab.com/pycqa/flake8
+      rev: 3.8.0a2
+      hooks:
+        - id: flake8
+          additional_dependencies:
+            - flake8-import-order

.travis.yml

@@ -0,0 +1,107 @@
+# For full coverage, we'd have to test all supported Python, MongoDB, and
+# PyMongo combinations. However, that would result in an overly long build
+# with a very large number of jobs, hence we only test a subset of all the
+# combinations.
+# * Python3.7, MongoDB v3.4 & the latest PyMongo v3.x is currently the "main" setup,
+# Other combinations are tested. See below for the details or check the travis jobs
+
+# We should periodically check MongoDB Server versions supported by MongoDB
+# Inc., add newly released versions to the test matrix, and remove versions
+# which have reached their End of Life. See:
+# 1. https://www.mongodb.com/support-policy.
+# 2. https://docs.mongodb.com/ecosystem/drivers/driver-compatibility-reference/#python-driver-compatibility
+#
+# Reminder: Update README.rst if you change MongoDB versions we test.
+
+language: python
+dist: xenial
+python:
+- 3.5
+- 3.6
+- 3.7
+- 3.8
+- pypy3
+
+env:
+  global:
+    - MONGODB_3_4=3.4.17
+    - MONGODB_3_6=3.6.12
+    - MONGODB_4_0=4.0.13
+
+    - PYMONGO_3_4=3.4
+    - PYMONGO_3_6=3.6
+    - PYMONGO_3_9=3.9
+    - PYMONGO_3_10=3.10
+
+    - MAIN_PYTHON_VERSION=3.7
+  matrix:
+    - MONGODB=${MONGODB_3_4} PYMONGO=${PYMONGO_3_10}
+
+matrix:
+  # Finish the build as soon as one job fails
+  fast_finish: true
+
+  include:
+  - python: 3.7
+    env: MONGODB=${MONGODB_3_6} PYMONGO=${PYMONGO_3_6}
+  - python: 3.7
+    env: MONGODB=${MONGODB_3_6} PYMONGO=${PYMONGO_3_9}
+  - python: 3.7
+    env: MONGODB=${MONGODB_3_6} PYMONGO=${PYMONGO_3_10}
+  - python: 3.8
+    env: MONGODB=${MONGODB_4_0} PYMONGO=${PYMONGO_3_10}
+
+install:
+  # Install Mongo
+  - wget http://fastdl.mongodb.org/linux/mongodb-linux-x86_64-${MONGODB}.tgz
+  - tar xzf mongodb-linux-x86_64-${MONGODB}.tgz
+  - ${PWD}/mongodb-linux-x86_64-${MONGODB}/bin/mongod --version
+  # Install Python dependencies.
+  - pip install --upgrade pip
+  - pip install coveralls
+  - pip install pre-commit
+  - pip install tox
+  # tox dryrun to setup the tox venv (we run a mock test).
+  - tox -e $(echo py$TRAVIS_PYTHON_VERSION-mg$PYMONGO | tr -d . | sed -e 's/pypypy/pypy/') -- -a "-k=test_ci_placeholder"
+
+before_script:
+  - mkdir ${PWD}/mongodb-linux-x86_64-${MONGODB}/data
+  - ${PWD}/mongodb-linux-x86_64-${MONGODB}/bin/mongod --dbpath ${PWD}/mongodb-linux-x86_64-${MONGODB}/data --logpath ${PWD}/mongodb-linux-x86_64-${MONGODB}/mongodb.log --fork
+  # Run pre-commit hooks (black, flake8, etc) on entire codebase
+  - if [[ $TRAVIS_PYTHON_VERSION == $MAIN_PYTHON_VERSION ]]; then pre-commit run -a; else echo "pre-commit checks only runs on py37"; fi
+  - mongo --eval 'db.version();'    # Make sure mongo is awake
+
+script:
+  - tox -e $(echo py$TRAVIS_PYTHON_VERSION-mg$PYMONGO | tr -d . | sed -e 's/pypypy/pypy/') -- -a "--cov=mongoengine"
+
+after_success:
+- - if [[ $TRAVIS_PYTHON_VERSION == $MAIN_PYTHON_VERSION ]]; then coveralls --verbose; else echo "coveralls only sent for py37"; fi
+
+notifications:
+  irc: irc.freenode.org#mongoengine
+
+# Only run builds on the master branch and GitHub releases (tagged as vX.Y.Z)
+branches:
+  only:
+  - master
+  - /^v.*$/
+
+# Whenever a new release is created via GitHub, publish it on PyPI.
+deploy:
+  provider: pypi
+  user: the_drow
+  password:
+    secure: QMyatmWBnC6ZN3XLW2+fTBDU4LQcp1m/LjR2/0uamyeUzWKdlOoh/Wx5elOgLwt/8N9ppdPeG83ose1jOz69l5G0MUMjv8n/RIcMFSpCT59tGYqn3kh55b0cIZXFT9ar+5cxlif6a5rS72IHm5li7QQyxexJIII6Uxp0kpvUmek=
+
+  # Create a source distribution and a pure python wheel for faster installs.
+  distributions: "sdist bdist_wheel"
+
+  # Only deploy on tagged commits (aka GitHub releases) and only for the parent
+  # repo's builds running Python v3.7 along with PyMongo v3.x and MongoDB v3.4.
+  # We run Travis against many different Python, PyMongo, and MongoDB versions
+  # and we don't want the deploy to occur multiple times).
+  on:
+    tags: true
+    repo: MongoEngine/mongoengine
+    condition: ($PYMONGO = ${PYMONGO_3_10}) && ($MONGODB = ${MONGODB_3_4})
+    python: 3.7

AUTHORS

@@ -0,0 +1,259 @@
+The PRIMARY AUTHORS are (and/or have been):
+
+Ross Lawley <ross.lawley@gmail.com>
+Harry Marr <harry@hmarr.com>
+Matt Dennewitz <mattdennewitz@gmail.com>
+Deepak Thukral <iapain@yahoo.com>
+Florian Schlachter <flori@n-schlachter.de>
+Steve Challis <steve@stevechallis.com>
+Wilson Júnior <wilsonpjunior@gmail.com>
+Dan Crosta https://github.com/dcrosta
+Laine Herron https://github.com/LaineHerron
+
+CONTRIBUTORS
+
+Derived from the git logs, inevitably incomplete but all of whom and others
+have submitted patches, reported bugs and generally helped make MongoEngine
+that much better:
+
+ * blackbrrr
+ * Florian Schlachter
+ * Vincent Driessen
+ * Steve Challis
+ * flosch
+ * Deepak Thukral
+ * Colin Howe
+ * Wilson Júnior (https://github.com/wpjunior)
+ * Alistair Roche
+ * Dan Crosta
+ * Viktor Kerkez
+ * Stephan Jaekel
+ * Rached Ben Mustapha
+ * Greg Turner
+ * Daniel Hasselrot
+ * Mircea Pasoi
+ * Matt Chisholm
+ * James Punteney
+ * TimothéePeignier
+ * Stuart Rackham
+ * Serge Matveenko
+ * Matt Dennewitz
+ * Don Spaulding
+ * Ales Zoulek
+ * sshwsfc
+ * sib
+ * Samuel Clay
+ * Nick Vlku
+ * martin
+ * Flavio Amieiro
+ * Анхбаяр Лхагвадорж
+ * Zak Johnson
+ * Victor Farazdagi
+ * vandersonmota
+ * Theo Julienne
+ * sp
+ * Slavi Pantaleev
+ * Richard Henry
+ * Nicolas Perriault
+ * Nick Vlku Jr
+ * Michael Henson
+ * Leo Honkanen
+ * kuno
+ * Josh Ourisman
+ * Jaime
+ * Igor Ivanov
+ * Gregg Lind
+ * Gareth Lloyd
+ * Albert Choi
+ * John Arnfield
+ * grubberr
+ * Paul Aliagas
+ * Paul Cunnane
+ * Julien Rebetez
+ * Marc Tamlyn
+ * Karim Allah
+ * Adam Parrish
+ * jpfarias
+ * jonrscott
+ * Alice Zoë Bevan-McGregor (https://github.com/amcgregor/)
+ * Stephen Young
+ * tkloc
+ * aid
+ * yamaneko1212
+ * dave mankoff
+ * Alexander G. Morano
+ * jwilder
+ * Joe Shaw
+ * Adam Flynn
+ * Ankhbayar
+ * Jan Schrewe
+ * David Koblas
+ * Crittercism
+ * Alvin Liang
+ * andrewmlevy
+ * Chris Faulkner
+ * Ashwin Purohit
+ * Shalabh Aggarwal
+ * Chris Williams
+ * Robert Kajic
+ * Jacob Peddicord
+ * Nils Hasenbanck
+ * mostlystatic
+ * Greg Banks
+ * swashbuckler
+ * Adam Reeve
+ * Anthony Nemitz
+ * deignacio
+ * Shaun Duncan
+ * Meir Kriheli
+ * Andrey Fedoseev
+ * aparajita
+ * Tristan Escalada
+ * Alexander Koshelev
+ * Jaime Irurzun
+ * Alexandre González
+ * Thomas Steinacher
+ * Tommi Komulainen
+ * Peter Landry
+ * biszkoptwielki
+ * Anton Kolechkin
+ * Sergey Nikitin
+ * psychogenic
+ * Stefan Wójcik (https://github.com/wojcikstefan)
+ * dimonb
+ * Garry Polley
+ * James Slagle
+ * Adrian Scott
+ * Peter Teichman
+ * Jakub Kot
+ * Jorge Bastida
+ * Aleksandr Sorokoumov
+ * Yohan Graterol
+ * bool-dev
+ * Russ Weeks
+ * Paul Swartz
+ * Sundar Raman
+ * Benoit Louy
+ * Loic Raucy (https://github.com/lraucy)
+ * hellysmile
+ * Jaepil Jeong
+ * Daniil Sharou
+ * Pete Campton
+ * Martyn Smith
+ * Marcelo Anton
+ * Aleksey Porfirov (https://github.com/lexqt)
+ * Nicolas Trippar
+ * Manuel Hermann
+ * Gustavo Gawryszewski
+ * Max Countryman
+ * caitifbrito
+ * lcya86 刘春洋
+ * Martin Alderete (https://github.com/malderete)
+ * Nick Joyce
+ * Jared Forsyth
+ * Kenneth Falck
+ * Lukasz Balcerzak
+ * Nicolas Cortot
+ * Alex (https://github.com/kelsta)
+ * Jin Zhang
+ * Daniel Axtens
+ * Leo-Naeka
+ * Ryan Witt (https://github.com/ryanwitt)
+ * Jiequan (https://github.com/Jiequan)
+ * hensom (https://github.com/hensom)
+ * zhy0216 (https://github.com/zhy0216)
+ * istinspring (https://github.com/istinspring)
+ * Massimo Santini (https://github.com/mapio)
+ * Nigel McNie (https://github.com/nigelmcnie)
+ * ygbourhis (https://github.com/ygbourhis)
+ * Bob Dickinson (https://github.com/BobDickinson)
+ * Michael Bartnett (https://github.com/michaelbartnett)
+ * Alon Horev (https://github.com/alonho)
+ * Kelvin Hammond (https://github.com/kelvinhammond)
+ * Jatin Chopra (https://github.com/jatin)
+ * Paul Uithol (https://github.com/PaulUithol)
+ * Thom Knowles (https://github.com/fleat)
+ * Paul (https://github.com/squamous)
+ * Olivier Cortès (https://github.com/Karmak23)
+ * crazyzubr (https://github.com/crazyzubr)
+ * FrankSomething (https://github.com/FrankSomething)
+ * Alexandr Morozov (https://github.com/LK4D4)
+ * mishudark (https://github.com/mishudark)
+ * Joe Friedl (https://github.com/grampajoe)
+ * Daniel Ward (https://github.com/danielward)
+ * Aniket Deshpande (https://github.com/anicake)
+ * rfkrocktk (https://github.com/rfkrocktk)
+ * Gustavo Andrés Angulo (https://github.com/woakas)
+ * Dmytro Popovych (https://github.com/drudim)
+ * Tom (https://github.com/tomprimozic)
+ * j0hnsmith (https://github.com/j0hnsmith)
+ * Damien Churchill (https://github.com/damoxc)
+ * Jonathan Simon Prates (https://github.com/jonathansp)
+ * Thiago Papageorgiou (https://github.com/tmpapageorgiou)
+ * Omer Katz (https://github.com/thedrow)
+ * Falcon Dai (https://github.com/falcondai)
+ * Polyrabbit (https://github.com/polyrabbit)
+ * Sagiv Malihi (https://github.com/sagivmalihi)
+ * Dmitry Konishchev (https://github.com/KonishchevDmitry)
+ * Martyn Smith (https://github.com/martynsmith)
+ * Andrei Zbikowski (https://github.com/b1naryth1ef)
+ * Ronald van Rij (https://github.com/ronaldvanrij)
+ * François Schmidts (https://github.com/jaesivsm)
+ * Eric Plumb (https://github.com/professorplumb)
+ * Damien Churchill (https://github.com/damoxc)
+ * Aleksandr Sorokoumov (https://github.com/Gerrrr)
+ * Clay McClure (https://github.com/claymation)
+ * Bruno Rocha (https://github.com/rochacbruno)
+ * Norberto Leite (https://github.com/nleite)
+ * Bob Cribbs (https://github.com/bocribbz)
+ * Jay Shirley (https://github.com/jshirley)
+ * David Bordeynik (https://github.com/DavidBord)
+ * Axel Haustant (https://github.com/noirbizarre)
+ * David Czarnecki (https://github.com/czarneckid)
+ * Vyacheslav Murashkin (https://github.com/a4tunado)
+ * André Ericson https://github.com/aericson)
+ * Mikhail Moshnogorsky (https://github.com/mikhailmoshnogorsky)
+ * Diego Berrocal (https://github.com/cestdiego)
+ * Matthew Ellison (https://github.com/seglberg)
+ * Jimmy Shen (https://github.com/jimmyshen)
+ * J. Fernando Sánchez (https://github.com/balkian)
+ * Michael Chase (https://github.com/rxsegrxup)
+ * Eremeev Danil (https://github.com/elephanter)
+ * Catstyle Lee (https://github.com/Catstyle)
+ * Kiryl Yermakou (https://github.com/rma4ok)
+ * Matthieu Rigal (https://github.com/MRigal)
+ * Charanpal Dhanjal (https://github.com/charanpald)
+ * Emmanuel Leblond (https://github.com/touilleMan)
+ * Breeze.Kay (https://github.com/9nix00)
+ * Vicki Donchenko (https://github.com/kivistein)
+ * Emile Caron (https://github.com/emilecaron)
+ * Amit Lichtenberg (https://github.com/amitlicht)
+ * Gang Li (https://github.com/iici-gli)
+ * Lars Butler (https://github.com/larsbutler)
+ * George Macon (https://github.com/gmacon)
+ * Ashley Whetter (https://github.com/AWhetter)
+ * Paul-Armand Verhaegen (https://github.com/paularmand)
+ * Steven Rossiter (https://github.com/BeardedSteve)
+ * Luo Peng (https://github.com/RussellLuo)
+ * Bryan Bennett (https://github.com/bbenne10)
+ * Gilb's Gilb's (https://github.com/gilbsgilbs)
+ * Joshua Nedrud (https://github.com/Neurostack)
+ * Shu Shen (https://github.com/shushen)
+ * xiaost7 (https://github.com/xiaost7)
+ * Victor Varvaryuk
+ * Stanislav Kaledin (https://github.com/sallyruthstruik)
+ * Dmitry Yantsen (https://github.com/mrTable)
+ * Renjianxin (https://github.com/Davidrjx)
+ * Erdenezul Batmunkh (https://github.com/erdenezul)
+ * Andy Yankovsky (https://github.com/werat)
+ * Bastien Gérard (https://github.com/bagerard)
+ * Trevor Hall (https://github.com/tjhall13)
+ * Gleb Voropaev (https://github.com/buggyspace)
+ * Paulo Amaral (https://github.com/pauloAmaral)
+ * Gaurav Dadhania (https://github.com/GVRV)
+ * Yurii Andrieiev (https://github.com/yandrieiev)
+ * Filip Kucharczyk (https://github.com/Pacu2)
+ * Eric Timmons (https://github.com/daewok)
+ * Matthew Simpson (https://github.com/mcsimps2)
+ * Leonardo Domingues (https://github.com/leodmgs)
+ * Agustin Barto (https://github.com/abarto)

CONTRIBUTING.rst

@@ -0,0 +1,105 @@
+Contributing to MongoEngine
+===========================
+
+MongoEngine has a large `community
+<https://raw.github.com/MongoEngine/mongoengine/master/AUTHORS>`_ and
+contributions are always encouraged. Contributions can be as simple as
+minor tweaks to the documentation. Please read these guidelines before
+sending a pull request.
+
+Bugfixes and New Features
+-------------------------
+
+Before starting to write code, look for existing `tickets
+<https://github.com/MongoEngine/mongoengine/issues?state=open>`_ or `create one
+<https://github.com/MongoEngine/mongoengine/issues>`_ for your specific
+issue or feature request. That way you avoid working on something
+that might not be of interest or that has already been addressed. If in doubt
+post to the `user group <http://groups.google.com/group/mongoengine-users>`
+
+Supported Interpreters
+----------------------
+
+MongoEngine supports CPython 3.5 and newer as well as Pypy3.
+Language features not supported by all interpreters can not be used.
+
+Python3 codebase
+----------------------
+
+Since 0.20, the codebase is exclusively Python 3.
+
+Earlier versions were exclusively Python2, and were relying on 2to3 to support Python3 installs.
+Travis runs the tests against the main Python 3.x versions.
+
+
+Style Guide
+-----------
+
+MongoEngine's codebase is formatted with `black <https://github.com/python/black>`_, other tools like
+flake8 are also used. Those tools will run as part of the CI and will fail in case the code is not formatted properly.
+
+To install all development tools, simply run the following commands:
+
+.. code-block:: console
+
+    $ python -m pip install -r requirements-dev.txt
+
+
+You can install `pre-commit <https://pre-commit.com/>`_ into your git hooks,
+to automatically check and fix any formatting issue before creating a
+git commit.
+
+To enable ``pre-commit`` simply run:
+
+.. code-block:: console
+
+    $ pre-commit install
+
+See the ``.pre-commit-config.yaml`` configuration file for more information
+on how it works.
+
+Testing
+-------
+
+All tests are run on `Travis <http://travis-ci.org/MongoEngine/mongoengine>`_
+and any pull requests are automatically tested. Any pull requests without
+tests will take longer to be integrated and might be refused.
+
+You may also submit a simple failing test as a pull request if you don't know
+how to fix it, it will be easier for other people to work on it and it may get
+fixed faster.
+
+General Guidelines
+------------------
+
+- Avoid backward breaking changes if at all possible.
+- If you *have* to introduce a breaking change, make it very clear in your
+  pull request's description. Also, describe how users of this package
+  should adapt to the breaking change in docs/upgrade.rst.
+- Write inline documentation for new classes and methods.
+- Write tests and make sure they pass (make sure you have a mongod
+  running on the default port, then execute ``python setup.py test``
+  from the cmd line to run the test suite).
+- Ensure tests pass on all supported Python, PyMongo, and MongoDB versions.
+  You can test various Python and PyMongo versions locally by executing
+  ``tox``. For different MongoDB versions, you can rely on our automated
+  Travis tests.
+- Add enhancements or problematic bug fixes to docs/changelog.rst.
+- Add yourself to AUTHORS :)
+
+Documentation
+-------------
+
+To contribute to the `API documentation
+<http://docs.mongoengine.org/en/latest/apireference.html>`_
+just make your changes to the inline documentation of the appropriate
+`source code <https://github.com/MongoEngine/mongoengine>`_ or `rst file
+<https://github.com/MongoEngine/mongoengine/tree/master/docs>`_ in a
+branch and submit a `pull request <https://help.github.com/articles/using-pull-requests>`_.
+You might also use the github `Edit <https://github.com/blog/844-forking-with-the-edit-button>`_
+button.
+
+If you want to test your documentation changes locally, you need to install
+the ``sphinx`` and ``sphinx_rtd_theme`` packages. Once these are installed,
+go to the ``docs`` directory, run ``make html`` and inspect the updated docs
+by running ``open _build/html/index.html``.

LICENSE

@@ -1,4 +1,4 @@
-Copyright (c) 2009-2010 Harry Marr
+Copyright (c) 2009 See AUTHORS
 
 Permission is hereby granted, free of charge, to any person
 obtaining a copy of this software and associated documentation

MANIFEST.in

@@ -1,6 +1,6 @@
+include MANIFEST.in
 include README.rst
 include LICENSE
+include AUTHORS
 recursive-include docs *
-prune docs/_build/*
-recursive-include tests *
-recursive-exclude * *.pyc *.swp
+prune docs/_build

README.rst

@@ -2,36 +2,81 @@
 MongoEngine
 ===========
 :Info: MongoEngine is an ORM-like layer on top of PyMongo.
+:Repository: https://github.com/MongoEngine/mongoengine
 :Author: Harry Marr (http://github.com/hmarr)
+:Maintainer: Stefan Wójcik (http://github.com/wojcikstefan)
+
+.. image:: https://travis-ci.org/MongoEngine/mongoengine.svg?branch=master
+  :target: https://travis-ci.org/MongoEngine/mongoengine
+
+.. image:: https://coveralls.io/repos/github/MongoEngine/mongoengine/badge.svg?branch=master
+  :target: https://coveralls.io/github/MongoEngine/mongoengine?branch=master
+
+.. image:: https://landscape.io/github/MongoEngine/mongoengine/master/landscape.svg?style=flat
+  :target: https://landscape.io/github/MongoEngine/mongoengine/master
+  :alt: Code Health
 
 About
 =====
 MongoEngine is a Python Object-Document Mapper for working with MongoDB.
-Documentation available at http://hmarr.com/mongoengine/ - there is currently 
-a `tutorial <http://hmarr.com/mongoengine/tutorial.html>`_, a `user guide 
-<http://hmarr.com/mongoengine/userguide.html>`_ and an `API reference
-<http://hmarr.com/mongoengine/apireference.html>`_.
+Documentation is available at https://mongoengine-odm.readthedocs.io - there
+is currently a `tutorial <https://mongoengine-odm.readthedocs.io/tutorial.html>`_,
+a `user guide <https://mongoengine-odm.readthedocs.io/guide/index.html>`_, and
+an `API reference <https://mongoengine-odm.readthedocs.io/apireference.html>`_.
+
+Supported MongoDB Versions
+==========================
+MongoEngine is currently tested against MongoDB v3.4, v3.6 and v4.0. Future versions
+should be supported as well, but aren't actively tested at the moment. Make
+sure to open an issue or submit a pull request if you experience any problems
+with MongoDB version > 4.0.
 
 Installation
 ============
-If you have `setuptools <http://peak.telecommunity.com/DevCenter/setuptools>`_
-you can use ``easy_install mongoengine``. Otherwise, you can download the
-source from `GitHub <http://github.com/hmarr/mongoengine>`_ and run ``python
-setup.py install``.
+We recommend the use of `virtualenv <https://virtualenv.pypa.io/>`_ and of
+`pip <https://pip.pypa.io/>`_. You can then use ``python -m pip install -U mongoengine``.
+You may also have `setuptools <http://peak.telecommunity.com/DevCenter/setuptools>`_
+and thus you can use ``easy_install -U mongoengine``. Another option is
+`pipenv <https://docs.pipenv.org/>`_. You can then use ``pipenv install mongoengine``
+to both create the virtual environment and install the package. Otherwise, you can
+download the source from `GitHub <http://github.com/MongoEngine/mongoengine>`_ and
+run ``python setup.py install``.
+
+The support for Python2 was dropped with MongoEngine 0.20.0
 
 Dependencies
 ============
-- pymongo 1.1+
-- sphinx (optional - for documentation generation)
+All of the dependencies can easily be installed via `python -m pip <https://pip.pypa.io/>`_.
+At the very least, you'll need these two packages to use MongoEngine:
+
+- pymongo>=3.4
+
+If you utilize a ``DateTimeField``, you might also use a more flexible date parser:
+
+- dateutil>=2.1.0
+
+If you need to use an ``ImageField`` or ``ImageGridFsProxy``:
+
+- Pillow>=2.0.0
+
+If you need to use signals:
+
+- blinker>=1.3
 
 Examples
 ========
-Some simple examples of what MongoEngine code looks like::
+Some simple examples of what MongoEngine code looks like:
+
+.. code :: python
+
+    from mongoengine import *
+    connect('mydb')
 
     class BlogPost(Document):
         title = StringField(required=True, max_length=200)
-        posted = DateTimeField(default=datetime.datetime.now)
+        posted = DateTimeField(default=datetime.datetime.utcnow)
         tags = ListField(StringField(max_length=50))
+        meta = {'allow_inheritance': True}
 
     class TextPost(BlogPost):
         content = StringField(required=True)
@@ -51,39 +96,60 @@ Some simple examples of what MongoEngine code looks like::
 
     # Iterate over all posts using the BlogPost superclass
     >>> for post in BlogPost.objects:
-    ...     print '===', post.title, '==='
+    ...     print('===', post.title, '===')
     ...     if isinstance(post, TextPost):
-    ...         print post.content
+    ...         print(post.content)
     ...     elif isinstance(post, LinkPost):
-    ...         print 'Link:', post.url
-    ...     print
+    ...         print('Link:', post.url)
     ...
-    === Using MongoEngine ===
-    See the tutorial
 
-    === MongoEngine Docs ===
-    Link: hmarr.com/mongoengine
-
-    >>> len(BlogPost.objects)
+    # Count all blog posts and its subtypes
+    >>> BlogPost.objects.count()
     2
-    >>> len(HtmlPost.objects)
+    >>> TextPost.objects.count()
     1
-    >>> len(LinkPost.objects)
+    >>> LinkPost.objects.count()
     1
 
-    # Find tagged posts
-    >>> len(BlogPost.objects(tags='mongoengine'))
+    # Count tagged posts
+    >>> BlogPost.objects(tags='mongoengine').count()
     2
-    >>> len(BlogPost.objects(tags='mongodb'))
+    >>> BlogPost.objects(tags='mongodb').count()
     1
 
 Tests
 =====
 To run the test suite, ensure you are running a local instance of MongoDB on
-the standard port, and run ``python setup.py test``.
+the standard port and have ``pytest`` installed. Then, run ``python setup.py test``
+or simply ``pytest``.
+
+To run the test suite on every supported Python and PyMongo version, you can
+use ``tox``. You'll need to make sure you have each supported Python version
+installed in your environment and then:
+
+.. code-block:: shell
+
+    # Install tox
+    $ python -m pip install tox
+    # Run the test suites
+    $ tox
+
+If you wish to run a subset of tests, use the pytest convention:
+
+.. code-block:: shell
+
+    # Run all the tests in a particular test file
+    $ pytest tests/fields/test_fields.py
+    # Run only particular test class in that file
+    $ pytest tests/fields/test_fields.py::TestField
+
+Community
+=========
+- `MongoEngine Users mailing list
+  <http://groups.google.com/group/mongoengine-users>`_
+- `MongoEngine Developers mailing list
+  <http://groups.google.com/group/mongoengine-dev>`_
 
 Contributing
 ============
-The source is available on `GitHub <http://github.com/hmarr/mongoengine>`_ - to
-contribute to the project, fork it on GitHub and send a pull request, all
-contributions and suggestions are welcome!
+We welcome contributions! See the `Contribution guidelines <https://github.com/MongoEngine/mongoengine/blob/master/CONTRIBUTING.rst>`_

benchmarks/test_basic_doc_ops.py

@@ -0,0 +1,142 @@
+from timeit import repeat
+
+import mongoengine
+from mongoengine import (
+    BooleanField,
+    Document,
+    EmailField,
+    EmbeddedDocument,
+    EmbeddedDocumentField,
+    IntField,
+    ListField,
+    StringField,
+)
+
+mongoengine.connect(db="mongoengine_benchmark_test")
+
+
+def timeit(f, n=10000):
+    return min(repeat(f, repeat=3, number=n)) / float(n)
+
+
+def test_basic():
+    class Book(Document):
+        name = StringField()
+        pages = IntField()
+        tags = ListField(StringField())
+        is_published = BooleanField()
+        author_email = EmailField()
+
+    Book.drop_collection()
+
+    def init_book():
+        return Book(
+            name="Always be closing",
+            pages=100,
+            tags=["self-help", "sales"],
+            is_published=True,
+            author_email="alec@example.com",
+        )
+
+    print("Doc initialization: %.3fus" % (timeit(init_book, 1000) * 10 ** 6))
+
+    b = init_book()
+    print("Doc getattr: %.3fus" % (timeit(lambda: b.name, 10000) * 10 ** 6))
+
+    print(
+        "Doc setattr: %.3fus"
+        % (timeit(lambda: setattr(b, "name", "New name"), 10000) * 10 ** 6)
+    )
+
+    print("Doc to mongo: %.3fus" % (timeit(b.to_mongo, 1000) * 10 ** 6))
+
+    print("Doc validation: %.3fus" % (timeit(b.validate, 1000) * 10 ** 6))
+
+    def save_book():
+        b._mark_as_changed("name")
+        b._mark_as_changed("tags")
+        b.save()
+
+    print("Save to database: %.3fus" % (timeit(save_book, 100) * 10 ** 6))
+
+    son = b.to_mongo()
+    print(
+        "Load from SON: %.3fus" % (timeit(lambda: Book._from_son(son), 1000) * 10 ** 6)
+    )
+
+    print(
+        "Load from database: %.3fus" % (timeit(lambda: Book.objects[0], 100) * 10 ** 6)
+    )
+
+    def create_and_delete_book():
+        b = init_book()
+        b.save()
+        b.delete()
+
+    print(
+        "Init + save to database + delete: %.3fms"
+        % (timeit(create_and_delete_book, 10) * 10 ** 3)
+    )
+
+
+def test_big_doc():
+    class Contact(EmbeddedDocument):
+        name = StringField()
+        title = StringField()
+        address = StringField()
+
+    class Company(Document):
+        name = StringField()
+        contacts = ListField(EmbeddedDocumentField(Contact))
+
+    Company.drop_collection()
+
+    def init_company():
+        return Company(
+            name="MongoDB, Inc.",
+            contacts=[
+                Contact(name="Contact %d" % x, title="CEO", address="Address %d" % x)
+                for x in range(1000)
+            ],
+        )
+
+    company = init_company()
+    print("Big doc to mongo: %.3fms" % (timeit(company.to_mongo, 100) * 10 ** 3))
+
+    print("Big doc validation: %.3fms" % (timeit(company.validate, 1000) * 10 ** 3))
+
+    company.save()
+
+    def save_company():
+        company._mark_as_changed("name")
+        company._mark_as_changed("contacts")
+        company.save()
+
+    print("Save to database: %.3fms" % (timeit(save_company, 100) * 10 ** 3))
+
+    son = company.to_mongo()
+    print(
+        "Load from SON: %.3fms"
+        % (timeit(lambda: Company._from_son(son), 100) * 10 ** 3)
+    )
+
+    print(
+        "Load from database: %.3fms"
+        % (timeit(lambda: Company.objects[0], 100) * 10 ** 3)
+    )
+
+    def create_and_delete_company():
+        c = init_company()
+        c.save()
+        c.delete()
+
+    print(
+        "Init + save to database + delete: %.3fms"
+        % (timeit(create_and_delete_company, 10) * 10 ** 3)
+    )
+
+
+if __name__ == "__main__":
+    test_basic()
+    print("-" * 100)
+    test_big_doc()

benchmarks/test_inserts.py

@@ -0,0 +1,161 @@
+import timeit
+
+
+def main():
+    setup = """
+from pymongo import MongoClient
+
+connection = MongoClient()
+connection.drop_database('mongoengine_benchmark_test')
+"""
+
+    stmt = """
+from pymongo import MongoClient
+
+connection = MongoClient()
+
+db = connection.mongoengine_benchmark_test
+noddy = db.noddy
+
+for i in range(10000):
+    example = {'fields': {}}
+    for j in range(20):
+        example['fields']["key"+str(j)] = "value "+str(j)
+
+    noddy.insert_one(example)
+
+myNoddys = noddy.find()
+[n for n in myNoddys]  # iterate
+"""
+
+    print("-" * 100)
+    print("PyMongo: Creating 10000 dictionaries.")
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    stmt = """
+from pymongo import MongoClient, WriteConcern
+connection = MongoClient()
+
+db = connection.mongoengine_benchmark_test
+noddy = db.noddy.with_options(write_concern=WriteConcern(w=0))
+
+for i in range(10000):
+    example = {'fields': {}}
+    for j in range(20):
+        example['fields']["key"+str(j)] = "value "+str(j)
+
+    noddy.insert_one(example)
+
+myNoddys = noddy.find()
+[n for n in myNoddys]  # iterate
+"""
+
+    print("-" * 100)
+    print('PyMongo: Creating 10000 dictionaries (write_concern={"w": 0}).')
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    setup = """
+from pymongo import MongoClient
+
+connection = MongoClient()
+connection.drop_database('mongoengine_benchmark_test')
+connection.close()
+
+from mongoengine import Document, DictField, connect
+connect("mongoengine_benchmark_test")
+
+class Noddy(Document):
+    fields = DictField()
+"""
+
+    stmt = """
+for i in range(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save()
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys]  # iterate
+"""
+
+    print("-" * 100)
+    print("MongoEngine: Creating 10000 dictionaries.")
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    stmt = """
+for i in range(10000):
+    noddy = Noddy()
+    fields = {}
+    for j in range(20):
+        fields["key"+str(j)] = "value "+str(j)
+    noddy.fields = fields
+    noddy.save()
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys]  # iterate
+"""
+
+    print("-" * 100)
+    print("MongoEngine: Creating 10000 dictionaries (using a single field assignment).")
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    stmt = """
+for i in range(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(write_concern={"w": 0})
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print("-" * 100)
+    print('MongoEngine: Creating 10000 dictionaries (write_concern={"w": 0}).')
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    stmt = """
+for i in range(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(write_concern={"w": 0}, validate=False)
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print("-" * 100)
+    print(
+        'MongoEngine: Creating 10000 dictionaries (write_concern={"w": 0}, validate=False).'
+    )
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+    stmt = """
+for i in range(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(force_insert=True, write_concern={"w": 0}, validate=False)
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print("-" * 100)
+    print(
+        'MongoEngine: Creating 10000 dictionaries (force_insert=True, write_concern={"w": 0}, validate=False).'
+    )
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print("{}s".format(t.timeit(1)))
+
+
+if __name__ == "__main__":
+    main()

docs/_themes/nature/static/nature.css_t

@@ -1,229 +0,0 @@
-/**
- * Sphinx stylesheet -- default theme
- * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- */
- 
-@import url("basic.css");
- 
-/* -- page layout ----------------------------------------------------------- */
- 
-body {
-    font-family: Arial, sans-serif;
-    font-size: 100%;
-    background-color: #111;
-    color: #555;
-    margin: 0;
-    padding: 0;
-}
-
-div.documentwrapper {
-    float: left;
-    width: 100%;
-}
-
-div.bodywrapper {
-    margin: 0 0 0 230px;
-}
-
-hr{
-    border: 1px solid #B1B4B6;
-}
- 
-div.document {
-    background-color: #eee;
-}
- 
-div.body {
-    background-color: #ffffff;
-    color: #3E4349;
-    padding: 0 30px 30px 30px;
-    font-size: 0.8em;
-}
- 
-div.footer {
-    color: #555;
-    width: 100%;
-    padding: 13px 0;
-    text-align: center;
-    font-size: 75%;
-}
- 
-div.footer a {
-    color: #444;
-    text-decoration: underline;
-}
- 
-div.related {
-    background-color: #6BA81E;
-    line-height: 32px;
-    color: #fff;
-    text-shadow: 0px 1px 0 #444;
-    font-size: 0.80em;
-}
- 
-div.related a {
-    color: #E2F3CC;
-}
- 
-div.sphinxsidebar {
-    font-size: 0.75em;
-    line-height: 1.5em;
-}
-
-div.sphinxsidebarwrapper{
-    padding: 20px 0;
-}
- 
-div.sphinxsidebar h3,
-div.sphinxsidebar h4 {
-    font-family: Arial, sans-serif;
-    color: #222;
-    font-size: 1.2em;
-    font-weight: normal;
-    margin: 0;
-    padding: 5px 10px;
-    background-color: #ddd;
-    text-shadow: 1px 1px 0 white
-}
-
-div.sphinxsidebar h4{
-    font-size: 1.1em;
-}
- 
-div.sphinxsidebar h3 a {
-    color: #444;
-}
- 
- 
-div.sphinxsidebar p {
-    color: #888;
-    padding: 5px 20px;
-}
- 
-div.sphinxsidebar p.topless {
-}
- 
-div.sphinxsidebar ul {
-    margin: 10px 20px;
-    padding: 0;
-    color: #000;
-}
- 
-div.sphinxsidebar a {
-    color: #444;
-}
- 
-div.sphinxsidebar input {
-    border: 1px solid #ccc;
-    font-family: sans-serif;
-    font-size: 1em;
-}
-
-div.sphinxsidebar input[type=text]{
-    margin-left: 20px;
-}
- 
-/* -- body styles ----------------------------------------------------------- */
- 
-a {
-    color: #005B81;
-    text-decoration: none;
-}
- 
-a:hover {
-    color: #E32E00;
-    text-decoration: underline;
-}
- 
-div.body h1,
-div.body h2,
-div.body h3,
-div.body h4,
-div.body h5,
-div.body h6 {
-    font-family: Arial, sans-serif;
-    background-color: #BED4EB;
-    font-weight: normal;
-    color: #212224;
-    margin: 30px 0px 10px 0px;
-    padding: 5px 0 5px 10px;
-    text-shadow: 0px 1px 0 white
-}
- 
-div.body h1 { border-top: 20px solid white; margin-top: 0; font-size: 200%; }
-div.body h2 { font-size: 150%; background-color: #C8D5E3; }
-div.body h3 { font-size: 120%; background-color: #D8DEE3; }
-div.body h4 { font-size: 110%; background-color: #D8DEE3; }
-div.body h5 { font-size: 100%; background-color: #D8DEE3; }
-div.body h6 { font-size: 100%; background-color: #D8DEE3; }
- 
-a.headerlink {
-    color: #c60f0f;
-    font-size: 0.8em;
-    padding: 0 4px 0 4px;
-    text-decoration: none;
-}
- 
-a.headerlink:hover {
-    background-color: #c60f0f;
-    color: white;
-}
- 
-div.body p, div.body dd, div.body li {
-    line-height: 1.5em;
-}
- 
-div.admonition p.admonition-title + p {
-    display: inline;
-}
-
-div.highlight{
-    background-color: white;
-}
-
-div.note {
-    background-color: #eee;
-    border: 1px solid #ccc;
-}
- 
-div.seealso {
-    background-color: #ffc;
-    border: 1px solid #ff6;
-}
- 
-div.topic {
-    background-color: #eee;
-}
- 
-div.warning {
-    background-color: #ffe4e4;
-    border: 1px solid #f66;
-}
- 
-p.admonition-title {
-    display: inline;
-}
- 
-p.admonition-title:after {
-    content: ":";
-}
- 
-pre {
-    padding: 10px;
-    background-color: White;
-    color: #222;
-    line-height: 1.2em;
-    border: 1px solid #C6C9CB;
-    font-size: 1.2em;
-    margin: 1.5em 0 1.5em 0;
-    -webkit-box-shadow: 1px 1px 1px #d8d8d8;
-    -moz-box-shadow: 1px 1px 1px #d8d8d8;
-}
- 
-tt {
-    background-color: #ecf0f3;
-    color: #222;
-    padding: 1px 2px;
-    font-size: 1.2em;
-    font-family: monospace;
-}

docs/_themes/nature/static/pygments.css

@@ -1,54 +0,0 @@
-.c { color: #999988; font-style: italic } /* Comment */
-.k { font-weight: bold } /* Keyword */
-.o { font-weight: bold } /* Operator */
-.cm { color: #999988; font-style: italic } /* Comment.Multiline */
-.cp { color: #999999; font-weight: bold } /* Comment.preproc */
-.c1 { color: #999988; font-style: italic } /* Comment.Single */
-.gd { color: #000000; background-color: #ffdddd } /* Generic.Deleted */
-.ge { font-style: italic } /* Generic.Emph */
-.gr { color: #aa0000 } /* Generic.Error */
-.gh { color: #999999 } /* Generic.Heading */
-.gi { color: #000000; background-color: #ddffdd } /* Generic.Inserted */
-.go { color: #111 } /* Generic.Output */
-.gp { color: #555555 } /* Generic.Prompt */
-.gs { font-weight: bold } /* Generic.Strong */
-.gu { color: #aaaaaa } /* Generic.Subheading */
-.gt { color: #aa0000 } /* Generic.Traceback */
-.kc { font-weight: bold } /* Keyword.Constant */
-.kd { font-weight: bold } /* Keyword.Declaration */
-.kp { font-weight: bold } /* Keyword.Pseudo */
-.kr { font-weight: bold } /* Keyword.Reserved */
-.kt { color: #445588; font-weight: bold } /* Keyword.Type */
-.m { color: #009999 } /* Literal.Number */
-.s { color: #bb8844 } /* Literal.String */
-.na { color: #008080 } /* Name.Attribute */
-.nb { color: #999999 } /* Name.Builtin */
-.nc { color: #445588; font-weight: bold } /* Name.Class */
-.no { color: #ff99ff } /* Name.Constant */
-.ni { color: #800080 } /* Name.Entity */
-.ne { color: #990000; font-weight: bold } /* Name.Exception */
-.nf { color: #990000; font-weight: bold } /* Name.Function */
-.nn { color: #555555 } /* Name.Namespace */
-.nt { color: #000080 } /* Name.Tag */
-.nv { color: purple } /* Name.Variable */
-.ow { font-weight: bold } /* Operator.Word */
-.mf { color: #009999 } /* Literal.Number.Float */
-.mh { color: #009999 } /* Literal.Number.Hex */
-.mi { color: #009999 } /* Literal.Number.Integer */
-.mo { color: #009999 } /* Literal.Number.Oct */
-.sb { color: #bb8844 } /* Literal.String.Backtick */
-.sc { color: #bb8844 } /* Literal.String.Char */
-.sd { color: #bb8844 } /* Literal.String.Doc */
-.s2 { color: #bb8844 } /* Literal.String.Double */
-.se { color: #bb8844 } /* Literal.String.Escape */
-.sh { color: #bb8844 } /* Literal.String.Heredoc */
-.si { color: #bb8844 } /* Literal.String.Interpol */
-.sx { color: #bb8844 } /* Literal.String.Other */
-.sr { color: #808000 } /* Literal.String.Regex */
-.s1 { color: #bb8844 } /* Literal.String.Single */
-.ss { color: #bb8844 } /* Literal.String.Symbol */
-.bp { color: #999999 } /* Name.Builtin.Pseudo */
-.vc { color: #ff99ff } /* Name.Variable.Class */
-.vg { color: #ff99ff } /* Name.Variable.Global */
-.vi { color: #ff99ff } /* Name.Variable.Instance */
-.il { color: #009999 } /* Literal.Number.Integer.Long */

docs/_themes/nature/theme.conf

@@ -1,4 +0,0 @@
-[theme]
-inherit = basic
-stylesheet = nature.css
-pygments_style = tango

docs/apireference.rst

@@ -6,12 +6,14 @@ Connecting
 ==========
 
 .. autofunction:: mongoengine.connect
+.. autofunction:: mongoengine.register_connection
 
 Documents
 =========
 
 .. autoclass:: mongoengine.Document
    :members:
+   :inherited-members:
 
    .. attribute:: objects
 
@@ -20,34 +22,113 @@ Documents
 
 .. autoclass:: mongoengine.EmbeddedDocument
    :members:
+   :inherited-members:
+
+.. autoclass:: mongoengine.DynamicDocument
+   :members:
+   :inherited-members:
+
+.. autoclass:: mongoengine.DynamicEmbeddedDocument
+   :members:
+   :inherited-members:
+
+.. autoclass:: mongoengine.document.MapReduceDocument
+   :members:
+
+.. autoclass:: mongoengine.ValidationError
+  :members:
+
+.. autoclass:: mongoengine.FieldDoesNotExist
+
+
+Context Managers
+================
+
+.. autoclass:: mongoengine.context_managers.switch_db
+.. autoclass:: mongoengine.context_managers.switch_collection
+.. autoclass:: mongoengine.context_managers.no_dereference
+.. autoclass:: mongoengine.context_managers.query_counter
 
 Querying
 ========
 
-.. autoclass:: mongoengine.queryset.QuerySet
+.. automodule:: mongoengine.queryset
+    :synopsis: Queryset level operations
+
+    .. autoclass:: mongoengine.queryset.QuerySet
+      :members:
+      :inherited-members:
+
+      .. automethod:: QuerySet.__call__
+
+    .. autoclass:: mongoengine.queryset.QuerySetNoCache
       :members:
 
-   .. automethod:: mongoengine.queryset.QuerySet.__call__
+       .. automethod:: mongoengine.queryset.QuerySetNoCache.__call__
 
-.. autofunction:: mongoengine.queryset.queryset_manager
+    .. autofunction:: mongoengine.queryset.queryset_manager
 
 Fields
 ======
 
-.. autoclass:: mongoengine.StringField
+.. autoclass:: mongoengine.base.fields.BaseField
+.. autoclass:: mongoengine.fields.StringField
+.. autoclass:: mongoengine.fields.URLField
+.. autoclass:: mongoengine.fields.EmailField
+.. autoclass:: mongoengine.fields.IntField
+.. autoclass:: mongoengine.fields.LongField
+.. autoclass:: mongoengine.fields.FloatField
+.. autoclass:: mongoengine.fields.DecimalField
+.. autoclass:: mongoengine.fields.BooleanField
+.. autoclass:: mongoengine.fields.DateTimeField
+.. autoclass:: mongoengine.fields.ComplexDateTimeField
+.. autoclass:: mongoengine.fields.EmbeddedDocumentField
+.. autoclass:: mongoengine.fields.GenericEmbeddedDocumentField
+.. autoclass:: mongoengine.fields.DynamicField
+.. autoclass:: mongoengine.fields.ListField
+.. autoclass:: mongoengine.fields.EmbeddedDocumentListField
+.. autoclass:: mongoengine.fields.SortedListField
+.. autoclass:: mongoengine.fields.DictField
+.. autoclass:: mongoengine.fields.MapField
+.. autoclass:: mongoengine.fields.ReferenceField
+.. autoclass:: mongoengine.fields.LazyReferenceField
+.. autoclass:: mongoengine.fields.GenericReferenceField
+.. autoclass:: mongoengine.fields.GenericLazyReferenceField
+.. autoclass:: mongoengine.fields.CachedReferenceField
+.. autoclass:: mongoengine.fields.BinaryField
+.. autoclass:: mongoengine.fields.FileField
+.. autoclass:: mongoengine.fields.ImageField
+.. autoclass:: mongoengine.fields.SequenceField
+.. autoclass:: mongoengine.fields.ObjectIdField
+.. autoclass:: mongoengine.fields.UUIDField
+.. autoclass:: mongoengine.fields.GeoPointField
+.. autoclass:: mongoengine.fields.PointField
+.. autoclass:: mongoengine.fields.LineStringField
+.. autoclass:: mongoengine.fields.PolygonField
+.. autoclass:: mongoengine.fields.MultiPointField
+.. autoclass:: mongoengine.fields.MultiLineStringField
+.. autoclass:: mongoengine.fields.MultiPolygonField
+.. autoclass:: mongoengine.fields.GridFSError
+.. autoclass:: mongoengine.fields.GridFSProxy
+.. autoclass:: mongoengine.fields.ImageGridFsProxy
+.. autoclass:: mongoengine.fields.ImproperlyConfigured
 
-.. autoclass:: mongoengine.IntField
+Embedded Document Querying
+==========================
 
-.. autoclass:: mongoengine.FloatField
+.. versionadded:: 0.9
 
-.. autoclass:: mongoengine.BooleanField
+Additional queries for Embedded Documents are available when using the
+:class:`~mongoengine.EmbeddedDocumentListField` to store a list of embedded
+documents.
 
-.. autoclass:: mongoengine.DateTimeField
+A list of embedded documents is returned as a special list with the
+following methods:
 
-.. autoclass:: mongoengine.EmbeddedDocumentField
+.. autoclass:: mongoengine.base.datastructures.EmbeddedDocumentList
+    :members:
 
-.. autoclass:: mongoengine.ListField
+Misc
+====
 
-.. autoclass:: mongoengine.ObjectIdField
-
-.. autoclass:: mongoengine.ReferenceField
+.. autofunction:: mongoengine.common._import_class

docs/code/tumblelog.py

@@ -1,66 +1,77 @@
 from mongoengine import *
 
-connect('tumblelog')
+connect("tumblelog")
+
 
 class Comment(EmbeddedDocument):
     content = StringField()
     name = StringField(max_length=120)
 
+
 class User(Document):
     email = StringField(required=True)
     first_name = StringField(max_length=50)
     last_name = StringField(max_length=50)
 
+
 class Post(Document):
     title = StringField(max_length=120, required=True)
     author = ReferenceField(User)
     tags = ListField(StringField(max_length=30))
     comments = ListField(EmbeddedDocumentField(Comment))
 
+    # bugfix
+    meta = {"allow_inheritance": True}
+
+
 class TextPost(Post):
     content = StringField()
 
+
 class ImagePost(Post):
     image_path = StringField()
 
+
 class LinkPost(Post):
     link_url = StringField()
 
+
 Post.drop_collection()
 
-john = User(email='jdoe@example.com', first_name='John', last_name='Doe')
+john = User(email="jdoe@example.com", first_name="John", last_name="Doe")
 john.save()
 
-post1 = TextPost(title='Fun with MongoEngine', author=john)
-post1.content = 'Took a look at MongoEngine today, looks pretty cool.'
-post1.tags = ['mongodb', 'mongoengine']
+post1 = TextPost(title="Fun with MongoEngine", author=john)
+post1.content = "Took a look at MongoEngine today, looks pretty cool."
+post1.tags = ["mongodb", "mongoengine"]
 post1.save()
 
-post2 = LinkPost(title='MongoEngine Documentation', author=john)
-post2.link_url = 'http://tractiondigital.com/labs/mongoengine/docs'
-post2.tags = ['mongoengine']
+post2 = LinkPost(title="MongoEngine Documentation", author=john)
+post2.link_url = "http://tractiondigital.com/labs/mongoengine/docs"
+post2.tags = ["mongoengine"]
 post2.save()
 
-print 'ALL POSTS'
-print
+print("ALL POSTS")
+print()
 for post in Post.objects:
-    print post.title
-    print '=' * len(post.title)
+    print(post.title)
+    # print '=' * post.title.count()
+    print("=" * 20)
 
     if isinstance(post, TextPost):
-        print post.content
+        print(post.content)
 
     if isinstance(post, LinkPost):
-        print 'Link:', post.link_url
+        print("Link:", post.link_url)
 
-    print
-print
+    print()
+print()
 
-print 'POSTS TAGGED \'MONGODB\''
-print
-for post in Post.objects(tags='mongodb'):
-    print post.title
-print
+print("POSTS TAGGED 'MONGODB'")
+print()
+for post in Post.objects(tags="mongodb"):
+    print(post.title)
+print()
 
-num_posts = Post.objects(tags='mongodb').count()
-print 'Found %d posts with tag "mongodb"' % num_posts
+num_posts = Post.objects(tags="mongodb").count()
+print('Found %d posts with tag "mongodb"' % num_posts)

docs/conf.py

@@ -11,40 +11,44 @@
 # All configuration values have a default; values that are commented out
 # serve to show the default.
 
-import sys, os
+import os
+import sys
+
+import sphinx_rtd_theme
+
+import mongoengine
 
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
-sys.path.append(os.path.abspath('..'))
+sys.path.insert(0, os.path.abspath(".."))
 
 # -- General configuration -----------------------------------------------------
 
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
-extensions = ['sphinx.ext.autodoc']
+extensions = ["sphinx.ext.autodoc", "sphinx.ext.todo"]
 
 # Add any paths that contain templates here, relative to this directory.
-templates_path = ['_templates']
+templates_path = ["_templates"]
 
 # The suffix of source filenames.
-source_suffix = '.rst'
+source_suffix = ".rst"
 
 # The encoding of source files.
-#source_encoding = 'utf-8'
+# source_encoding = 'utf-8'
 
 # The master toctree document.
-master_doc = 'index'
+master_doc = "index"
 
 # General information about the project.
-project = u'MongoEngine'
-copyright = u'2009, Harry Marr'
+project = u"MongoEngine"
+copyright = u"2009, MongoEngine Authors"
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
 # built documents.
 #
-import mongoengine
 # The short X.Y version.
 version = mongoengine.get_version()
 # The full version, including alpha/beta/rc tags.
@@ -52,144 +56,149 @@ release = mongoengine.get_version()
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.
-#language = None
+# language = None
 
 # There are two options for replacing |today|: either, you set today to some
 # non-false value, then it is used:
-#today = ''
+# today = ''
 # Else, today_fmt is used as the format for a strftime call.
-#today_fmt = '%B %d, %Y'
+# today_fmt = '%B %d, %Y'
 
 # List of documents that shouldn't be included in the build.
-#unused_docs = []
+# unused_docs = []
 
 # List of directories, relative to source directory, that shouldn't be searched
 # for source files.
-exclude_trees = ['_build']
+exclude_trees = ["_build"]
 
 # The reST default role (used for this markup: `text`) to use for all documents.
-#default_role = None
+# default_role = None
 
 # If true, '()' will be appended to :func: etc. cross-reference text.
-#add_function_parentheses = True
+# add_function_parentheses = True
 
 # If true, the current module name will be prepended to all description
 # unit titles (such as .. function::).
-#add_module_names = True
+# add_module_names = True
 
 # If true, sectionauthor and moduleauthor directives will be shown in the
 # output. They are ignored by default.
-#show_authors = False
+# show_authors = False
 
 # The name of the Pygments (syntax highlighting) style to use.
-pygments_style = 'sphinx'
+pygments_style = "sphinx"
 
 # A list of ignored prefixes for module index sorting.
-#modindex_common_prefix = []
+# modindex_common_prefix = []
 
 
 # -- Options for HTML output ---------------------------------------------------
 
 # The theme to use for HTML and HTML Help pages.  Major themes that come with
 # Sphinx are currently 'default' and 'sphinxdoc'.
-html_theme = 'nature'
+html_theme = "sphinx_rtd_theme"
 
 # Theme options are theme-specific and customize the look and feel of a theme
 # further.  For a list of options available for each theme, see the
 # documentation.
-#html_theme_options = {}
+html_theme_options = {"canonical_url": "http://docs.mongoengine.org/en/latest/"}
 
 # Add any paths that contain custom themes here, relative to this directory.
-html_theme_path = ['_themes']
+html_theme_path = [sphinx_rtd_theme.get_html_theme_path()]
 
 # The name for this set of Sphinx documents.  If None, it defaults to
 # "<project> v<release> documentation".
-#html_title = None
+# html_title = None
 
 # A shorter title for the navigation bar.  Default is the same as html_title.
-#html_short_title = None
+# html_short_title = None
 
 # The name of an image file (relative to this directory) to place at the top
 # of the sidebar.
-#html_logo = None
+# html_logo = None
 
 # The name of an image file (within the static path) to use as favicon of the
 # docs.  This file should be a Windows icon file (.ico) being 16x16 or 32x32
 # pixels large.
-#html_favicon = None
+html_favicon = "favicon.ico"
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+# html_static_path = ['_static']
 
 # If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
 # using the given strftime format.
-#html_last_updated_fmt = '%b %d, %Y'
+# html_last_updated_fmt = '%b %d, %Y'
 
 # If true, SmartyPants will be used to convert quotes and dashes to
 # typographically correct entities.
 html_use_smartypants = True
 
 # Custom sidebar templates, maps document names to template names.
-#html_sidebars = {}
+html_sidebars = {
+    "index": ["globaltoc.html", "searchbox.html"],
+    "**": ["localtoc.html", "relations.html", "searchbox.html"],
+}
+
 
 # Additional templates that should be rendered to pages, maps page names to
 # template names.
-#html_additional_pages = {}
+# html_additional_pages = {}
 
 # If false, no module index is generated.
-#html_use_modindex = True
+# html_use_modindex = True
 
 # If false, no index is generated.
-#html_use_index = True
+# html_use_index = True
 
 # If true, the index is split into individual pages for each letter.
-#html_split_index = False
+# html_split_index = False
 
 # If true, links to the reST sources are added to the pages.
-#html_show_sourcelink = True
+# html_show_sourcelink = True
 
 # If true, an OpenSearch description file will be output, and all pages will
 # contain a <link> tag referring to it.  The value of this option must be the
 # base URL from which the finished HTML is served.
-#html_use_opensearch = ''
+# html_use_opensearch = ''
 
 # If nonempty, this is the file name suffix for HTML files (e.g. ".xhtml").
-#html_file_suffix = ''
+# html_file_suffix = ''
 
 # Output file base name for HTML help builder.
-htmlhelp_basename = 'MongoEnginedoc'
+htmlhelp_basename = "MongoEnginedoc"
 
 
 # -- Options for LaTeX output --------------------------------------------------
 
 # The paper size ('letter' or 'a4').
-latex_paper_size = 'a4'
+latex_paper_size = "a4"
 
 # The font size ('10pt', '11pt' or '12pt').
-#latex_font_size = '10pt'
+# latex_font_size = '10pt'
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
 latex_documents = [
-  ('index', 'MongoEngine.tex', u'MongoEngine Documentation',
-   u'Harry Marr', 'manual'),
+    ("index", "MongoEngine.tex", "MongoEngine Documentation", "Ross Lawley", "manual")
 ]
 
 # The name of an image file (relative to this directory) to place at the top of
 # the title page.
-#latex_logo = None
+# latex_logo = None
 
 # For "manual" documents, if this is true, then toplevel headings are parts,
 # not chapters.
-#latex_use_parts = False
+# latex_use_parts = False
 
 # Additional stuff for the LaTeX preamble.
-#latex_preamble = ''
+# latex_preamble = ''
 
 # Documents to append as an appendix to all manuals.
-#latex_appendices = []
+# latex_appendices = []
 
 # If false, no module index is generated.
-#latex_use_modindex = True
+# latex_use_modindex = True
+
+autoclass_content = "both"

docs/django.rst

@@ -1,46 +1,19 @@
-=============================
-Using MongoEngine with Django
-=============================
-
-Connecting
-==========
-In your **settings.py** file, ignore the standard database settings (unless you
-also plan to use the ORM in your project), and instead call 
-:func:`~mongoengine.connect` somewhere in the settings module.
-
-Authentication
 ==============
-MongoEngine includes a Django authentication backend, which uses MongoDB. The
-:class:`~mongoengine.django.auth.User` model is a MongoEngine 
-:class:`~mongoengine.Document`, but implements most of the methods and 
-attributes that the standard Django :class:`User` model does - so the two are
-moderately compatible. Using this backend will allow you to store users in 
-MongoDB but still use many of the Django authentication infrastucture (such as
-the :func:`login_required` decorator and the :func:`authenticate` function). To
-enable the MongoEngine auth backend, add the following to you **settings.py**
-file::
+Django Support
+==============
 
-    AUTHENTICATION_BACKENDS = (
-        'mongoengine.django.auth.MongoEngineBackend',
-    )
+.. note:: Django support has been split from the main MongoEngine
+    repository. The *legacy* Django extension may be found bundled with the
+    0.9 release of MongoEngine.
 
-The :mod:`~mongoengine.django.auth` module also contains a 
-:func:`~mongoengine.django.auth.get_user` helper function, that takes a user's
-:attr:`id` and returns a :class:`~mongoengine.django.auth.User` object.
 
-.. versionadded:: 0.1.3
 
-Sessions
-========
-Django allows the use of different backend stores for its sessions. MongoEngine
-provides a MongoDB-based session backend for Django, which allows you to use
-sessions in you Django application with just MongoDB. To enable the MongoEngine
-session backend, ensure that your settings module has
-``'django.contrib.sessions.middleware.SessionMiddleware'`` in the
-``MIDDLEWARE_CLASSES`` field  and ``'django.contrib.sessions'`` in your
-``INSTALLED_APPS``. From there, all you need to do is add the following line
-into you settings module::
+Help Wanted!
+------------
 
-    SESSION_ENGINE = 'mongoengine.django.sessions'
-
-.. versionadded:: 0.2.1
+The MongoEngine team is looking for help contributing and maintaining a new
+Django extension for MongoEngine! If you have Django experience and would like
+to help contribute to the project, please get in touch on the
+`mailing list <http://groups.google.com/group/mongoengine-users>`_ or by
+simply contributing on
+`GitHub <https://github.com/MongoEngine/django-mongoengine>`_.

docs/faq.rst

@@ -0,0 +1,12 @@
+==========================
+Frequently Asked Questions
+==========================
+
+Does MongoEngine support asynchronous drivers (Motor, TxMongo)?
+---------------------------------------------------------------
+
+No, MongoEngine is exclusively based on PyMongo and isn't designed to support other driver.
+If this is a requirement for your project, check the alternative:  `uMongo`_ and `MotorEngine`_.
+
+.. _uMongo: https://umongo.readthedocs.io/
+.. _MotorEngine: https://motorengine.readthedocs.io/

docs/guide/connecting.rst

@@ -3,18 +3,175 @@
 =====================
 Connecting to MongoDB
 =====================
-To connect to a running instance of :program:`mongod`, use the
-:func:`~mongoengine.connect` function. The first argument is the name of the
-database to connect to. If the database does not exist, it will be created. If
-the database requires authentication, :attr:`username` and :attr:`password`
-arguments may be provided::
+
+Connections in MongoEngine are registered globally and are identified with aliases.
+If no `alias` is provided during the connection, it will use "default" as alias.
+
+To connect to a running instance of :program:`mongod`, use the :func:`~mongoengine.connect`
+function. The first argument is the name of the database to connect to::
 
     from mongoengine import connect
-    connect('project1', username='webapp', password='pwd123')
+    connect('project1')
 
 By default, MongoEngine assumes that the :program:`mongod` instance is running
-on **localhost** on port **27017**. If MongoDB is running elsewhere, you may
-provide :attr:`host` and :attr:`port` arguments to
+on **localhost** on port **27017**. If MongoDB is running elsewhere, you should
+provide the :attr:`host` and :attr:`port` arguments to
 :func:`~mongoengine.connect`::
 
     connect('project1', host='192.168.1.35', port=12345)
+
+If the database requires authentication, :attr:`username`, :attr:`password`
+and :attr:`authentication_source` arguments should be provided::
+
+    connect('project1', username='webapp', password='pwd123', authentication_source='admin')
+
+URI style connections are also supported -- just supply the URI as
+the :attr:`host` to
+:func:`~mongoengine.connect`::
+
+    connect('project1', host='mongodb://localhost/database_name')
+
+.. note:: Database, username and password from URI string overrides
+    corresponding parameters in :func:`~mongoengine.connect`: ::
+
+        connect(
+            db='test',
+            username='user',
+            password='12345',
+            host='mongodb://admin:qwerty@localhost/production'
+        )
+
+    will establish connection to ``production`` database using
+    ``admin`` username and ``qwerty`` password.
+
+.. note:: Calling :func:`~mongoengine.connect` without argument will establish
+    a connection to the "test" database by default
+
+Replica Sets
+============
+
+MongoEngine supports connecting to replica sets::
+
+    from mongoengine import connect
+
+    # Regular connect
+    connect('dbname', replicaset='rs-name')
+
+    # MongoDB URI-style connect
+    connect(host='mongodb://localhost/dbname?replicaSet=rs-name')
+
+Read preferences are supported through the connection or via individual
+queries by passing the read_preference ::
+
+    Bar.objects().read_preference(ReadPreference.PRIMARY)
+    Bar.objects(read_preference=ReadPreference.PRIMARY)
+
+Multiple Databases
+==================
+
+To use multiple databases you can use :func:`~mongoengine.connect` and provide
+an `alias` name for the connection - if no `alias` is provided then "default"
+is used.
+
+In the background this uses :func:`~mongoengine.register_connection` to
+store the data and you can register all aliases up front if required.
+
+Documents defined in different database
+---------------------------------------
+Individual documents can be attached to different databases by providing a
+`db_alias` in their meta data. This allows :class:`~pymongo.dbref.DBRef`
+objects to point across databases and collections. Below is an example schema,
+using 3 different databases to store data::
+
+        connect(alias='user-db-alias', db='user-db')
+        connect(alias='book-db-alias', db='book-db')
+        connect(alias='users-books-db-alias', db='users-books-db')
+
+        class User(Document):
+            name = StringField()
+
+            meta = {'db_alias': 'user-db-alias'}
+
+        class Book(Document):
+            name = StringField()
+
+            meta = {'db_alias': 'book-db-alias'}
+
+        class AuthorBooks(Document):
+            author = ReferenceField(User)
+            book = ReferenceField(Book)
+
+            meta = {'db_alias': 'users-books-db-alias'}
+
+
+Disconnecting an existing connection
+------------------------------------
+The function :func:`~mongoengine.disconnect` can be used to
+disconnect a particular connection. This can be used to change a
+connection globally::
+
+        from mongoengine import connect, disconnect
+        connect('a_db', alias='db1')
+
+        class User(Document):
+            name = StringField()
+            meta = {'db_alias': 'db1'}
+
+        disconnect(alias='db1')
+
+        connect('another_db', alias='db1')
+
+.. note:: Calling :func:`~mongoengine.disconnect` without argument
+    will disconnect the "default" connection
+
+.. note:: Since connections gets registered globally, it is important
+    to use the `disconnect` function from MongoEngine and not the
+    `disconnect()` method of an existing connection (pymongo.MongoClient)
+
+.. note:: :class:`~mongoengine.Document` are caching the pymongo collection.
+    using `disconnect` ensures that it gets cleaned as well
+
+Context Managers
+================
+Sometimes you may want to switch the database or collection to query against.
+For example, archiving older data into a separate database for performance
+reasons or writing functions that dynamically choose collections to write
+a document to.
+
+Switch Database
+---------------
+The :class:`~mongoengine.context_managers.switch_db` context manager allows
+you to change the database alias for a given class allowing quick and easy
+access to the same User document across databases::
+
+    from mongoengine.context_managers import switch_db
+
+    class User(Document):
+        name = StringField()
+
+        meta = {'db_alias': 'user-db'}
+
+    with switch_db(User, 'archive-user-db') as User:
+        User(name='Ross').save()  # Saves the 'archive-user-db'
+
+
+Switch Collection
+-----------------
+The :func:`~mongoengine.context_managers.switch_collection` context manager
+allows you to change the collection for a given class allowing quick and easy
+access to the same Group document across collection::
+
+        from mongoengine.context_managers import switch_collection
+
+        class Group(Document):
+            name = StringField()
+
+        Group(name='test').save()  # Saves in the default db
+
+        with switch_collection(Group, 'group2000') as Group:
+            Group(name='hello Group 2000 collection!').save()  # Saves in group2000 collection
+
+
+.. note:: Make sure any aliases have been registered with
+    :func:`~mongoengine.register_connection` or :func:`~mongoengine.connect`
+    before using the context manager.

docs/guide/defining-documents.rst

@@ -4,7 +4,7 @@ Defining documents
 In MongoDB, a **document** is roughly equivalent to a **row** in an RDBMS. When
 working with relational databases, rows are stored in **tables**, which have a
 strict **schema** that the rows follow. MongoDB stores documents in
-**collections** rather than tables - the principle difference is that no schema 
+**collections** rather than tables --- the principal difference is that no schema
 is enforced at a database level.
 
 Defining a document's schema
@@ -22,7 +22,39 @@ objects** as class attributes to the document class::
 
     class Page(Document):
         title = StringField(max_length=200, required=True)
-        date_modified = DateTimeField(default=datetime.now)
+        date_modified = DateTimeField(default=datetime.datetime.utcnow)
+
+As BSON (the binary format for storing data in mongodb) is order dependent,
+documents are serialized based on their field order.
+
+Dynamic document schemas
+========================
+One of the benefits of MongoDB is dynamic schemas for a collection, whilst data
+should be planned and organised (after all explicit is better than implicit!)
+there are scenarios where having dynamic / expando style documents is desirable.
+
+:class:`~mongoengine.DynamicDocument` documents work in the same way as
+:class:`~mongoengine.Document` but any data / attributes set to them will also
+be saved ::
+
+    from mongoengine import *
+
+    class Page(DynamicDocument):
+        title = StringField(max_length=200, required=True)
+
+    # Create a new page and add tags
+    >>> page = Page(title='Using MongoEngine')
+    >>> page.tags = ['mongodb', 'mongoengine']
+    >>> page.save()
+
+    >>> Page.objects(tags='mongoengine').count()
+    >>> 1
+
+.. note::
+
+   There is one caveat on Dynamic Documents: fields cannot start with `_`
+
+Dynamic fields are stored in creation order *after* any declared fields.
 
 Fields
 ======
@@ -34,27 +66,150 @@ not provided. Default values may optionally be a callable, which will be called
 to retrieve the value (such as in the above example). The field types available
 are as follows:
 
-* :class:`~mongoengine.StringField`
-* :class:`~mongoengine.IntField`
-* :class:`~mongoengine.FloatField`
-* :class:`~mongoengine.DateTimeField`
-* :class:`~mongoengine.ListField`
-* :class:`~mongoengine.ObjectIdField`
-* :class:`~mongoengine.EmbeddedDocumentField`
-* :class:`~mongoengine.ReferenceField`
+* :class:`~mongoengine.fields.BinaryField`
+* :class:`~mongoengine.fields.BooleanField`
+* :class:`~mongoengine.fields.ComplexDateTimeField`
+* :class:`~mongoengine.fields.DateTimeField`
+* :class:`~mongoengine.fields.DecimalField`
+* :class:`~mongoengine.fields.DictField`
+* :class:`~mongoengine.fields.DynamicField`
+* :class:`~mongoengine.fields.EmailField`
+* :class:`~mongoengine.fields.EmbeddedDocumentField`
+* :class:`~mongoengine.fields.EmbeddedDocumentListField`
+* :class:`~mongoengine.fields.FileField`
+* :class:`~mongoengine.fields.FloatField`
+* :class:`~mongoengine.fields.GenericEmbeddedDocumentField`
+* :class:`~mongoengine.fields.GenericReferenceField`
+* :class:`~mongoengine.fields.GenericLazyReferenceField`
+* :class:`~mongoengine.fields.GeoPointField`
+* :class:`~mongoengine.fields.ImageField`
+* :class:`~mongoengine.fields.IntField`
+* :class:`~mongoengine.fields.ListField`
+* :class:`~mongoengine.fields.LongField`
+* :class:`~mongoengine.fields.MapField`
+* :class:`~mongoengine.fields.ObjectIdField`
+* :class:`~mongoengine.fields.ReferenceField`
+* :class:`~mongoengine.fields.LazyReferenceField`
+* :class:`~mongoengine.fields.SequenceField`
+* :class:`~mongoengine.fields.SortedListField`
+* :class:`~mongoengine.fields.StringField`
+* :class:`~mongoengine.fields.URLField`
+* :class:`~mongoengine.fields.UUIDField`
+* :class:`~mongoengine.fields.PointField`
+* :class:`~mongoengine.fields.LineStringField`
+* :class:`~mongoengine.fields.PolygonField`
+* :class:`~mongoengine.fields.MultiPointField`
+* :class:`~mongoengine.fields.MultiLineStringField`
+* :class:`~mongoengine.fields.MultiPolygonField`
+
+Field arguments
+---------------
+Each field type can be customized by keyword arguments.  The following keyword
+arguments can be set on all fields:
+
+:attr:`db_field` (Default: None)
+    The MongoDB field name.
+
+:attr:`required` (Default: False)
+    If set to True and the field is not set on the document instance, a
+    :class:`~mongoengine.ValidationError` will be raised when the document is
+    validated.
+
+:attr:`default` (Default: None)
+    A value to use when no value is set for this field.
+
+    The definition of default parameters follow `the general rules on Python
+    <http://docs.python.org/reference/compound_stmts.html#function-definitions>`__,
+    which means that some care should be taken when dealing with default mutable objects
+    (like in :class:`~mongoengine.fields.ListField` or :class:`~mongoengine.fields.DictField`)::
+
+        class ExampleFirst(Document):
+            # Default an empty list
+            values = ListField(IntField(), default=list)
+
+        class ExampleSecond(Document):
+            # Default a set of values
+            values = ListField(IntField(), default=lambda: [1,2,3])
+
+        class ExampleDangerous(Document):
+            # This can make an .append call to  add values to the default (and all the following objects),
+            # instead to just an object
+            values = ListField(IntField(), default=[1,2,3])
+
+    .. note:: Unsetting a field with a default value will revert back to the default.
+
+:attr:`unique` (Default: False)
+    When True, no documents in the collection will have the same value for this
+    field.
+
+:attr:`unique_with` (Default: None)
+    A field name (or list of field names) that when taken together with this
+    field, will not have two documents in the collection with the same value.
+
+:attr:`primary_key` (Default: False)
+    When True, use this field as a primary key for the collection.  `DictField`
+    and `EmbeddedDocuments` both support being the primary key for a document.
+
+    .. note:: If set, this field is also accessible through the `pk` field.
+
+:attr:`choices` (Default: None)
+    An iterable (e.g. list, tuple or set) of choices to which the value of this
+    field should be limited.
+
+    Can either be nested tuples of value (stored in mongo) and a
+    human readable key ::
+
+        SIZE = (('S', 'Small'),
+                ('M', 'Medium'),
+                ('L', 'Large'),
+                ('XL', 'Extra Large'),
+                ('XXL', 'Extra Extra Large'))
+
+
+        class Shirt(Document):
+            size = StringField(max_length=3, choices=SIZE)
+
+    Or a flat iterable just containing values ::
+
+        SIZE = ('S', 'M', 'L', 'XL', 'XXL')
+
+        class Shirt(Document):
+            size = StringField(max_length=3, choices=SIZE)
+
+:attr:`validation` (Optional)
+    A callable to validate the value of the field.
+    The callable takes the value as parameter and should raise a ValidationError
+    if validation fails
+
+    e.g ::
+
+        def _not_empty(val):
+            if not val:
+                raise ValidationError('value can not be empty')
+
+        class Person(Document):
+            name = StringField(validation=_not_empty)
+
+
+:attr:`**kwargs` (Optional)
+    You can supply additional metadata as arbitrary additional keyword
+    arguments.  You can not override existing attributes, however.  Common
+    choices include `help_text` and `verbose_name`, commonly used by form and
+    widget libraries.
+
 
 List fields
-^^^^^^^^^^^
-MongoDB allows the storage of lists of items. To add a list of items to a
-:class:`~mongoengine.Document`, use the :class:`~mongoengine.ListField` field
-type. :class:`~mongoengine.ListField` takes another field object as its first
+-----------
+MongoDB allows storing lists of items. To add a list of items to a
+:class:`~mongoengine.Document`, use the :class:`~mongoengine.fields.ListField` field
+type. :class:`~mongoengine.fields.ListField` takes another field object as its first
 argument, which specifies which type elements may be stored within the list::
 
     class Page(Document):
         tags = ListField(StringField(max_length=50))
 
 Embedded documents
-^^^^^^^^^^^^^^^^^^
+------------------
 MongoDB has the ability to embed documents within other documents. Schemata may
 be defined for these embedded documents, just as they may be for regular
 documents. To create an embedded document, just define a document as usual, but
@@ -65,20 +220,40 @@ inherit from :class:`~mongoengine.EmbeddedDocument` rather than
         content = StringField()
 
 To embed the document within another document, use the
-:class:`~mongoengine.EmbeddedDocumentField` field type, providing the embedded
+:class:`~mongoengine.fields.EmbeddedDocumentField` field type, providing the embedded
 document class as the first argument::
 
     class Page(Document):
         comments = ListField(EmbeddedDocumentField(Comment))
 
-    comment1 = Comment('Good work!')
-    comment2 = Comment('Nice article!')
+    comment1 = Comment(content='Good work!')
+    comment2 = Comment(content='Nice article!')
     page = Page(comments=[comment1, comment2])
 
+Dictionary Fields
+-----------------
+Often, an embedded document may be used instead of a dictionary – generally
+embedded documents are recommended as dictionaries don’t support validation
+or custom field types. However, sometimes you will not know the structure of what you want to
+store; in this situation a :class:`~mongoengine.fields.DictField` is appropriate::
+
+    class SurveyResponse(Document):
+        date = DateTimeField()
+        user = ReferenceField(User)
+        answers = DictField()
+
+    survey_response = SurveyResponse(date=datetime.utcnow(), user=request.user)
+    response_form = ResponseForm(request.POST)
+    survey_response.answers = response_form.cleaned_data()
+    survey_response.save()
+
+Dictionaries can store complex data, other dictionaries, lists, references to
+other objects, so are the most flexible field type available.
+
 Reference fields
-^^^^^^^^^^^^^^^^
+----------------
 References may be stored to other documents in the database using the
-:class:`~mongoengine.ReferenceField`. Pass in another document class as the
+:class:`~mongoengine.fields.ReferenceField`. Pass in another document class as the
 first argument to the constructor, then simply assign document objects to the
 field::
 
@@ -99,26 +274,177 @@ field::
 The :class:`User` object is automatically turned into a reference behind the
 scenes, and dereferenced when the :class:`Page` object is retrieved.
 
+To add a :class:`~mongoengine.fields.ReferenceField` that references the document
+being defined, use the string ``'self'`` in place of the document class as the
+argument to :class:`~mongoengine.fields.ReferenceField`'s constructor. To reference a
+document that has not yet been defined, use the name of the undefined document
+as the constructor's argument::
+
+    class Employee(Document):
+        name = StringField()
+        boss = ReferenceField('self')
+        profile_page = ReferenceField('ProfilePage')
+
+    class ProfilePage(Document):
+        content = StringField()
+
+
+.. _one-to-many-with-listfields:
+
+One to Many with ListFields
+'''''''''''''''''''''''''''
+
+If you are implementing a one to many relationship via a list of references,
+then the references are stored as DBRefs and to query you need to pass an
+instance of the object to the query::
+
+    class User(Document):
+        name = StringField()
+
+    class Page(Document):
+        content = StringField()
+        authors = ListField(ReferenceField(User))
+
+    bob = User(name="Bob Jones").save()
+    john = User(name="John Smith").save()
+
+    Page(content="Test Page", authors=[bob, john]).save()
+    Page(content="Another Page", authors=[john]).save()
+
+    # Find all pages Bob authored
+    Page.objects(authors__in=[bob])
+
+    # Find all pages that both Bob and John have authored
+    Page.objects(authors__all=[bob, john])
+
+    # Remove Bob from the authors for a page.
+    Page.objects(id='...').update_one(pull__authors=bob)
+
+    # Add John to the authors for a page.
+    Page.objects(id='...').update_one(push__authors=john)
+
+
+Dealing with deletion of referred documents
+'''''''''''''''''''''''''''''''''''''''''''
+By default, MongoDB doesn't check the integrity of your data, so deleting
+documents that other documents still hold references to will lead to consistency
+issues.  Mongoengine's :class:`ReferenceField` adds some functionality to
+safeguard against these kinds of database integrity problems, providing each
+reference with a delete rule specification.  A delete rule is specified by
+supplying the :attr:`reverse_delete_rule` attributes on the
+:class:`ReferenceField` definition, like this::
+
+    class ProfilePage(Document):
+        ...
+        employee = ReferenceField('Employee', reverse_delete_rule=mongoengine.CASCADE)
+
+The declaration in this example means that when an :class:`Employee` object is
+removed, the :class:`ProfilePage` that references that employee is removed as
+well.  If a whole batch of employees is removed, all profile pages that are
+linked are removed as well.
+
+Its value can take any of the following constants:
+
+:const:`mongoengine.DO_NOTHING`
+  This is the default and won't do anything.  Deletes are fast, but may cause
+  database inconsistency or dangling references.
+:const:`mongoengine.DENY`
+  Deletion is denied if there still exist references to the object being
+  deleted.
+:const:`mongoengine.NULLIFY`
+  Any object's fields still referring to the object being deleted are set to None
+  (using MongoDB's "unset" operation), effectively nullifying the relationship.
+:const:`mongoengine.CASCADE`
+  Any object containing fields that are referring to the object being deleted
+  are deleted first.
+:const:`mongoengine.PULL`
+  Removes the reference to the object (using MongoDB's "pull" operation)
+  from any object's fields of
+  :class:`~mongoengine.fields.ListField` (:class:`~mongoengine.fields.ReferenceField`).
+
+
+.. warning::
+   A safety note on setting up these delete rules!  Since the delete rules are
+   not recorded on the database level by MongoDB itself, but instead at runtime,
+   in-memory, by the MongoEngine module, it is of the upmost importance
+   that the module that declares the relationship is loaded **BEFORE** the
+   delete is invoked.
+
+   If, for example, the :class:`Employee` object lives in the
+   :mod:`payroll` app, and the :class:`ProfilePage` in the :mod:`people`
+   app, it is extremely important that the :mod:`people` app is loaded
+   before any employee is removed, because otherwise, MongoEngine could
+   never know this relationship exists.
+
+   In Django, be sure to put all apps that have such delete rule declarations in
+   their :file:`models.py` in the :const:`INSTALLED_APPS` tuple.
+
+Generic reference fields
+''''''''''''''''''''''''
+A second kind of reference field also exists,
+:class:`~mongoengine.fields.GenericReferenceField`. This allows you to reference any
+kind of :class:`~mongoengine.Document`, and hence doesn't take a
+:class:`~mongoengine.Document` subclass as a constructor argument::
+
+    class Link(Document):
+        url = StringField()
+
+    class Post(Document):
+        title = StringField()
+
+    class Bookmark(Document):
+        bookmark_object = GenericReferenceField()
+
+    link = Link(url='http://hmarr.com/mongoengine/')
+    link.save()
+
+    post = Post(title='Using MongoEngine')
+    post.save()
+
+    Bookmark(bookmark_object=link).save()
+    Bookmark(bookmark_object=post).save()
+
+.. note::
+
+   Using :class:`~mongoengine.fields.GenericReferenceField`\ s is slightly less
+   efficient than the standard :class:`~mongoengine.fields.ReferenceField`\ s, so if
+   you will only be referencing one document type, prefer the standard
+   :class:`~mongoengine.fields.ReferenceField`.
+
 Uniqueness constraints
-^^^^^^^^^^^^^^^^^^^^^^
+----------------------
 MongoEngine allows you to specify that a field should be unique across a
-collection by providing ``unique=True`` to a :class:`~mongoengine.Field`\ 's
+collection by providing ``unique=True`` to a :class:`~mongoengine.fields.Field`\ 's
 constructor. If you try to save a document that has the same value for a unique
 field as a document that is already in the database, a
-:class:`~mongoengine.OperationError` will be raised. You may also specify
+:class:`~mongoengine.NotUniqueError` will be raised. You may also specify
 multi-field uniqueness constraints by using :attr:`unique_with`, which may be
 either a single field name, or a list or tuple of field names::
 
     class User(Document):
         username = StringField(unique=True)
         first_name = StringField()
-        last_name = StringField(unique_with='last_name')
+        last_name = StringField(unique_with='first_name')
+
+Skipping Document validation on save
+------------------------------------
+You can also skip the whole document validation process by setting
+``validate=False`` when calling the :meth:`~mongoengine.document.Document.save`
+method::
+
+    class Recipient(Document):
+        name = StringField()
+        email = EmailField()
+
+    recipient = Recipient(name='admin', email='root@localhost')
+    recipient.save()               # will raise a ValidationError while
+    recipient.save(validate=False) # won't
 
 Document collections
 ====================
 Document classes that inherit **directly** from :class:`~mongoengine.Document`
 will have their own **collection** in the database. The name of the collection
-is by default the name of the class, coverted to lowercase (so in the example
+is by default the name of the class, converted to lowercase (so in the example
 above, the collection would be called `page`). If you need to change the name
 of the collection (e.g. to use MongoEngine with an existing database), then
 create a class dictionary attribute called :attr:`meta` on your document, and
@@ -130,13 +456,15 @@ document class to use::
         meta = {'collection': 'cmsPage'}
 
 Capped collections
-^^^^^^^^^^^^^^^^^^
+------------------
 A :class:`~mongoengine.Document` may use a **Capped Collection** by specifying
 :attr:`max_documents` and :attr:`max_size` in the :attr:`meta` dictionary.
 :attr:`max_documents` is the maximum number of documents that is allowed to be
 stored in the collection, and :attr:`max_size` is the maximum size of the
-collection in bytes. If :attr:`max_size` is not specified and
-:attr:`max_documents` is, :attr:`max_size` defaults to 10000000 bytes (10MB).
+collection in bytes. :attr:`max_size` is rounded up to the next multiple of 256
+by MongoDB internally and mongoengine before. Use also a multiple of 256 to
+avoid confusions. If :attr:`max_size` is not specified and
+:attr:`max_documents` is, :attr:`max_size` defaults to 10485760 bytes (10MB).
 The following example shows a :class:`Log` document that will be limited to
 1000 entries and 2MB of disk space::
 
@@ -144,22 +472,199 @@ The following example shows a :class:`Log` document that will be limited to
         ip_address = StringField()
         meta = {'max_documents': 1000, 'max_size': 2000000}
 
+.. defining-indexes_
+
 Indexes
 =======
+
 You can specify indexes on collections to make querying faster. This is done
 by creating a list of index specifications called :attr:`indexes` in the
 :attr:`~mongoengine.Document.meta` dictionary, where an index specification may
-either be a single field name, or a tuple containing multiple field names. A
-direction may be specified on fields by prefixing the field name with a **+**
-or a **-** sign. Note that direction only matters on multi-field indexes. ::
+either be a single field name, a tuple containing multiple field names, or a
+dictionary containing a full index definition.
+
+A direction may be specified on fields by prefixing the field name with a
+**+** (for ascending) or a **-** sign (for descending). Note that direction
+only matters on multi-field indexes. Text indexes may be specified by prefixing
+the field name with a **$**. Hashed indexes may be specified by prefixing
+the field name with a **#**::
+
+    class Page(Document):
+        category = IntField()
+        title = StringField()
+        rating = StringField()
+        created = DateTimeField()
+        meta = {
+            'indexes': [
+                'title',
+                '$title',  # text index
+                '#title',  # hashed index
+                ('title', '-rating'),
+                ('category', '_cls'),
+                {
+                    'fields': ['created'],
+                    'expireAfterSeconds': 3600
+                }
+            ]
+        }
+
+If a dictionary is passed then additional options become available. Valid options include,
+but are not limited to:
+
+
+:attr:`fields` (Default: None)
+    The fields to index. Specified in the same format as described above.
+
+:attr:`cls` (Default: True)
+    If you have polymorphic models that inherit and have
+    :attr:`allow_inheritance` turned on, you can configure whether the index
+    should have the :attr:`_cls` field added automatically to the start of the
+    index.
+
+:attr:`sparse` (Default: False)
+    Whether the index should be sparse.
+
+:attr:`unique` (Default: False)
+    Whether the index should be unique.
+
+:attr:`expireAfterSeconds` (Optional)
+    Allows you to automatically expire data from a collection by setting the
+    time in seconds to expire the a field.
+
+:attr:`name` (Optional)
+    Allows you to specify a name for the index
+
+:attr:`collation` (Optional)
+    Allows to create case insensitive indexes (MongoDB v3.4+ only)
+
+.. note::
+
+    Additional options are forwarded as **kwargs to pymongo's create_index method.
+    Inheritance adds extra fields indices see: :ref:`document-inheritance`.
+
+Global index default options
+----------------------------
+
+There are a few top level defaults for all indexes that can be set::
 
     class Page(Document):
         title = StringField()
         rating = StringField()
         meta = {
-            'indexes': ['title', ('title', '-rating')]
+            'index_opts': {},
+            'index_background': True,
+            'index_cls': False,
+            'auto_create_index': True,
         }
 
+
+:attr:`index_opts` (Optional)
+    Set any default index options - see the `full options list <https://docs.mongodb.com/manual/reference/method/db.collection.createIndex/#db.collection.createIndex>`_
+
+:attr:`index_background` (Optional)
+    Set the default value for if an index should be indexed in the background
+
+:attr:`index_cls` (Optional)
+    A way to turn off a specific index for _cls.
+
+:attr:`auto_create_index` (Optional)
+    When this is True (default), MongoEngine will ensure that the correct
+    indexes exist in MongoDB each time a command is run. This can be disabled
+    in systems where indexes are managed separately. Disabling this will improve
+    performance.
+
+
+Compound Indexes and Indexing sub documents
+-------------------------------------------
+
+Compound indexes can be created by adding the Embedded field or dictionary
+field name to the index definition.
+
+Sometimes its more efficient to index parts of Embedded / dictionary fields,
+in this case use 'dot' notation to identify the value to index eg: `rank.title`
+
+.. _geospatial-indexes:
+
+Geospatial indexes
+------------------
+
+The best geo index for mongodb is the new "2dsphere", which has an improved
+spherical model and provides better performance and more options when querying.
+The following fields will explicitly add a "2dsphere" index:
+
+    - :class:`~mongoengine.fields.PointField`
+    - :class:`~mongoengine.fields.LineStringField`
+    - :class:`~mongoengine.fields.PolygonField`
+    - :class:`~mongoengine.fields.MultiPointField`
+    - :class:`~mongoengine.fields.MultiLineStringField`
+    - :class:`~mongoengine.fields.MultiPolygonField`
+
+As "2dsphere" indexes can be part of a compound index, you may not want the
+automatic index but would prefer a compound index.  In this example we turn off
+auto indexing and explicitly declare a compound index on ``location`` and ``datetime``::
+
+    class Log(Document):
+        location = PointField(auto_index=False)
+        datetime = DateTimeField()
+
+        meta = {
+            'indexes': [[("location", "2dsphere"), ("datetime", 1)]]
+        }
+
+
+Pre MongoDB 2.4 Geo
+'''''''''''''''''''
+
+.. note:: For MongoDB < 2.4 this is still current, however the new 2dsphere
+    index is a big improvement over the previous 2D model - so upgrading is
+    advised.
+
+Geospatial indexes will be automatically created for all
+:class:`~mongoengine.fields.GeoPointField`\ s
+
+It is also possible to explicitly define geospatial indexes. This is
+useful if you need to define a geospatial index on a subfield of a
+:class:`~mongoengine.fields.DictField` or a custom field that contains a
+point. To create a geospatial index you must prefix the field with the
+***** sign. ::
+
+    class Place(Document):
+        location = DictField()
+        meta = {
+            'indexes': [
+                '*location.point',
+            ],
+        }
+
+Time To Live indexes
+--------------------
+
+A special index type that allows you to automatically expire data from a
+collection after a given period. See the official
+`ttl <http://docs.mongodb.org/manual/tutorial/expire-data/#expire-data-from-collections-by-setting-ttl>`_
+documentation for more information.  A common usecase might be session data::
+
+    class Session(Document):
+        created = DateTimeField(default=datetime.utcnow)
+        meta = {
+            'indexes': [
+                {'fields': ['created'], 'expireAfterSeconds': 3600}
+            ]
+        }
+
+.. warning:: TTL indexes happen on the MongoDB server and not in the application
+    code, therefore no signals will be fired on document deletion.
+    If you need signals to be fired on deletion, then you must handle the
+    deletion of Documents in your application code.
+
+Comparing Indexes
+-----------------
+
+Use :func:`mongoengine.Document.compare_indexes` to compare actual indexes in
+the database to those that your document definitions define.  This is useful
+for maintenance purposes and ensuring you have the correct indexes for your
+schema.
+
 Ordering
 ========
 A default ordering can be specified for your
@@ -179,13 +684,13 @@ subsequent calls to :meth:`~mongoengine.queryset.QuerySet.order_by`. ::
         }
 
     blog_post_1 = BlogPost(title="Blog Post #1")
-    blog_post_1.published_date = datetime(2010, 1, 5, 0, 0 ,0))
+    blog_post_1.published_date = datetime(2010, 1, 5, 0, 0 ,0)
 
     blog_post_2 = BlogPost(title="Blog Post #2")
-    blog_post_2.published_date = datetime(2010, 1, 6, 0, 0 ,0))
+    blog_post_2.published_date = datetime(2010, 1, 6, 0, 0 ,0)
 
     blog_post_3 = BlogPost(title="Blog Post #3")
-    blog_post_3.published_date = datetime(2010, 1, 7, 0, 0 ,0))
+    blog_post_3.published_date = datetime(2010, 1, 7, 0, 0 ,0)
 
     blog_post_1.save()
     blog_post_2.save()
@@ -194,45 +699,127 @@ subsequent calls to :meth:`~mongoengine.queryset.QuerySet.order_by`. ::
     # get the "first" BlogPost using default ordering
     # from BlogPost.meta.ordering
     latest_post = BlogPost.objects.first()
-    self.assertEqual(latest_post.title, "Blog Post #3")
+    assert latest_post.title == "Blog Post #3"
 
     # override default ordering, order BlogPosts by "published_date"
     first_post = BlogPost.objects.order_by("+published_date").first()
-    self.assertEqual(first_post.title, "Blog Post #1")
+    assert first_post.title == "Blog Post #1"
+
+Shard keys
+==========
+
+If your collection is sharded by multiple keys, then you can improve shard
+routing (and thus the performance of your application) by specifying the shard
+key, using the :attr:`shard_key` attribute of
+:attr:`~mongoengine.Document.meta`. The shard key should be defined as a tuple.
+
+This ensures that the full shard key is sent with the query when calling
+methods such as :meth:`~mongoengine.document.Document.save`,
+:meth:`~mongoengine.document.Document.update`,
+:meth:`~mongoengine.document.Document.modify`, or
+:meth:`~mongoengine.document.Document.delete` on an existing
+:class:`~mongoengine.Document` instance::
+
+    class LogEntry(Document):
+        machine = StringField()
+        app = StringField()
+        timestamp = DateTimeField()
+        data = StringField()
+
+        meta = {
+            'shard_key': ('machine', 'timestamp'),
+            'indexes': ('machine', 'timestamp'),
+        }
+
+.. _document-inheritance:
 
 Document inheritance
 ====================
+
 To create a specialised type of a :class:`~mongoengine.Document` you have
 defined, you may subclass it and add any extra fields or methods you may need.
-As this is new class is not a direct subclass of
+As this new class is not a direct subclass of
 :class:`~mongoengine.Document`, it will not be stored in its own collection; it
 will use the same collection as its superclass uses. This allows for more
-convenient and efficient retrieval of related documents::
+convenient and efficient retrieval of related documents -- all you need do is
+set :attr:`allow_inheritance` to True in the :attr:`meta` data for a
+document.::
 
     # Stored in a collection named 'page'
     class Page(Document):
         title = StringField(max_length=200, required=True)
 
+        meta = {'allow_inheritance': True}
+
     # Also stored in the collection named 'page'
     class DatedPage(Page):
         date = DateTimeField()
 
+.. note:: From 0.8 onwards :attr:`allow_inheritance` defaults
+          to False, meaning you must set it to True to use inheritance.
+
+          Setting :attr:`allow_inheritance` to True should also be used in
+          :class:`~mongoengine.EmbeddedDocument` class in case you need to subclass it
+
+When it comes to querying using :attr:`.objects()`, querying `Page.objects()` will query
+both `Page` and `DatedPage` whereas querying `DatedPage` will only query the `DatedPage` documents.
+Behind the scenes, MongoEngine deals with inheritance by adding a :attr:`_cls` attribute that contains
+the class name in every documents. When a document is loaded, MongoEngine checks
+it's :attr:`_cls` attribute and use that class to construct the instance.::
+
+    Page(title='a funky title').save()
+    DatedPage(title='another title', date=datetime.utcnow()).save()
+
+    print(Page.objects().count())         # 2
+    print(DatedPage.objects().count())    # 1
+
+    # print documents in their native form
+    # we remove 'id' to avoid polluting the output with unnecessary detail
+    qs = Page.objects.exclude('id').as_pymongo()
+    print(list(qs))
+    # [
+    #   {'_cls': u 'Page', 'title': 'a funky title'},
+    #   {'_cls': u 'Page.DatedPage', 'title': u 'another title', 'date': datetime.datetime(2019, 12, 13, 20, 16, 59, 993000)}
+    # ]
+
 Working with existing data
-^^^^^^^^^^^^^^^^^^^^^^^^^^
-To enable correct retrieval of documents involved in this kind of heirarchy,
-two extra attributes are stored on each document in the database: :attr:`_cls`
-and :attr:`_types`. These are hidden from the user through the MongoEngine
-interface, but may not be present if you are trying to use MongoEngine with 
-an existing database. For this reason, you may disable this inheritance
-mechansim, removing the dependency of :attr:`_cls` and :attr:`_types`, enabling
-you to work with existing databases. To disable inheritance on a document
-class, set :attr:`allow_inheritance` to ``False`` in the :attr:`meta`
-dictionary::
+--------------------------
+As MongoEngine no longer defaults to needing :attr:`_cls`, you can quickly and
+easily get working with existing data.  Just define the document to match
+the expected schema in your database ::
 
     # Will work with data in an existing collection named 'cmsPage'
     class Page(Document):
         title = StringField(max_length=200, required=True)
         meta = {
-            'collection': 'cmsPage',
-            'allow_inheritance': False,
+            'collection': 'cmsPage'
         }
+
+If you have wildly varying schemas then using a
+:class:`~mongoengine.DynamicDocument` might be more appropriate, instead of
+defining all possible field types.
+
+If you use :class:`~mongoengine.Document` and the database contains data that
+isn't defined then that data will be stored in the `document._data` dictionary.
+
+Abstract classes
+================
+
+If you want to add some extra functionality to a group of Document classes but
+you don't need or want the overhead of inheritance you can use the
+:attr:`abstract` attribute of :attr:`~mongoengine.Document.meta`.
+This won't turn on :ref:`document-inheritance` but will allow you to keep your
+code DRY::
+
+        class BaseDocument(Document):
+            meta = {
+                'abstract': True,
+            }
+            def check_permissions(self):
+                ...
+
+        class User(BaseDocument):
+           ...
+
+Now the User class will have access to the inherited `check_permissions` method
+and won't store any of the extra `_cls` information.

docs/guide/document-instances.rst

@@ -2,7 +2,7 @@
 Documents instances
 ===================
 To create a new document object, create an instance of the relevant document
-class, providing values for its fields as its constructor keyword arguments.
+class, providing values for its fields as constructor keyword arguments.
 You may provide values for any of the fields on the document::
 
     >>> page = Page(title="Test Page")
@@ -17,21 +17,76 @@ attribute syntax::
     'Example Page'
 
 Saving and deleting documents
------------------------------
-To save the document to the database, call the
-:meth:`~mongoengine.Document.save` method. If the document does not exist in
-the database, it will be created. If it does already exist, it will be
-updated.
+=============================
+MongoEngine tracks changes to documents to provide efficient saving.  To save
+the document to the database, call the :meth:`~mongoengine.Document.save` method.
+If the document does not exist in the database, it will be created. If it does
+already exist, then any changes will be updated atomically.  For example::
 
-To delete a document, call the :meth:`~mongoengine.Document.delete` method.
-Note that this will only work if the document exists in the database and has a
-valide :attr:`id`.
+    >>> page = Page(title="Test Page")
+    >>> page.save()  # Performs an insert
+    >>> page.title = "My Page"
+    >>> page.save()  # Performs an atomic set on the title field.
+
+.. note::
+
+    Changes to documents are tracked and on the whole perform ``set`` operations.
+
+    * ``list_field.push(0)`` --- *sets* the resulting list
+    * ``del(list_field)``   --- *unsets* whole list
+
+    With lists its preferable to use ``Doc.update(push__list_field=0)`` as
+    this stops the whole list being updated --- stopping any race conditions.
 
 .. seealso::
     :ref:`guide-atomic-updates`
 
+Pre save data validation and cleaning
+-------------------------------------
+MongoEngine allows you to create custom cleaning rules for your documents when
+calling :meth:`~mongoengine.Document.save`.  By providing a custom
+:meth:`~mongoengine.Document.clean` method you can do any pre validation / data
+cleaning.
+
+This might be useful if you want to ensure a default value based on other
+document values for example::
+
+    class Essay(Document):
+        status = StringField(choices=('Published', 'Draft'), required=True)
+        pub_date = DateTimeField()
+
+        def clean(self):
+            """Ensures that only published essays have a `pub_date` and
+            automatically sets `pub_date` if essay is published and `pub_date`
+            is not set"""
+            if self.status == 'Draft' and self.pub_date is not None:
+                msg = 'Draft entries should not have a publication date.'
+                raise ValidationError(msg)
+            # Set the pub_date for published items if not set.
+            if self.status == 'Published' and self.pub_date is None:
+                self.pub_date = datetime.now()
+
+.. note::
+    Cleaning is only called if validation is turned on and when calling
+    :meth:`~mongoengine.Document.save`.
+
+Cascading Saves
+---------------
+If your document contains :class:`~mongoengine.fields.ReferenceField` or
+:class:`~mongoengine.fields.GenericReferenceField` objects, then by default the
+:meth:`~mongoengine.Document.save` method will not save any changes to
+those objects.  If you want all references to be saved also, noting each
+save is a separate query, then passing :attr:`cascade` as True
+to the save method will cascade any saves.
+
+Deleting documents
+------------------
+To delete a document, call the :meth:`~mongoengine.Document.delete` method.
+Note that this will only work if the document exists in the database and has a
+valid :attr:`id`.
+
 Document IDs
-------------
+============
 Each document in the database has a unique id. This may be accessed through the
 :attr:`id` attribute on :class:`~mongoengine.Document` objects. Usually, the id
 will be generated automatically by the database server when the object is save,
@@ -59,7 +114,16 @@ you may still use :attr:`id` to access the primary key if you want::
     >>> bob.id == bob.email == 'bob@example.com'
     True
 
+You can also access the document's "primary key" using the :attr:`pk` field,
+it's an alias to :attr:`id`::
+
+    >>> page = Page(title="Another Test Page")
+    >>> page.save()
+    >>> page.id == page.pk
+    True
+
 .. note::
+
    If you define your own primary key field, the field implicitly becomes
-   required, so a :class:`ValidationError` will be thrown if you don't provide
-   it.
+   required, so a :class:`~mongoengine.ValidationError` will be thrown if
+   you don't provide it.

docs/guide/gridfs.rst

@@ -0,0 +1,91 @@
+======
+GridFS
+======
+
+.. versionadded:: 0.4
+
+Writing
+-------
+
+GridFS support comes in the form of the :class:`~mongoengine.fields.FileField` field
+object. This field acts as a file-like object and provides a couple of
+different ways of inserting and retrieving data. Arbitrary metadata such as
+content type can also be stored alongside the files. The object returned when accessing a
+FileField is a proxy to `Pymongo's GridFS <https://api.mongodb.com/python/current/examples/gridfs.html#gridfs-example>`_
+In the following example, a document is created to store details about animals, including a photo::
+
+    class Animal(Document):
+        genus = StringField()
+        family = StringField()
+        photo = FileField()
+
+    marmot = Animal(genus='Marmota', family='Sciuridae')
+
+    with open('marmot.jpg', 'rb') as fd:
+        marmot.photo.put(fd, content_type = 'image/jpeg')
+    marmot.save()
+
+Retrieval
+---------
+
+So using the :class:`~mongoengine.fields.FileField` is just like using any other
+field. The file can also be retrieved just as easily::
+
+    marmot = Animal.objects(genus='Marmota').first()
+    photo = marmot.photo.read()
+    content_type = marmot.photo.content_type
+
+.. note:: If you need to read() the content of a file multiple times, you'll need to "rewind"
+    the file-like object using `seek`::
+
+        marmot = Animal.objects(genus='Marmota').first()
+        content1 = marmot.photo.read()
+        assert content1 != ""
+
+        content2 = marmot.photo.read()    # will be empty
+        assert content2 == ""
+
+        marmot.photo.seek(0)              # rewind the file by setting the current position of the cursor in the file to 0
+        content3 = marmot.photo.read()
+        assert content3 == content1
+
+Streaming
+---------
+
+Streaming data into a :class:`~mongoengine.fields.FileField` is achieved in a
+slightly different manner.  First, a new file must be created by calling the
+:func:`new_file` method. Data can then be written using :func:`write`::
+
+    marmot.photo.new_file()
+    marmot.photo.write('some_image_data')
+    marmot.photo.write('some_more_image_data')
+    marmot.photo.close()
+
+    marmot.save()
+
+Deletion
+--------
+
+Deleting stored files is achieved with the :func:`delete` method::
+
+    marmot.photo.delete()    # Deletes the GridFS document
+    marmot.save()            # Saves the GridFS reference (being None) contained in the marmot instance
+
+.. warning::
+
+    The FileField in a Document actually only stores the ID of a file in a
+    separate GridFS collection. This means that deleting a document
+    with a defined FileField does not actually delete the file. You must be
+    careful to delete any files in a Document as above before deleting the
+    Document itself.
+
+
+Replacing files
+---------------
+
+Files can be replaced with the :func:`replace` method. This works just like
+the :func:`put` method so even metadata can (and should) be replaced::
+
+    another_marmot = open('another_marmot.png', 'rb')
+    marmot.photo.replace(another_marmot, content_type='image/png')  # Replaces the GridFS document
+    marmot.save()                                                   # Replaces the GridFS reference contained in marmot instance

docs/guide/index.rst

@@ -10,3 +10,8 @@ User Guide
    defining-documents
    document-instances
    querying
+   gridfs
+   signals
+   text-indexes
+   logging-monitoring
+   mongomock

docs/guide/installing.rst

@@ -1,31 +1,31 @@
 ======================
 Installing MongoEngine
 ======================
-To use MongoEngine, you will need to download `MongoDB <http://mongodb.org/>`_
+
+To use MongoEngine, you will need to download `MongoDB <http://mongodb.com/>`_
 and ensure it is running in an accessible location. You will also need
 `PyMongo <http://api.mongodb.org/python>`_ to use MongoEngine, but if you
 install MongoEngine using setuptools, then the dependencies will be handled for
 you.
 
-MongoEngine is available on PyPI, so to use it you can use 
-:program:`easy_install`:
+MongoEngine is available on PyPI, so you can use :program:`pip`:
 
 .. code-block:: console
 
-    # easy_install mongoengine
+    $ python -m pip install mongoengine
 
 Alternatively, if you don't have setuptools installed, `download it from PyPi
 <http://pypi.python.org/pypi/mongoengine/>`_ and run
 
 .. code-block:: console
 
-    # python setup.py install
+    $ python setup.py install
 
 To use the bleeding-edge version of MongoEngine, you can get the source from
-`GitHub <http://github.com/hmarr/mongoengine/>`_ and install it as above:
+`GitHub <http://github.com/mongoengine/mongoengine/>`_ and install it as above:
 
 .. code-block:: console
 
-    # git clone git://github.com/hmarr/mongoengine
-    # cd mongoengine
-    # python setup.py install
+    $ git clone git://github.com/mongoengine/mongoengine
+    $ cd mongoengine
+    $ python setup.py install

docs/guide/logging-monitoring.rst

@@ -0,0 +1,80 @@
+==================
+Logging/Monitoring
+==================
+
+It is possible to use `pymongo.monitoring <https://api.mongodb.com/python/current/api/pymongo/monitoring.html>`_ to monitor
+the driver events (e.g: queries, connections, etc). This can be handy if you want to monitor the queries issued by
+MongoEngine to the driver.
+
+To use `pymongo.monitoring` with MongoEngine, you need to make sure that you are registering the listeners
+**before** establishing the database connection (i.e calling `connect`):
+
+The following snippet provides a basic logging of all command events:
+
+.. code-block:: python
+
+    import logging
+    from pymongo import monitoring
+    from mongoengine import *
+
+    log = logging.getLogger()
+    log.setLevel(logging.DEBUG)
+    logging.basicConfig(level=logging.DEBUG)
+
+
+    class CommandLogger(monitoring.CommandListener):
+
+        def started(self, event):
+            log.debug("Command {0.command_name} with request id "
+                     "{0.request_id} started on server "
+                     "{0.connection_id}".format(event))
+
+        def succeeded(self, event):
+            log.debug("Command {0.command_name} with request id "
+                     "{0.request_id} on server {0.connection_id} "
+                     "succeeded in {0.duration_micros} "
+                     "microseconds".format(event))
+
+        def failed(self, event):
+            log.debug("Command {0.command_name} with request id "
+                     "{0.request_id} on server {0.connection_id} "
+                     "failed in {0.duration_micros} "
+                     "microseconds".format(event))
+
+    monitoring.register(CommandLogger())
+
+
+    class Jedi(Document):
+        name = StringField()
+
+
+    connect()
+
+
+    log.info('GO!')
+
+    log.info('Saving an item through MongoEngine...')
+    Jedi(name='Obi-Wan Kenobii').save()
+
+    log.info('Querying through MongoEngine...')
+    obiwan = Jedi.objects.first()
+
+    log.info('Updating through MongoEngine...')
+    obiwan.name = 'Obi-Wan Kenobi'
+    obiwan.save()
+
+
+Executing this prints the following output::
+
+    INFO:root:GO!
+    INFO:root:Saving an item through MongoEngine...
+    DEBUG:root:Command insert with request id 1681692777 started on server ('localhost', 27017)
+    DEBUG:root:Command insert with request id 1681692777 on server ('localhost', 27017) succeeded in 562 microseconds
+    INFO:root:Querying through MongoEngine...
+    DEBUG:root:Command find with request id 1714636915 started on server ('localhost', 27017)
+    DEBUG:root:Command find with request id 1714636915 on server ('localhost', 27017) succeeded in 341 microseconds
+    INFO:root:Updating through MongoEngine...
+    DEBUG:root:Command update with request id 1957747793 started on server ('localhost', 27017)
+    DEBUG:root:Command update with request id 1957747793 on server ('localhost', 27017) succeeded in 455 microseconds
+
+More details can of course be obtained by checking the `event` argument from the `CommandListener`.

docs/guide/mongomock.rst

@@ -0,0 +1,48 @@
+==============================
+Use mongomock for testing
+==============================
+
+`mongomock <https://github.com/vmalloc/mongomock/>`_ is a package to do just
+what the name implies, mocking a mongo database.
+
+To use with mongoengine, simply specify mongomock when connecting with
+mongoengine:
+
+.. code-block:: python
+
+    connect('mongoenginetest', host='mongomock://localhost')
+    conn = get_connection()
+
+or with an alias:
+
+.. code-block:: python
+
+    connect('mongoenginetest', host='mongomock://localhost', alias='testdb')
+    conn = get_connection('testdb')
+
+Example of test file:
+---------------------
+.. code-block:: python
+
+    import unittest
+    from mongoengine import connect, disconnect
+
+    class Person(Document):
+        name = StringField()
+
+    class TestPerson(unittest.TestCase):
+
+        @classmethod
+        def setUpClass(cls):
+            connect('mongoenginetest', host='mongomock://localhost')
+
+        @classmethod
+        def tearDownClass(cls):
+           disconnect()
+
+        def test_thing(self):
+            pers = Person(name='John')
+            pers.save()
+
+            fresh_pers = Person.objects().first()
+            assert fresh_pers.name ==  'John'

docs/guide/querying.rst

@@ -5,16 +5,23 @@ Querying the database
 is used for accessing the objects in the database associated with the class.
 The :attr:`objects` attribute is actually a
 :class:`~mongoengine.queryset.QuerySetManager`, which creates and returns a new
-a new :class:`~mongoengine.queryset.QuerySet` object on access. The
-:class:`~mongoengine.queryset.QuerySet` object may may be iterated over to
+:class:`~mongoengine.queryset.QuerySet` object on access. The
+:class:`~mongoengine.queryset.QuerySet` object may be iterated over to
 fetch documents from the database::
 
     # Prints out the names of all the users in the database
     for user in User.objects:
         print user.name
 
+.. note::
+
+    As of MongoEngine 0.8 the querysets utilise a local cache.  So iterating
+    it multiple times will only cause a single query.  If this is not the
+    desired behaviour you can call :class:`~mongoengine.QuerySet.no_cache`
+    (version **0.8.3+**) to return a non-caching queryset.
+
 Filtering queries
------------------
+=================
 The query may be filtered by calling the
 :class:`~mongoengine.queryset.QuerySet` object with field lookup keyword
 arguments. The keys in the keyword arguments correspond to fields on the
@@ -32,11 +39,144 @@ syntax::
     # been written by a user whose 'country' field is set to 'uk'
     uk_pages = Page.objects(author__country='uk')
 
+.. note::
+
+   (version **0.9.1+**) if your field name is like mongodb operator name (for example
+   type, lte, lt...) and you want to place it at the end of lookup keyword
+   mongoengine automatically  prepend $ to it. To avoid this use  __ at the end of
+   your lookup keyword. For example if your field name is ``type`` and you want to
+   query by this field you must use ``.objects(user__type__="admin")`` instead of
+   ``.objects(user__type="admin")``
+
+Query operators
+===============
+Operators other than equality may also be used in queries --- just attach the
+operator name to a key with a double-underscore::
+
+    # Only find users whose age is 18 or less
+    young_users = Users.objects(age__lte=18)
+
+Available operators are as follows:
+
+* ``ne`` -- not equal to
+* ``lt`` -- less than
+* ``lte`` -- less than or equal to
+* ``gt`` -- greater than
+* ``gte`` -- greater than or equal to
+* ``not`` -- negate a standard check, may be used before other operators (e.g.
+  ``Q(age__not__mod=(5, 0))``)
+* ``in`` -- value is in list (a list of values should be provided)
+* ``nin`` -- value is not in list (a list of values should be provided)
+* ``mod`` -- ``value % x == y``, where ``x`` and ``y`` are two provided values
+* ``all`` -- every item in list of values provided is in array
+* ``size`` -- the size of the array is
+* ``exists`` -- value for field exists
+
+String queries
+--------------
+
+The following operators are available as shortcuts to querying with regular
+expressions:
+
+* ``exact`` -- string field exactly matches value
+* ``iexact`` -- string field exactly matches value (case insensitive)
+* ``contains`` -- string field contains value
+* ``icontains`` -- string field contains value (case insensitive)
+* ``startswith`` -- string field starts with value
+* ``istartswith`` -- string field starts with value (case insensitive)
+* ``endswith`` -- string field ends with value
+* ``iendswith`` -- string field ends with value (case insensitive)
+* ``match``  -- performs an $elemMatch so you can match an entire document within an array
+
+
+Geo queries
+-----------
+
+There are a few special operators for performing geographical queries.
+The following were added in MongoEngine 0.8 for
+:class:`~mongoengine.fields.PointField`,
+:class:`~mongoengine.fields.LineStringField` and
+:class:`~mongoengine.fields.PolygonField`:
+
+* ``geo_within`` -- check if a geometry is within a polygon. For ease of use
+  it accepts either a geojson geometry or just the polygon coordinates eg::
+
+        loc.objects(point__geo_within=[[[40, 5], [40, 6], [41, 6], [40, 5]]])
+        loc.objects(point__geo_within={"type": "Polygon",
+                                 "coordinates": [[[40, 5], [40, 6], [41, 6], [40, 5]]]})
+
+* ``geo_within_box`` -- simplified geo_within searching with a box eg::
+
+        loc.objects(point__geo_within_box=[(-125.0, 35.0), (-100.0, 40.0)])
+        loc.objects(point__geo_within_box=[<bottom left coordinates>, <upper right coordinates>])
+
+* ``geo_within_polygon`` -- simplified geo_within searching within a simple polygon eg::
+
+        loc.objects(point__geo_within_polygon=[[40, 5], [40, 6], [41, 6], [40, 5]])
+        loc.objects(point__geo_within_polygon=[ [ <x1> , <y1> ] ,
+                                                [ <x2> , <y2> ] ,
+                                                [ <x3> , <y3> ] ])
+
+* ``geo_within_center`` -- simplified geo_within the flat circle radius of a point eg::
+
+        loc.objects(point__geo_within_center=[(-125.0, 35.0), 1])
+        loc.objects(point__geo_within_center=[ [ <x>, <y> ] , <radius> ])
+
+* ``geo_within_sphere`` -- simplified geo_within the spherical circle radius of a point eg::
+
+        loc.objects(point__geo_within_sphere=[(-125.0, 35.0), 1])
+        loc.objects(point__geo_within_sphere=[ [ <x>, <y> ] , <radius> ])
+
+* ``geo_intersects`` -- selects all locations that intersect with a geometry eg::
+
+        # Inferred from provided points lists:
+        loc.objects(poly__geo_intersects=[40, 6])
+        loc.objects(poly__geo_intersects=[[40, 5], [40, 6]])
+        loc.objects(poly__geo_intersects=[[[40, 5], [40, 6], [41, 6], [41, 5], [40, 5]]])
+
+        # With geoJson style objects
+        loc.objects(poly__geo_intersects={"type": "Point", "coordinates": [40, 6]})
+        loc.objects(poly__geo_intersects={"type": "LineString",
+                                          "coordinates": [[40, 5], [40, 6]]})
+        loc.objects(poly__geo_intersects={"type": "Polygon",
+                                          "coordinates": [[[40, 5], [40, 6], [41, 6], [41, 5], [40, 5]]]})
+
+* ``near`` -- find all the locations near a given point::
+
+        loc.objects(point__near=[40, 5])
+        loc.objects(point__near={"type": "Point", "coordinates": [40, 5]})
+
+  You can also set the maximum and/or the minimum distance in meters as well::
+
+        loc.objects(point__near=[40, 5], point__max_distance=1000)
+        loc.objects(point__near=[40, 5], point__min_distance=100)
+
+The older 2D indexes are still supported with the
+:class:`~mongoengine.fields.GeoPointField`:
+
+* ``within_distance`` -- provide a list containing a point and a maximum
+  distance (e.g. [(41.342, -87.653), 5])
+* ``within_spherical_distance`` -- same as above but using the spherical geo model
+  (e.g. [(41.342, -87.653), 5/earth_radius])
+* ``near`` -- order the documents by how close they are to a given point
+* ``near_sphere`` -- Same as above but using the spherical geo model
+* ``within_box`` -- filter documents to those within a given bounding box (e.g.
+  [(35.0, -125.0), (40.0, -100.0)])
+* ``within_polygon`` -- filter documents to those within a given polygon (e.g.
+  [(41.91,-87.69), (41.92,-87.68), (41.91,-87.65), (41.89,-87.65)]).
+
+  .. note:: Requires Mongo Server 2.0
+
+* ``max_distance`` -- can be added to your location queries to set a maximum
+  distance.
+* ``min_distance`` -- can be added to your location queries to set a minimum
+  distance.
+
 Querying lists
-^^^^^^^^^^^^^^
+--------------
 On most fields, this syntax will look up documents where the field specified
 matches the given value exactly, but when the field refers to a
-:class:`~mongoengine.ListField`, a single item may be provided, in which case
+:class:`~mongoengine.fields.ListField`, a single item may be provided, in which case
 lists that contain that item will be matched::
 
     class Page(Document):
@@ -46,36 +186,62 @@ lists that contain that item will be matched::
     # 'tags' list
     Page.objects(tags='coding')
 
-Query operators
----------------
-Operators other than equality may also be used in queries; just attach the
-operator name to a key with a double-underscore::
+It is possible to query by position in a list by using a numerical value as a
+query operator. So if you wanted to find all pages whose first tag was ``db``,
+you could use the following query::
 
-    # Only find users whose age is 18 or less
-    young_users = Users.objects(age__lte=18)
+    Page.objects(tags__0='db')
 
-Available operators are as follows:
+If you only want to fetch part of a list eg: you want to paginate a list, then
+the `slice` operator is required::
+
+    # comments - skip 5, limit 10
+    Page.objects.fields(slice__comments=[5, 10])
+
+For updating documents, if you don't know the position in a list, you can use
+the $ positional operator ::
+
+    Post.objects(comments__by="joe").update(**{'inc__comments__$__votes': 1})
+
+However, this doesn't map well to the syntax so you can also use a capital S instead ::
+
+    Post.objects(comments__by="joe").update(inc__comments__S__votes=1)
+
+.. note::
+    Due to :program:`Mongo`, currently the $ operator only applies to the
+    first matched item in the query.
+
+
+Raw queries
+-----------
+It is possible to provide a raw :mod:`PyMongo` query as a query parameter, which will
+be integrated directly into the query. This is done using the ``__raw__``
+keyword argument::
+
+    Page.objects(__raw__={'tags': 'coding'})
+
+.. versionadded:: 0.4
+
+Sorting/Ordering results
+========================
+It is possible to order the results by 1 or more keys using :meth:`~mongoengine.queryset.QuerySet.order_by`.
+The order may be specified by prepending each of the keys by "+" or "-". Ascending order is assumed if there's no prefix.::
+
+    # Order by ascending date
+    blogs = BlogPost.objects().order_by('date')    # equivalent to .order_by('+date')
+
+    # Order by ascending date first, then descending title
+    blogs = BlogPost.objects().order_by('+date', '-title')
 
-* ``neq`` -- not equal to
-* ``lt`` -- less than
-* ``lte`` -- less than or equal to
-* ``gt`` -- greater than
-* ``gte`` -- greater than or equal to
-* ``in`` -- value is in list (a list of values should be provided)
-* ``nin`` -- value is not in list (a list of values should be provided)
-* ``mod`` -- ``value % x == y``, where ``x`` and ``y`` are two provided values
-* ``all`` -- every item in array is in list of values provided
-* ``size`` -- the size of the array is 
-* ``exists`` -- value for field exists
 
 Limiting and skipping results
------------------------------
-Just as with traditional ORMs, you may limit the number of results returned, or
+=============================
+Just as with traditional ORMs, you may limit the number of results returned or
 skip a number or results in you query.
 :meth:`~mongoengine.queryset.QuerySet.limit` and
 :meth:`~mongoengine.queryset.QuerySet.skip` and methods are available on
-:class:`~mongoengine.queryset.QuerySet` objects, but the prefered syntax for
-achieving this is using array-slicing syntax::
+:class:`~mongoengine.queryset.QuerySet` objects, but the `array-slicing` syntax
+is preferred for achieving this::
 
     # Only the first 5 people
     users = User.objects[:5]
@@ -83,33 +249,132 @@ achieving this is using array-slicing syntax::
     # All except for the first 5 people
     users = User.objects[5:]
 
-    # 5 users, starting from the 10th user found
+    # 5 users, starting from the 11th user found
     users = User.objects[10:15]
 
+You may also index the query to retrieve a single result. If an item at that
+index does not exists, an :class:`IndexError` will be raised. A shortcut for
+retrieving the first result and returning :attr:`None` if no result exists is
+provided (:meth:`~mongoengine.queryset.QuerySet.first`)::
+
+    >>> # Make sure there are no users
+    >>> User.drop_collection()
+    >>> User.objects[0]
+    IndexError: list index out of range
+    >>> User.objects.first() == None
+    True
+    >>> User(name='Test User').save()
+    >>> User.objects[0] == User.objects.first()
+    True
+
+Retrieving unique results
+-------------------------
+To retrieve a result that should be unique in the collection, use
+:meth:`~mongoengine.queryset.QuerySet.get`. This will raise
+:class:`~mongoengine.queryset.DoesNotExist` if
+no document matches the query, and
+:class:`~mongoengine.queryset.MultipleObjectsReturned`
+if more than one document matched the query.  These exceptions are merged into
+your document definitions eg: `MyDoc.DoesNotExist`
+
+A variation of this method, get_or_create() existed, but it was unsafe. It
+could not be made safe, because there are no transactions in mongoDB. Other
+approaches should be investigated, to ensure you don't accidentally duplicate
+data when using something similar to this method. Therefore it was deprecated
+in 0.8 and removed in 0.10.
+
+Default Document queries
+========================
+By default, the objects :attr:`~Document.objects` attribute on a
+document returns a :class:`~mongoengine.queryset.QuerySet` that doesn't filter
+the collection -- it returns all objects. This may be changed by defining a
+method on a document that modifies a queryset. The method should accept two
+arguments -- :attr:`doc_cls` and :attr:`queryset`. The first argument is the
+:class:`~mongoengine.Document` class that the method is defined on (in this
+sense, the method is more like a :func:`classmethod` than a regular method),
+and the second argument is the initial queryset. The method needs to be
+decorated with :func:`~mongoengine.queryset.queryset_manager` in order for it
+to be recognised. ::
+
+    class BlogPost(Document):
+        title = StringField()
+        date = DateTimeField()
+
+        @queryset_manager
+        def objects(doc_cls, queryset):
+            # This may actually also be done by defining a default ordering for
+            # the document, but this illustrates the use of manager methods
+            return queryset.order_by('-date')
+
+You don't need to call your method :attr:`objects` -- you may define as many
+custom manager methods as you like::
+
+    class BlogPost(Document):
+        title = StringField()
+        published = BooleanField()
+
+        @queryset_manager
+        def live_posts(doc_cls, queryset):
+            return queryset.filter(published=True)
+
+    BlogPost(title='test1', published=False).save()
+    BlogPost(title='test2', published=True).save()
+    assert len(BlogPost.objects) == 2
+    assert len(BlogPost.live_posts()) == 1
+
+Custom QuerySets
+================
+Should you want to add custom methods for interacting with or filtering
+documents, extending the :class:`~mongoengine.queryset.QuerySet` class may be
+the way to go. To use a custom :class:`~mongoengine.queryset.QuerySet` class on
+a document, set ``queryset_class`` to the custom class in a
+:class:`~mongoengine.Document`'s ``meta`` dictionary::
+
+    class AwesomerQuerySet(QuerySet):
+
+        def get_awesome(self):
+            return self.filter(awesome=True)
+
+    class Page(Document):
+        meta = {'queryset_class': AwesomerQuerySet}
+
+    # To call:
+    Page.objects.get_awesome()
+
+.. versionadded:: 0.4
+
 Aggregation
------------
+===========
 MongoDB provides some aggregation methods out of the box, but there are not as
 many as you typically get with an RDBMS. MongoEngine provides a wrapper around
 the built-in methods and provides some of its own, which are implemented as
 Javascript code that is executed on the database server.
 
 Counting results
-^^^^^^^^^^^^^^^^
-Just as with limiting and skipping results, there is a method on
-:class:`~mongoengine.queryset.QuerySet` objects -- 
-:meth:`~mongoengine.queryset.QuerySet.count`, but there is also a more Pythonic
-way of achieving this::
+----------------
+Just as with limiting and skipping results, there is a method on a
+:class:`~mongoengine.queryset.QuerySet` object --
+:meth:`~mongoengine.queryset.QuerySet.count`::
 
-    num_users = len(User.objects)
+    num_users = User.objects.count()
+
+You could technically use ``len(User.objects)`` to get the same result, but it
+would be significantly slower than :meth:`~mongoengine.queryset.QuerySet.count`.
+When you execute a server-side count query, you let MongoDB do the heavy
+lifting and you receive a single integer over the wire. Meanwhile, ``len()``
+retrieves all the results, places them in a local cache, and finally counts
+them. If we compare the performance of the two operations, ``len()`` is much slower
+than :meth:`~mongoengine.queryset.QuerySet.count`.
 
 Further aggregation
-^^^^^^^^^^^^^^^^^^^
+-------------------
 You may sum over the values of a specific field on documents using
 :meth:`~mongoengine.queryset.QuerySet.sum`::
 
     yearly_expense = Employee.objects.sum('salary')
 
 .. note::
+
    If the field isn't present on a document, that document will be ignored from
    the sum.
 
@@ -132,8 +397,112 @@ would be generating "tag-clouds"::
     from operator import itemgetter
     top_tags = sorted(tag_freqs.items(), key=itemgetter(1), reverse=True)[:10]
 
+
+MongoDB aggregation API
+-----------------------
+If you need to run aggregation pipelines, MongoEngine provides an entry point to `Pymongo's aggregation framework <https://api.mongodb.com/python/current/examples/aggregation.html#aggregation-framework>`_
+through :meth:`~mongoengine.queryset.QuerySet.aggregate`. Check out Pymongo's documentation for the syntax and pipeline.
+An example of its use would be::
+
+        class Person(Document):
+            name = StringField()
+
+        Person(name='John').save()
+        Person(name='Bob').save()
+
+        pipeline = [
+            {"$sort" : {"name" : -1}},
+            {"$project": {"_id": 0, "name": {"$toUpper": "$name"}}}
+            ]
+        data = Person.objects().aggregate(pipeline)
+        assert data == [{'name': 'BOB'}, {'name': 'JOHN'}]
+
+Query efficiency and performance
+================================
+
+There are a couple of methods to improve efficiency when querying, reducing the
+information returned by the query or efficient dereferencing .
+
+Retrieving a subset of fields
+-----------------------------
+
+Sometimes a subset of fields on a :class:`~mongoengine.Document` is required,
+and for efficiency only these should be retrieved from the database. This issue
+is especially important for MongoDB, as fields may often be extremely large
+(e.g. a :class:`~mongoengine.fields.ListField` of
+:class:`~mongoengine.EmbeddedDocument`\ s, which represent the comments on a
+blog post. To select only a subset of fields, use
+:meth:`~mongoengine.queryset.QuerySet.only`, specifying the fields you want to
+retrieve as its arguments. Note that if fields that are not downloaded are
+accessed, their default value (or :attr:`None` if no default value is provided)
+will be given::
+
+    >>> class Film(Document):
+    ...     title = StringField()
+    ...     year = IntField()
+    ...     rating = IntField(default=3)
+    ...
+    >>> Film(title='The Shawshank Redemption', year=1994, rating=5).save()
+    >>> f = Film.objects.only('title').first()
+    >>> f.title
+    'The Shawshank Redemption'
+    >>> f.year   # None
+    >>> f.rating # default value
+    3
+
+.. note::
+
+    The :meth:`~mongoengine.queryset.QuerySet.exclude` is the opposite of
+    :meth:`~mongoengine.queryset.QuerySet.only` if you want to exclude a field.
+
+If you later need the missing fields, just call
+:meth:`~mongoengine.Document.reload` on your document.
+
+Getting related data
+--------------------
+
+When iterating the results of :class:`~mongoengine.fields.ListField` or
+:class:`~mongoengine.fields.DictField` we automatically dereference any
+:class:`~pymongo.dbref.DBRef` objects as efficiently as possible, reducing the
+number the queries to mongo.
+
+There are times when that efficiency is not enough, documents that have
+:class:`~mongoengine.fields.ReferenceField` objects or
+:class:`~mongoengine.fields.GenericReferenceField` objects at the top level are
+expensive as the number of queries to MongoDB can quickly rise.
+
+To limit the number of queries use
+:func:`~mongoengine.queryset.QuerySet.select_related` which converts the
+QuerySet to a list and dereferences as efficiently as possible.  By default
+:func:`~mongoengine.queryset.QuerySet.select_related` only dereferences any
+references to the depth of 1 level.  If you have more complicated documents and
+want to dereference more of the object at once then increasing the :attr:`max_depth`
+will dereference more levels of the document.
+
+Turning off dereferencing
+-------------------------
+
+Sometimes for performance reasons you don't want to automatically dereference
+data. To turn off dereferencing of the results of a query use
+:func:`~mongoengine.queryset.QuerySet.no_dereference` on the queryset like so::
+
+    post = Post.objects.no_dereference().first()
+    assert(isinstance(post.author, DBRef))
+
+You can also turn off all dereferencing for a fixed period by using the
+:class:`~mongoengine.context_managers.no_dereference` context manager::
+
+    with no_dereference(Post) as Post:
+        post = Post.objects.first()
+        assert(isinstance(post.author, DBRef))
+
+    # Outside the context manager dereferencing occurs.
+    assert(isinstance(post.author, User))
+
+
 Advanced queries
-----------------
+================
+
 Sometimes calling a :class:`~mongoengine.queryset.QuerySet` object with keyword
 arguments can't fully express the query you want to use -- for example if you
 need to combine a number of constraints using *and* and *or*. This is made
@@ -142,40 +511,49 @@ A :class:`~mongoengine.queryset.Q` object represents part of a query, and
 can be initialised using the same keyword-argument syntax you use to query
 documents. To build a complex query, you may combine
 :class:`~mongoengine.queryset.Q` objects using the ``&`` (and) and ``|`` (or)
-operators. To use :class:`~mongoengine.queryset.Q` objects, pass them in
-as positional arguments to :attr:`Document.objects` when you filter it by
+operators. To use a :class:`~mongoengine.queryset.Q` object, pass it in as the
+first positional argument to :attr:`Document.objects` when you filter it by
 calling it with keyword arguments::
 
+    from mongoengine.queryset.visitor import Q
+
     # Get published posts
     Post.objects(Q(published=True) | Q(publish_date__lte=datetime.now()))
 
     # Get top posts
     Post.objects((Q(featured=True) & Q(hits__gte=1000)) | Q(hits__gte=5000))
 
-.. warning::
-   Only use these advanced queries if absolutely necessary as they will execute
-   significantly slower than regular queries. This is because they are not
-   natively supported by MongoDB -- they are compiled to Javascript and sent
-   to the server for execution.
+.. warning:: You have to use bitwise operators.  You cannot use ``or``, ``and``
+    to combine queries as ``Q(a=a) or Q(b=b)`` is not the same as
+    ``Q(a=a) | Q(b=b)``. As ``Q(a=a)`` equates to true ``Q(a=a) or Q(b=b)`` is
+    the same as ``Q(a=a)``.
 
 .. _guide-atomic-updates:
 
 Atomic updates
---------------
+==============
 Documents may be updated atomically by using the
-:meth:`~mongoengine.queryset.QuerySet.update_one` and
-:meth:`~mongoengine.queryset.QuerySet.update` methods on a 
-:meth:`~mongoengine.queryset.QuerySet`. There are several different "modifiers"
-that you may use with these methods:
+:meth:`~mongoengine.queryset.QuerySet.update_one`,
+:meth:`~mongoengine.queryset.QuerySet.update` and
+:meth:`~mongoengine.queryset.QuerySet.modify` methods on a
+:class:`~mongoengine.queryset.QuerySet` or
+:meth:`~mongoengine.Document.modify` and
+:meth:`~mongoengine.Document.save` (with :attr:`save_condition` argument) on a
+:class:`~mongoengine.Document`.
+There are several different "modifiers" that you may use with these methods:
 
 * ``set`` -- set a particular value
-* ``unset`` -- delete a particular value (since MongoDB v1.3+)
+* ``unset`` -- delete a particular value (since MongoDB v1.3)
 * ``inc`` -- increment a value by a given amount
 * ``dec`` -- decrement a value by a given amount
 * ``push`` -- append a value to a list
 * ``push_all`` -- append several values to a list
+* ``pop`` -- remove the first or last element of a list `depending on the value`_
 * ``pull`` -- remove a value from a list
 * ``pull_all`` -- remove several values from a list
+* ``add_to_set`` -- add value to a list only if its not in the list already
+
+.. _depending on the value: http://docs.mongodb.org/manual/reference/operator/update/pop/
 
 The syntax for atomic updates is similar to the querying syntax, but the
 modifier comes before the field, not after it::
@@ -194,3 +572,144 @@ modifier comes before the field, not after it::
     >>> post.reload()
     >>> post.tags
     ['database', 'nosql']
+
+.. note::
+
+    If no modifier operator is specified the default will be ``$set``. So the following sentences are identical::
+
+        >>> BlogPost.objects(id=post.id).update(title='Example Post')
+        >>> BlogPost.objects(id=post.id).update(set__title='Example Post')
+
+.. note::
+
+    In version 0.5 the :meth:`~mongoengine.Document.save` runs atomic updates
+    on changed documents by tracking changes to that document.
+
+The positional operator allows you to update list items without knowing the
+index position, therefore making the update a single atomic operation.  As we
+cannot use the `$` syntax in keyword arguments it has been mapped to `S`::
+
+    >>> post = BlogPost(title='Test', page_views=0, tags=['database', 'mongo'])
+    >>> post.save()
+    >>> BlogPost.objects(id=post.id, tags='mongo').update(set__tags__S='mongodb')
+    >>> post.reload()
+    >>> post.tags
+    ['database', 'mongodb']
+
+From MongoDB version 2.6, push operator supports $position value which allows
+to push values with index::
+
+    >>> post = BlogPost(title="Test", tags=["mongo"])
+    >>> post.save()
+    >>> post.update(push__tags__0=["database", "code"])
+    >>> post.reload()
+    >>> post.tags
+    ['database', 'code', 'mongo']
+
+.. note::
+    Currently only top level lists are handled, future versions of mongodb /
+    pymongo plan to support nested positional operators.  See `The $ positional
+    operator <http://www.mongodb.org/display/DOCS/Updating#Updating-The%24positionaloperator>`_.
+
+Server-side javascript execution
+================================
+Javascript functions may be written and sent to the server for execution. The
+result of this is the return value of the Javascript function. This
+functionality is accessed through the
+:meth:`~mongoengine.queryset.QuerySet.exec_js` method on
+:meth:`~mongoengine.queryset.QuerySet` objects. Pass in a string containing a
+Javascript function as the first argument.
+
+The remaining positional arguments are names of fields that will be passed into
+you Javascript function as its arguments. This allows functions to be written
+that may be executed on any field in a collection (e.g. the
+:meth:`~mongoengine.queryset.QuerySet.sum` method, which accepts the name of
+the field to sum over as its argument). Note that field names passed in in this
+manner are automatically translated to the names used on the database (set
+using the :attr:`name` keyword argument to a field constructor).
+
+Keyword arguments to :meth:`~mongoengine.queryset.QuerySet.exec_js` are
+combined into an object called :attr:`options`, which is available in the
+Javascript function. This may be used for defining specific parameters for your
+function.
+
+Some variables are made available in the scope of the Javascript function:
+
+* ``collection`` -- the name of the collection that corresponds to the
+  :class:`~mongoengine.Document` class that is being used; this should be
+  used to get the :class:`Collection` object from :attr:`db` in Javascript
+  code
+* ``query`` -- the query that has been generated by the
+  :class:`~mongoengine.queryset.QuerySet` object; this may be passed into
+  the :meth:`find` method on a :class:`Collection` object in the Javascript
+  function
+* ``options`` -- an object containing the keyword arguments passed into
+  :meth:`~mongoengine.queryset.QuerySet.exec_js`
+
+The following example demonstrates the intended usage of
+:meth:`~mongoengine.queryset.QuerySet.exec_js` by defining a function that sums
+over a field on a document (this functionality is already available through
+:meth:`~mongoengine.queryset.QuerySet.sum` but is shown here for sake of
+example)::
+
+    def sum_field(document, field_name, include_negatives=True):
+        code = """
+        function(sumField) {
+            var total = 0.0;
+            db[collection].find(query).forEach(function(doc) {
+                var val = doc[sumField];
+                if (val >= 0.0 || options.includeNegatives) {
+                    total += val;
+                }
+            });
+            return total;
+        }
+        """
+        options = {'includeNegatives': include_negatives}
+        return document.objects.exec_js(code, field_name, **options)
+
+As fields in MongoEngine may use different names in the database (set using the
+:attr:`db_field` keyword argument to a :class:`Field` constructor), a mechanism
+exists for replacing MongoEngine field names with the database field names in
+Javascript code. When accessing a field on a collection object, use
+square-bracket notation, and prefix the MongoEngine field name with a tilde.
+The field name that follows the tilde will be translated to the name used in
+the database. Note that when referring to fields on embedded documents,
+the name of the :class:`~mongoengine.fields.EmbeddedDocumentField`, followed by a dot,
+should be used before the name of the field on the embedded document. The
+following example shows how the substitutions are made::
+
+    class Comment(EmbeddedDocument):
+        content = StringField(db_field='body')
+
+    class BlogPost(Document):
+        title = StringField(db_field='doctitle')
+        comments = ListField(EmbeddedDocumentField(Comment), name='cs')
+
+    # Returns a list of dictionaries. Each dictionary contains a value named
+    # "document", which corresponds to the "title" field on a BlogPost, and
+    # "comment", which corresponds to an individual comment. The substitutions
+    # made are shown in the comments.
+    BlogPost.objects.exec_js("""
+    function() {
+        var comments = [];
+        db[collection].find(query).forEach(function(doc) {
+            // doc[~comments] -> doc["cs"]
+            var docComments = doc[~comments];
+
+            for (var i = 0; i < docComments.length; i++) {
+                // doc[~comments][i] -> doc["cs"][i]
+                var comment = doc[~comments][i];
+
+                comments.push({
+                    // doc[~title] -> doc["doctitle"]
+                    'document': doc[~title],
+
+                    // comment[~comments.content] -> comment["body"]
+                    'comment': comment[~comments.content]
+                });
+            }
+        });
+        return comments;
+    }
+    """)

docs/guide/signals.rst

@@ -0,0 +1,149 @@
+.. _signals:
+
+=======
+Signals
+=======
+
+.. versionadded:: 0.5
+
+.. note::
+
+  Signal support is provided by the excellent `blinker`_ library. If you wish
+  to enable signal support this library must be installed, though it is not
+  required for MongoEngine to function.
+
+Overview
+--------
+
+Signals are found within the `mongoengine.signals` module.  Unless
+specified signals receive no additional arguments beyond the `sender` class and
+`document` instance.  Post-signals are only called if there were no exceptions
+raised during the processing of their related function.
+
+Available signals include:
+
+`pre_init`
+  Called during the creation of a new :class:`~mongoengine.Document` or
+  :class:`~mongoengine.EmbeddedDocument` instance, after the constructor
+  arguments have been collected but before any additional processing has been
+  done to them.  (I.e. assignment of default values.)  Handlers for this signal
+  are passed the dictionary of arguments using the `values` keyword argument
+  and may modify this dictionary prior to returning.
+
+`post_init`
+  Called after all processing of a new :class:`~mongoengine.Document` or
+  :class:`~mongoengine.EmbeddedDocument` instance has been completed.
+
+`pre_save`
+  Called within :meth:`~mongoengine.Document.save` prior to performing
+  any actions.
+
+`pre_save_post_validation`
+  Called within :meth:`~mongoengine.Document.save` after validation
+  has taken place but before saving.
+
+`post_save`
+  Called within :meth:`~mongoengine.Document.save` after most actions
+  (validation, insert/update, and cascades, but not clearing dirty flags) have
+  completed successfully.  Passed the additional boolean keyword argument
+  `created` to indicate if the save was an insert or an update.
+
+`pre_delete`
+  Called within :meth:`~mongoengine.Document.delete` prior to
+  attempting the delete operation.
+
+`post_delete`
+  Called within :meth:`~mongoengine.Document.delete` upon successful
+  deletion of the record.
+
+`pre_bulk_insert`
+  Called after validation of the documents to insert, but prior to any data
+  being written. In this case, the `document` argument is replaced by a
+  `documents` argument representing the list of documents being inserted.
+
+`post_bulk_insert`
+  Called after a successful bulk insert operation.  As per `pre_bulk_insert`,
+  the `document` argument is omitted and replaced with a `documents` argument.
+  An additional boolean argument, `loaded`, identifies the contents of
+  `documents` as either :class:`~mongoengine.Document` instances when `True` or
+  simply a list of primary key values for the inserted records if `False`.
+
+Attaching Events
+----------------
+
+After writing a handler function like the following::
+
+    import logging
+    from datetime import datetime
+
+    from mongoengine import *
+    from mongoengine import signals
+
+    def update_modified(sender, document):
+        document.modified = datetime.utcnow()
+
+You attach the event handler to your :class:`~mongoengine.Document` or
+:class:`~mongoengine.EmbeddedDocument` subclass::
+
+    class Record(Document):
+        modified = DateTimeField()
+
+    signals.pre_save.connect(update_modified)
+
+While this is not the most elaborate document model, it does demonstrate the
+concepts involved.  As a more complete demonstration you can also define your
+handlers within your subclass::
+
+    class Author(Document):
+        name = StringField()
+
+        @classmethod
+        def pre_save(cls, sender, document, **kwargs):
+            logging.debug("Pre Save: %s" % document.name)
+
+        @classmethod
+        def post_save(cls, sender, document, **kwargs):
+            logging.debug("Post Save: %s" % document.name)
+            if 'created' in kwargs:
+                if kwargs['created']:
+                    logging.debug("Created")
+                else:
+                    logging.debug("Updated")
+
+    signals.pre_save.connect(Author.pre_save, sender=Author)
+    signals.post_save.connect(Author.post_save, sender=Author)
+
+.. warning::
+
+    Note that EmbeddedDocument only supports pre/post_init signals. pre/post_save, etc should be attached to Document's class only. Attaching pre_save to an EmbeddedDocument is ignored silently.
+
+Finally, you can also use this small decorator to quickly create a number of
+signals and attach them to your :class:`~mongoengine.Document` or
+:class:`~mongoengine.EmbeddedDocument` subclasses as class decorators::
+
+    def handler(event):
+        """Signal decorator to allow use of callback functions as class decorators."""
+
+        def decorator(fn):
+            def apply(cls):
+                event.connect(fn, sender=cls)
+                return cls
+
+            fn.apply = apply
+            return fn
+
+        return decorator
+
+Using the first example of updating a modification time the code is now much
+cleaner looking while still allowing manual execution of the callback::
+
+    @handler(signals.pre_save)
+    def update_modified(sender, document):
+        document.modified = datetime.utcnow()
+
+    @update_modified.apply
+    class Record(Document):
+        modified = DateTimeField()
+
+
+.. _blinker: http://pypi.python.org/pypi/blinker

docs/guide/text-indexes.rst

@@ -0,0 +1,51 @@
+===========
+Text Search
+===========
+
+After MongoDB 2.4 version, supports search documents by text indexes.
+
+
+Defining a Document with text index
+===================================
+Use the *$* prefix to set a text index, Look the declaration::
+
+  class News(Document):
+      title = StringField()
+      content = StringField()
+      is_active = BooleanField()
+
+      meta = {'indexes': [
+          {'fields': ['$title', "$content"],
+           'default_language': 'english',
+           'weights': {'title': 10, 'content': 2}
+          }
+      ]}
+
+
+
+Querying
+========
+
+Saving a document::
+
+  News(title="Using mongodb text search",
+       content="Testing text search").save()
+
+  News(title="MongoEngine 0.9 released",
+       content="Various improvements").save()
+
+Next, start a text search using :attr:`QuerySet.search_text` method::
+
+  document = News.objects.search_text('testing').first()
+  document.title # may be: "Using mongodb text search"
+
+  document = News.objects.search_text('released').first()
+  document.title # may be: "MongoEngine 0.9 released"
+
+
+Ordering by text score
+======================
+
+::
+
+  objects = News.objects.search_text('mongo').order_by('$text_score')

docs/index.rst

@@ -2,27 +2,92 @@
 MongoEngine User Documentation
 ==============================
 
-MongoEngine is an Object-Document Mapper, written in Python for working with 
+**MongoEngine** is an Object-Document Mapper, written in Python for working with
 MongoDB. To install it, simply run
 
 .. code-block:: console
 
-    # easy_install mongoengine
+    $ python -m pip install -U mongoengine
+
+:doc:`tutorial`
+  A quick tutorial building a tumblelog to get you up and running with
+  MongoEngine.
+
+:doc:`guide/index`
+  The Full guide to MongoEngine --- from modeling documents to storing files,
+  from querying for data to firing signals and *everything* between.
+
+:doc:`apireference`
+  The complete API documentation --- the innards of documents, querysets and fields.
+
+:doc:`upgrade`
+  How to upgrade MongoEngine.
+
+:doc:`faq`
+  Frequently Asked Questions
+
+:doc:`django`
+  Using MongoEngine and Django
+
+MongoDB and driver support
+--------------------------
+
+MongoEngine is based on the PyMongo driver and tested against multiple versions of MongoDB.
+For further details, please refer to the `readme <https://github.com/MongoEngine/mongoengine#mongoengine>`_.
+
+Community
+---------
+
+To get help with using MongoEngine, use the `MongoEngine Users mailing list
+<http://groups.google.com/group/mongoengine-users>`_ or the ever popular
+`stackoverflow <http://www.stackoverflow.com>`_.
+
+Contributing
+------------
+
+**Yes please!**  We are always looking for contributions, additions and improvements.
+
+The source is available on `GitHub <http://github.com/MongoEngine/mongoengine>`_
+and contributions are always encouraged. Contributions can be as simple as
+minor tweaks to this documentation, the website or the core.
+
+To contribute, fork the project on
+`GitHub <http://github.com/MongoEngine/mongoengine>`_ and send a
+pull request.
+
+Changes
+-------
+
+See the :doc:`changelog` for a full list of changes to MongoEngine and
+:doc:`upgrade` for upgrade information.
+
+.. note::  Always read and test the `upgrade <upgrade>`_ documentation before
+    putting updates live in production **;)**
+
+Offline Reading
+---------------
+
+Download the docs in `pdf <https://media.readthedocs.org/pdf/mongoengine-odm/latest/mongoengine-odm.pdf>`_
+or `epub <https://media.readthedocs.org/epub/mongoengine-odm/latest/mongoengine-odm.epub>`_
+formats for offline reading.
 
-The source is available on `GitHub <http://github.com/hmarr/mongoengine>`_.
 
 .. toctree::
-   :maxdepth: 2
+    :maxdepth: 1
+    :numbered:
+    :hidden:
 
     tutorial
     guide/index
     apireference
-   django
     changelog
+    upgrade
+    faq
+    django
 
 Indices and tables
-==================
+------------------
 
 * :ref:`genindex`
+* :ref:`modindex`
 * :ref:`search`
-

docs/tutorial.rst

@@ -1,68 +1,78 @@
 ========
 Tutorial
 ========
+
 This tutorial introduces **MongoEngine** by means of example --- we will walk
-through how to create a simple **Tumblelog** application. A Tumblelog is a type
-of blog where posts are not constrained to being conventional text-based posts.
-As well as text-based entries, users may post images, links, videos, etc. For
-simplicity's sake, we'll stick to text, image and link entries in our
-application. As the purpose of this tutorial is to introduce MongoEngine, we'll
+through how to create a simple **Tumblelog** application. A tumblelog is a
+blog that supports mixed media content, including text, images, links, video,
+audio, etc. For simplicity's sake, we'll stick to text, image, and link
+entries. As the purpose of this tutorial is to introduce MongoEngine, we'll
 focus on the data-modelling side of the application, leaving out a user
 interface.
 
 Getting started
 ===============
+
 Before we start, make sure that a copy of MongoDB is running in an accessible
 location --- running it locally will be easier, but if that is not an option
-then it may be run on a remote server.
+then it may be run on a remote server. If you haven't installed MongoEngine,
+simply use pip to install it like so::
+
+    $ python -m pip install mongoengine
 
 Before we can start using MongoEngine, we need to tell it how to connect to our
 instance of :program:`mongod`. For this we use the :func:`~mongoengine.connect`
-function. The only argument we need to provide is the name of the MongoDB
-database to use::
+function. If running locally, the only argument we need to provide is the name
+of the MongoDB database to use::
 
     from mongoengine import *
 
     connect('tumblelog')
 
-For more information about connecting to MongoDB see :ref:`guide-connecting`.
+There are lots of options for connecting to MongoDB, for more information about
+them see the :ref:`guide-connecting` guide.
 
 Defining our documents
 ======================
+
 MongoDB is *schemaless*, which means that no schema is enforced by the database
 --- we may add and remove fields however we want and MongoDB won't complain.
 This makes life a lot easier in many regards, especially when there is a change
-to the data model. However, defining schemata for our documents can help to
-iron out bugs involving incorrect types or missing fields, and also allow us to
+to the data model. However, defining schemas for our documents can help to iron
+out bugs involving incorrect types or missing fields, and also allow us to
 define utility methods on our documents in the same way that traditional
 :abbr:`ORMs (Object-Relational Mappers)` do.
 
 In our Tumblelog application we need to store several different types of
 information. We will need to have a collection of **users**, so that we may
-link posts to an individual. We also need to store our different types
-**posts** (text, image and link) in the database. To aid navigation of our
+link posts to an individual. We also need to store our different types of
+**posts** (eg: text, image and link) in the database. To aid navigation of our
 Tumblelog, posts may have **tags** associated with them, so that the list of
 posts shown to the user may be limited to posts that have been assigned a
-specified tag.  Finally, it would be nice if **comments** could be added to
-posts. We'll start with **users**, as the others are slightly more involved.
+specific tag. Finally, it would be nice if **comments** could be added to
+posts. We'll start with **users**, as the other document models are slightly
+more involved.
 
 Users
 -----
+
 Just as if we were using a relational database with an ORM, we need to define
-which fields a :class:`User` may have, and what their types will be::
+which fields a :class:`User` may have, and what types of data they might store::
 
     class User(Document):
         email = StringField(required=True)
         first_name = StringField(max_length=50)
         last_name = StringField(max_length=50)
 
-This looks similar to how a the structure of a table would be defined in a
+This looks similar to how the structure of a table would be defined in a
 regular ORM. The key difference is that this schema will never be passed on to
-MongoDB --- this will only be enforced at the application level. Also, the User
-documents will be stored in a MongoDB *collection* rather than a table.
+MongoDB --- this will only be enforced at the application level, making future
+changes easy to manage. Also, the User documents will be stored in a
+MongoDB *collection* rather than a table.
 
 Posts, Comments and Tags
 ------------------------
+
 Now we'll think about how to store the rest of the information. If we were
 using a relational database, we would most likely have a table of **posts**, a
 table of **comments** and a table of **tags**.  To associate the comments with
@@ -75,21 +85,25 @@ of them stand out as particularly intuitive solutions.
 
 Posts
 ^^^^^
-But MongoDB *isn't* a relational database, so we're not going to do it that
+
+Happily MongoDB *isn't* a relational database, so we're not going to do it that
 way. As it turns out, we can use MongoDB's schemaless nature to provide us with
-a much nicer solution. We will store all of the posts in *one collection* ---
-each post type will just have the fields it needs. If we later want to add
+a much nicer solution. We will store all of the posts in *one collection* and
+each post type will only store the fields it needs. If we later want to add
 video posts, we don't have to modify the collection at all, we just *start
 using* the new fields we need to support video posts. This fits with the
 Object-Oriented principle of *inheritance* nicely. We can think of
 :class:`Post` as a base class, and :class:`TextPost`, :class:`ImagePost` and
 :class:`LinkPost` as subclasses of :class:`Post`. In fact, MongoEngine supports
-this kind of modelling out of the box::
+this kind of modeling out of the box --- all you need do is turn on inheritance
+by setting :attr:`allow_inheritance` to True in the :attr:`meta`::
 
     class Post(Document):
         title = StringField(max_length=120, required=True)
         author = ReferenceField(User)
 
+        meta = {'allow_inheritance': True}
+
     class TextPost(Post):
         content = StringField()
 
@@ -100,20 +114,21 @@ this kind of modelling out of the box::
         link_url = StringField()
 
 We are storing a reference to the author of the posts using a
-:class:`~mongoengine.ReferenceField` object. These are similar to foreign key
+:class:`~mongoengine.fields.ReferenceField` object. These are similar to foreign key
 fields in traditional ORMs, and are automatically translated into references
 when they are saved, and dereferenced when they are loaded.
 
 Tags
 ^^^^
+
 Now that we have our Post models figured out, how will we attach tags to them?
 MongoDB allows us to store lists of items natively, so rather than having a
 link table, we can just store a list of tags in each post. So, for both
 efficiency and simplicity's sake, we'll store the tags as strings directly
 within the post, rather than storing references to tags in a separate
 collection. Especially as tags are generally very short (often even shorter
-than a document's id), this denormalisation won't impact very strongly on the 
-size of our database. So let's take a look that the code our modified
+than a document's id), this denormalization won't impact the size of the
+database very strongly. Let's take a look at the code of our modified
 :class:`Post` class::
 
     class Post(Document):
@@ -121,21 +136,24 @@ size of our database. So let's take a look that the code our modified
         author = ReferenceField(User)
         tags = ListField(StringField(max_length=30))
 
-The :class:`~mongoengine.ListField` object that is used to define a Post's tags
+The :class:`~mongoengine.fields.ListField` object that is used to define a Post's tags
 takes a field object as its first argument --- this means that you can have
-lists of any type of field (including lists). Note that we don't need to
-modify the specialised post types as they all inherit from :class:`Post`.
+lists of any type of field (including lists).
+
+.. note:: We don't need to modify the specialized post types as they all
+    inherit from :class:`Post`.
 
 Comments
 ^^^^^^^^
+
 A comment is typically associated with *one* post. In a relational database, to
 display a post with its comments, we would have to retrieve the post from the
-database, then query the database again for the comments associated with the
+database and then query the database again for the comments associated with the
 post. This works, but there is no real reason to be storing the comments
 separately from their associated posts, other than to work around the
 relational model. Using MongoDB we can store the comments as a list of
 *embedded documents* directly on a post document. An embedded document should
-be treated no differently that a regular document; it just doesn't have its own
+be treated no differently than a regular document; it just doesn't have its own
 collection in the database. Using MongoEngine, we can define the structure of
 embedded documents, along with utility methods, in exactly the same way we do
 with regular documents::
@@ -152,39 +170,63 @@ We can then store a list of comment documents in our post document::
         tags = ListField(StringField(max_length=30))
         comments = ListField(EmbeddedDocumentField(Comment))
 
+Handling deletions of references
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The :class:`~mongoengine.fields.ReferenceField` object takes a keyword
+`reverse_delete_rule` for handling deletion rules if the reference is deleted.
+To delete all the posts if a user is deleted set the rule::
+
+    class Post(Document):
+        title = StringField(max_length=120, required=True)
+        author = ReferenceField(User, reverse_delete_rule=CASCADE)
+        tags = ListField(StringField(max_length=30))
+        comments = ListField(EmbeddedDocumentField(Comment))
+
+See :class:`~mongoengine.fields.ReferenceField` for more information.
+
+.. note::
+    MapFields and DictFields currently don't support automatic handling of
+    deleted references
+
+
 Adding data to our Tumblelog
 ============================
 Now that we've defined how our documents will be structured, let's start adding
 some documents to the database. Firstly, we'll need to create a :class:`User`
 object::
 
-    john = User(email='jdoe@example.com', first_name='John', last_name='Doe')
-    john.save()
+    ross = User(email='ross@example.com', first_name='Ross', last_name='Lawley').save()
 
-Note that we could have also defined our user using attribute syntax::
+.. note::
+    We could have also defined our user using attribute syntax::
 
-    john = User(email='jdoe@example.com')
-    john.first_name = 'John'
-    john.last_name = 'Doe'
-    john.save()
+        ross = User(email='ross@example.com')
+        ross.first_name = 'Ross'
+        ross.last_name = 'Lawley'
+        ross.save()
 
-Now that we've got our user in the database, let's add a couple of posts::
+Assign another user to a variable called ``john``, just like we did above with
+``ross``.
+
+Now that we've got our users in the database, let's add a couple of posts::
 
     post1 = TextPost(title='Fun with MongoEngine', author=john)
     post1.content = 'Took a look at MongoEngine today, looks pretty cool.'
     post1.tags = ['mongodb', 'mongoengine']
     post1.save()
 
-    post2 = LinkPost(title='MongoEngine Documentation', author=john)
-    post2.link_url = 'http://tractiondigital.com/labs/mongoengine/docs'
+    post2 = LinkPost(title='MongoEngine Documentation', author=ross)
+    post2.link_url = 'http://docs.mongoengine.com/'
     post2.tags = ['mongoengine']
     post2.save()
 
-Note that if you change a field on a object that has already been saved, then
-call :meth:`save` again, the document will be updated.
+.. note:: If you change a field on an object that has already been saved and
+    then call :meth:`save` again, the document will be updated.
 
 Accessing our data
 ==================
+
 So now we've got a couple of posts in our database, how do we display them?
 Each document class (i.e. any class that inherits either directly or indirectly
 from :class:`~mongoengine.Document`) has an :attr:`objects` attribute, which is
@@ -192,16 +234,17 @@ used to access the documents in the database collection associated with that
 class. So let's see how we can get our posts' titles::
 
     for post in Post.objects:
-        print post.title
+        print(post.title)
 
 Retrieving type-specific information
 ------------------------------------
-This will print the titles of our posts, one on each line. But What if we want
+
+This will print the titles of our posts, one on each line. But what if we want
 to access the type-specific data (link_url, content, etc.)? One way is simply
 to use the :attr:`objects` attribute of a subclass of :class:`Post`::
 
     for post in TextPost.objects:
-        print post.content
+        print(post.content)
 
 Using TextPost's :attr:`objects` attribute only returns documents that were
 created using :class:`TextPost`. Actually, there is a more general rule here:
@@ -218,22 +261,21 @@ instances of :class:`Post` --- they were instances of the subclass of
 practice::
 
     for post in Post.objects:
-        print post.title
-        print '=' * len(post.title)
+        print(post.title)
+        print('=' * len(post.title))
 
         if isinstance(post, TextPost):
-            print post.content
+            print(post.content)
 
         if isinstance(post, LinkPost):
-            print 'Link:', post.link_url
-
-        print
+            print('Link: {}'.format(post.link_url))
 
 This would print the title of each post, followed by the content if it was a
 text post, and "Link: <url>" if it was a link post.
 
 Searching our posts by tag
 --------------------------
+
 The :attr:`objects` attribute of a :class:`~mongoengine.Document` is actually a
 :class:`~mongoengine.queryset.QuerySet` object. This lazily queries the
 database only when you need the data. It may also be filtered to narrow down
@@ -241,7 +283,7 @@ your query.  Let's adjust our query so that only posts with the tag "mongodb"
 are returned::
 
     for post in Post.objects(tags='mongodb'):
-        print post.title
+        print(post.title)
 
 There are also methods available on :class:`~mongoengine.queryset.QuerySet`
 objects that allow different results to be returned, for example, calling
@@ -250,5 +292,11 @@ the first matched by the query you provide. Aggregation functions may also be
 used on :class:`~mongoengine.queryset.QuerySet` objects::
 
     num_posts = Post.objects(tags='mongodb').count()
-    print 'Found % posts with tag "mongodb"' % num_posts
+    print('Found {} posts with tag "mongodb"'.format(num_posts))
 
+Learning more about MongoEngine
+-------------------------------
+
+If you got this far you've made a great start, so well done! The next step on
+your MongoEngine journey is the `full user guide <guide/index.html>`_, where
+you can learn in-depth about how to use MongoEngine and MongoDB.

docs/upgrade.rst

@@ -0,0 +1,628 @@
+#########
+Upgrading
+#########
+
+Development
+***********
+(Fill this out whenever you introduce breaking changes to MongoEngine)
+
+URLField's constructor no longer takes `verify_exists`
+
+0.15.0
+******
+
+0.14.0
+******
+This release includes a few bug fixes and a significant code cleanup. The most
+important change is that `QuerySet.as_pymongo` no longer supports a
+`coerce_types` mode. If you used it in the past, a) please let us know of your
+use case, b) you'll need to override `as_pymongo` to get the desired outcome.
+
+This release also makes the EmbeddedDocument not hashable by default. If you
+use embedded documents in sets or dictionaries, you might have to override
+`__hash__` and implement a hashing logic specific to your use case. See #1528
+for the reason behind this change.
+
+0.13.0
+******
+This release adds Unicode support to the `EmailField` and changes its
+structure significantly. Previously, email addresses containing Unicode
+characters didn't work at all. Starting with v0.13.0, domains with Unicode
+characters are supported out of the box, meaning some emails that previously
+didn't pass validation now do. Make sure the rest of your application can
+accept such email addresses. Additionally, if you subclassed the `EmailField`
+in your application and overrode `EmailField.EMAIL_REGEX`, you will have to
+adjust your code to override `EmailField.USER_REGEX`, `EmailField.DOMAIN_REGEX`,
+and potentially `EmailField.UTF8_USER_REGEX`.
+
+0.12.0
+******
+This release includes various fixes for the `BaseQuerySet` methods and how they
+are chained together. Since version 0.10.1 applying limit/skip/hint/batch_size
+to an already-existing queryset wouldn't modify the underlying PyMongo cursor.
+This has been fixed now, so you'll need to make sure that your code didn't rely
+on the broken implementation.
+
+Additionally, a public `BaseQuerySet.clone_into` has been renamed to a private
+`_clone_into`. If you directly used that method in your code, you'll need to
+rename its occurrences.
+
+0.11.0
+******
+This release includes a major rehaul of MongoEngine's code quality and
+introduces a few breaking changes. It also touches many different parts of
+the package and although all the changes have been tested and scrutinized,
+you're encouraged to thoroughly test the upgrade.
+
+First breaking change involves renaming `ConnectionError` to `MongoEngineConnectionError`.
+If you import or catch this exception, you'll need to rename it in your code.
+
+Second breaking change drops Python v2.6 support. If you run MongoEngine on
+that Python version, you'll need to upgrade it first.
+
+Third breaking change drops an old backward compatibility measure where
+`from mongoengine.base import ErrorClass` would work on top of
+`from mongoengine.errors import ErrorClass` (where `ErrorClass` is e.g.
+`ValidationError`). If you import any exceptions from `mongoengine.base`,
+change it to `mongoengine.errors`.
+
+0.10.8
+******
+This version fixed an issue where specifying a MongoDB URI host would override
+more information than it should. These changes are minor, but they still
+subtly modify the connection logic and thus you're encouraged to test your
+MongoDB connection before shipping v0.10.8 in production.
+
+0.10.7
+******
+
+`QuerySet.aggregate_sum` and `QuerySet.aggregate_average` are dropped. Use
+`QuerySet.sum` and `QuerySet.average` instead which use the aggreation framework
+by default from now on.
+
+0.9.0
+*****
+
+The 0.8.7 package on pypi was corrupted.  If upgrading from 0.8.7 to 0.9.0 please follow: ::
+
+    python -m pip uninstall pymongo
+    python -m pip uninstall mongoengine
+    python -m pip install pymongo==2.8
+    python -m pip install mongoengine
+
+0.8.7
+*****
+
+Calling reload on deleted / nonexistent documents now raises a DoesNotExist
+exception.
+
+
+0.8.2 to 0.8.3
+**************
+
+Minor change that may impact users:
+
+DynamicDocument fields are now stored in creation order after any declared
+fields.  Previously they were stored alphabetically.
+
+
+0.7 to 0.8
+**********
+
+There have been numerous backwards breaking changes in 0.8.  The reasons for
+these are to ensure that MongoEngine has sane defaults going forward and that it
+performs the best it can out of the box.  Where possible there have been
+FutureWarnings to help get you ready for the change, but that hasn't been
+possible for the whole of the release.
+
+.. warning:: Breaking changes - test upgrading on a test system before putting
+    live. There maybe multiple manual steps in migrating and these are best honed
+    on a staging / test system.
+
+Python and PyMongo
+==================
+
+MongoEngine requires python 2.6 (or above) and pymongo 2.5 (or above)
+
+Data Model
+==========
+
+Inheritance
+-----------
+
+The inheritance model has changed, we no longer need to store an array of
+:attr:`types` with the model we can just use the classname in :attr:`_cls`.
+This means that you will have to update your indexes for each of your
+inherited classes like so: ::
+
+    # 1. Declaration of the class
+    class Animal(Document):
+        name = StringField()
+        meta = {
+            'allow_inheritance': True,
+            'indexes': ['name']
+        }
+
+    # 2. Remove _types
+    collection = Animal._get_collection()
+    collection.update({}, {"$unset": {"_types": 1}}, multi=True)
+
+    # 3. Confirm extra data is removed
+    count = collection.find({'_types': {"$exists": True}}).count()
+    assert count == 0
+
+    # 4. Remove indexes
+    info = collection.index_information()
+    indexes_to_drop = [key for key, value in info.items()
+                       if '_types' in dict(value['key'])]
+    for index in indexes_to_drop:
+        collection.drop_index(index)
+
+    # 5. Recreate indexes
+    Animal.ensure_indexes()
+
+
+Document Definition
+-------------------
+
+The default for inheritance has changed - it is now off by default and
+:attr:`_cls` will not be stored automatically with the class.  So if you extend
+your :class:`~mongoengine.Document` or :class:`~mongoengine.EmbeddedDocuments`
+you will need to declare :attr:`allow_inheritance` in the meta data like so: ::
+
+    class Animal(Document):
+        name = StringField()
+
+        meta = {'allow_inheritance': True}
+
+Previously, if you had data in the database that wasn't defined in the Document
+definition, it would set it as an attribute on the document.  This is no longer
+the case and the data is set only in the ``document._data`` dictionary: ::
+
+    >>> from mongoengine import *
+    >>> class Animal(Document):
+    ...    name = StringField()
+    ...
+    >>> cat = Animal(name="kit", size="small")
+
+    # 0.7
+    >>> cat.size
+    u'small'
+
+    # 0.8
+    >>> cat.size
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    AttributeError: 'Animal' object has no attribute 'size'
+
+The Document class has introduced a reserved function `clean()`, which will be
+called before saving the document. If your document class happens to have a method
+with the same name, please try to rename it.
+
+    def clean(self):
+        pass
+
+ReferenceField
+--------------
+
+ReferenceFields now store ObjectIds by default - this is more efficient than
+DBRefs as we already know what Document types they reference::
+
+    # Old code
+    class Animal(Document):
+        name = ReferenceField('self')
+
+    # New code to keep dbrefs
+    class Animal(Document):
+        name = ReferenceField('self', dbref=True)
+
+To migrate all the references you need to touch each object and mark it as dirty
+eg::
+
+    # Doc definition
+    class Person(Document):
+        name = StringField()
+        parent = ReferenceField('self')
+        friends = ListField(ReferenceField('self'))
+
+    # Mark all ReferenceFields as dirty and save
+    for p in Person.objects:
+        p._mark_as_changed('parent')
+        p._mark_as_changed('friends')
+        p.save()
+
+`An example test migration for ReferenceFields is available on github
+<https://github.com/MongoEngine/mongoengine/blob/master/tests/migration/refrencefield_dbref_to_object_id.py>`_.
+
+.. Note:: Internally mongoengine handles ReferenceFields the same, so they are
+   converted to DBRef on loading and ObjectIds or DBRefs depending on settings
+   on storage.
+
+UUIDField
+---------
+
+UUIDFields now default to storing binary values::
+
+    # Old code
+    class Animal(Document):
+        uuid = UUIDField()
+
+    # New code
+    class Animal(Document):
+        uuid = UUIDField(binary=False)
+
+To migrate all the uuids you need to touch each object and mark it as dirty
+eg::
+
+    # Doc definition
+    class Animal(Document):
+        uuid = UUIDField()
+
+    # Mark all UUIDFields as dirty and save
+    for a in Animal.objects:
+        a._mark_as_changed('uuid')
+        a.save()
+
+`An example test migration for UUIDFields is available on github
+<https://github.com/MongoEngine/mongoengine/blob/master/tests/migration/uuidfield_to_binary.py>`_.
+
+DecimalField
+------------
+
+DecimalFields now store floats - previously it was storing strings and that
+made it impossible to do comparisons when querying correctly.::
+
+    # Old code
+    class Person(Document):
+        balance = DecimalField()
+
+    # New code
+    class Person(Document):
+        balance = DecimalField(force_string=True)
+
+To migrate all the DecimalFields you need to touch each object and mark it as dirty
+eg::
+
+    # Doc definition
+    class Person(Document):
+        balance = DecimalField()
+
+    # Mark all DecimalField's as dirty and save
+    for p in Person.objects:
+        p._mark_as_changed('balance')
+        p.save()
+
+.. note:: DecimalFields have also been improved with the addition of precision
+    and rounding.  See :class:`~mongoengine.fields.DecimalField` for more information.
+
+`An example test migration for DecimalFields is available on github
+<https://github.com/MongoEngine/mongoengine/blob/master/tests/migration/decimalfield_as_float.py>`_.
+
+Cascading Saves
+---------------
+To improve performance document saves will no longer automatically cascade.
+Any changes to a Document's references will either have to be saved manually or
+you will have to explicitly tell it to cascade on save::
+
+    # At the class level:
+    class Person(Document):
+        meta = {'cascade': True}
+
+    # Or on save:
+    my_document.save(cascade=True)
+
+Storage
+-------
+
+Document and Embedded Documents are now serialized based on declared field order.
+Previously, the data was passed to mongodb as a dictionary and which meant that
+order wasn't guaranteed - so things like ``$addToSet`` operations on
+:class:`~mongoengine.EmbeddedDocument` could potentially fail in unexpected
+ways.
+
+If this impacts you, you may want to rewrite the objects using the
+``doc.mark_as_dirty('field')`` pattern described above.  If you are using a
+compound primary key then you will need to ensure the order is fixed and match
+your EmbeddedDocument to that order.
+
+Querysets
+=========
+
+Attack of the clones
+--------------------
+
+Querysets now return clones and should no longer be considered editable in
+place.  This brings us in line with how Django's querysets work and removes a
+long running gotcha.  If you edit your querysets inplace you will have to
+update your code like so: ::
+
+    # Old code:
+    mammals = Animal.objects(type="mammal")
+    mammals.filter(order="Carnivora")       # Returns a cloned queryset that isn't assigned to anything - so this will break in 0.8
+    [m for m in mammals]                    # This will return all mammals in 0.8 as the 2nd filter returned a new queryset
+
+    # Update example a) assign queryset after a change:
+    mammals = Animal.objects(type="mammal")
+    carnivores = mammals.filter(order="Carnivora") # Reassign the new queryset so filter can be applied
+    [m for m in carnivores]                        # This will return all carnivores
+
+    # Update example b) chain the queryset:
+    mammals = Animal.objects(type="mammal").filter(order="Carnivora")  # The final queryset is assgined to mammals
+    [m for m in mammals]                                               # This will return all carnivores
+
+Len iterates the queryset
+-------------------------
+
+If you ever did `len(queryset)` it previously did a `count()` under the covers,
+this caused some unusual issues.  As `len(queryset)` is most often used by
+`list(queryset)` we now cache the queryset results and use that for the length.
+
+This isn't as performant as a `count()` and if you aren't iterating the
+queryset you should upgrade to use count::
+
+    # Old code
+    len(Animal.objects(type="mammal"))
+
+    # New code
+    Animal.objects(type="mammal").count()
+
+
+.only() now inline with .exclude()
+----------------------------------
+
+The behaviour of `.only()` was highly ambiguous, now it works in mirror fashion
+to `.exclude()`.  Chaining `.only()` calls will increase the fields required::
+
+    # Old code
+    Animal.objects().only(['type', 'name']).only('name', 'order')  # Would have returned just `name`
+
+    # New code
+    Animal.objects().only('name')
+
+    # Note:
+    Animal.objects().only(['name']).only('order')  # Now returns `name` *and* `order`
+
+
+Client
+======
+PyMongo 2.4 came with a new connection client; MongoClient_ and started the
+depreciation of the old :class:`~pymongo.connection.Connection`. MongoEngine
+now uses the latest `MongoClient` for connections.  By default operations were
+`safe` but if you turned them off or used the connection directly this will
+impact your queries.
+
+Querysets
+---------
+
+Safe
+^^^^
+
+`safe` has been depreciated in the new MongoClient connection.  Please use
+`write_concern` instead.  As `safe` always defaulted as `True` normally no code
+change is required. To disable confirmation of the write just pass `{"w": 0}`
+eg: ::
+
+   # Old
+   Animal(name="Dinasour").save(safe=False)
+
+   # new code:
+   Animal(name="Dinasour").save(write_concern={"w": 0})
+
+Write Concern
+^^^^^^^^^^^^^
+
+`write_options` has been replaced with `write_concern` to bring it inline with
+pymongo. To upgrade simply rename any instances where you used the `write_option`
+keyword  to `write_concern` like so::
+
+   # Old code:
+   Animal(name="Dinasour").save(write_options={"w": 2})
+
+   # new code:
+   Animal(name="Dinasour").save(write_concern={"w": 2})
+
+
+Indexes
+=======
+
+Index methods are no longer tied to querysets but rather to the document class.
+Although `QuerySet._ensure_indexes` and `QuerySet.ensure_index` still exist.
+They should be replaced with :func:`~mongoengine.Document.ensure_indexes` /
+:func:`~mongoengine.Document.ensure_index`.
+
+SequenceFields
+==============
+
+:class:`~mongoengine.fields.SequenceField` now inherits from `BaseField` to
+allow flexible storage of the calculated value.  As such MIN and MAX settings
+are no longer handled.
+
+.. _MongoClient: http://blog.mongodb.org/post/36666163412/introducing-mongoclient
+
+0.6 to 0.7
+**********
+
+Cascade saves
+=============
+
+Saves will raise a `FutureWarning` if they cascade and cascade hasn't been set
+to True.  This is because in 0.8 it will default to False.  If you require
+cascading saves then either set it in the `meta` or pass
+via `save` eg ::
+
+    # At the class level:
+    class Person(Document):
+        meta = {'cascade': True}
+
+    # Or in code:
+    my_document.save(cascade=True)
+
+.. note::
+    Remember: cascading saves **do not** cascade through lists.
+
+ReferenceFields
+===============
+
+ReferenceFields now can store references as ObjectId strings instead of DBRefs.
+This will become the default in 0.8 and if `dbref` is not set a `FutureWarning`
+will be raised.
+
+
+To explicitly continue to use DBRefs change the `dbref` flag
+to True ::
+
+   class Person(Document):
+       groups = ListField(ReferenceField(Group, dbref=True))
+
+To migrate to using strings instead of DBRefs you will have to manually
+migrate ::
+
+        # Step 1 - Migrate the model definition
+        class Group(Document):
+            author = ReferenceField(User, dbref=False)
+            members = ListField(ReferenceField(User, dbref=False))
+
+        # Step 2 - Migrate the data
+        for g in Group.objects():
+            g.author = g.author
+            g.members = g.members
+            g.save()
+
+
+item_frequencies
+================
+
+In the 0.6 series we added support for null / zero / false values in
+item_frequencies.  A side effect was to return keys in the value they are
+stored in rather than as string representations.  Your code may need to be
+updated to handle native types rather than strings keys for the results of
+item frequency queries.
+
+BinaryFields
+============
+
+Binary fields have been updated so that they are native binary types.  If you
+previously were doing `str` comparisons with binary field values you will have
+to update and wrap the value in a `str`.
+
+0.5 to 0.6
+**********
+
+Embedded Documents - if you had a `pk` field you will have to rename it from
+`_id` to `pk` as pk is no longer a property of Embedded Documents.
+
+Reverse Delete Rules in Embedded Documents, MapFields and DictFields now throw
+an InvalidDocument error as they aren't currently supported.
+
+Document._get_subclasses - Is no longer used and the class method has been
+removed.
+
+Document.objects.with_id - now raises an InvalidQueryError if used with a
+filter.
+
+FutureWarning - A future warning has been added to all inherited classes that
+don't define :attr:`allow_inheritance` in their meta.
+
+You may need to update pyMongo to 2.0 for use with Sharding.
+
+0.4 to 0.5
+**********
+
+There have been the following backwards incompatibilities from 0.4 to 0.5.  The
+main areas of changed are: choices in fields, map_reduce and collection names.
+
+Choice options:
+===============
+
+Are now expected to be an iterable of tuples, with the first element in each
+tuple being the actual value to be stored. The second element is the
+human-readable name for the option.
+
+
+PyMongo / MongoDB
+=================
+
+map reduce now requires pymongo 1.11+- The pymongo `merge_output` and
+`reduce_output` parameters, have been depreciated.
+
+More methods now use map_reduce as db.eval is not supported for sharding as
+such the following have been changed:
+
+    * :meth:`~mongoengine.queryset.QuerySet.sum`
+    * :meth:`~mongoengine.queryset.QuerySet.average`
+    * :meth:`~mongoengine.queryset.QuerySet.item_frequencies`
+
+
+Default collection naming
+=========================
+
+Previously it was just lowercase, it's now much more pythonic and readable as
+it's lowercase and underscores, previously ::
+
+    class MyAceDocument(Document):
+        pass
+
+    MyAceDocument._meta['collection'] == myacedocument
+
+In 0.5 this will change to ::
+
+    class MyAceDocument(Document):
+        pass
+
+    MyAceDocument._get_collection_name() == my_ace_document
+
+To upgrade use a Mixin class to set meta like so ::
+
+    class BaseMixin(object):
+        meta = {
+            'collection': lambda c: c.__name__.lower()
+        }
+
+    class MyAceDocument(Document, BaseMixin):
+        pass
+
+    MyAceDocument._get_collection_name() == "myacedocument"
+
+Alternatively, you can rename your collections eg ::
+
+    from mongoengine.connection import _get_db
+    from mongoengine.base import _document_registry
+
+    def rename_collections():
+        db = _get_db()
+
+        failure = False
+
+        collection_names = [d._get_collection_name()
+                            for d in _document_registry.values()]
+
+        for new_style_name in collection_names:
+            if not new_style_name:  # embedded documents don't have collections
+                continue
+            old_style_name = new_style_name.replace('_', '')
+
+            if old_style_name == new_style_name:
+                continue  # Nothing to do
+
+            existing = db.collection_names()
+            if old_style_name in existing:
+                if new_style_name in existing:
+                    failure = True
+                    print "FAILED to rename: %s to %s (already exists)" % (
+                        old_style_name, new_style_name)
+                else:
+                    db[old_style_name].rename(new_style_name)
+                    print "Renamed:  %s to %s" % (old_style_name,
+                                                  new_style_name)
+
+        if failure:
+            print "Upgrading  collection names failed"
+        else:
+            print "Upgraded collection names"
+
+
+mongodb 1.8 > 2.0 +
+===================
+
+It's been reported that indexes may need to be recreated to the newer version of indexes.
+To do this drop indexes and call ``ensure_indexes`` on each model.

mongoengine/__init__.py

@@ -1,24 +1,42 @@
-import document
-from document import *
-import fields
-from fields import *
-import connection
-from connection import *
-import queryset
-from queryset import *
+# Import submodules so that we can expose their __all__
+from mongoengine import connection
+from mongoengine import document
+from mongoengine import errors
+from mongoengine import fields
+from mongoengine import queryset
+from mongoengine import signals
 
-__all__ = (document.__all__ + fields.__all__ + connection.__all__ +
-           queryset.__all__)
+# Import everything from each submodule so that it can be accessed via
+# mongoengine, e.g. instead of `from mongoengine.connection import connect`,
+# users can simply use `from mongoengine import connect`, or even
+# `from mongoengine import *` and then `connect('testdb')`.
+from mongoengine.connection import *
+from mongoengine.document import *
+from mongoengine.errors import *
+from mongoengine.fields import *
+from mongoengine.queryset import *
+from mongoengine.signals import *
 
-__author__ = 'Harry Marr'
 
-VERSION = (0, 2, 2)
+__all__ = (
+    list(document.__all__)
+    + list(fields.__all__)
+    + list(connection.__all__)
+    + list(queryset.__all__)
+    + list(signals.__all__)
+    + list(errors.__all__)
+)
+
+
+VERSION = (0, 20, 0)
+
 
 def get_version():
-    version = '%s.%s' % (VERSION[0], VERSION[1])
-    if VERSION[2]:
-        version = '%s.%s' % (version, VERSION[2])
-    return version
+    """Return the VERSION as a string.
+
+    For example, if `VERSION == (0, 10, 7)`, return '0.10.7'.
+    """
+    return ".".join(map(str, VERSION))
+
 
 __version__ = get_version()
-

mongoengine/base.py

@@ -1,375 +0,0 @@
-from queryset import QuerySet, QuerySetManager
-
-import pymongo
-
-
-class ValidationError(Exception):
-    pass
-
-
-class BaseField(object):
-    """A base class for fields in a MongoDB document. Instances of this class
-    may be added to subclasses of `Document` to define a document's schema.
-    """
-
-    # Fields may have _types inserted into indexes by default 
-    _index_with_types = True
-    
-    def __init__(self, name=None, required=False, default=None, unique=False,
-                 unique_with=None, primary_key=False):
-        self.name = name if not primary_key else '_id'
-        self.required = required or primary_key
-        self.default = default
-        self.unique = bool(unique or unique_with)
-        self.unique_with = unique_with
-        self.primary_key = primary_key
-
-    def __get__(self, instance, owner):
-        """Descriptor for retrieving a value from a field in a document. Do 
-        any necessary conversion between Python and MongoDB types.
-        """
-        if instance is None:
-            # Document class being used rather than a document object
-            return self
-
-        # Get value from document instance if available, if not use default
-        value = instance._data.get(self.name)
-        if value is None:
-            value = self.default
-            # Allow callable default values
-            if callable(value):
-                value = value()
-        return value
-
-    def __set__(self, instance, value):
-        """Descriptor for assigning a value to a field in a document.
-        """
-        instance._data[self.name] = value
-
-    def to_python(self, value):
-        """Convert a MongoDB-compatible type to a Python type.
-        """
-        return value
-
-    def to_mongo(self, value):
-        """Convert a Python type to a MongoDB-compatible type.
-        """
-        return self.to_python(value)
-
-    def prepare_query_value(self, value):
-        """Prepare a value that is being used in a query for PyMongo.
-        """
-        return value
-
-    def validate(self, value):
-        """Perform validation on a value.
-        """
-        pass
-
-
-class ObjectIdField(BaseField):
-    """An field wrapper around MongoDB's ObjectIds.
-    """
-    
-    def to_python(self, value):
-        return unicode(value)
-
-    def to_mongo(self, value):
-        if not isinstance(value, pymongo.objectid.ObjectId):
-            return pymongo.objectid.ObjectId(str(value))
-        return value
-
-    def prepare_query_value(self, value):
-        return self.to_mongo(value)
-
-    def validate(self, value):
-        try:
-            pymongo.objectid.ObjectId(str(value))
-        except:
-            raise ValidationError('Invalid Object ID')
-
-
-class DocumentMetaclass(type):
-    """Metaclass for all documents.
-    """
-
-    def __new__(cls, name, bases, attrs):
-        metaclass = attrs.get('__metaclass__')
-        super_new = super(DocumentMetaclass, cls).__new__
-        if metaclass and issubclass(metaclass, DocumentMetaclass):
-            return super_new(cls, name, bases, attrs)
-
-        doc_fields = {}
-        class_name = [name]
-        superclasses = {}
-        for base in bases:
-            # Include all fields present in superclasses
-            if hasattr(base, '_fields'):
-                doc_fields.update(base._fields)
-                class_name.append(base._class_name)
-                # Get superclasses from superclass
-                superclasses[base._class_name] = base
-                superclasses.update(base._superclasses)
-        attrs['_class_name'] = '.'.join(reversed(class_name))
-        attrs['_superclasses'] = superclasses
-
-        # Add the document's fields to the _fields attribute
-        for attr_name, attr_value in attrs.items():
-            if hasattr(attr_value, "__class__") and \
-                issubclass(attr_value.__class__, BaseField):
-                if not attr_value.name:
-                    attr_value.name = attr_name
-                doc_fields[attr_name] = attr_value
-        attrs['_fields'] = doc_fields
-
-        return super_new(cls, name, bases, attrs)
-
-
-class TopLevelDocumentMetaclass(DocumentMetaclass):
-    """Metaclass for top-level documents (i.e. documents that have their own
-    collection in the database.
-    """
-
-    def __new__(cls, name, bases, attrs):
-        super_new = super(TopLevelDocumentMetaclass, cls).__new__
-        # Classes defined in this package are abstract and should not have 
-        # their own metadata with DB collection, etc.
-        # __metaclass__ is only set on the class with the __metaclass__ 
-        # attribute (i.e. it is not set on subclasses). This differentiates
-        # 'real' documents from the 'Document' class
-        if attrs.get('__metaclass__') == TopLevelDocumentMetaclass:
-            return super_new(cls, name, bases, attrs)
-
-        collection = name.lower()
-        
-        simple_class = True
-        id_field = None
-        base_indexes = []
-
-        # Subclassed documents inherit collection from superclass
-        for base in bases:
-            if hasattr(base, '_meta') and 'collection' in base._meta:
-                # Ensure that the Document class may be subclassed - 
-                # inheritance may be disabled to remove dependency on 
-                # additional fields _cls and _types
-                if base._meta.get('allow_inheritance', True) == False:
-                    raise ValueError('Document %s may not be subclassed' %
-                                     base.__name__)
-                else:
-                    simple_class = False
-                collection = base._meta['collection']
-
-                id_field = id_field or base._meta.get('id_field')
-                base_indexes += base._meta.get('indexes', [])
-
-        meta = {
-            'collection': collection,
-            'allow_inheritance': True,
-            'max_documents': None,
-            'max_size': None,
-            'ordering': [], # default ordering applied at runtime
-            'indexes': [], # indexes to be ensured at runtime
-            'id_field': id_field,
-        }
-
-        # Apply document-defined meta options
-        meta.update(attrs.get('meta', {}))
-
-        # Only simple classes - direct subclasses of Document - may set
-        # allow_inheritance to False
-        if not simple_class and not meta['allow_inheritance']:
-            raise ValueError('Only direct subclasses of Document may set '
-                             '"allow_inheritance" to False')
-        attrs['_meta'] = meta
-
-        # Set up collection manager, needs the class to have fields so use
-        # DocumentMetaclass before instantiating CollectionManager object
-        new_class = super_new(cls, name, bases, attrs)
-        new_class.objects = QuerySetManager()
-
-        user_indexes = [QuerySet._build_index_spec(new_class, spec)
-                        for spec in meta['indexes']] + base_indexes
-        new_class._meta['indexes'] = user_indexes
-        
-        unique_indexes = []
-        for field_name, field in new_class._fields.items():
-            # Generate a list of indexes needed by uniqueness constraints
-            if field.unique:
-                field.required = True
-                unique_fields = [field_name]
-
-                # Add any unique_with fields to the back of the index spec
-                if field.unique_with:
-                    if isinstance(field.unique_with, basestring):
-                        field.unique_with = [field.unique_with]
-
-                    # Convert unique_with field names to real field names
-                    unique_with = []
-                    for other_name in field.unique_with:
-                        parts = other_name.split('.')
-                        # Lookup real name
-                        parts = QuerySet._lookup_field(new_class, parts)
-                        name_parts = [part.name for part in parts]
-                        unique_with.append('.'.join(name_parts))
-                        # Unique field should be required
-                        parts[-1].required = True
-                    unique_fields += unique_with
-
-                # Add the new index to the list
-                index = [(f, pymongo.ASCENDING) for f in unique_fields]
-                unique_indexes.append(index)
-
-            # Check for custom primary key
-            if field.primary_key:
-                if not new_class._meta['id_field']:
-                    new_class._meta['id_field'] = field_name
-                    # Make 'Document.id' an alias to the real primary key field
-                    new_class.id = field
-                    #new_class._fields['id'] = field
-                else:
-                    raise ValueError('Cannot override primary key field')
-
-        new_class._meta['unique_indexes'] = unique_indexes
-
-        if not new_class._meta['id_field']:
-            new_class._meta['id_field'] = 'id'
-            new_class.id = new_class._fields['id'] = ObjectIdField(name='_id')
-
-        return new_class
-
-
-class BaseDocument(object):
-
-    def __init__(self, **values):
-        self._data = {}
-        # Assign initial values to instance
-        for attr_name, attr_value in self._fields.items():
-            if attr_name in values:
-                setattr(self, attr_name, values.pop(attr_name))
-            else:
-                # Use default value if present
-                value = getattr(self, attr_name, None)
-                setattr(self, attr_name, value)
-
-    def validate(self):
-        """Ensure that all fields' values are valid and that required fields
-        are present.
-        """
-        # Get a list of tuples of field names and their current values
-        fields = [(field, getattr(self, name)) 
-                  for name, field in self._fields.items()]
-
-        # Ensure that each field is matched to a valid value
-        for field, value in fields:
-            if value is not None:
-                try:
-                    field.validate(value)
-                except (ValueError, AttributeError, AssertionError), e:
-                    raise ValidationError('Invalid value for field of type "' +
-                                          field.__class__.__name__ + '"')
-            elif field.required:
-                raise ValidationError('Field "%s" is required' % field.name)
-
-    @classmethod
-    def _get_subclasses(cls):
-        """Return a dictionary of all subclasses (found recursively).
-        """
-        try:
-            subclasses = cls.__subclasses__()
-        except:
-            subclasses = cls.__subclasses__(cls)
-
-        all_subclasses = {}
-        for subclass in subclasses:
-            all_subclasses[subclass._class_name] = subclass
-            all_subclasses.update(subclass._get_subclasses())
-        return all_subclasses
-
-    def __iter__(self):
-        return iter(self._fields)
-
-    def __getitem__(self, name):
-        """Dictionary-style field access, return a field's value if present.
-        """
-        try:
-            if name in self._fields:
-                return getattr(self, name)
-        except AttributeError:
-            pass
-        raise KeyError(name)
-
-    def __setitem__(self, name, value):
-        """Dictionary-style field access, set a field's value.
-        """
-        # Ensure that the field exists before settings its value
-        if name not in self._fields:
-            raise KeyError(name)
-        return setattr(self, name, value)
-
-    def __contains__(self, name):
-        try:
-            val = getattr(self, name)
-            return val is not None
-        except AttributeError:
-            return False
-
-    def __len__(self):
-        return len(self._data)
-
-    def __repr__(self):
-        try:
-            u = unicode(self)
-        except (UnicodeEncodeError, UnicodeDecodeError):
-            u = '[Bad Unicode data]'
-        return u'<%s: %s>' % (self.__class__.__name__, u)
-
-    def __str__(self):
-        if hasattr(self, '__unicode__'):
-            return unicode(self).encode('utf-8')
-        return '%s object' % self.__class__.__name__
-
-    def to_mongo(self):
-        """Return data dictionary ready for use with MongoDB.
-        """
-        data = {}
-        for field_name, field in self._fields.items():
-            value = getattr(self, field_name, None)
-            if value is not None:
-                data[field.name] = field.to_mongo(value)
-        # Only add _cls and _types if allow_inheritance is not False
-        if not (hasattr(self, '_meta') and
-                self._meta.get('allow_inheritance', True) == False):
-            data['_cls'] = self._class_name
-            data['_types'] = self._superclasses.keys() + [self._class_name]
-        return data
-    
-    @classmethod
-    def _from_son(cls, son):
-        """Create an instance of a Document (subclass) from a PyMongo SOM.
-        """
-        # get the class name from the document, falling back to the given
-        # class if unavailable
-        class_name = son.get(u'_cls', cls._class_name)
-
-        data = dict((str(key), value) for key, value in son.items())
-
-        if '_types' in data:
-            del data['_types']
-
-        if '_cls' in data:
-            del data['_cls']
-
-        # Return correct subclass for document type
-        if class_name != cls._class_name:
-            subclasses = cls._get_subclasses()
-            if class_name not in subclasses:
-                # Type of document is probably more generic than the class
-                # that has been queried to return this SON
-                return None
-            cls = subclasses[class_name]
-
-        for field_name, field in cls._fields.items():
-            if field.name in data:
-                data[field_name] = field.to_python(data[field.name])
-
-        return cls(**data)

mongoengine/base/__init__.py

@@ -0,0 +1,33 @@
+# Base module is split into several files for convenience. Files inside of
+# this module should import from a specific submodule (e.g.
+# `from mongoengine.base.document import BaseDocument`), but all of the
+# other modules should import directly from the top-level module (e.g.
+# `from mongoengine.base import BaseDocument`). This approach is cleaner and
+# also helps with cyclical import errors.
+from mongoengine.base.common import *
+from mongoengine.base.datastructures import *
+from mongoengine.base.document import *
+from mongoengine.base.fields import *
+from mongoengine.base.metaclasses import *
+
+__all__ = (
+    # common
+    "UPDATE_OPERATORS",
+    "_document_registry",
+    "get_document",
+    # datastructures
+    "BaseDict",
+    "BaseList",
+    "EmbeddedDocumentList",
+    "LazyReference",
+    # document
+    "BaseDocument",
+    # fields
+    "BaseField",
+    "ComplexBaseField",
+    "ObjectIdField",
+    "GeoJsonBaseField",
+    # metaclasses
+    "DocumentMetaclass",
+    "TopLevelDocumentMetaclass",
+)

mongoengine/base/common.py

@@ -0,0 +1,62 @@
+from mongoengine.errors import NotRegistered
+
+__all__ = ("UPDATE_OPERATORS", "get_document", "_document_registry")
+
+
+UPDATE_OPERATORS = {
+    "set",
+    "unset",
+    "inc",
+    "dec",
+    "mul",
+    "pop",
+    "push",
+    "push_all",
+    "pull",
+    "pull_all",
+    "add_to_set",
+    "set_on_insert",
+    "min",
+    "max",
+    "rename",
+}
+
+
+_document_registry = {}
+
+
+def get_document(name):
+    """Get a registered Document class by name."""
+    doc = _document_registry.get(name, None)
+    if not doc:
+        # Possible old style name
+        single_end = name.split(".")[-1]
+        compound_end = ".%s" % single_end
+        possible_match = [
+            k for k in _document_registry if k.endswith(compound_end) or k == single_end
+        ]
+        if len(possible_match) == 1:
+            doc = _document_registry.get(possible_match.pop(), None)
+    if not doc:
+        raise NotRegistered(
+            """
+            `%s` has not been registered in the document registry.
+            Importing the document class automatically registers it, has it
+            been imported?
+        """.strip()
+            % name
+        )
+    return doc
+
+
+def _get_documents_by_db(connection_alias, default_connection_alias):
+    """Get all registered Documents class attached to a given database"""
+
+    def get_doc_alias(doc_cls):
+        return doc_cls._meta.get("db_alias", default_connection_alias)
+
+    return [
+        doc_cls
+        for doc_cls in _document_registry.values()
+        if get_doc_alias(doc_cls) == connection_alias
+    ]

mongoengine/base/datastructures.py

@@ -0,0 +1,475 @@
+import weakref
+
+from bson import DBRef
+
+from mongoengine.common import _import_class
+from mongoengine.errors import DoesNotExist, MultipleObjectsReturned
+
+__all__ = (
+    "BaseDict",
+    "StrictDict",
+    "BaseList",
+    "EmbeddedDocumentList",
+    "LazyReference",
+)
+
+
+def mark_as_changed_wrapper(parent_method):
+    """Decorator that ensures _mark_as_changed method gets called."""
+
+    def wrapper(self, *args, **kwargs):
+        # Can't use super() in the decorator.
+        result = parent_method(self, *args, **kwargs)
+        self._mark_as_changed()
+        return result
+
+    return wrapper
+
+
+def mark_key_as_changed_wrapper(parent_method):
+    """Decorator that ensures _mark_as_changed method gets called with the key argument"""
+
+    def wrapper(self, key, *args, **kwargs):
+        # Can't use super() in the decorator.
+        result = parent_method(self, key, *args, **kwargs)
+        self._mark_as_changed(key)
+        return result
+
+    return wrapper
+
+
+class BaseDict(dict):
+    """A special dict so we can watch any changes."""
+
+    _dereferenced = False
+    _instance = None
+    _name = None
+
+    def __init__(self, dict_items, instance, name):
+        BaseDocument = _import_class("BaseDocument")
+
+        if isinstance(instance, BaseDocument):
+            self._instance = weakref.proxy(instance)
+        self._name = name
+        super().__init__(dict_items)
+
+    def get(self, key, default=None):
+        # get does not use __getitem__ by default so we must override it as well
+        try:
+            return self.__getitem__(key)
+        except KeyError:
+            return default
+
+    def __getitem__(self, key):
+        value = super().__getitem__(key)
+
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        if isinstance(value, EmbeddedDocument) and value._instance is None:
+            value._instance = self._instance
+        elif isinstance(value, dict) and not isinstance(value, BaseDict):
+            value = BaseDict(value, None, "{}.{}".format(self._name, key))
+            super().__setitem__(key, value)
+            value._instance = self._instance
+        elif isinstance(value, list) and not isinstance(value, BaseList):
+            value = BaseList(value, None, "{}.{}".format(self._name, key))
+            super().__setitem__(key, value)
+            value._instance = self._instance
+        return value
+
+    def __getstate__(self):
+        self.instance = None
+        self._dereferenced = False
+        return self
+
+    def __setstate__(self, state):
+        self = state
+        return self
+
+    __setitem__ = mark_key_as_changed_wrapper(dict.__setitem__)
+    __delattr__ = mark_key_as_changed_wrapper(dict.__delattr__)
+    __delitem__ = mark_key_as_changed_wrapper(dict.__delitem__)
+    pop = mark_as_changed_wrapper(dict.pop)
+    clear = mark_as_changed_wrapper(dict.clear)
+    update = mark_as_changed_wrapper(dict.update)
+    popitem = mark_as_changed_wrapper(dict.popitem)
+    setdefault = mark_as_changed_wrapper(dict.setdefault)
+
+    def _mark_as_changed(self, key=None):
+        if hasattr(self._instance, "_mark_as_changed"):
+            if key:
+                self._instance._mark_as_changed("{}.{}".format(self._name, key))
+            else:
+                self._instance._mark_as_changed(self._name)
+
+
+class BaseList(list):
+    """A special list so we can watch any changes."""
+
+    _dereferenced = False
+    _instance = None
+    _name = None
+
+    def __init__(self, list_items, instance, name):
+        BaseDocument = _import_class("BaseDocument")
+
+        if isinstance(instance, BaseDocument):
+            self._instance = weakref.proxy(instance)
+        self._name = name
+        super().__init__(list_items)
+
+    def __getitem__(self, key):
+        # change index to positive value because MongoDB does not support negative one
+        if isinstance(key, int) and key < 0:
+            key = len(self) + key
+        value = super().__getitem__(key)
+
+        if isinstance(key, slice):
+            # When receiving a slice operator, we don't convert the structure and bind
+            # to parent's instance. This is buggy for now but would require more work to be handled properly
+            return value
+
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        if isinstance(value, EmbeddedDocument) and value._instance is None:
+            value._instance = self._instance
+        elif isinstance(value, dict) and not isinstance(value, BaseDict):
+            # Replace dict by BaseDict
+            value = BaseDict(value, None, "{}.{}".format(self._name, key))
+            super().__setitem__(key, value)
+            value._instance = self._instance
+        elif isinstance(value, list) and not isinstance(value, BaseList):
+            # Replace list by BaseList
+            value = BaseList(value, None, "{}.{}".format(self._name, key))
+            super().__setitem__(key, value)
+            value._instance = self._instance
+        return value
+
+    def __iter__(self):
+        yield from super().__iter__()
+
+    def __getstate__(self):
+        self.instance = None
+        self._dereferenced = False
+        return self
+
+    def __setstate__(self, state):
+        self = state
+        return self
+
+    def __setitem__(self, key, value):
+        changed_key = key
+        if isinstance(key, slice):
+            # In case of slice, we don't bother to identify the exact elements being updated
+            # instead, we simply marks the whole list as changed
+            changed_key = None
+
+        result = super().__setitem__(key, value)
+        self._mark_as_changed(changed_key)
+        return result
+
+    append = mark_as_changed_wrapper(list.append)
+    extend = mark_as_changed_wrapper(list.extend)
+    insert = mark_as_changed_wrapper(list.insert)
+    pop = mark_as_changed_wrapper(list.pop)
+    remove = mark_as_changed_wrapper(list.remove)
+    reverse = mark_as_changed_wrapper(list.reverse)
+    sort = mark_as_changed_wrapper(list.sort)
+    __delitem__ = mark_as_changed_wrapper(list.__delitem__)
+    __iadd__ = mark_as_changed_wrapper(list.__iadd__)
+    __imul__ = mark_as_changed_wrapper(list.__imul__)
+
+    def _mark_as_changed(self, key=None):
+        if hasattr(self._instance, "_mark_as_changed"):
+            if key:
+                self._instance._mark_as_changed(
+                    "{}.{}".format(self._name, key % len(self))
+                )
+            else:
+                self._instance._mark_as_changed(self._name)
+
+
+class EmbeddedDocumentList(BaseList):
+    def __init__(self, list_items, instance, name):
+        super().__init__(list_items, instance, name)
+        self._instance = instance
+
+    @classmethod
+    def __match_all(cls, embedded_doc, kwargs):
+        """Return True if a given embedded doc matches all the filter
+        kwargs. If it doesn't return False.
+        """
+        for key, expected_value in kwargs.items():
+            doc_val = getattr(embedded_doc, key)
+            if doc_val != expected_value and str(doc_val) != expected_value:
+                return False
+        return True
+
+    @classmethod
+    def __only_matches(cls, embedded_docs, kwargs):
+        """Return embedded docs that match the filter kwargs."""
+        if not kwargs:
+            return embedded_docs
+        return [doc for doc in embedded_docs if cls.__match_all(doc, kwargs)]
+
+    def filter(self, **kwargs):
+        """
+        Filters the list by only including embedded documents with the
+        given keyword arguments.
+
+        This method only supports simple comparison (e.g: .filter(name='John Doe'))
+        and does not support operators like __gte, __lte, __icontains like queryset.filter does
+
+        :param kwargs: The keyword arguments corresponding to the fields to
+         filter on. *Multiple arguments are treated as if they are ANDed
+         together.*
+        :return: A new ``EmbeddedDocumentList`` containing the matching
+         embedded documents.
+
+        Raises ``AttributeError`` if a given keyword is not a valid field for
+        the embedded document class.
+        """
+        values = self.__only_matches(self, kwargs)
+        return EmbeddedDocumentList(values, self._instance, self._name)
+
+    def exclude(self, **kwargs):
+        """
+        Filters the list by excluding embedded documents with the given
+        keyword arguments.
+
+        :param kwargs: The keyword arguments corresponding to the fields to
+         exclude on. *Multiple arguments are treated as if they are ANDed
+         together.*
+        :return: A new ``EmbeddedDocumentList`` containing the non-matching
+         embedded documents.
+
+        Raises ``AttributeError`` if a given keyword is not a valid field for
+        the embedded document class.
+        """
+        exclude = self.__only_matches(self, kwargs)
+        values = [item for item in self if item not in exclude]
+        return EmbeddedDocumentList(values, self._instance, self._name)
+
+    def count(self):
+        """
+        The number of embedded documents in the list.
+
+        :return: The length of the list, equivalent to the result of ``len()``.
+        """
+        return len(self)
+
+    def get(self, **kwargs):
+        """
+        Retrieves an embedded document determined by the given keyword
+        arguments.
+
+        :param kwargs: The keyword arguments corresponding to the fields to
+         search on. *Multiple arguments are treated as if they are ANDed
+         together.*
+        :return: The embedded document matched by the given keyword arguments.
+
+        Raises ``DoesNotExist`` if the arguments used to query an embedded
+        document returns no results. ``MultipleObjectsReturned`` if more
+        than one result is returned.
+        """
+        values = self.__only_matches(self, kwargs)
+        if len(values) == 0:
+            raise DoesNotExist("%s matching query does not exist." % self._name)
+        elif len(values) > 1:
+            raise MultipleObjectsReturned(
+                "%d items returned, instead of 1" % len(values)
+            )
+
+        return values[0]
+
+    def first(self):
+        """Return the first embedded document in the list, or ``None``
+        if empty.
+        """
+        if len(self) > 0:
+            return self[0]
+
+    def create(self, **values):
+        """
+        Creates a new instance of the EmbeddedDocument and appends it to this EmbeddedDocumentList.
+
+        .. note::
+            the instance of the EmbeddedDocument is not automatically saved to the database.
+            You still need to call .save() on the parent Document.
+
+        :param values: A dictionary of values for the embedded document.
+        :return: The new embedded document instance.
+        """
+        name = self._name
+        EmbeddedClass = self._instance._fields[name].field.document_type_obj
+        self._instance[self._name].append(EmbeddedClass(**values))
+
+        return self._instance[self._name][-1]
+
+    def save(self, *args, **kwargs):
+        """
+        Saves the ancestor document.
+
+        :param args: Arguments passed up to the ancestor Document's save
+         method.
+        :param kwargs: Keyword arguments passed up to the ancestor Document's
+         save method.
+        """
+        self._instance.save(*args, **kwargs)
+
+    def delete(self):
+        """
+        Deletes the embedded documents from the database.
+
+        .. note::
+            The embedded document changes are not automatically saved
+            to the database after calling this method.
+
+        :return: The number of entries deleted.
+        """
+        values = list(self)
+        for item in values:
+            self._instance[self._name].remove(item)
+
+        return len(values)
+
+    def update(self, **update):
+        """
+        Updates the embedded documents with the given replacement values. This
+        function does not support mongoDB update operators such as ``inc__``.
+
+        .. note::
+            The embedded document changes are not automatically saved
+            to the database after calling this method.
+
+        :param update: A dictionary of update values to apply to each
+         embedded document.
+        :return: The number of entries updated.
+        """
+        if len(update) == 0:
+            return 0
+        values = list(self)
+        for item in values:
+            for k, v in update.items():
+                setattr(item, k, v)
+
+        return len(values)
+
+
+class StrictDict:
+    __slots__ = ()
+    _special_fields = {"get", "pop", "iteritems", "items", "keys", "create"}
+    _classes = {}
+
+    def __init__(self, **kwargs):
+        for k, v in kwargs.items():
+            setattr(self, k, v)
+
+    def __getitem__(self, key):
+        key = "_reserved_" + key if key in self._special_fields else key
+        try:
+            return getattr(self, key)
+        except AttributeError:
+            raise KeyError(key)
+
+    def __setitem__(self, key, value):
+        key = "_reserved_" + key if key in self._special_fields else key
+        return setattr(self, key, value)
+
+    def __contains__(self, key):
+        return hasattr(self, key)
+
+    def get(self, key, default=None):
+        try:
+            return self[key]
+        except KeyError:
+            return default
+
+    def pop(self, key, default=None):
+        v = self.get(key, default)
+        try:
+            delattr(self, key)
+        except AttributeError:
+            pass
+        return v
+
+    def iteritems(self):
+        for key in self:
+            yield key, self[key]
+
+    def items(self):
+        return [(k, self[k]) for k in iter(self)]
+
+    def iterkeys(self):
+        return iter(self)
+
+    def keys(self):
+        return list(iter(self))
+
+    def __iter__(self):
+        return (key for key in self.__slots__ if hasattr(self, key))
+
+    def __len__(self):
+        return len(list(self.items()))
+
+    def __eq__(self, other):
+        return list(self.items()) == list(other.items())
+
+    def __ne__(self, other):
+        return not (self == other)
+
+    @classmethod
+    def create(cls, allowed_keys):
+        allowed_keys_tuple = tuple(
+            ("_reserved_" + k if k in cls._special_fields else k) for k in allowed_keys
+        )
+        allowed_keys = frozenset(allowed_keys_tuple)
+        if allowed_keys not in cls._classes:
+
+            class SpecificStrictDict(cls):
+                __slots__ = allowed_keys_tuple
+
+                def __repr__(self):
+                    return "{%s}" % ", ".join(
+                        '"{!s}": {!r}'.format(k, v) for k, v in self.items()
+                    )
+
+            cls._classes[allowed_keys] = SpecificStrictDict
+        return cls._classes[allowed_keys]
+
+
+class LazyReference(DBRef):
+    __slots__ = ("_cached_doc", "passthrough", "document_type")
+
+    def fetch(self, force=False):
+        if not self._cached_doc or force:
+            self._cached_doc = self.document_type.objects.get(pk=self.pk)
+            if not self._cached_doc:
+                raise DoesNotExist("Trying to dereference unknown document %s" % (self))
+        return self._cached_doc
+
+    @property
+    def pk(self):
+        return self.id
+
+    def __init__(self, document_type, pk, cached_doc=None, passthrough=False):
+        self.document_type = document_type
+        self._cached_doc = cached_doc
+        self.passthrough = passthrough
+        super().__init__(self.document_type._get_collection_name(), pk)
+
+    def __getitem__(self, name):
+        if not self.passthrough:
+            raise KeyError()
+        document = self.fetch()
+        return document[name]
+
+    def __getattr__(self, name):
+        if not object.__getattribute__(self, "passthrough"):
+            raise AttributeError()
+        document = self.fetch()
+        try:
+            return document[name]
+        except KeyError:
+            raise AttributeError()
+
+    def __repr__(self):
+        return "<LazyReference({}, {!r})>".format(self.document_type, self.pk)

mongoengine/base/fields.py

@@ -0,0 +1,682 @@
+import operator
+import warnings
+import weakref
+
+from bson import DBRef, ObjectId, SON
+import pymongo
+
+from mongoengine.base.common import UPDATE_OPERATORS
+from mongoengine.base.datastructures import BaseDict, BaseList, EmbeddedDocumentList
+from mongoengine.common import _import_class
+from mongoengine.errors import DeprecatedError, ValidationError
+
+__all__ = ("BaseField", "ComplexBaseField", "ObjectIdField", "GeoJsonBaseField")
+
+
+class BaseField:
+    """A base class for fields in a MongoDB document. Instances of this class
+    may be added to subclasses of `Document` to define a document's schema.
+
+    .. versionchanged:: 0.5 - added verbose and help text
+    """
+
+    name = None
+    _geo_index = False
+    _auto_gen = False  # Call `generate` to generate a value
+    _auto_dereference = True
+
+    # These track each time a Field instance is created. Used to retain order.
+    # The auto_creation_counter is used for fields that MongoEngine implicitly
+    # creates, creation_counter is used for all user-specified fields.
+    creation_counter = 0
+    auto_creation_counter = -1
+
+    def __init__(
+        self,
+        db_field=None,
+        required=False,
+        default=None,
+        unique=False,
+        unique_with=None,
+        primary_key=False,
+        validation=None,
+        choices=None,
+        null=False,
+        sparse=False,
+        **kwargs
+    ):
+        """
+        :param db_field: The database field to store this field in
+            (defaults to the name of the field)
+        :param required: If the field is required. Whether it has to have a
+            value or not. Defaults to False.
+        :param default: (optional) The default value for this field if no value
+            has been set (or if the value has been unset).  It can be a
+            callable.
+        :param unique: Is the field value unique or not.  Defaults to False.
+        :param unique_with: (optional) The other field this field should be
+            unique with.
+        :param primary_key: Mark this field as the primary key. Defaults to False.
+        :param validation: (optional) A callable to validate the value of the
+            field.  The callable takes the value as parameter and should raise
+            a ValidationError if validation fails
+        :param choices: (optional) The valid choices
+        :param null: (optional) If the field value can be null. If no and there is a default value
+            then the default value is set
+        :param sparse: (optional) `sparse=True` combined with `unique=True` and `required=False`
+            means that uniqueness won't be enforced for `None` values
+        :param **kwargs: (optional) Arbitrary indirection-free metadata for
+            this field can be supplied as additional keyword arguments and
+            accessed as attributes of the field. Must not conflict with any
+            existing attributes. Common metadata includes `verbose_name` and
+            `help_text`.
+        """
+        self.db_field = db_field if not primary_key else "_id"
+
+        self.required = required or primary_key
+        self.default = default
+        self.unique = bool(unique or unique_with)
+        self.unique_with = unique_with
+        self.primary_key = primary_key
+        self.validation = validation
+        self.choices = choices
+        self.null = null
+        self.sparse = sparse
+        self._owner_document = None
+
+        # Make sure db_field is a string (if it's explicitly defined).
+        if self.db_field is not None and not isinstance(self.db_field, str):
+            raise TypeError("db_field should be a string.")
+
+        # Make sure db_field doesn't contain any forbidden characters.
+        if isinstance(self.db_field, str) and (
+            "." in self.db_field
+            or "\0" in self.db_field
+            or self.db_field.startswith("$")
+        ):
+            raise ValueError(
+                'field names cannot contain dots (".") or null characters '
+                '("\\0"), and they must not start with a dollar sign ("$").'
+            )
+
+        # Detect and report conflicts between metadata and base properties.
+        conflicts = set(dir(self)) & set(kwargs)
+        if conflicts:
+            raise TypeError(
+                "%s already has attribute(s): %s"
+                % (self.__class__.__name__, ", ".join(conflicts))
+            )
+
+        # Assign metadata to the instance
+        # This efficient method is available because no __slots__ are defined.
+        self.__dict__.update(kwargs)
+
+        # Adjust the appropriate creation counter, and save our local copy.
+        if self.db_field == "_id":
+            self.creation_counter = BaseField.auto_creation_counter
+            BaseField.auto_creation_counter -= 1
+        else:
+            self.creation_counter = BaseField.creation_counter
+            BaseField.creation_counter += 1
+
+    def __get__(self, instance, owner):
+        """Descriptor for retrieving a value from a field in a document.
+        """
+        if instance is None:
+            # Document class being used rather than a document object
+            return self
+
+        # Get value from document instance if available
+        return instance._data.get(self.name)
+
+    def __set__(self, instance, value):
+        """Descriptor for assigning a value to a field in a document."""
+        # If setting to None and there is a default value provided for this
+        # field, then set the value to the default value.
+        if value is None:
+            if self.null:
+                value = None
+            elif self.default is not None:
+                value = self.default
+                if callable(value):
+                    value = value()
+
+        if instance._initialised:
+            try:
+                value_has_changed = (
+                    self.name not in instance._data
+                    or instance._data[self.name] != value
+                )
+                if value_has_changed:
+                    instance._mark_as_changed(self.name)
+            except Exception:
+                # Some values can't be compared and throw an error when we
+                # attempt to do so (e.g. tz-naive and tz-aware datetimes).
+                # Mark the field as changed in such cases.
+                instance._mark_as_changed(self.name)
+
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        if isinstance(value, EmbeddedDocument):
+            value._instance = weakref.proxy(instance)
+        elif isinstance(value, (list, tuple)):
+            for v in value:
+                if isinstance(v, EmbeddedDocument):
+                    v._instance = weakref.proxy(instance)
+
+        instance._data[self.name] = value
+
+    def error(self, message="", errors=None, field_name=None):
+        """Raise a ValidationError."""
+        field_name = field_name if field_name else self.name
+        raise ValidationError(message, errors=errors, field_name=field_name)
+
+    def to_python(self, value):
+        """Convert a MongoDB-compatible type to a Python type."""
+        return value
+
+    def to_mongo(self, value):
+        """Convert a Python type to a MongoDB-compatible type."""
+        return self.to_python(value)
+
+    def _to_mongo_safe_call(self, value, use_db_field=True, fields=None):
+        """Helper method to call to_mongo with proper inputs."""
+        f_inputs = self.to_mongo.__code__.co_varnames
+        ex_vars = {}
+        if "fields" in f_inputs:
+            ex_vars["fields"] = fields
+
+        if "use_db_field" in f_inputs:
+            ex_vars["use_db_field"] = use_db_field
+
+        return self.to_mongo(value, **ex_vars)
+
+    def prepare_query_value(self, op, value):
+        """Prepare a value that is being used in a query for PyMongo."""
+        if op in UPDATE_OPERATORS:
+            self.validate(value)
+        return value
+
+    def validate(self, value, clean=True):
+        """Perform validation on a value."""
+        pass
+
+    def _validate_choices(self, value):
+        Document = _import_class("Document")
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+
+        choice_list = self.choices
+        if isinstance(next(iter(choice_list)), (list, tuple)):
+            # next(iter) is useful for sets
+            choice_list = [k for k, _ in choice_list]
+
+        # Choices which are other types of Documents
+        if isinstance(value, (Document, EmbeddedDocument)):
+            if not any(isinstance(value, c) for c in choice_list):
+                self.error("Value must be an instance of %s" % (choice_list))
+        # Choices which are types other than Documents
+        else:
+            values = value if isinstance(value, (list, tuple)) else [value]
+            if len(set(values) - set(choice_list)):
+                self.error("Value must be one of %s" % str(choice_list))
+
+    def _validate(self, value, **kwargs):
+        # Check the Choices Constraint
+        if self.choices:
+            self._validate_choices(value)
+
+        # check validation argument
+        if self.validation is not None:
+            if callable(self.validation):
+                try:
+                    # breaking change of 0.18
+                    # Get rid of True/False-type return for the validation method
+                    # in favor of having validation raising a ValidationError
+                    ret = self.validation(value)
+                    if ret is not None:
+                        raise DeprecatedError(
+                            "validation argument for `%s` must not return anything, "
+                            "it should raise a ValidationError if validation fails"
+                            % self.name
+                        )
+                except ValidationError as ex:
+                    self.error(str(ex))
+            else:
+                raise ValueError(
+                    'validation argument for `"%s"` must be a ' "callable." % self.name
+                )
+
+        self.validate(value, **kwargs)
+
+    @property
+    def owner_document(self):
+        return self._owner_document
+
+    def _set_owner_document(self, owner_document):
+        self._owner_document = owner_document
+
+    @owner_document.setter
+    def owner_document(self, owner_document):
+        self._set_owner_document(owner_document)
+
+
+class ComplexBaseField(BaseField):
+    """Handles complex fields, such as lists / dictionaries.
+
+    Allows for nesting of embedded documents inside complex types.
+    Handles the lazy dereferencing of a queryset by lazily dereferencing all
+    items in a list / dict rather than one at a time.
+
+    .. versionadded:: 0.5
+    """
+
+    field = None
+
+    def __get__(self, instance, owner):
+        """Descriptor to automatically dereference references."""
+        if instance is None:
+            # Document class being used rather than a document object
+            return self
+
+        ReferenceField = _import_class("ReferenceField")
+        GenericReferenceField = _import_class("GenericReferenceField")
+        EmbeddedDocumentListField = _import_class("EmbeddedDocumentListField")
+
+        auto_dereference = instance._fields[self.name]._auto_dereference
+
+        dereference = auto_dereference and (
+            self.field is None
+            or isinstance(self.field, (GenericReferenceField, ReferenceField))
+        )
+
+        _dereference = _import_class("DeReference")()
+
+        if (
+            instance._initialised
+            and dereference
+            and instance._data.get(self.name)
+            and not getattr(instance._data[self.name], "_dereferenced", False)
+        ):
+            instance._data[self.name] = _dereference(
+                instance._data.get(self.name),
+                max_depth=1,
+                instance=instance,
+                name=self.name,
+            )
+            if hasattr(instance._data[self.name], "_dereferenced"):
+                instance._data[self.name]._dereferenced = True
+
+        value = super().__get__(instance, owner)
+
+        # Convert lists / values so we can watch for any changes on them
+        if isinstance(value, (list, tuple)):
+            if issubclass(type(self), EmbeddedDocumentListField) and not isinstance(
+                value, EmbeddedDocumentList
+            ):
+                value = EmbeddedDocumentList(value, instance, self.name)
+            elif not isinstance(value, BaseList):
+                value = BaseList(value, instance, self.name)
+            instance._data[self.name] = value
+        elif isinstance(value, dict) and not isinstance(value, BaseDict):
+            value = BaseDict(value, instance, self.name)
+            instance._data[self.name] = value
+
+        if (
+            auto_dereference
+            and instance._initialised
+            and isinstance(value, (BaseList, BaseDict))
+            and not value._dereferenced
+        ):
+            value = _dereference(value, max_depth=1, instance=instance, name=self.name)
+            value._dereferenced = True
+            instance._data[self.name] = value
+
+        return value
+
+    def to_python(self, value):
+        """Convert a MongoDB-compatible type to a Python type."""
+        if isinstance(value, str):
+            return value
+
+        if hasattr(value, "to_python"):
+            return value.to_python()
+
+        BaseDocument = _import_class("BaseDocument")
+        if isinstance(value, BaseDocument):
+            # Something is wrong, return the value as it is
+            return value
+
+        is_list = False
+        if not hasattr(value, "items"):
+            try:
+                is_list = True
+                value = {idx: v for idx, v in enumerate(value)}
+            except TypeError:  # Not iterable return the value
+                return value
+
+        if self.field:
+            self.field._auto_dereference = self._auto_dereference
+            value_dict = {
+                key: self.field.to_python(item) for key, item in value.items()
+            }
+        else:
+            Document = _import_class("Document")
+            value_dict = {}
+            for k, v in value.items():
+                if isinstance(v, Document):
+                    # We need the id from the saved object to create the DBRef
+                    if v.pk is None:
+                        self.error(
+                            "You can only reference documents once they"
+                            " have been saved to the database"
+                        )
+                    collection = v._get_collection_name()
+                    value_dict[k] = DBRef(collection, v.pk)
+                elif hasattr(v, "to_python"):
+                    value_dict[k] = v.to_python()
+                else:
+                    value_dict[k] = self.to_python(v)
+
+        if is_list:  # Convert back to a list
+            return [
+                v for _, v in sorted(value_dict.items(), key=operator.itemgetter(0))
+            ]
+        return value_dict
+
+    def to_mongo(self, value, use_db_field=True, fields=None):
+        """Convert a Python type to a MongoDB-compatible type."""
+        Document = _import_class("Document")
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        GenericReferenceField = _import_class("GenericReferenceField")
+
+        if isinstance(value, str):
+            return value
+
+        if hasattr(value, "to_mongo"):
+            if isinstance(value, Document):
+                return GenericReferenceField().to_mongo(value)
+            cls = value.__class__
+            val = value.to_mongo(use_db_field, fields)
+            # If it's a document that is not inherited add _cls
+            if isinstance(value, EmbeddedDocument):
+                val["_cls"] = cls.__name__
+            return val
+
+        is_list = False
+        if not hasattr(value, "items"):
+            try:
+                is_list = True
+                value = {k: v for k, v in enumerate(value)}
+            except TypeError:  # Not iterable return the value
+                return value
+
+        if self.field:
+            value_dict = {
+                key: self.field._to_mongo_safe_call(item, use_db_field, fields)
+                for key, item in value.items()
+            }
+        else:
+            value_dict = {}
+            for k, v in value.items():
+                if isinstance(v, Document):
+                    # We need the id from the saved object to create the DBRef
+                    if v.pk is None:
+                        self.error(
+                            "You can only reference documents once they"
+                            " have been saved to the database"
+                        )
+
+                    # If its a document that is not inheritable it won't have
+                    # any _cls data so make it a generic reference allows
+                    # us to dereference
+                    meta = getattr(v, "_meta", {})
+                    allow_inheritance = meta.get("allow_inheritance")
+                    if not allow_inheritance and not self.field:
+                        value_dict[k] = GenericReferenceField().to_mongo(v)
+                    else:
+                        collection = v._get_collection_name()
+                        value_dict[k] = DBRef(collection, v.pk)
+                elif hasattr(v, "to_mongo"):
+                    cls = v.__class__
+                    val = v.to_mongo(use_db_field, fields)
+                    # If it's a document that is not inherited add _cls
+                    if isinstance(v, (Document, EmbeddedDocument)):
+                        val["_cls"] = cls.__name__
+                    value_dict[k] = val
+                else:
+                    value_dict[k] = self.to_mongo(v, use_db_field, fields)
+
+        if is_list:  # Convert back to a list
+            return [
+                v for _, v in sorted(value_dict.items(), key=operator.itemgetter(0))
+            ]
+        return value_dict
+
+    def validate(self, value):
+        """If field is provided ensure the value is valid."""
+        errors = {}
+        if self.field:
+            if hasattr(value, "items"):
+                sequence = value.items()
+            else:
+                sequence = enumerate(value)
+            for k, v in sequence:
+                try:
+                    self.field._validate(v)
+                except ValidationError as error:
+                    errors[k] = error.errors or error
+                except (ValueError, AssertionError) as error:
+                    errors[k] = error
+
+            if errors:
+                field_class = self.field.__class__.__name__
+                self.error(
+                    "Invalid {} item ({})".format(field_class, value), errors=errors
+                )
+        # Don't allow empty values if required
+        if self.required and not value:
+            self.error("Field is required and cannot be empty")
+
+    def prepare_query_value(self, op, value):
+        return self.to_mongo(value)
+
+    def lookup_member(self, member_name):
+        if self.field:
+            return self.field.lookup_member(member_name)
+        return None
+
+    def _set_owner_document(self, owner_document):
+        if self.field:
+            self.field.owner_document = owner_document
+        self._owner_document = owner_document
+
+
+class ObjectIdField(BaseField):
+    """A field wrapper around MongoDB's ObjectIds."""
+
+    def to_python(self, value):
+        try:
+            if not isinstance(value, ObjectId):
+                value = ObjectId(value)
+        except Exception:
+            pass
+        return value
+
+    def to_mongo(self, value):
+        if not isinstance(value, ObjectId):
+            try:
+                return ObjectId(str(value))
+            except Exception as e:
+                self.error(str(e))
+        return value
+
+    def prepare_query_value(self, op, value):
+        return self.to_mongo(value)
+
+    def validate(self, value):
+        try:
+            ObjectId(str(value))
+        except Exception:
+            self.error("Invalid ObjectID")
+
+
+class GeoJsonBaseField(BaseField):
+    """A geo json field storing a geojson style object.
+
+    .. versionadded:: 0.8
+    """
+
+    _geo_index = pymongo.GEOSPHERE
+    _type = "GeoBase"
+
+    def __init__(self, auto_index=True, *args, **kwargs):
+        """
+        :param bool auto_index: Automatically create a '2dsphere' index.\
+            Defaults to `True`.
+        """
+        self._name = "%sField" % self._type
+        if not auto_index:
+            self._geo_index = False
+        super().__init__(*args, **kwargs)
+
+    def validate(self, value):
+        """Validate the GeoJson object based on its type."""
+        if isinstance(value, dict):
+            if set(value.keys()) == {"type", "coordinates"}:
+                if value["type"] != self._type:
+                    self.error('{} type must be "{}"'.format(self._name, self._type))
+                return self.validate(value["coordinates"])
+            else:
+                self.error(
+                    "%s can only accept a valid GeoJson dictionary"
+                    " or lists of (x, y)" % self._name
+                )
+                return
+        elif not isinstance(value, (list, tuple)):
+            self.error("%s can only accept lists of [x, y]" % self._name)
+            return
+
+        validate = getattr(self, "_validate_%s" % self._type.lower())
+        error = validate(value)
+        if error:
+            self.error(error)
+
+    def _validate_polygon(self, value, top_level=True):
+        if not isinstance(value, (list, tuple)):
+            return "Polygons must contain list of linestrings"
+
+        # Quick and dirty validator
+        try:
+            value[0][0][0]
+        except (TypeError, IndexError):
+            return "Invalid Polygon must contain at least one valid linestring"
+
+        errors = []
+        for val in value:
+            error = self._validate_linestring(val, False)
+            if not error and val[0] != val[-1]:
+                error = "LineStrings must start and end at the same point"
+            if error and error not in errors:
+                errors.append(error)
+        if errors:
+            if top_level:
+                return "Invalid Polygon:\n%s" % ", ".join(errors)
+            else:
+                return "%s" % ", ".join(errors)
+
+    def _validate_linestring(self, value, top_level=True):
+        """Validate a linestring."""
+        if not isinstance(value, (list, tuple)):
+            return "LineStrings must contain list of coordinate pairs"
+
+        # Quick and dirty validator
+        try:
+            value[0][0]
+        except (TypeError, IndexError):
+            return "Invalid LineString must contain at least one valid point"
+
+        errors = []
+        for val in value:
+            error = self._validate_point(val)
+            if error and error not in errors:
+                errors.append(error)
+        if errors:
+            if top_level:
+                return "Invalid LineString:\n%s" % ", ".join(errors)
+            else:
+                return "%s" % ", ".join(errors)
+
+    def _validate_point(self, value):
+        """Validate each set of coords"""
+        if not isinstance(value, (list, tuple)):
+            return "Points must be a list of coordinate pairs"
+        elif not len(value) == 2:
+            return "Value (%s) must be a two-dimensional point" % repr(value)
+        elif not isinstance(value[0], (float, int)) or not isinstance(
+            value[1], (float, int)
+        ):
+            return "Both values (%s) in point must be float or int" % repr(value)
+
+    def _validate_multipoint(self, value):
+        if not isinstance(value, (list, tuple)):
+            return "MultiPoint must be a list of Point"
+
+        # Quick and dirty validator
+        try:
+            value[0][0]
+        except (TypeError, IndexError):
+            return "Invalid MultiPoint must contain at least one valid point"
+
+        errors = []
+        for point in value:
+            error = self._validate_point(point)
+            if error and error not in errors:
+                errors.append(error)
+
+        if errors:
+            return "%s" % ", ".join(errors)
+
+    def _validate_multilinestring(self, value, top_level=True):
+        if not isinstance(value, (list, tuple)):
+            return "MultiLineString must be a list of LineString"
+
+        # Quick and dirty validator
+        try:
+            value[0][0][0]
+        except (TypeError, IndexError):
+            return "Invalid MultiLineString must contain at least one valid linestring"
+
+        errors = []
+        for linestring in value:
+            error = self._validate_linestring(linestring, False)
+            if error and error not in errors:
+                errors.append(error)
+
+        if errors:
+            if top_level:
+                return "Invalid MultiLineString:\n%s" % ", ".join(errors)
+            else:
+                return "%s" % ", ".join(errors)
+
+    def _validate_multipolygon(self, value):
+        if not isinstance(value, (list, tuple)):
+            return "MultiPolygon must be a list of Polygon"
+
+        # Quick and dirty validator
+        try:
+            value[0][0][0][0]
+        except (TypeError, IndexError):
+            return "Invalid MultiPolygon must contain at least one valid Polygon"
+
+        errors = []
+        for polygon in value:
+            error = self._validate_polygon(polygon, False)
+            if error and error not in errors:
+                errors.append(error)
+
+        if errors:
+            return "Invalid MultiPolygon:\n%s" % ", ".join(errors)
+
+    def to_mongo(self, value):
+        if isinstance(value, dict):
+            return value
+        return SON([("type", self._type), ("coordinates", value)])

mongoengine/base/metaclasses.py

@@ -0,0 +1,466 @@
+import itertools
+import warnings
+
+from mongoengine.base.common import _document_registry
+from mongoengine.base.fields import BaseField, ComplexBaseField, ObjectIdField
+from mongoengine.common import _import_class
+from mongoengine.errors import InvalidDocumentError
+from mongoengine.queryset import (
+    DO_NOTHING,
+    DoesNotExist,
+    MultipleObjectsReturned,
+    QuerySetManager,
+)
+
+
+__all__ = ("DocumentMetaclass", "TopLevelDocumentMetaclass")
+
+
+class DocumentMetaclass(type):
+    """Metaclass for all documents."""
+
+    # TODO lower complexity of this method
+    def __new__(mcs, name, bases, attrs):
+        flattened_bases = mcs._get_bases(bases)
+        super_new = super().__new__
+
+        # If a base class just call super
+        metaclass = attrs.get("my_metaclass")
+        if metaclass and issubclass(metaclass, DocumentMetaclass):
+            return super_new(mcs, name, bases, attrs)
+
+        attrs["_is_document"] = attrs.get("_is_document", False)
+        attrs["_cached_reference_fields"] = []
+
+        # EmbeddedDocuments could have meta data for inheritance
+        if "meta" in attrs:
+            attrs["_meta"] = attrs.pop("meta")
+
+        # EmbeddedDocuments should inherit meta data
+        if "_meta" not in attrs:
+            meta = MetaDict()
+            for base in flattened_bases[::-1]:
+                # Add any mixin metadata from plain objects
+                if hasattr(base, "meta"):
+                    meta.merge(base.meta)
+                elif hasattr(base, "_meta"):
+                    meta.merge(base._meta)
+            attrs["_meta"] = meta
+            attrs["_meta"][
+                "abstract"
+            ] = False  # 789: EmbeddedDocument shouldn't inherit abstract
+
+        # If allow_inheritance is True, add a "_cls" string field to the attrs
+        if attrs["_meta"].get("allow_inheritance"):
+            StringField = _import_class("StringField")
+            attrs["_cls"] = StringField()
+
+        # Handle document Fields
+
+        # Merge all fields from subclasses
+        doc_fields = {}
+        for base in flattened_bases[::-1]:
+            if hasattr(base, "_fields"):
+                doc_fields.update(base._fields)
+
+            # Standard object mixin - merge in any Fields
+            if not hasattr(base, "_meta"):
+                base_fields = {}
+                for attr_name, attr_value in base.__dict__.items():
+                    if not isinstance(attr_value, BaseField):
+                        continue
+                    attr_value.name = attr_name
+                    if not attr_value.db_field:
+                        attr_value.db_field = attr_name
+                    base_fields[attr_name] = attr_value
+
+                doc_fields.update(base_fields)
+
+        # Discover any document fields
+        field_names = {}
+        for attr_name, attr_value in attrs.items():
+            if not isinstance(attr_value, BaseField):
+                continue
+            attr_value.name = attr_name
+            if not attr_value.db_field:
+                attr_value.db_field = attr_name
+            doc_fields[attr_name] = attr_value
+
+            # Count names to ensure no db_field redefinitions
+            field_names[attr_value.db_field] = (
+                field_names.get(attr_value.db_field, 0) + 1
+            )
+
+        # Ensure no duplicate db_fields
+        duplicate_db_fields = [k for k, v in field_names.items() if v > 1]
+        if duplicate_db_fields:
+            msg = "Multiple db_fields defined for: %s " % ", ".join(duplicate_db_fields)
+            raise InvalidDocumentError(msg)
+
+        # Set _fields and db_field maps
+        attrs["_fields"] = doc_fields
+        attrs["_db_field_map"] = {
+            k: getattr(v, "db_field", k) for k, v in doc_fields.items()
+        }
+        attrs["_reverse_db_field_map"] = {
+            v: k for k, v in attrs["_db_field_map"].items()
+        }
+
+        attrs["_fields_ordered"] = tuple(
+            i[1]
+            for i in sorted((v.creation_counter, v.name) for v in doc_fields.values())
+        )
+
+        #
+        # Set document hierarchy
+        #
+        superclasses = ()
+        class_name = [name]
+        for base in flattened_bases:
+            if not getattr(base, "_is_base_cls", True) and not getattr(
+                base, "_meta", {}
+            ).get("abstract", True):
+                # Collate hierarchy for _cls and _subclasses
+                class_name.append(base.__name__)
+
+            if hasattr(base, "_meta"):
+                # Warn if allow_inheritance isn't set and prevent
+                # inheritance of classes where inheritance is set to False
+                allow_inheritance = base._meta.get("allow_inheritance")
+                if not allow_inheritance and not base._meta.get("abstract"):
+                    raise ValueError(
+                        "Document %s may not be subclassed. "
+                        'To enable inheritance, use the "allow_inheritance" meta attribute.'
+                        % base.__name__
+                    )
+
+        # Get superclasses from last base superclass
+        document_bases = [b for b in flattened_bases if hasattr(b, "_class_name")]
+        if document_bases:
+            superclasses = document_bases[0]._superclasses
+            superclasses += (document_bases[0]._class_name,)
+
+        _cls = ".".join(reversed(class_name))
+        attrs["_class_name"] = _cls
+        attrs["_superclasses"] = superclasses
+        attrs["_subclasses"] = (_cls,)
+        attrs["_types"] = attrs["_subclasses"]  # TODO depreciate _types
+
+        # Create the new_class
+        new_class = super_new(mcs, name, bases, attrs)
+
+        # Set _subclasses
+        for base in document_bases:
+            if _cls not in base._subclasses:
+                base._subclasses += (_cls,)
+            base._types = base._subclasses  # TODO depreciate _types
+
+        (
+            Document,
+            EmbeddedDocument,
+            DictField,
+            CachedReferenceField,
+        ) = mcs._import_classes()
+
+        if issubclass(new_class, Document):
+            new_class._collection = None
+
+        # Add class to the _document_registry
+        _document_registry[new_class._class_name] = new_class
+
+        # Handle delete rules
+        for field in new_class._fields.values():
+            f = field
+            if f.owner_document is None:
+                f.owner_document = new_class
+            delete_rule = getattr(f, "reverse_delete_rule", DO_NOTHING)
+            if isinstance(f, CachedReferenceField):
+
+                if issubclass(new_class, EmbeddedDocument):
+                    raise InvalidDocumentError(
+                        "CachedReferenceFields is not allowed in EmbeddedDocuments"
+                    )
+
+                if f.auto_sync:
+                    f.start_listener()
+
+                f.document_type._cached_reference_fields.append(f)
+
+            if isinstance(f, ComplexBaseField) and hasattr(f, "field"):
+                delete_rule = getattr(f.field, "reverse_delete_rule", DO_NOTHING)
+                if isinstance(f, DictField) and delete_rule != DO_NOTHING:
+                    msg = (
+                        "Reverse delete rules are not supported "
+                        "for %s (field: %s)" % (field.__class__.__name__, field.name)
+                    )
+                    raise InvalidDocumentError(msg)
+
+                f = field.field
+
+            if delete_rule != DO_NOTHING:
+                if issubclass(new_class, EmbeddedDocument):
+                    msg = (
+                        "Reverse delete rules are not supported for "
+                        "EmbeddedDocuments (field: %s)" % field.name
+                    )
+                    raise InvalidDocumentError(msg)
+                f.document_type.register_delete_rule(new_class, field.name, delete_rule)
+
+            if (
+                field.name
+                and hasattr(Document, field.name)
+                and EmbeddedDocument not in new_class.mro()
+            ):
+                msg = "%s is a document method and not a valid field name" % field.name
+                raise InvalidDocumentError(msg)
+
+        return new_class
+
+    @classmethod
+    def _get_bases(mcs, bases):
+        if isinstance(bases, BasesTuple):
+            return bases
+        seen = []
+        bases = mcs.__get_bases(bases)
+        unique_bases = (b for b in bases if not (b in seen or seen.append(b)))
+        return BasesTuple(unique_bases)
+
+    @classmethod
+    def __get_bases(mcs, bases):
+        for base in bases:
+            if base is object:
+                continue
+            yield base
+            yield from mcs.__get_bases(base.__bases__)
+
+    @classmethod
+    def _import_classes(mcs):
+        Document = _import_class("Document")
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        DictField = _import_class("DictField")
+        CachedReferenceField = _import_class("CachedReferenceField")
+        return Document, EmbeddedDocument, DictField, CachedReferenceField
+
+
+class TopLevelDocumentMetaclass(DocumentMetaclass):
+    """Metaclass for top-level documents (i.e. documents that have their own
+    collection in the database.
+    """
+
+    def __new__(mcs, name, bases, attrs):
+        flattened_bases = mcs._get_bases(bases)
+        super_new = super().__new__
+
+        # Set default _meta data if base class, otherwise get user defined meta
+        if attrs.get("my_metaclass") == TopLevelDocumentMetaclass:
+            # defaults
+            attrs["_meta"] = {
+                "abstract": True,
+                "max_documents": None,
+                "max_size": None,
+                "ordering": [],  # default ordering applied at runtime
+                "indexes": [],  # indexes to be ensured at runtime
+                "id_field": None,
+                "index_background": False,
+                "index_opts": None,
+                "delete_rules": None,
+                # allow_inheritance can be True, False, and None. True means
+                # "allow inheritance", False means "don't allow inheritance",
+                # None means "do whatever your parent does, or don't allow
+                # inheritance if you're a top-level class".
+                "allow_inheritance": None,
+            }
+            attrs["_is_base_cls"] = True
+            attrs["_meta"].update(attrs.get("meta", {}))
+        else:
+            attrs["_meta"] = attrs.get("meta", {})
+            # Explicitly set abstract to false unless set
+            attrs["_meta"]["abstract"] = attrs["_meta"].get("abstract", False)
+            attrs["_is_base_cls"] = False
+
+        # Set flag marking as document class - as opposed to an object mixin
+        attrs["_is_document"] = True
+
+        # Ensure queryset_class is inherited
+        if "objects" in attrs:
+            manager = attrs["objects"]
+            if hasattr(manager, "queryset_class"):
+                attrs["_meta"]["queryset_class"] = manager.queryset_class
+
+        # Clean up top level meta
+        if "meta" in attrs:
+            del attrs["meta"]
+
+        # Find the parent document class
+        parent_doc_cls = [
+            b for b in flattened_bases if b.__class__ == TopLevelDocumentMetaclass
+        ]
+        parent_doc_cls = None if not parent_doc_cls else parent_doc_cls[0]
+
+        # Prevent classes setting collection different to their parents
+        # If parent wasn't an abstract class
+        if (
+            parent_doc_cls
+            and "collection" in attrs.get("_meta", {})
+            and not parent_doc_cls._meta.get("abstract", True)
+        ):
+            msg = "Trying to set a collection on a subclass (%s)" % name
+            warnings.warn(msg, SyntaxWarning)
+            del attrs["_meta"]["collection"]
+
+        # Ensure abstract documents have abstract bases
+        if attrs.get("_is_base_cls") or attrs["_meta"].get("abstract"):
+            if parent_doc_cls and not parent_doc_cls._meta.get("abstract", False):
+                msg = "Abstract document cannot have non-abstract base"
+                raise ValueError(msg)
+            return super_new(mcs, name, bases, attrs)
+
+        # Merge base class metas.
+        # Uses a special MetaDict that handles various merging rules
+        meta = MetaDict()
+        for base in flattened_bases[::-1]:
+            # Add any mixin metadata from plain objects
+            if hasattr(base, "meta"):
+                meta.merge(base.meta)
+            elif hasattr(base, "_meta"):
+                meta.merge(base._meta)
+
+            # Set collection in the meta if its callable
+            if getattr(base, "_is_document", False) and not base._meta.get("abstract"):
+                collection = meta.get("collection", None)
+                if callable(collection):
+                    meta["collection"] = collection(base)
+
+        meta.merge(attrs.get("_meta", {}))  # Top level meta
+
+        # Only simple classes (i.e. direct subclasses of Document) may set
+        # allow_inheritance to False. If the base Document allows inheritance,
+        # none of its subclasses can override allow_inheritance to False.
+        simple_class = all(
+            [b._meta.get("abstract") for b in flattened_bases if hasattr(b, "_meta")]
+        )
+        if (
+            not simple_class
+            and meta["allow_inheritance"] is False
+            and not meta["abstract"]
+        ):
+            raise ValueError(
+                "Only direct subclasses of Document may set "
+                '"allow_inheritance" to False'
+            )
+
+        # Set default collection name
+        if "collection" not in meta:
+            meta["collection"] = (
+                "".join("_%s" % c if c.isupper() else c for c in name)
+                .strip("_")
+                .lower()
+            )
+        attrs["_meta"] = meta
+
+        # Call super and get the new class
+        new_class = super_new(mcs, name, bases, attrs)
+
+        meta = new_class._meta
+
+        # Set index specifications
+        meta["index_specs"] = new_class._build_index_specs(meta["indexes"])
+
+        # If collection is a callable - call it and set the value
+        collection = meta.get("collection")
+        if callable(collection):
+            new_class._meta["collection"] = collection(new_class)
+
+        # Provide a default queryset unless exists or one has been set
+        if "objects" not in dir(new_class):
+            new_class.objects = QuerySetManager()
+
+        # Validate the fields and set primary key if needed
+        for field_name, field in new_class._fields.items():
+            if field.primary_key:
+                # Ensure only one primary key is set
+                current_pk = new_class._meta.get("id_field")
+                if current_pk and current_pk != field_name:
+                    raise ValueError("Cannot override primary key field")
+
+                # Set primary key
+                if not current_pk:
+                    new_class._meta["id_field"] = field_name
+                    new_class.id = field
+
+        # If the document doesn't explicitly define a primary key field, create
+        # one. Make it an ObjectIdField and give it a non-clashing name ("id"
+        # by default, but can be different if that one's taken).
+        if not new_class._meta.get("id_field"):
+            id_name, id_db_name = mcs.get_auto_id_names(new_class)
+            new_class._meta["id_field"] = id_name
+            new_class._fields[id_name] = ObjectIdField(db_field=id_db_name)
+            new_class._fields[id_name].name = id_name
+            new_class.id = new_class._fields[id_name]
+            new_class._db_field_map[id_name] = id_db_name
+            new_class._reverse_db_field_map[id_db_name] = id_name
+
+            # Prepend the ID field to _fields_ordered (so that it's *always*
+            # the first field).
+            new_class._fields_ordered = (id_name,) + new_class._fields_ordered
+
+        # Merge in exceptions with parent hierarchy.
+        exceptions_to_merge = (DoesNotExist, MultipleObjectsReturned)
+        module = attrs.get("__module__")
+        for exc in exceptions_to_merge:
+            name = exc.__name__
+            parents = tuple(
+                getattr(base, name) for base in flattened_bases if hasattr(base, name)
+            ) or (exc,)
+
+            # Create a new exception and set it as an attribute on the new
+            # class.
+            exception = type(name, parents, {"__module__": module})
+            setattr(new_class, name, exception)
+
+        return new_class
+
+    @classmethod
+    def get_auto_id_names(mcs, new_class):
+        """Find a name for the automatic ID field for the given new class.
+
+        Return a two-element tuple where the first item is the field name (i.e.
+        the attribute name on the object) and the second element is the DB
+        field name (i.e. the name of the key stored in MongoDB).
+
+        Defaults to ('id', '_id'), or generates a non-clashing name in the form
+        of ('auto_id_X', '_auto_id_X') if the default name is already taken.
+        """
+        id_name, id_db_name = ("id", "_id")
+        existing_fields = {field_name for field_name in new_class._fields}
+        existing_db_fields = {v.db_field for v in new_class._fields.values()}
+        if id_name not in existing_fields and id_db_name not in existing_db_fields:
+            return id_name, id_db_name
+
+        id_basename, id_db_basename, i = ("auto_id", "_auto_id", 0)
+        for i in itertools.count():
+            id_name = "{}_{}".format(id_basename, i)
+            id_db_name = "{}_{}".format(id_db_basename, i)
+            if id_name not in existing_fields and id_db_name not in existing_db_fields:
+                return id_name, id_db_name
+
+
+class MetaDict(dict):
+    """Custom dictionary for meta classes.
+    Handles the merging of set indexes
+    """
+
+    _merge_options = ("indexes",)
+
+    def merge(self, new_options):
+        for k, v in new_options.items():
+            if k in self._merge_options:
+                self[k] = self.get(k, []) + v
+            else:
+                self[k] = v
+
+
+class BasesTuple(tuple):
+    """Special class to handle introspection of bases tuple in __new__"""
+
+    pass

mongoengine/base/utils.py

@@ -0,0 +1,22 @@
+import re
+
+
+class LazyRegexCompiler:
+    """Descriptor to allow lazy compilation of regex"""
+
+    def __init__(self, pattern, flags=0):
+        self._pattern = pattern
+        self._flags = flags
+        self._compiled_regex = None
+
+    @property
+    def compiled_regex(self):
+        if self._compiled_regex is None:
+            self._compiled_regex = re.compile(self._pattern, self._flags)
+        return self._compiled_regex
+
+    def __get__(self, instance, owner):
+        return self.compiled_regex
+
+    def __set__(self, instance, value):
+        raise AttributeError("Can not set attribute LazyRegexCompiler")

mongoengine/common.py

@@ -0,0 +1,64 @@
+_class_registry_cache = {}
+_field_list_cache = []
+
+
+def _import_class(cls_name):
+    """Cache mechanism for imports.
+
+    Due to complications of circular imports mongoengine needs to do lots of
+    inline imports in functions.  This is inefficient as classes are
+    imported repeated throughout the mongoengine code.  This is
+    compounded by some recursive functions requiring inline imports.
+
+    :mod:`mongoengine.common` provides a single point to import all these
+    classes.  Circular imports aren't an issue as it dynamically imports the
+    class when first needed.  Subsequent calls to the
+    :func:`~mongoengine.common._import_class` can then directly retrieve the
+    class from the :data:`mongoengine.common._class_registry_cache`.
+    """
+    if cls_name in _class_registry_cache:
+        return _class_registry_cache.get(cls_name)
+
+    doc_classes = (
+        "Document",
+        "DynamicEmbeddedDocument",
+        "EmbeddedDocument",
+        "MapReduceDocument",
+    )
+
+    # Field Classes
+    if not _field_list_cache:
+        from mongoengine.fields import __all__ as fields
+
+        _field_list_cache.extend(fields)
+        from mongoengine.base.fields import __all__ as fields
+
+        _field_list_cache.extend(fields)
+
+    field_classes = _field_list_cache
+
+    deref_classes = ("DeReference",)
+
+    if cls_name == "BaseDocument":
+        from mongoengine.base import document as module
+
+        import_classes = ["BaseDocument"]
+    elif cls_name in doc_classes:
+        from mongoengine import document as module
+
+        import_classes = doc_classes
+    elif cls_name in field_classes:
+        from mongoengine import fields as module
+
+        import_classes = field_classes
+    elif cls_name in deref_classes:
+        from mongoengine import dereference as module
+
+        import_classes = deref_classes
+    else:
+        raise ValueError("No import set for: %s" % cls_name)
+
+    for cls in import_classes:
+        _class_registry_cache[cls] = getattr(module, cls)
+
+    return _class_registry_cache.get(cls_name)

mongoengine/connection.py

@@ -1,62 +1,410 @@
-from pymongo import Connection
+from pymongo import MongoClient, ReadPreference, uri_parser
+from pymongo.database import _check_name
+
+__all__ = [
+    "DEFAULT_CONNECTION_NAME",
+    "DEFAULT_DATABASE_NAME",
+    "ConnectionFailure",
+    "connect",
+    "disconnect",
+    "disconnect_all",
+    "get_connection",
+    "get_db",
+    "register_connection",
+]
 
 
-__all__ = ['ConnectionError', 'connect']
+DEFAULT_CONNECTION_NAME = "default"
+DEFAULT_DATABASE_NAME = "test"
+DEFAULT_HOST = "localhost"
+DEFAULT_PORT = 27017
+
+_connection_settings = {}
+_connections = {}
+_dbs = {}
+
+READ_PREFERENCE = ReadPreference.PRIMARY
 
 
-_connection_settings = {
-    'host': 'localhost',
-    'port': 27017,
-    'pool_size': 1,
-}
-_connection = None
+class ConnectionFailure(Exception):
+    """Error raised when the database connection can't be established or
+    when a connection with a requested alias can't be retrieved.
+    """
 
-_db_name = None
-_db_username = None
-_db_password = None
-_db = None
-
-
-class ConnectionError(Exception):
     pass
 
 
-def _get_connection():
-    global _connection
-    # Connect to the database if not already connected
-    if _connection is None:
-        try:
-            _connection = Connection(**_connection_settings)
-        except:
-            raise ConnectionError('Cannot connect to the database')
-    return _connection
-
-def _get_db():
-    global _db, _connection
-    # Connect if not already connected
-    if _connection is None:
-        _connection = _get_connection()
-
-    if _db is None:
-        # _db_name will be None if the user hasn't called connect()
-        if _db_name is None:
-            raise ConnectionError('Not connected to the database')
-
-        # Get DB from current connection and authenticate if necessary
-        _db = _connection[_db_name]
-        if _db_username and _db_password:
-            _db.authenticate(_db_username, _db_password)
-
-    return _db
-
-def connect(db, username=None, password=None, **kwargs):
-    """Connect to the database specified by the 'db' argument. Connection 
-    settings may be provided here as well if the database is not running on
-    the default port on localhost. If authentication is needed, provide
-    username and password arguments as well.
+def _check_db_name(name):
+    """Check if a database name is valid.
+    This functionality is copied from pymongo Database class constructor.
     """
-    global _connection_settings, _db_name, _db_username, _db_password
-    _connection_settings.update(kwargs)
-    _db_name = db
-    _db_username = username
-    _db_password = password
+    if not isinstance(name, str):
+        raise TypeError("name must be an instance of %s" % str)
+    elif name != "$external":
+        _check_name(name)
+
+
+def _get_connection_settings(
+    db=None,
+    name=None,
+    host=None,
+    port=None,
+    read_preference=READ_PREFERENCE,
+    username=None,
+    password=None,
+    authentication_source=None,
+    authentication_mechanism=None,
+    **kwargs
+):
+    """Get the connection settings as a dict
+
+    : param db: the name of the database to use, for compatibility with connect
+    : param name: the name of the specific database to use
+    : param host: the host name of the: program: `mongod` instance to connect to
+    : param port: the port that the: program: `mongod` instance is running on
+    : param read_preference: The read preference for the collection
+    : param username: username to authenticate with
+    : param password: password to authenticate with
+    : param authentication_source: database to authenticate against
+    : param authentication_mechanism: database authentication mechanisms.
+        By default, use SCRAM-SHA-1 with MongoDB 3.0 and later,
+        MONGODB-CR (MongoDB Challenge Response protocol) for older servers.
+    : param is_mock: explicitly use mongomock for this connection
+        (can also be done by using `mongomock: // ` as db host prefix)
+    : param kwargs: ad-hoc parameters to be passed into the pymongo driver,
+        for example maxpoolsize, tz_aware, etc. See the documentation
+        for pymongo's `MongoClient` for a full list.
+
+    .. versionchanged:: 0.10.6 - added mongomock support
+    """
+    conn_settings = {
+        "name": name or db or DEFAULT_DATABASE_NAME,
+        "host": host or DEFAULT_HOST,
+        "port": port or DEFAULT_PORT,
+        "read_preference": read_preference,
+        "username": username,
+        "password": password,
+        "authentication_source": authentication_source,
+        "authentication_mechanism": authentication_mechanism,
+    }
+
+    _check_db_name(conn_settings["name"])
+    conn_host = conn_settings["host"]
+
+    # Host can be a list or a string, so if string, force to a list.
+    if isinstance(conn_host, str):
+        conn_host = [conn_host]
+
+    resolved_hosts = []
+    for entity in conn_host:
+
+        # Handle Mongomock
+        if entity.startswith("mongomock://"):
+            conn_settings["is_mock"] = True
+            # `mongomock://` is not a valid url prefix and must be replaced by `mongodb://`
+            new_entity = entity.replace("mongomock://", "mongodb://", 1)
+            resolved_hosts.append(new_entity)
+
+            uri_dict = uri_parser.parse_uri(new_entity)
+
+            database = uri_dict.get("database")
+            if database:
+                conn_settings["name"] = database
+
+        # Handle URI style connections, only updating connection params which
+        # were explicitly specified in the URI.
+        elif "://" in entity:
+            uri_dict = uri_parser.parse_uri(entity)
+            resolved_hosts.append(entity)
+
+            database = uri_dict.get("database")
+            if database:
+                conn_settings["name"] = database
+
+            for param in ("read_preference", "username", "password"):
+                if uri_dict.get(param):
+                    conn_settings[param] = uri_dict[param]
+
+            uri_options = uri_dict["options"]
+            if "replicaset" in uri_options:
+                conn_settings["replicaSet"] = uri_options["replicaset"]
+            if "authsource" in uri_options:
+                conn_settings["authentication_source"] = uri_options["authsource"]
+            if "authmechanism" in uri_options:
+                conn_settings["authentication_mechanism"] = uri_options["authmechanism"]
+            if "readpreference" in uri_options:
+                read_preferences = (
+                    ReadPreference.NEAREST,
+                    ReadPreference.PRIMARY,
+                    ReadPreference.PRIMARY_PREFERRED,
+                    ReadPreference.SECONDARY,
+                    ReadPreference.SECONDARY_PREFERRED,
+                )
+
+                # Starting with PyMongo v3.5, the "readpreference" option is
+                # returned as a string (e.g. "secondaryPreferred") and not an
+                # int (e.g. 3).
+                # TODO simplify the code below once we drop support for
+                # PyMongo v3.4.
+                read_pf_mode = uri_options["readpreference"]
+                if isinstance(read_pf_mode, str):
+                    read_pf_mode = read_pf_mode.lower()
+                for preference in read_preferences:
+                    if (
+                        preference.name.lower() == read_pf_mode
+                        or preference.mode == read_pf_mode
+                    ):
+                        conn_settings["read_preference"] = preference
+                        break
+        else:
+            resolved_hosts.append(entity)
+    conn_settings["host"] = resolved_hosts
+
+    # Deprecated parameters that should not be passed on
+    kwargs.pop("slaves", None)
+    kwargs.pop("is_slave", None)
+
+    conn_settings.update(kwargs)
+    return conn_settings
+
+
+def register_connection(
+    alias,
+    db=None,
+    name=None,
+    host=None,
+    port=None,
+    read_preference=READ_PREFERENCE,
+    username=None,
+    password=None,
+    authentication_source=None,
+    authentication_mechanism=None,
+    **kwargs
+):
+    """Register the connection settings.
+
+    : param alias: the name that will be used to refer to this connection
+        throughout MongoEngine
+    : param db: the name of the database to use, for compatibility with connect
+    : param name: the name of the specific database to use
+    : param host: the host name of the: program: `mongod` instance to connect to
+    : param port: the port that the: program: `mongod` instance is running on
+    : param read_preference: The read preference for the collection
+    : param username: username to authenticate with
+    : param password: password to authenticate with
+    : param authentication_source: database to authenticate against
+    : param authentication_mechanism: database authentication mechanisms.
+        By default, use SCRAM-SHA-1 with MongoDB 3.0 and later,
+        MONGODB-CR (MongoDB Challenge Response protocol) for older servers.
+    : param is_mock: explicitly use mongomock for this connection
+        (can also be done by using `mongomock: // ` as db host prefix)
+    : param kwargs: ad-hoc parameters to be passed into the pymongo driver,
+        for example maxpoolsize, tz_aware, etc. See the documentation
+        for pymongo's `MongoClient` for a full list.
+
+    .. versionchanged:: 0.10.6 - added mongomock support
+    """
+    conn_settings = _get_connection_settings(
+        db=db,
+        name=name,
+        host=host,
+        port=port,
+        read_preference=read_preference,
+        username=username,
+        password=password,
+        authentication_source=authentication_source,
+        authentication_mechanism=authentication_mechanism,
+        **kwargs
+    )
+    _connection_settings[alias] = conn_settings
+
+
+def disconnect(alias=DEFAULT_CONNECTION_NAME):
+    """Close the connection with a given alias."""
+    from mongoengine.base.common import _get_documents_by_db
+    from mongoengine import Document
+
+    if alias in _connections:
+        get_connection(alias=alias).close()
+        del _connections[alias]
+
+    if alias in _dbs:
+        # Detach all cached collections in Documents
+        for doc_cls in _get_documents_by_db(alias, DEFAULT_CONNECTION_NAME):
+            if issubclass(doc_cls, Document):  # Skip EmbeddedDocument
+                doc_cls._disconnect()
+
+        del _dbs[alias]
+
+    if alias in _connection_settings:
+        del _connection_settings[alias]
+
+
+def disconnect_all():
+    """Close all registered database."""
+    for alias in list(_connections.keys()):
+        disconnect(alias)
+
+
+def get_connection(alias=DEFAULT_CONNECTION_NAME, reconnect=False):
+    """Return a connection with a given alias."""
+
+    # Connect to the database if not already connected
+    if reconnect:
+        disconnect(alias)
+
+    # If the requested alias already exists in the _connections list, return
+    # it immediately.
+    if alias in _connections:
+        return _connections[alias]
+
+    # Validate that the requested alias exists in the _connection_settings.
+    # Raise ConnectionFailure if it doesn't.
+    if alias not in _connection_settings:
+        if alias == DEFAULT_CONNECTION_NAME:
+            msg = "You have not defined a default connection"
+        else:
+            msg = 'Connection with alias "%s" has not been defined' % alias
+        raise ConnectionFailure(msg)
+
+    def _clean_settings(settings_dict):
+        irrelevant_fields_set = {
+            "name",
+            "username",
+            "password",
+            "authentication_source",
+            "authentication_mechanism",
+        }
+        return {
+            k: v for k, v in settings_dict.items() if k not in irrelevant_fields_set
+        }
+
+    raw_conn_settings = _connection_settings[alias].copy()
+
+    # Retrieve a copy of the connection settings associated with the requested
+    # alias and remove the database name and authentication info (we don't
+    # care about them at this point).
+    conn_settings = _clean_settings(raw_conn_settings)
+
+    # Determine if we should use PyMongo's or mongomock's MongoClient.
+    is_mock = conn_settings.pop("is_mock", False)
+    if is_mock:
+        try:
+            import mongomock
+        except ImportError:
+            raise RuntimeError("You need mongomock installed to mock MongoEngine.")
+        connection_class = mongomock.MongoClient
+    else:
+        connection_class = MongoClient
+
+    # Re-use existing connection if one is suitable.
+    existing_connection = _find_existing_connection(raw_conn_settings)
+    if existing_connection:
+        connection = existing_connection
+    else:
+        connection = _create_connection(
+            alias=alias, connection_class=connection_class, **conn_settings
+        )
+    _connections[alias] = connection
+    return _connections[alias]
+
+
+def _create_connection(alias, connection_class, **connection_settings):
+    """
+    Create the new connection for this alias. Raise
+    ConnectionFailure if it can't be established.
+    """
+    try:
+        return connection_class(**connection_settings)
+    except Exception as e:
+        raise ConnectionFailure("Cannot connect to database {} :\n{}".format(alias, e))
+
+
+def _find_existing_connection(connection_settings):
+    """
+    Check if an existing connection could be reused
+
+    Iterate over all of the connection settings and if an existing connection
+    with the same parameters is suitable, return it
+
+    :param connection_settings: the settings of the new connection
+    :return: An existing connection or None
+    """
+    connection_settings_bis = (
+        (db_alias, settings.copy())
+        for db_alias, settings in _connection_settings.items()
+    )
+
+    def _clean_settings(settings_dict):
+        # Only remove the name but it's important to
+        # keep the username/password/authentication_source/authentication_mechanism
+        # to identify if the connection could be shared (cfr https://github.com/MongoEngine/mongoengine/issues/2047)
+        return {k: v for k, v in settings_dict.items() if k != "name"}
+
+    cleaned_conn_settings = _clean_settings(connection_settings)
+    for db_alias, connection_settings in connection_settings_bis:
+        db_conn_settings = _clean_settings(connection_settings)
+        if cleaned_conn_settings == db_conn_settings and _connections.get(db_alias):
+            return _connections[db_alias]
+
+
+def get_db(alias=DEFAULT_CONNECTION_NAME, reconnect=False):
+    if reconnect:
+        disconnect(alias)
+
+    if alias not in _dbs:
+        conn = get_connection(alias)
+        conn_settings = _connection_settings[alias]
+        db = conn[conn_settings["name"]]
+        auth_kwargs = {"source": conn_settings["authentication_source"]}
+        if conn_settings["authentication_mechanism"] is not None:
+            auth_kwargs["mechanism"] = conn_settings["authentication_mechanism"]
+        # Authenticate if necessary
+        if conn_settings["username"] and (
+            conn_settings["password"]
+            or conn_settings["authentication_mechanism"] == "MONGODB-X509"
+        ):
+            db.authenticate(
+                conn_settings["username"], conn_settings["password"], **auth_kwargs
+            )
+        _dbs[alias] = db
+    return _dbs[alias]
+
+
+def connect(db=None, alias=DEFAULT_CONNECTION_NAME, **kwargs):
+    """Connect to the database specified by the 'db' argument.
+
+    Connection settings may be provided here as well if the database is not
+    running on the default port on localhost. If authentication is needed,
+    provide username and password arguments as well.
+
+    Multiple databases are supported by using aliases. Provide a separate
+    `alias` to connect to a different instance of: program: `mongod`.
+
+    In order to replace a connection identified by a given alias, you'll
+    need to call ``disconnect`` first
+
+    See the docstring for `register_connection` for more details about all
+    supported kwargs.
+
+    .. versionchanged:: 0.6 - added multiple database support.
+    """
+    if alias in _connections:
+        prev_conn_setting = _connection_settings[alias]
+        new_conn_settings = _get_connection_settings(db, **kwargs)
+
+        if new_conn_settings != prev_conn_setting:
+            err_msg = (
+                "A different connection with alias `{}` was already "
+                "registered. Use disconnect() first"
+            ).format(alias)
+            raise ConnectionFailure(err_msg)
+    else:
+        register_connection(alias, db, **kwargs)
+
+    return get_connection(alias)
+
+
+# Support old naming convention
+_get_connection = get_connection
+_get_db = get_db

mongoengine/context_managers.py

@@ -0,0 +1,278 @@
+from contextlib import contextmanager
+
+from pymongo.read_concern import ReadConcern
+from pymongo.write_concern import WriteConcern
+
+from mongoengine.common import _import_class
+from mongoengine.connection import DEFAULT_CONNECTION_NAME, get_db
+from mongoengine.pymongo_support import count_documents
+
+__all__ = (
+    "switch_db",
+    "switch_collection",
+    "no_dereference",
+    "no_sub_classes",
+    "query_counter",
+    "set_write_concern",
+    "set_read_write_concern",
+)
+
+
+class switch_db:
+    """switch_db alias context manager.
+
+    Example ::
+
+        # Register connections
+        register_connection('default', 'mongoenginetest')
+        register_connection('testdb-1', 'mongoenginetest2')
+
+        class Group(Document):
+            name = StringField()
+
+        Group(name='test').save()  # Saves in the default db
+
+        with switch_db(Group, 'testdb-1') as Group:
+            Group(name='hello testdb!').save()  # Saves in testdb-1
+    """
+
+    def __init__(self, cls, db_alias):
+        """Construct the switch_db context manager
+
+        :param cls: the class to change the registered db
+        :param db_alias: the name of the specific database to use
+        """
+        self.cls = cls
+        self.collection = cls._get_collection()
+        self.db_alias = db_alias
+        self.ori_db_alias = cls._meta.get("db_alias", DEFAULT_CONNECTION_NAME)
+
+    def __enter__(self):
+        """Change the db_alias and clear the cached collection."""
+        self.cls._meta["db_alias"] = self.db_alias
+        self.cls._collection = None
+        return self.cls
+
+    def __exit__(self, t, value, traceback):
+        """Reset the db_alias and collection."""
+        self.cls._meta["db_alias"] = self.ori_db_alias
+        self.cls._collection = self.collection
+
+
+class switch_collection:
+    """switch_collection alias context manager.
+
+    Example ::
+
+        class Group(Document):
+            name = StringField()
+
+        Group(name='test').save()  # Saves in the default db
+
+        with switch_collection(Group, 'group1') as Group:
+            Group(name='hello testdb!').save()  # Saves in group1 collection
+    """
+
+    def __init__(self, cls, collection_name):
+        """Construct the switch_collection context manager.
+
+        :param cls: the class to change the registered db
+        :param collection_name: the name of the collection to use
+        """
+        self.cls = cls
+        self.ori_collection = cls._get_collection()
+        self.ori_get_collection_name = cls._get_collection_name
+        self.collection_name = collection_name
+
+    def __enter__(self):
+        """Change the _get_collection_name and clear the cached collection."""
+
+        @classmethod
+        def _get_collection_name(cls):
+            return self.collection_name
+
+        self.cls._get_collection_name = _get_collection_name
+        self.cls._collection = None
+        return self.cls
+
+    def __exit__(self, t, value, traceback):
+        """Reset the collection."""
+        self.cls._collection = self.ori_collection
+        self.cls._get_collection_name = self.ori_get_collection_name
+
+
+class no_dereference:
+    """no_dereference context manager.
+
+    Turns off all dereferencing in Documents for the duration of the context
+    manager::
+
+        with no_dereference(Group) as Group:
+            Group.objects.find()
+    """
+
+    def __init__(self, cls):
+        """Construct the no_dereference context manager.
+
+        :param cls: the class to turn dereferencing off on
+        """
+        self.cls = cls
+
+        ReferenceField = _import_class("ReferenceField")
+        GenericReferenceField = _import_class("GenericReferenceField")
+        ComplexBaseField = _import_class("ComplexBaseField")
+
+        self.deref_fields = [
+            k
+            for k, v in self.cls._fields.items()
+            if isinstance(v, (ReferenceField, GenericReferenceField, ComplexBaseField))
+        ]
+
+    def __enter__(self):
+        """Change the objects default and _auto_dereference values."""
+        for field in self.deref_fields:
+            self.cls._fields[field]._auto_dereference = False
+        return self.cls
+
+    def __exit__(self, t, value, traceback):
+        """Reset the default and _auto_dereference values."""
+        for field in self.deref_fields:
+            self.cls._fields[field]._auto_dereference = True
+        return self.cls
+
+
+class no_sub_classes:
+    """no_sub_classes context manager.
+
+    Only returns instances of this class and no sub (inherited) classes::
+
+        with no_sub_classes(Group) as Group:
+            Group.objects.find()
+    """
+
+    def __init__(self, cls):
+        """Construct the no_sub_classes context manager.
+
+        :param cls: the class to turn querying sub classes on
+        """
+        self.cls = cls
+        self.cls_initial_subclasses = None
+
+    def __enter__(self):
+        """Change the objects default and _auto_dereference values."""
+        self.cls_initial_subclasses = self.cls._subclasses
+        self.cls._subclasses = (self.cls._class_name,)
+        return self.cls
+
+    def __exit__(self, t, value, traceback):
+        """Reset the default and _auto_dereference values."""
+        self.cls._subclasses = self.cls_initial_subclasses
+
+
+class query_counter:
+    """Query_counter context manager to get the number of queries.
+    This works by updating the `profiling_level` of the database so that all queries get logged,
+    resetting the db.system.profile collection at the beginning of the context and counting the new entries.
+
+    This was designed for debugging purpose. In fact it is a global counter so queries issued by other threads/processes
+    can interfere with it
+
+    Be aware that:
+    - Iterating over large amount of documents (>101) makes pymongo issue `getmore` queries to fetch the next batch of
+        documents (https://docs.mongodb.com/manual/tutorial/iterate-a-cursor/#cursor-batches)
+    - Some queries are ignored by default by the counter (killcursors, db.system.indexes)
+    """
+
+    def __init__(self, alias=DEFAULT_CONNECTION_NAME):
+        """Construct the query_counter
+        """
+        self.db = get_db(alias=alias)
+        self.initial_profiling_level = None
+        self._ctx_query_counter = 0  # number of queries issued by the context
+
+        self._ignored_query = {
+            "ns": {"$ne": "%s.system.indexes" % self.db.name},
+            "op": {"$ne": "killcursors"},  # MONGODB < 3.2
+            "command.killCursors": {"$exists": False},  # MONGODB >= 3.2
+        }
+
+    def _turn_on_profiling(self):
+        self.initial_profiling_level = self.db.profiling_level()
+        self.db.set_profiling_level(0)
+        self.db.system.profile.drop()
+        self.db.set_profiling_level(2)
+
+    def _resets_profiling(self):
+        self.db.set_profiling_level(self.initial_profiling_level)
+
+    def __enter__(self):
+        self._turn_on_profiling()
+        return self
+
+    def __exit__(self, t, value, traceback):
+        self._resets_profiling()
+
+    def __eq__(self, value):
+        counter = self._get_count()
+        return value == counter
+
+    def __ne__(self, value):
+        return not self.__eq__(value)
+
+    def __lt__(self, value):
+        return self._get_count() < value
+
+    def __le__(self, value):
+        return self._get_count() <= value
+
+    def __gt__(self, value):
+        return self._get_count() > value
+
+    def __ge__(self, value):
+        return self._get_count() >= value
+
+    def __int__(self):
+        return self._get_count()
+
+    def __repr__(self):
+        """repr query_counter as the number of queries."""
+        return "%s" % self._get_count()
+
+    def _get_count(self):
+        """Get the number of queries by counting the current number of entries in db.system.profile
+        and substracting the queries issued by this context. In fact everytime this is called, 1 query is
+        issued so we need to balance that
+        """
+        count = (
+            count_documents(self.db.system.profile, self._ignored_query)
+            - self._ctx_query_counter
+        )
+        self._ctx_query_counter += (
+            1  # Account for the query we just issued to gather the information
+        )
+        return count
+
+
+@contextmanager
+def set_write_concern(collection, write_concerns):
+    combined_concerns = dict(collection.write_concern.document.items())
+    combined_concerns.update(write_concerns)
+    yield collection.with_options(write_concern=WriteConcern(**combined_concerns))
+
+
+@contextmanager
+def set_read_write_concern(collection, write_concerns, read_concerns):
+    combined_write_concerns = dict(collection.write_concern.document.items())
+
+    if write_concerns is not None:
+        combined_write_concerns.update(write_concerns)
+
+    combined_read_concerns = dict(collection.read_concern.document.items())
+
+    if read_concerns is not None:
+        combined_read_concerns.update(read_concerns)
+
+    yield collection.with_options(
+        write_concern=WriteConcern(**combined_write_concerns),
+        read_concern=ReadConcern(**combined_read_concerns),
+    )

mongoengine/dereference.py

@@ -0,0 +1,292 @@
+from bson import DBRef, SON
+
+from mongoengine.base import (
+    BaseDict,
+    BaseList,
+    EmbeddedDocumentList,
+    TopLevelDocumentMetaclass,
+    get_document,
+)
+from mongoengine.base.datastructures import LazyReference
+from mongoengine.connection import get_db
+from mongoengine.document import Document, EmbeddedDocument
+from mongoengine.fields import DictField, ListField, MapField, ReferenceField
+from mongoengine.queryset import QuerySet
+
+
+class DeReference:
+    def __call__(self, items, max_depth=1, instance=None, name=None):
+        """
+        Cheaply dereferences the items to a set depth.
+        Also handles the conversion of complex data types.
+
+        :param items: The iterable (dict, list, queryset) to be dereferenced.
+        :param max_depth: The maximum depth to recurse to
+        :param instance: The owning instance used for tracking changes by
+            :class:`~mongoengine.base.ComplexBaseField`
+        :param name: The name of the field, used for tracking changes by
+            :class:`~mongoengine.base.ComplexBaseField`
+        :param get: A boolean determining if being called by __get__
+        """
+        if items is None or isinstance(items, str):
+            return items
+
+        # cheapest way to convert a queryset to a list
+        # list(queryset) uses a count() query to determine length
+        if isinstance(items, QuerySet):
+            items = [i for i in items]
+
+        self.max_depth = max_depth
+        doc_type = None
+
+        if instance and isinstance(
+            instance, (Document, EmbeddedDocument, TopLevelDocumentMetaclass)
+        ):
+            doc_type = instance._fields.get(name)
+            while hasattr(doc_type, "field"):
+                doc_type = doc_type.field
+
+            if isinstance(doc_type, ReferenceField):
+                field = doc_type
+                doc_type = doc_type.document_type
+                is_list = not hasattr(items, "items")
+
+                if is_list and all([i.__class__ == doc_type for i in items]):
+                    return items
+                elif not is_list and all(
+                    [i.__class__ == doc_type for i in items.values()]
+                ):
+                    return items
+                elif not field.dbref:
+                    # We must turn the ObjectIds into DBRefs
+
+                    # Recursively dig into the sub items of a list/dict
+                    # to turn the ObjectIds into DBRefs
+                    def _get_items_from_list(items):
+                        new_items = []
+                        for v in items:
+                            value = v
+                            if isinstance(v, dict):
+                                value = _get_items_from_dict(v)
+                            elif isinstance(v, list):
+                                value = _get_items_from_list(v)
+                            elif not isinstance(v, (DBRef, Document)):
+                                value = field.to_python(v)
+                            new_items.append(value)
+                        return new_items
+
+                    def _get_items_from_dict(items):
+                        new_items = {}
+                        for k, v in items.items():
+                            value = v
+                            if isinstance(v, list):
+                                value = _get_items_from_list(v)
+                            elif isinstance(v, dict):
+                                value = _get_items_from_dict(v)
+                            elif not isinstance(v, (DBRef, Document)):
+                                value = field.to_python(v)
+                            new_items[k] = value
+                        return new_items
+
+                    if not hasattr(items, "items"):
+                        items = _get_items_from_list(items)
+                    else:
+                        items = _get_items_from_dict(items)
+
+        self.reference_map = self._find_references(items)
+        self.object_map = self._fetch_objects(doc_type=doc_type)
+        return self._attach_objects(items, 0, instance, name)
+
+    def _find_references(self, items, depth=0):
+        """
+        Recursively finds all db references to be dereferenced
+
+        :param items: The iterable (dict, list, queryset)
+        :param depth: The current depth of recursion
+        """
+        reference_map = {}
+        if not items or depth >= self.max_depth:
+            return reference_map
+
+        # Determine the iterator to use
+        if isinstance(items, dict):
+            iterator = items.values()
+        else:
+            iterator = items
+
+        # Recursively find dbreferences
+        depth += 1
+        for item in iterator:
+            if isinstance(item, (Document, EmbeddedDocument)):
+                for field_name, field in item._fields.items():
+                    v = item._data.get(field_name, None)
+                    if isinstance(v, LazyReference):
+                        # LazyReference inherits DBRef but should not be dereferenced here !
+                        continue
+                    elif isinstance(v, DBRef):
+                        reference_map.setdefault(field.document_type, set()).add(v.id)
+                    elif isinstance(v, (dict, SON)) and "_ref" in v:
+                        reference_map.setdefault(get_document(v["_cls"]), set()).add(
+                            v["_ref"].id
+                        )
+                    elif isinstance(v, (dict, list, tuple)) and depth <= self.max_depth:
+                        field_cls = getattr(
+                            getattr(field, "field", None), "document_type", None
+                        )
+                        references = self._find_references(v, depth)
+                        for key, refs in references.items():
+                            if isinstance(
+                                field_cls, (Document, TopLevelDocumentMetaclass)
+                            ):
+                                key = field_cls
+                            reference_map.setdefault(key, set()).update(refs)
+            elif isinstance(item, LazyReference):
+                # LazyReference inherits DBRef but should not be dereferenced here !
+                continue
+            elif isinstance(item, DBRef):
+                reference_map.setdefault(item.collection, set()).add(item.id)
+            elif isinstance(item, (dict, SON)) and "_ref" in item:
+                reference_map.setdefault(get_document(item["_cls"]), set()).add(
+                    item["_ref"].id
+                )
+            elif isinstance(item, (dict, list, tuple)) and depth - 1 <= self.max_depth:
+                references = self._find_references(item, depth - 1)
+                for key, refs in references.items():
+                    reference_map.setdefault(key, set()).update(refs)
+
+        return reference_map
+
+    def _fetch_objects(self, doc_type=None):
+        """Fetch all references and convert to their document objects
+        """
+        object_map = {}
+        for collection, dbrefs in self.reference_map.items():
+
+            # we use getattr instead of hasattr because hasattr swallows any exception under python2
+            # so it could hide nasty things without raising exceptions (cfr bug #1688))
+            ref_document_cls_exists = getattr(collection, "objects", None) is not None
+
+            if ref_document_cls_exists:
+                col_name = collection._get_collection_name()
+                refs = [
+                    dbref for dbref in dbrefs if (col_name, dbref) not in object_map
+                ]
+                references = collection.objects.in_bulk(refs)
+                for key, doc in references.items():
+                    object_map[(col_name, key)] = doc
+            else:  # Generic reference: use the refs data to convert to document
+                if isinstance(doc_type, (ListField, DictField, MapField)):
+                    continue
+
+                refs = [
+                    dbref for dbref in dbrefs if (collection, dbref) not in object_map
+                ]
+
+                if doc_type:
+                    references = doc_type._get_db()[collection].find(
+                        {"_id": {"$in": refs}}
+                    )
+                    for ref in references:
+                        doc = doc_type._from_son(ref)
+                        object_map[(collection, doc.id)] = doc
+                else:
+                    references = get_db()[collection].find({"_id": {"$in": refs}})
+                    for ref in references:
+                        if "_cls" in ref:
+                            doc = get_document(ref["_cls"])._from_son(ref)
+                        elif doc_type is None:
+                            doc = get_document(
+                                "".join(x.capitalize() for x in collection.split("_"))
+                            )._from_son(ref)
+                        else:
+                            doc = doc_type._from_son(ref)
+                        object_map[(collection, doc.id)] = doc
+        return object_map
+
+    def _attach_objects(self, items, depth=0, instance=None, name=None):
+        """
+        Recursively finds all db references to be dereferenced
+
+        :param items: The iterable (dict, list, queryset)
+        :param depth: The current depth of recursion
+        :param instance: The owning instance used for tracking changes by
+            :class:`~mongoengine.base.ComplexBaseField`
+        :param name: The name of the field, used for tracking changes by
+            :class:`~mongoengine.base.ComplexBaseField`
+        """
+        if not items:
+            if isinstance(items, (BaseDict, BaseList)):
+                return items
+
+            if instance:
+                if isinstance(items, dict):
+                    return BaseDict(items, instance, name)
+                else:
+                    return BaseList(items, instance, name)
+
+        if isinstance(items, (dict, SON)):
+            if "_ref" in items:
+                return self.object_map.get(
+                    (items["_ref"].collection, items["_ref"].id), items
+                )
+            elif "_cls" in items:
+                doc = get_document(items["_cls"])._from_son(items)
+                _cls = doc._data.pop("_cls", None)
+                del items["_cls"]
+                doc._data = self._attach_objects(doc._data, depth, doc, None)
+                if _cls is not None:
+                    doc._data["_cls"] = _cls
+                return doc
+
+        if not hasattr(items, "items"):
+            is_list = True
+            list_type = BaseList
+            if isinstance(items, EmbeddedDocumentList):
+                list_type = EmbeddedDocumentList
+            as_tuple = isinstance(items, tuple)
+            iterator = enumerate(items)
+            data = []
+        else:
+            is_list = False
+            iterator = items.items()
+            data = {}
+
+        depth += 1
+        for k, v in iterator:
+            if is_list:
+                data.append(v)
+            else:
+                data[k] = v
+
+            if k in self.object_map and not is_list:
+                data[k] = self.object_map[k]
+            elif isinstance(v, (Document, EmbeddedDocument)):
+                for field_name in v._fields:
+                    v = data[k]._data.get(field_name, None)
+                    if isinstance(v, DBRef):
+                        data[k]._data[field_name] = self.object_map.get(
+                            (v.collection, v.id), v
+                        )
+                    elif isinstance(v, (dict, SON)) and "_ref" in v:
+                        data[k]._data[field_name] = self.object_map.get(
+                            (v["_ref"].collection, v["_ref"].id), v
+                        )
+                    elif isinstance(v, (dict, list, tuple)) and depth <= self.max_depth:
+                        item_name = "{}.{}.{}".format(name, k, field_name)
+                        data[k]._data[field_name] = self._attach_objects(
+                            v, depth, instance=instance, name=item_name
+                        )
+            elif isinstance(v, (dict, list, tuple)) and depth <= self.max_depth:
+                item_name = "{}.{}".format(name, k) if name else name
+                data[k] = self._attach_objects(
+                    v, depth - 1, instance=instance, name=item_name
+                )
+            elif isinstance(v, DBRef) and hasattr(v, "id"):
+                data[k] = self.object_map.get((v.collection, v.id), v)
+
+        if instance and name:
+            if is_list:
+                return tuple(data) if as_tuple else list_type(data, instance, name)
+            return BaseDict(data, instance, name)
+        depth += 1
+        return data

mongoengine/django/auth.py

@@ -1,99 +0,0 @@
-from mongoengine import *
-
-from django.utils.hashcompat import md5_constructor, sha_constructor
-from django.utils.encoding import smart_str
-from django.contrib.auth.models import AnonymousUser
-
-import datetime
-
-REDIRECT_FIELD_NAME = 'next'
-
-def get_hexdigest(algorithm, salt, raw_password):
-    raw_password, salt = smart_str(raw_password), smart_str(salt)
-    if algorithm == 'md5':
-        return md5_constructor(salt + raw_password).hexdigest()
-    elif algorithm == 'sha1':
-        return sha_constructor(salt + raw_password).hexdigest()
-    raise ValueError('Got unknown password algorithm type in password')
-
-
-class User(Document):
-    """A User document that aims to mirror most of the API specified by Django
-    at http://docs.djangoproject.com/en/dev/topics/auth/#users
-    """
-    username = StringField(max_length=30, required=True)
-    first_name = StringField(max_length=30)
-    last_name = StringField(max_length=30)
-    email = StringField()
-    password = StringField(max_length=128)
-    is_staff = BooleanField(default=False)
-    is_active = BooleanField(default=True)
-    is_superuser = BooleanField(default=False)
-    last_login = DateTimeField(default=datetime.datetime.now)
-
-    def get_full_name(self):
-        """Returns the users first and last names, separated by a space.
-        """
-        full_name = u'%s %s' % (self.first_name or '', self.last_name or '')
-        return full_name.strip()
-
-    def is_anonymous(self):
-        return False
-
-    def is_authenticated(self):
-        return True
-
-    def set_password(self, raw_password):
-        """Sets the user's password - always use this rather than directly
-        assigning to :attr:`~mongoengine.django.auth.User.password` as the
-        password is hashed before storage.
-        """
-        from random import random
-        algo = 'sha1'
-        salt = get_hexdigest(algo, str(random()), str(random()))[:5]
-        hash = get_hexdigest(algo, salt, raw_password)
-        self.password = '%s$%s$%s' % (algo, salt, hash)
-
-    def check_password(self, raw_password):
-        """Checks the user's password against a provided password - always use
-        this rather than directly comparing to
-        :attr:`~mongoengine.django.auth.User.password` as the password is
-        hashed before storage.
-        """
-        algo, salt, hash = self.password.split('$')
-        return hash == get_hexdigest(algo, salt, raw_password)
-
-    @classmethod
-    def create_user(cls, username, password, email=None):
-        """Create (and save) a new user with the given username, password and
-        email address.
-        """
-        user = User(username=username, email=email)
-        user.set_password(password)
-        user.save()
-        return user
-
-
-class MongoEngineBackend(object):
-    """Authenticate using MongoEngine and mongoengine.django.auth.User.
-    """
-
-    def authenticate(self, username=None, password=None):
-        user = User.objects(username=username).first()
-        if user:
-            if password and user.check_password(password):
-                return user
-        return None
-
-    def get_user(self, user_id):
-        return User.objects.with_id(user_id)
-
-
-def get_user(userid):
-    """Returns a User object from an id (User.id). Django's equivalent takes
-    request, but taking an id instead leaves it up to the developer to store
-    the id in any way they want (session, signed cookie, etc.)
-    """
-    if not userid:
-        return AnonymousUser()
-    return MongoEngineBackend().get_user(userid) or AnonymousUser()

mongoengine/django/sessions.py

@@ -1,63 +0,0 @@
-from django.contrib.sessions.backends.base import SessionBase, CreateError
-from django.core.exceptions import SuspiciousOperation
-from django.utils.encoding import force_unicode
-
-from mongoengine.document import Document
-from mongoengine import fields
-from mongoengine.queryset import OperationError
-
-from datetime import datetime
-
-
-class MongoSession(Document):
-    session_key = fields.StringField(primary_key=True, max_length=40)
-    session_data = fields.StringField()
-    expire_date = fields.DateTimeField()
-    
-    meta = {'collection': 'django_session', 'allow_inheritance': False}
-
-
-class SessionStore(SessionBase):
-    """A MongoEngine-based session store for Django.
-    """
-
-    def load(self):
-        try:
-            s = MongoSession.objects(session_key=self.session_key,
-                                     expire_date__gt=datetime.now())[0]
-            return self.decode(force_unicode(s.session_data))
-        except (IndexError, SuspiciousOperation):
-            self.create()
-            return {}
-
-    def exists(self, session_key):
-        return bool(MongoSession.objects(session_key=session_key).first())
-
-    def create(self):
-        while True:
-            self.session_key = self._get_new_session_key()
-            try:
-                self.save(must_create=True)
-            except CreateError:
-                continue
-            self.modified = True
-            self._session_cache = {}
-            return
-
-    def save(self, must_create=False):
-        s = MongoSession(session_key=self.session_key)
-        s.session_data = self.encode(self._get_session(no_load=must_create))
-        s.expire_date = self.get_expiry_date()
-        try:
-            s.save(force_insert=must_create, safe=True)
-        except OperationError:
-            if must_create:
-                raise CreateError
-            raise
-
-    def delete(self, session_key=None):
-        if session_key is None:
-            if self.session_key is None:
-                return
-            session_key = self.session_key
-        MongoSession.objects(session_key=session_key).delete()

mongoengine/errors.py

@@ -0,0 +1,162 @@
+from collections import defaultdict
+
+
+__all__ = (
+    "NotRegistered",
+    "InvalidDocumentError",
+    "LookUpError",
+    "DoesNotExist",
+    "MultipleObjectsReturned",
+    "InvalidQueryError",
+    "OperationError",
+    "NotUniqueError",
+    "BulkWriteError",
+    "FieldDoesNotExist",
+    "ValidationError",
+    "SaveConditionError",
+    "DeprecatedError",
+)
+
+
+class NotRegistered(Exception):
+    pass
+
+
+class InvalidDocumentError(Exception):
+    pass
+
+
+class LookUpError(AttributeError):
+    pass
+
+
+class DoesNotExist(Exception):
+    pass
+
+
+class MultipleObjectsReturned(Exception):
+    pass
+
+
+class InvalidQueryError(Exception):
+    pass
+
+
+class OperationError(Exception):
+    pass
+
+
+class NotUniqueError(OperationError):
+    pass
+
+
+class BulkWriteError(OperationError):
+    pass
+
+
+class SaveConditionError(OperationError):
+    pass
+
+
+class FieldDoesNotExist(Exception):
+    """Raised when trying to set a field
+    not declared in a :class:`~mongoengine.Document`
+    or an :class:`~mongoengine.EmbeddedDocument`.
+
+    To avoid this behavior on data loading,
+    you should set the :attr:`strict` to ``False``
+    in the :attr:`meta` dictionary.
+    """
+
+
+class ValidationError(AssertionError):
+    """Validation exception.
+
+    May represent an error validating a field or a
+    document containing fields with validation errors.
+
+    :ivar errors: A dictionary of errors for fields within this
+        document or list, or None if the error is for an
+        individual field.
+    """
+
+    errors = {}
+    field_name = None
+    _message = None
+
+    def __init__(self, message="", **kwargs):
+        super().__init__(message)
+        self.errors = kwargs.get("errors", {})
+        self.field_name = kwargs.get("field_name")
+        self.message = message
+
+    def __str__(self):
+        return str(self.message)
+
+    def __repr__(self):
+        return "{}({},)".format(self.__class__.__name__, self.message)
+
+    def __getattribute__(self, name):
+        message = super().__getattribute__(name)
+        if name == "message":
+            if self.field_name:
+                message = "%s" % message
+            if self.errors:
+                message = "{}({})".format(message, self._format_errors())
+        return message
+
+    def _get_message(self):
+        return self._message
+
+    def _set_message(self, message):
+        self._message = message
+
+    message = property(_get_message, _set_message)
+
+    def to_dict(self):
+        """Returns a dictionary of all errors within a document
+
+        Keys are field names or list indices and values are the
+        validation error messages, or a nested dictionary of
+        errors for an embedded document or list.
+        """
+
+        def build_dict(source):
+            errors_dict = {}
+            if isinstance(source, dict):
+                for field_name, error in source.items():
+                    errors_dict[field_name] = build_dict(error)
+            elif isinstance(source, ValidationError) and source.errors:
+                return build_dict(source.errors)
+            else:
+                return str(source)
+
+            return errors_dict
+
+        if not self.errors:
+            return {}
+
+        return build_dict(self.errors)
+
+    def _format_errors(self):
+        """Returns a string listing all errors within a document"""
+
+        def generate_key(value, prefix=""):
+            if isinstance(value, list):
+                value = " ".join([generate_key(k) for k in value])
+            elif isinstance(value, dict):
+                value = " ".join([generate_key(v, k) for k, v in value.items()])
+
+            results = "{}.{}".format(prefix, value) if prefix else value
+            return results
+
+        error_dict = defaultdict(list)
+        for k, v in self.to_dict().items():
+            error_dict[generate_key(v)].append(k)
+        return " ".join(["{}: {}".format(k, v) for k, v in error_dict.items()])
+
+
+class DeprecatedError(Exception):
+    """Raise when a user uses a feature that has been Deprecated"""
+
+    pass

mongoengine/mongodb_support.py

@@ -0,0 +1,19 @@
+"""
+Helper functions, constants, and types to aid with MongoDB version support
+"""
+from mongoengine.connection import get_connection
+
+
+# Constant that can be used to compare the version retrieved with
+# get_mongodb_version()
+MONGODB_34 = (3, 4)
+MONGODB_36 = (3, 6)
+
+
+def get_mongodb_version():
+    """Return the version of the default connected mongoDB (first 2 digits)
+
+    :return: tuple(int, int)
+    """
+    version_list = get_connection().server_info()["versionArray"][:2]  # e.g: (3, 2)
+    return tuple(version_list)

mongoengine/pymongo_support.py

@@ -0,0 +1,32 @@
+"""
+Helper functions, constants, and types to aid with PyMongo v2.7 - v3.x support.
+"""
+import pymongo
+
+_PYMONGO_37 = (3, 7)
+
+PYMONGO_VERSION = tuple(pymongo.version_tuple[:2])
+
+IS_PYMONGO_GTE_37 = PYMONGO_VERSION >= _PYMONGO_37
+
+
+def count_documents(collection, filter):
+    """Pymongo>3.7 deprecates count in favour of count_documents"""
+    if IS_PYMONGO_GTE_37:
+        return collection.count_documents(filter)
+    else:
+        count = collection.find(filter).count()
+    return count
+
+
+def list_collection_names(db, include_system_collections=False):
+    """Pymongo>3.7 deprecates collection_names in favour of list_collection_names"""
+    if IS_PYMONGO_GTE_37:
+        collections = db.list_collection_names()
+    else:
+        collections = db.collection_names()
+
+    if not include_system_collections:
+        collections = [c for c in collections if not c.startswith("system.")]
+
+    return collections

mongoengine/queryset.py

@@ -1,665 +0,0 @@
-from connection import _get_db
-
-import pymongo
-import copy
-
-
-__all__ = ['queryset_manager', 'Q', 'InvalidQueryError', 
-           'InvalidCollectionError']
-
-# The maximum number of items to display in a QuerySet.__repr__
-REPR_OUTPUT_SIZE = 20
-
-
-class InvalidQueryError(Exception):
-    pass
-
-
-class OperationError(Exception):
-    pass
-
-
-class Q(object):
-    
-    OR = '||'
-    AND = '&&'
-    OPERATORS = {
-        'eq': 'this.%(field)s == %(value)s',
-        'neq': 'this.%(field)s != %(value)s',
-        'gt': 'this.%(field)s > %(value)s',
-        'gte': 'this.%(field)s >= %(value)s',
-        'lt': 'this.%(field)s < %(value)s',
-        'lte': 'this.%(field)s <= %(value)s',
-        'lte': 'this.%(field)s <= %(value)s',
-        'in': 'this.%(field)s.indexOf(%(value)s) != -1',
-        'nin': 'this.%(field)s.indexOf(%(value)s) == -1',
-        'mod': '%(field)s %% %(value)s',
-        'all': ('%(value)s.every(function(a){'
-                'return this.%(field)s.indexOf(a) != -1 })'),
-        'size': 'this.%(field)s.length == %(value)s',
-        'exists': 'this.%(field)s != null',
-    }
-    
-    def __init__(self, **query):
-        self.query = [query]
-
-    def _combine(self, other, op):
-        obj = Q()
-        obj.query = ['('] + copy.deepcopy(self.query) + [op]
-        obj.query += copy.deepcopy(other.query) + [')']
-        return obj
-
-    def __or__(self, other):
-        return self._combine(other, self.OR)
-
-    def __and__(self, other):
-        return self._combine(other, self.AND)
-
-    def as_js(self, document):
-        js = []
-        js_scope = {}
-        for i, item in enumerate(self.query):
-            if isinstance(item, dict):
-                item_query = QuerySet._transform_query(document, **item)
-                # item_query will values will either be a value or a dict
-                js.append(self._item_query_as_js(item_query, js_scope, i))
-            else:
-                js.append(item)
-        return pymongo.code.Code(' '.join(js), js_scope)
-
-    def _item_query_as_js(self, item_query, js_scope, item_num):
-        # item_query will be in one of the following forms
-        #    {'age': 25, 'name': 'Test'}
-        #    {'age': {'$lt': 25}, 'name': {'$in': ['Test', 'Example']}
-        #    {'age': {'$lt': 25, '$gt': 18}}
-        js = []
-        for i, (key, value) in enumerate(item_query.items()):
-            op = 'eq'
-            # Construct a variable name for the value in the JS
-            value_name = 'i%sf%s' % (item_num, i)
-            if isinstance(value, dict):
-                # Multiple operators for this field
-                for j, (op, value) in enumerate(value.items()):
-                    # Create a custom variable name for this operator
-                    op_value_name = '%so%s' % (value_name, j)
-                    # Update the js scope with the value for this op
-                    js_scope[op_value_name] = value
-                    # Construct the JS that uses this op
-                    operation_js = Q.OPERATORS[op.strip('$')] % {
-                        'field': key, 
-                        'value': op_value_name
-                    }
-                    js.append(operation_js)
-            else:
-                js_scope[value_name] = value
-                # Construct the JS for this field
-                field_js = Q.OPERATORS[op.strip('$')] % {
-                    'field': key, 
-                    'value': value_name
-                }
-                js.append(field_js)
-        return ' && '.join(js)
-
-
-class QuerySet(object):
-    """A set of results returned from a query. Wraps a MongoDB cursor, 
-    providing :class:`~mongoengine.Document` objects as the results.
-    """
-    
-    def __init__(self, document, collection):
-        self._document = document
-        self._collection_obj = collection
-        self._accessed_collection = False
-        self._query = {}
-        self._where_clauses = []
-
-        # If inheritance is allowed, only return instances and instances of
-        # subclasses of the class being used
-        if document._meta.get('allow_inheritance'):
-            self._query = {'_types': self._document._class_name}
-        self._cursor_obj = None
-        
-    def ensure_index(self, key_or_list):
-        """Ensure that the given indexes are in place.
-
-        :param key_or_list: a single index key or a list of index keys (to
-            construct a multi-field index); keys may be prefixed with a **+**
-            or a **-** to determine the index ordering
-        """
-        index_list = QuerySet._build_index_spec(self._document, key_or_list)
-        self._collection.ensure_index(index_list)
-        return self
-
-    @classmethod
-    def _build_index_spec(cls, doc_cls, key_or_list):
-        """Build a PyMongo index spec from a MongoEngine index spec.
-        """
-        if isinstance(key_or_list, basestring):
-            key_or_list = [key_or_list]
-
-        index_list = []
-        use_types = doc_cls._meta.get('allow_inheritance', True)
-        for key in key_or_list:
-            # Get direction from + or -
-            direction = pymongo.ASCENDING
-            if key.startswith("-"):
-                direction = pymongo.DESCENDING
-            if key.startswith(("+", "-")):
-                    key = key[1:]
-
-            # Use real field name, do it manually because we need field
-            # objects for the next part (list field checking)
-            parts = key.split('.')
-            fields = QuerySet._lookup_field(doc_cls, parts)
-            parts = [field.name for field in fields]
-            key = '.'.join(parts)
-            index_list.append((key, direction))
-
-            # Check if a list field is being used, don't use _types if it is
-            if use_types and not all(f._index_with_types for f in fields):
-                use_types = False
-
-        # If _types is being used, prepend it to every specified index
-        if doc_cls._meta.get('allow_inheritance') and use_types:
-            index_list.insert(0, ('_types', 1))
-
-        return index_list
-
-    def __call__(self, *q_objs, **query):
-        """Filter the selected documents by calling the 
-        :class:`~mongoengine.queryset.QuerySet` with a query.
-
-        :param q_objs: :class:`~mongoengine.queryset.Q` objects to be used in
-            the query 
-        :param query: Django-style query keyword arguments
-        """
-        for q in q_objs:
-            self._where_clauses.append(q.as_js(self._document))
-        query = QuerySet._transform_query(_doc_cls=self._document, **query)
-        self._query.update(query)
-        return self
-        
-    def filter(self, *q_objs, **query):
-        """An alias of :meth:`~mongoengine.queryset.QuerySet.__call__`
-        """
-        return self.__call__(*q_objs, **query)
-
-    @property
-    def _collection(self):
-        """Property that returns the collection object. This allows us to
-        perform operations only if the collection is accessed.
-        """
-        if not self._accessed_collection:
-            self._accessed_collection = True
-
-            # Ensure document-defined indexes are created
-            if self._document._meta['indexes']:
-                for key_or_list in self._document._meta['indexes']:
-                    #self.ensure_index(key_or_list)
-                    self._collection.ensure_index(key_or_list)
-
-            # Ensure indexes created by uniqueness constraints
-            for index in self._document._meta['unique_indexes']:
-                self._collection.ensure_index(index, unique=True)
-
-            # If _types is being used (for polymorphism), it needs an index
-            if '_types' in self._query:
-                self._collection.ensure_index('_types')
-        return self._collection_obj
-
-    @property
-    def _cursor(self):
-        if not self._cursor_obj:
-            self._cursor_obj = self._collection.find(self._query)
-            # Apply where clauses to cursor
-            for js in self._where_clauses:
-                self._cursor_obj.where(js)
-            
-            # apply default ordering
-            if self._document._meta['ordering']:
-                self.order_by(*self._document._meta['ordering'])
-            
-        return self._cursor_obj
-
-    @classmethod
-    def _lookup_field(cls, document, parts):
-        """Lookup a field based on its attribute and return a list containing
-        the field's parents and the field.
-        """
-        if not isinstance(parts, (list, tuple)):
-            parts = [parts]
-        fields = []
-        field = None
-        for field_name in parts:
-            if field is None:
-                # Look up first field from the document
-                field = document._fields[field_name]
-            else:
-                # Look up subfield on the previous field
-                field = field.lookup_member(field_name)
-                if field is None:
-                    raise InvalidQueryError('Cannot resolve field "%s"'
-                                            % field_name)
-            fields.append(field)
-        return fields
-
-    @classmethod
-    def _translate_field_name(cls, doc_cls, field, sep='.'):
-        """Translate a field attribute name to a database field name.
-        """
-        parts = field.split(sep)
-        parts = [f.name for f in QuerySet._lookup_field(doc_cls, parts)]
-        return '.'.join(parts)
-
-    @classmethod
-    def _transform_query(cls, _doc_cls=None, **query):
-        """Transform a query from Django-style format to Mongo format.
-        """
-        operators = ['neq', 'gt', 'gte', 'lt', 'lte', 'in', 'nin', 'mod',
-                     'all', 'size', 'exists']
-
-        mongo_query = {}
-        for key, value in query.items():
-            parts = key.split('__')
-            # Check for an operator and transform to mongo-style if there is
-            op = None
-            if parts[-1] in operators:
-                op = parts.pop()
-
-            if _doc_cls:
-                # Switch field names to proper names [set in Field(name='foo')]
-                fields = QuerySet._lookup_field(_doc_cls, parts)
-                parts = [field.name for field in fields]
-
-                # Convert value to proper value
-                field = fields[-1]
-                if op in (None, 'neq', 'gt', 'gte', 'lt', 'lte'):
-                    value = field.prepare_query_value(value)
-                elif op in ('in', 'nin', 'all'):
-                    # 'in', 'nin' and 'all' require a list of values
-                    value = [field.prepare_query_value(v) for v in value]
-
-            if op:
-                value = {'$' + op: value}
-
-            key = '.'.join(parts)
-            if op is None or key not in mongo_query:
-                mongo_query[key] = value
-            elif key in mongo_query and isinstance(mongo_query[key], dict):
-                mongo_query[key].update(value)
-
-        return mongo_query
-
-    def first(self):
-        """Retrieve the first object matching the query.
-        """
-        try:
-            result = self[0]
-        except IndexError:
-            result = None
-        return result
-
-    def with_id(self, object_id):
-        """Retrieve the object matching the id provided.
-
-        :param object_id: the value for the id of the document to look up
-        """
-        id_field = self._document._meta['id_field']
-        object_id = self._document._fields[id_field].to_mongo(object_id)
-
-        result = self._collection.find_one({'_id': object_id})
-        if result is not None:
-            result = self._document._from_son(result)
-        return result
-
-    def next(self):
-        """Wrap the result in a :class:`~mongoengine.Document` object.
-        """
-        return self._document._from_son(self._cursor.next())
-
-    def count(self):
-        """Count the selected elements in the query.
-        """
-        return self._cursor.count()
-
-    def __len__(self):
-        return self.count()
-
-    def limit(self, n):
-        """Limit the number of returned documents to `n`. This may also be
-        achieved using array-slicing syntax (e.g. ``User.objects[:5]``).
-
-        :param n: the maximum number of objects to return
-        """
-        self._cursor.limit(n)
-        # Return self to allow chaining
-        return self
-
-    def skip(self, n):
-        """Skip `n` documents before returning the results. This may also be
-        achieved using array-slicing syntax (e.g. ``User.objects[5:]``).
-
-        :param n: the number of objects to skip before returning results
-        """
-        self._cursor.skip(n)
-        return self
-
-    def __getitem__(self, key):
-        """Support skip and limit using getitem and slicing syntax.
-        """
-        # Slice provided
-        if isinstance(key, slice):
-            try:
-                self._cursor_obj = self._cursor[key]
-            except IndexError, err:
-                # PyMongo raises an error if key.start == key.stop, catch it,
-                # bin it, kill it. 
-                if key.start >=0 and key.stop >= 0 and key.step is None:
-                    if key.start == key.stop:
-                        self.limit(0)
-                        return self
-                raise err
-            # Allow further QuerySet modifications to be performed
-            return self
-        # Integer index provided
-        elif isinstance(key, int):
-            return self._document._from_son(self._cursor[key])
-
-    def order_by(self, *keys):
-        """Order the :class:`~mongoengine.queryset.QuerySet` by the keys. The
-        order may be specified by prepending each of the keys by a + or a -.
-        Ascending order is assumed.
-
-        :param keys: fields to order the query results by; keys may be
-            prefixed with **+** or **-** to determine the ordering direction
-        """
-        key_list = []
-        for key in keys:
-            direction = pymongo.ASCENDING
-            if key[0] == '-':
-                direction = pymongo.DESCENDING
-            if key[0] in ('-', '+'):
-                key = key[1:]
-            key_list.append((key, direction)) 
-
-        self._cursor.sort(key_list)
-        return self
-        
-    def explain(self, format=False):
-        """Return an explain plan record for the 
-        :class:`~mongoengine.queryset.QuerySet`\ 's cursor.
-
-        :param format: format the plan before returning it
-        """
-
-        plan = self._cursor.explain()
-        if format:
-            import pprint
-            plan = pprint.pformat(plan)
-        return plan
-        
-    def delete(self, safe=False):
-        """Delete the documents matched by the query.
-
-        :param safe: check if the operation succeeded before returning
-        """
-        self._collection.remove(self._query, safe=safe)
-
-    @classmethod
-    def _transform_update(cls, _doc_cls=None, **update):
-        """Transform an update spec from Django-style format to Mongo format.
-        """
-        operators = ['set', 'unset', 'inc', 'dec', 'push', 'push_all', 'pull', 
-                     'pull_all']
-
-        mongo_update = {}
-        for key, value in update.items():
-            parts = key.split('__')
-            # Check for an operator and transform to mongo-style if there is
-            op = None
-            if parts[0] in operators:
-                op = parts.pop(0)
-                # Convert Pythonic names to Mongo equivalents
-                if op in ('push_all', 'pull_all'):
-                    op = op.replace('_all', 'All')
-                elif op == 'dec':
-                    # Support decrement by flipping a positive value's sign
-                    # and using 'inc'
-                    op = 'inc'
-                    if value > 0:
-                        value = -value
-
-            if _doc_cls:
-                # Switch field names to proper names [set in Field(name='foo')]
-                fields = QuerySet._lookup_field(_doc_cls, parts)
-                parts = [field.name for field in fields]
-
-                # Convert value to proper value
-                field = fields[-1]
-                if op in (None, 'set', 'unset', 'push', 'pull'):
-                    value = field.prepare_query_value(value)
-                elif op in ('pushAll', 'pullAll'):
-                    value = [field.prepare_query_value(v) for v in value]
-
-            key = '.'.join(parts)
-
-            if op:
-                value = {key: value}
-                key = '$' + op
-
-            if op is None or key not in mongo_update:
-                mongo_update[key] = value
-            elif key in mongo_update and isinstance(mongo_update[key], dict):
-                mongo_update[key].update(value)
-
-        return mongo_update
-
-    def update(self, safe_update=True, **update):
-        """Perform an atomic update on the fields matched by the query.
-
-        :param safe: check if the operation succeeded before returning
-        :param update: Django-style update keyword arguments
-
-        .. versionadded:: 0.2
-        """
-        if pymongo.version < '1.1.1':
-            raise OperationError('update() method requires PyMongo 1.1.1+')
-
-        update = QuerySet._transform_update(self._document, **update)
-        try:
-            self._collection.update(self._query, update, safe=safe_update, 
-                                    multi=True)
-        except pymongo.errors.OperationFailure, err:
-            if str(err) == 'multi not coded yet':
-                raise OperationError('update() method requires MongoDB 1.1.3+')
-            raise OperationError('Update failed (%s)' % str(err))
-
-    def update_one(self, safe_update=True, **update):
-        """Perform an atomic update on first field matched by the query.
-
-        :param safe: check if the operation succeeded before returning
-        :param update: Django-style update keyword arguments
-
-        .. versionadded:: 0.2
-        """
-        update = QuerySet._transform_update(self._document, **update)
-        try:
-            # Explicitly provide 'multi=False' to newer versions of PyMongo
-            # as the default may change to 'True'
-            if pymongo.version >= '1.1.1':
-                self._collection.update(self._query, update, safe=safe_update, 
-                                        multi=False)
-            else:
-                # Older versions of PyMongo don't support 'multi'
-                self._collection.update(self._query, update, safe=safe_update)
-        except pymongo.errors.OperationFailure, e:
-            raise OperationError('Update failed [%s]' % str(e))
-
-    def __iter__(self):
-        return self
-
-    def exec_js(self, code, *fields, **options):
-        """Execute a Javascript function on the server. A list of fields may be
-        provided, which will be translated to their correct names and supplied
-        as the arguments to the function. A few extra variables are added to
-        the function's scope: ``collection``, which is the name of the 
-        collection in use; ``query``, which is an object representing the 
-        current query; and ``options``, which is an object containing any
-        options specified as keyword arguments.
-
-        :param code: a string of Javascript code to execute
-        :param fields: fields that you will be using in your function, which
-            will be passed in to your function as arguments
-        :param options: options that you want available to the function 
-            (accessed in Javascript through the ``options`` object)
-        """
-        fields = [QuerySet._translate_field_name(self._document, f)
-                  for f in fields]
-        collection = self._document._meta['collection']
-        scope = {
-            'collection': collection,
-            'query': self._query,
-            'options': options or {},
-        }
-        code = pymongo.code.Code(code, scope=scope)
-
-        db = _get_db()
-        return db.eval(code, *fields)
-
-    def sum(self, field):
-        """Sum over the values of the specified field.
-
-        :param field: the field to sum over; use dot-notation to refer to
-            embedded document fields
-        """
-        sum_func = """
-            function(sumField) {
-                var total = 0.0;
-                db[collection].find(query).forEach(function(doc) {
-                    total += (doc[sumField] || 0.0);
-                });
-                return total;
-            }
-        """
-        return self.exec_js(sum_func, field)
-
-    def average(self, field):
-        """Average over the values of the specified field.
-
-        :param field: the field to average over; use dot-notation to refer to
-            embedded document fields
-        """
-        average_func = """
-            function(averageField) {
-                var total = 0.0;
-                var num = 0;
-                db[collection].find(query).forEach(function(doc) {
-                    if (doc[averageField]) {
-                        total += doc[averageField];
-                        num += 1;
-                    }
-                });
-                return total / num;
-            }
-        """
-        return self.exec_js(average_func, field)
-
-    def item_frequencies(self, list_field, normalize=False):
-        """Returns a dictionary of all items present in a list field across
-        the whole queried set of documents, and their corresponding frequency.
-        This is useful for generating tag clouds, or searching documents. 
-
-        :param list_field: the list field to use
-        :param normalize: normalize the results so they add to 1.0
-        """
-        freq_func = """
-            function(listField) {
-                if (options.normalize) {
-                    var total = 0.0;
-                    db[collection].find(query).forEach(function(doc) {
-                        total += doc[listField].length;
-                    });
-                }
-
-                var frequencies = {};
-                var inc = 1.0;
-                if (options.normalize) {
-                    inc /= total;
-                }
-                db[collection].find(query).forEach(function(doc) {
-                    doc[listField].forEach(function(item) {
-                        frequencies[item] = inc + (frequencies[item] || 0);
-                    });
-                });
-                return frequencies;
-            }
-        """
-        return self.exec_js(freq_func, list_field, normalize=normalize)
-
-    def __repr__(self):
-        data = list(self[:REPR_OUTPUT_SIZE + 1])
-        if len(data) > REPR_OUTPUT_SIZE:
-            data[-1] = "...(remaining elements truncated)..."
-        return repr(data)
-
-class InvalidCollectionError(Exception):
-    pass
-
-
-class QuerySetManager(object):
-
-    def __init__(self, manager_func=None):
-        self._manager_func = manager_func
-        self._collection = None
-
-    def __get__(self, instance, owner):
-        """Descriptor for instantiating a new QuerySet object when 
-        Document.objects is accessed.
-        """
-        if instance is not None:
-            # Document class being used rather than a document object
-            return self
-
-        if self._collection is None:
-            db = _get_db()
-            collection = owner._meta['collection']
-
-            # Create collection as a capped collection if specified
-            if owner._meta['max_size'] or owner._meta['max_documents']:
-                # Get max document limit and max byte size from meta
-                max_size = owner._meta['max_size'] or 10000000 # 10MB default
-                max_documents = owner._meta['max_documents']
-
-                if collection in db.collection_names():
-                    self._collection = db[collection]
-                    # The collection already exists, check if its capped 
-                    # options match the specified capped options
-                    options = self._collection.options()
-                    if options.get('max') != max_documents or \
-                       options.get('size') != max_size:
-                        msg = ('Cannot create collection "%s" as a capped '
-                               'collection as it already exists') % collection
-                        raise InvalidCollectionError(msg)
-                else:
-                    # Create the collection as a capped collection
-                    opts = {'capped': True, 'size': max_size}
-                    if max_documents:
-                        opts['max'] = max_documents
-                    self._collection = db.create_collection(collection, opts)
-            else:
-                self._collection = db[collection]
-        
-        # owner is the document that contains the QuerySetManager
-        queryset = QuerySet(owner, self._collection)
-        if self._manager_func:
-            queryset = self._manager_func(queryset)
-        return queryset
-
-def queryset_manager(func):
-    """Decorator that allows you to define custom QuerySet managers on 
-    :class:`~mongoengine.Document` classes. The manager must be a function that
-    accepts a :class:`~mongoengine.queryset.QuerySet` as its only argument, and
-    returns a :class:`~mongoengine.queryset.QuerySet`, probably the same one 
-    but modified in some way.
-    """
-    return QuerySetManager(func)

mongoengine/queryset/__init__.py

@@ -0,0 +1,28 @@
+from mongoengine.errors import *
+from mongoengine.queryset.field_list import *
+from mongoengine.queryset.manager import *
+from mongoengine.queryset.queryset import *
+from mongoengine.queryset.transform import *
+from mongoengine.queryset.visitor import *
+
+# Expose just the public subset of all imported objects and constants.
+__all__ = (
+    "QuerySet",
+    "QuerySetNoCache",
+    "Q",
+    "queryset_manager",
+    "QuerySetManager",
+    "QueryFieldList",
+    "DO_NOTHING",
+    "NULLIFY",
+    "CASCADE",
+    "DENY",
+    "PULL",
+    # Errors that might be related to a queryset, mostly here for backward
+    # compatibility
+    "DoesNotExist",
+    "InvalidQueryError",
+    "MultipleObjectsReturned",
+    "NotUniqueError",
+    "OperationError",
+)

mongoengine/queryset/field_list.py

@@ -0,0 +1,88 @@
+__all__ = ("QueryFieldList",)
+
+
+class QueryFieldList:
+    """Object that handles combinations of .only() and .exclude() calls"""
+
+    ONLY = 1
+    EXCLUDE = 0
+
+    def __init__(
+        self, fields=None, value=ONLY, always_include=None, _only_called=False
+    ):
+        """The QueryFieldList builder
+
+        :param fields: A list of fields used in `.only()` or `.exclude()`
+        :param value: How to handle the fields; either `ONLY` or `EXCLUDE`
+        :param always_include: Any fields to always_include eg `_cls`
+        :param _only_called: Has `.only()` been called?  If so its a set of fields
+           otherwise it performs a union.
+        """
+        self.value = value
+        self.fields = set(fields or [])
+        self.always_include = set(always_include or [])
+        self._id = None
+        self._only_called = _only_called
+        self.slice = {}
+
+    def __add__(self, f):
+        if isinstance(f.value, dict):
+            for field in f.fields:
+                self.slice[field] = f.value
+            if not self.fields:
+                self.fields = f.fields
+        elif not self.fields:
+            self.fields = f.fields
+            self.value = f.value
+            self.slice = {}
+        elif self.value is self.ONLY and f.value is self.ONLY:
+            self._clean_slice()
+            if self._only_called:
+                self.fields = self.fields.union(f.fields)
+            else:
+                self.fields = f.fields
+        elif self.value is self.EXCLUDE and f.value is self.EXCLUDE:
+            self.fields = self.fields.union(f.fields)
+            self._clean_slice()
+        elif self.value is self.ONLY and f.value is self.EXCLUDE:
+            self.fields -= f.fields
+            self._clean_slice()
+        elif self.value is self.EXCLUDE and f.value is self.ONLY:
+            self.value = self.ONLY
+            self.fields = f.fields - self.fields
+            self._clean_slice()
+
+        if "_id" in f.fields:
+            self._id = f.value
+
+        if self.always_include:
+            if self.value is self.ONLY and self.fields:
+                if sorted(self.slice.keys()) != sorted(self.fields):
+                    self.fields = self.fields.union(self.always_include)
+            else:
+                self.fields -= self.always_include
+
+        if getattr(f, "_only_called", False):
+            self._only_called = True
+        return self
+
+    def __bool__(self):
+        return bool(self.fields)
+
+    def as_dict(self):
+        field_list = {field: self.value for field in self.fields}
+        if self.slice:
+            field_list.update(self.slice)
+        if self._id is not None:
+            field_list["_id"] = self._id
+        return field_list
+
+    def reset(self):
+        self.fields = set()
+        self.slice = {}
+        self.value = self.ONLY
+
+    def _clean_slice(self):
+        if self.slice:
+            for field in set(self.slice.keys()) - self.fields:
+                del self.slice[field]

mongoengine/queryset/manager.py

@@ -0,0 +1,57 @@
+from functools import partial
+from mongoengine.queryset.queryset import QuerySet
+
+__all__ = ("queryset_manager", "QuerySetManager")
+
+
+class QuerySetManager:
+    """
+    The default QuerySet Manager.
+
+    Custom QuerySet Manager functions can extend this class and users can
+    add extra queryset functionality.  Any custom manager methods must accept a
+    :class:`~mongoengine.Document` class as its first argument, and a
+    :class:`~mongoengine.queryset.QuerySet` as its second argument.
+
+    The method function should return a :class:`~mongoengine.queryset.QuerySet`
+    , probably the same one that was passed in, but modified in some way.
+    """
+
+    get_queryset = None
+    default = QuerySet
+
+    def __init__(self, queryset_func=None):
+        if queryset_func:
+            self.get_queryset = queryset_func
+
+    def __get__(self, instance, owner):
+        """Descriptor for instantiating a new QuerySet object when
+        Document.objects is accessed.
+        """
+        if instance is not None:
+            # Document object being used rather than a document class
+            return self
+
+        # owner is the document that contains the QuerySetManager
+        queryset_class = owner._meta.get("queryset_class", self.default)
+        queryset = queryset_class(owner, owner._get_collection())
+        if self.get_queryset:
+            arg_count = self.get_queryset.__code__.co_argcount
+            if arg_count == 1:
+                queryset = self.get_queryset(queryset)
+            elif arg_count == 2:
+                queryset = self.get_queryset(owner, queryset)
+            else:
+                queryset = partial(self.get_queryset, owner, queryset)
+        return queryset
+
+
+def queryset_manager(func):
+    """Decorator that allows you to define custom QuerySet managers on
+    :class:`~mongoengine.Document` classes. The manager must be a function that
+    accepts a :class:`~mongoengine.Document` class as its first argument, and a
+    :class:`~mongoengine.queryset.QuerySet` as its second argument. The method
+    function should return a :class:`~mongoengine.queryset.QuerySet`, probably
+    the same one that was passed in, but modified in some way.
+    """
+    return QuerySetManager(func)

mongoengine/queryset/queryset.py

@@ -0,0 +1,198 @@
+from mongoengine.errors import OperationError
+from mongoengine.queryset.base import (
+    BaseQuerySet,
+    CASCADE,
+    DENY,
+    DO_NOTHING,
+    NULLIFY,
+    PULL,
+)
+
+__all__ = (
+    "QuerySet",
+    "QuerySetNoCache",
+    "DO_NOTHING",
+    "NULLIFY",
+    "CASCADE",
+    "DENY",
+    "PULL",
+)
+
+# The maximum number of items to display in a QuerySet.__repr__
+REPR_OUTPUT_SIZE = 20
+ITER_CHUNK_SIZE = 100
+
+
+class QuerySet(BaseQuerySet):
+    """The default queryset, that builds queries and handles a set of results
+    returned from a query.
+
+    Wraps a MongoDB cursor, providing :class:`~mongoengine.Document` objects as
+    the results.
+    """
+
+    _has_more = True
+    _len = None
+    _result_cache = None
+
+    def __iter__(self):
+        """Iteration utilises a results cache which iterates the cursor
+        in batches of ``ITER_CHUNK_SIZE``.
+
+        If ``self._has_more`` the cursor hasn't been exhausted so cache then
+        batch. Otherwise iterate the result_cache.
+        """
+        self._iter = True
+
+        if self._has_more:
+            return self._iter_results()
+
+        # iterating over the cache.
+        return iter(self._result_cache)
+
+    def __len__(self):
+        """Since __len__ is called quite frequently (for example, as part of
+        list(qs)), we populate the result cache and cache the length.
+        """
+        if self._len is not None:
+            return self._len
+
+        # Populate the result cache with *all* of the docs in the cursor
+        if self._has_more:
+            list(self._iter_results())
+
+        # Cache the length of the complete result cache and return it
+        self._len = len(self._result_cache)
+        return self._len
+
+    def __repr__(self):
+        """Provide a string representation of the QuerySet"""
+        if self._iter:
+            return ".. queryset mid-iteration .."
+
+        self._populate_cache()
+        data = self._result_cache[: REPR_OUTPUT_SIZE + 1]
+        if len(data) > REPR_OUTPUT_SIZE:
+            data[-1] = "...(remaining elements truncated)..."
+        return repr(data)
+
+    def _iter_results(self):
+        """A generator for iterating over the result cache.
+
+        Also populates the cache if there are more possible results to
+        yield. Raises StopIteration when there are no more results.
+        """
+        if self._result_cache is None:
+            self._result_cache = []
+
+        pos = 0
+        while True:
+
+            # For all positions lower than the length of the current result
+            # cache, serve the docs straight from the cache w/o hitting the
+            # database.
+            # XXX it's VERY important to compute the len within the `while`
+            # condition because the result cache might expand mid-iteration
+            # (e.g. if we call len(qs) inside a loop that iterates over the
+            # queryset). Fortunately len(list) is O(1) in Python, so this
+            # doesn't cause performance issues.
+            while pos < len(self._result_cache):
+                yield self._result_cache[pos]
+                pos += 1
+
+            # return if we already established there were no more
+            # docs in the db cursor.
+            if not self._has_more:
+                return
+
+            # Otherwise, populate more of the cache and repeat.
+            if len(self._result_cache) <= pos:
+                self._populate_cache()
+
+    def _populate_cache(self):
+        """
+        Populates the result cache with ``ITER_CHUNK_SIZE`` more entries
+        (until the cursor is exhausted).
+        """
+        if self._result_cache is None:
+            self._result_cache = []
+
+        # Skip populating the cache if we already established there are no
+        # more docs to pull from the database.
+        if not self._has_more:
+            return
+
+        # Pull in ITER_CHUNK_SIZE docs from the database and store them in
+        # the result cache.
+        try:
+            for _ in range(ITER_CHUNK_SIZE):
+                self._result_cache.append(next(self))
+        except StopIteration:
+            # Getting this exception means there are no more docs in the
+            # db cursor. Set _has_more to False so that we can use that
+            # information in other places.
+            self._has_more = False
+
+    def count(self, with_limit_and_skip=False):
+        """Count the selected elements in the query.
+
+        :param with_limit_and_skip (optional): take any :meth:`limit` or
+            :meth:`skip` that has been applied to this cursor into account when
+            getting the count
+        """
+        if with_limit_and_skip is False:
+            return super().count(with_limit_and_skip)
+
+        if self._len is None:
+            self._len = super().count(with_limit_and_skip)
+
+        return self._len
+
+    def no_cache(self):
+        """Convert to a non-caching queryset
+
+        .. versionadded:: 0.8.3 Convert to non caching queryset
+        """
+        if self._result_cache is not None:
+            raise OperationError("QuerySet already cached")
+
+        return self._clone_into(QuerySetNoCache(self._document, self._collection))
+
+
+class QuerySetNoCache(BaseQuerySet):
+    """A non caching QuerySet"""
+
+    def cache(self):
+        """Convert to a caching queryset
+
+        .. versionadded:: 0.8.3 Convert to caching queryset
+        """
+        return self._clone_into(QuerySet(self._document, self._collection))
+
+    def __repr__(self):
+        """Provides the string representation of the QuerySet
+
+        .. versionchanged:: 0.6.13 Now doesnt modify the cursor
+        """
+        if self._iter:
+            return ".. queryset mid-iteration .."
+
+        data = []
+        for _ in range(REPR_OUTPUT_SIZE + 1):
+            try:
+                data.append(next(self))
+            except StopIteration:
+                break
+
+        if len(data) > REPR_OUTPUT_SIZE:
+            data[-1] = "...(remaining elements truncated)..."
+
+        self.rewind()
+        return repr(data)
+
+    def __iter__(self):
+        queryset = self
+        if queryset._iter:
+            queryset = self.clone()
+        queryset.rewind()
+        return queryset

mongoengine/queryset/transform.py

@@ -0,0 +1,502 @@
+from collections import defaultdict
+
+from bson import ObjectId, SON
+from bson.dbref import DBRef
+import pymongo
+
+from mongoengine.base import UPDATE_OPERATORS
+from mongoengine.common import _import_class
+from mongoengine.errors import InvalidQueryError
+
+__all__ = ("query", "update", "STRING_OPERATORS")
+
+COMPARISON_OPERATORS = (
+    "ne",
+    "gt",
+    "gte",
+    "lt",
+    "lte",
+    "in",
+    "nin",
+    "mod",
+    "all",
+    "size",
+    "exists",
+    "not",
+    "elemMatch",
+    "type",
+)
+GEO_OPERATORS = (
+    "within_distance",
+    "within_spherical_distance",
+    "within_box",
+    "within_polygon",
+    "near",
+    "near_sphere",
+    "max_distance",
+    "min_distance",
+    "geo_within",
+    "geo_within_box",
+    "geo_within_polygon",
+    "geo_within_center",
+    "geo_within_sphere",
+    "geo_intersects",
+)
+STRING_OPERATORS = (
+    "contains",
+    "icontains",
+    "startswith",
+    "istartswith",
+    "endswith",
+    "iendswith",
+    "exact",
+    "iexact",
+)
+CUSTOM_OPERATORS = ("match",)
+MATCH_OPERATORS = (
+    COMPARISON_OPERATORS + GEO_OPERATORS + STRING_OPERATORS + CUSTOM_OPERATORS
+)
+
+
+# TODO make this less complex
+def query(_doc_cls=None, **kwargs):
+    """Transform a query from Django-style format to Mongo format."""
+    mongo_query = {}
+    merge_query = defaultdict(list)
+    for key, value in sorted(kwargs.items()):
+        if key == "__raw__":
+            mongo_query.update(value)
+            continue
+
+        parts = key.rsplit("__")
+        indices = [(i, p) for i, p in enumerate(parts) if p.isdigit()]
+        parts = [part for part in parts if not part.isdigit()]
+        # Check for an operator and transform to mongo-style if there is
+        op = None
+        if len(parts) > 1 and parts[-1] in MATCH_OPERATORS:
+            op = parts.pop()
+
+        # Allow to escape operator-like field name by __
+        if len(parts) > 1 and parts[-1] == "":
+            parts.pop()
+
+        negate = False
+        if len(parts) > 1 and parts[-1] == "not":
+            parts.pop()
+            negate = True
+
+        if _doc_cls:
+            # Switch field names to proper names [set in Field(name='foo')]
+            try:
+                fields = _doc_cls._lookup_field(parts)
+            except Exception as e:
+                raise InvalidQueryError(e)
+            parts = []
+
+            CachedReferenceField = _import_class("CachedReferenceField")
+            GenericReferenceField = _import_class("GenericReferenceField")
+
+            cleaned_fields = []
+            for field in fields:
+                append_field = True
+                if isinstance(field, str):
+                    parts.append(field)
+                    append_field = False
+                # is last and CachedReferenceField
+                elif isinstance(field, CachedReferenceField) and fields[-1] == field:
+                    parts.append("%s._id" % field.db_field)
+                else:
+                    parts.append(field.db_field)
+
+                if append_field:
+                    cleaned_fields.append(field)
+
+            # Convert value to proper value
+            field = cleaned_fields[-1]
+
+            singular_ops = [None, "ne", "gt", "gte", "lt", "lte", "not"]
+            singular_ops += STRING_OPERATORS
+            if op in singular_ops:
+                value = field.prepare_query_value(op, value)
+
+                if isinstance(field, CachedReferenceField) and value:
+                    value = value["_id"]
+
+            elif op in ("in", "nin", "all", "near") and not isinstance(value, dict):
+                # Raise an error if the in/nin/all/near param is not iterable.
+                value = _prepare_query_for_iterable(field, op, value)
+
+            # If we're querying a GenericReferenceField, we need to alter the
+            # key depending on the value:
+            # * If the value is a DBRef, the key should be "field_name._ref".
+            # * If the value is an ObjectId, the key should be "field_name._ref.$id".
+            if isinstance(field, GenericReferenceField):
+                if isinstance(value, DBRef):
+                    parts[-1] += "._ref"
+                elif isinstance(value, ObjectId):
+                    parts[-1] += "._ref.$id"
+
+        # if op and op not in COMPARISON_OPERATORS:
+        if op:
+            if op in GEO_OPERATORS:
+                value = _geo_operator(field, op, value)
+            elif op in ("match", "elemMatch"):
+                ListField = _import_class("ListField")
+                EmbeddedDocumentField = _import_class("EmbeddedDocumentField")
+                if (
+                    isinstance(value, dict)
+                    and isinstance(field, ListField)
+                    and isinstance(field.field, EmbeddedDocumentField)
+                ):
+                    value = query(field.field.document_type, **value)
+                else:
+                    value = field.prepare_query_value(op, value)
+                value = {"$elemMatch": value}
+            elif op in CUSTOM_OPERATORS:
+                NotImplementedError(
+                    'Custom method "%s" has not ' "been implemented" % op
+                )
+            elif op not in STRING_OPERATORS:
+                value = {"$" + op: value}
+
+        if negate:
+            value = {"$not": value}
+
+        for i, part in indices:
+            parts.insert(i, part)
+
+        key = ".".join(parts)
+
+        if key not in mongo_query:
+            mongo_query[key] = value
+        else:
+            if isinstance(mongo_query[key], dict) and isinstance(value, dict):
+                mongo_query[key].update(value)
+                # $max/minDistance needs to come last - convert to SON
+                value_dict = mongo_query[key]
+                if ("$maxDistance" in value_dict or "$minDistance" in value_dict) and (
+                    "$near" in value_dict or "$nearSphere" in value_dict
+                ):
+                    value_son = SON()
+                    for k, v in value_dict.items():
+                        if k == "$maxDistance" or k == "$minDistance":
+                            continue
+                        value_son[k] = v
+                    # Required for MongoDB >= 2.6, may fail when combining
+                    # PyMongo 3+ and MongoDB < 2.6
+                    near_embedded = False
+                    for near_op in ("$near", "$nearSphere"):
+                        if isinstance(value_dict.get(near_op), dict):
+                            value_son[near_op] = SON(value_son[near_op])
+                            if "$maxDistance" in value_dict:
+                                value_son[near_op]["$maxDistance"] = value_dict[
+                                    "$maxDistance"
+                                ]
+                            if "$minDistance" in value_dict:
+                                value_son[near_op]["$minDistance"] = value_dict[
+                                    "$minDistance"
+                                ]
+                            near_embedded = True
+
+                    if not near_embedded:
+                        if "$maxDistance" in value_dict:
+                            value_son["$maxDistance"] = value_dict["$maxDistance"]
+                        if "$minDistance" in value_dict:
+                            value_son["$minDistance"] = value_dict["$minDistance"]
+                    mongo_query[key] = value_son
+            else:
+                # Store for manually merging later
+                merge_query[key].append(value)
+
+    # The queryset has been filter in such a way we must manually merge
+    for k, v in merge_query.items():
+        merge_query[k].append(mongo_query[k])
+        del mongo_query[k]
+        if isinstance(v, list):
+            value = [{k: val} for val in v]
+            if "$and" in mongo_query.keys():
+                mongo_query["$and"].extend(value)
+            else:
+                mongo_query["$and"] = value
+
+    return mongo_query
+
+
+def update(_doc_cls=None, **update):
+    """Transform an update spec from Django-style format to Mongo
+    format.
+    """
+    mongo_update = {}
+
+    for key, value in update.items():
+        if key == "__raw__":
+            mongo_update.update(value)
+            continue
+
+        parts = key.split("__")
+
+        # if there is no operator, default to 'set'
+        if len(parts) < 3 and parts[0] not in UPDATE_OPERATORS:
+            parts.insert(0, "set")
+
+        # Check for an operator and transform to mongo-style if there is
+        op = None
+        if parts[0] in UPDATE_OPERATORS:
+            op = parts.pop(0)
+            # Convert Pythonic names to Mongo equivalents
+            operator_map = {
+                "push_all": "pushAll",
+                "pull_all": "pullAll",
+                "dec": "inc",
+                "add_to_set": "addToSet",
+                "set_on_insert": "setOnInsert",
+            }
+            if op == "dec":
+                # Support decrement by flipping a positive value's sign
+                # and using 'inc'
+                value = -value
+            # If the operator doesn't found from operator map, the op value
+            # will stay unchanged
+            op = operator_map.get(op, op)
+
+        match = None
+        if parts[-1] in COMPARISON_OPERATORS:
+            match = parts.pop()
+
+        # Allow to escape operator-like field name by __
+        if len(parts) > 1 and parts[-1] == "":
+            parts.pop()
+
+        if _doc_cls:
+            # Switch field names to proper names [set in Field(name='foo')]
+            try:
+                fields = _doc_cls._lookup_field(parts)
+            except Exception as e:
+                raise InvalidQueryError(e)
+            parts = []
+
+            cleaned_fields = []
+            appended_sub_field = False
+            for field in fields:
+                append_field = True
+                if isinstance(field, str):
+                    # Convert the S operator to $
+                    if field == "S":
+                        field = "$"
+                    parts.append(field)
+                    append_field = False
+                else:
+                    parts.append(field.db_field)
+                if append_field:
+                    appended_sub_field = False
+                    cleaned_fields.append(field)
+                    if hasattr(field, "field"):
+                        cleaned_fields.append(field.field)
+                        appended_sub_field = True
+
+            # Convert value to proper value
+            if appended_sub_field:
+                field = cleaned_fields[-2]
+            else:
+                field = cleaned_fields[-1]
+
+            GeoJsonBaseField = _import_class("GeoJsonBaseField")
+            if isinstance(field, GeoJsonBaseField):
+                value = field.to_mongo(value)
+
+            if op == "pull":
+                if field.required or value is not None:
+                    if match in ("in", "nin") and not isinstance(value, dict):
+                        value = _prepare_query_for_iterable(field, op, value)
+                    else:
+                        value = field.prepare_query_value(op, value)
+            elif op == "push" and isinstance(value, (list, tuple, set)):
+                value = [field.prepare_query_value(op, v) for v in value]
+            elif op in (None, "set", "push"):
+                if field.required or value is not None:
+                    value = field.prepare_query_value(op, value)
+            elif op in ("pushAll", "pullAll"):
+                value = [field.prepare_query_value(op, v) for v in value]
+            elif op in ("addToSet", "setOnInsert"):
+                if isinstance(value, (list, tuple, set)):
+                    value = [field.prepare_query_value(op, v) for v in value]
+                elif field.required or value is not None:
+                    value = field.prepare_query_value(op, value)
+            elif op == "unset":
+                value = 1
+            elif op == "inc":
+                value = field.prepare_query_value(op, value)
+
+        if match:
+            match = "$" + match
+            value = {match: value}
+
+        key = ".".join(parts)
+
+        if "pull" in op and "." in key:
+            # Dot operators don't work on pull operations
+            # unless they point to a list field
+            # Otherwise it uses nested dict syntax
+            if op == "pullAll":
+                raise InvalidQueryError(
+                    "pullAll operations only support a single field depth"
+                )
+
+            # Look for the last list field and use dot notation until there
+            field_classes = [c.__class__ for c in cleaned_fields]
+            field_classes.reverse()
+            ListField = _import_class("ListField")
+            EmbeddedDocumentListField = _import_class("EmbeddedDocumentListField")
+            if ListField in field_classes or EmbeddedDocumentListField in field_classes:
+                # Join all fields via dot notation to the last ListField or EmbeddedDocumentListField
+                # Then process as normal
+                if ListField in field_classes:
+                    _check_field = ListField
+                else:
+                    _check_field = EmbeddedDocumentListField
+
+                last_listField = len(cleaned_fields) - field_classes.index(_check_field)
+                key = ".".join(parts[:last_listField])
+                parts = parts[last_listField:]
+                parts.insert(0, key)
+
+            parts.reverse()
+            for key in parts:
+                value = {key: value}
+        elif op == "addToSet" and isinstance(value, list):
+            value = {key: {"$each": value}}
+        elif op in ("push", "pushAll"):
+            if parts[-1].isdigit():
+                key = ".".join(parts[0:-1])
+                position = int(parts[-1])
+                # $position expects an iterable. If pushing a single value,
+                # wrap it in a list.
+                if not isinstance(value, (set, tuple, list)):
+                    value = [value]
+                value = {key: {"$each": value, "$position": position}}
+            else:
+                if op == "pushAll":
+                    op = "push"  # convert to non-deprecated keyword
+                    if not isinstance(value, (set, tuple, list)):
+                        value = [value]
+                    value = {key: {"$each": value}}
+                else:
+                    value = {key: value}
+        else:
+            value = {key: value}
+        key = "$" + op
+        if key not in mongo_update:
+            mongo_update[key] = value
+        elif key in mongo_update and isinstance(mongo_update[key], dict):
+            mongo_update[key].update(value)
+
+    return mongo_update
+
+
+def _geo_operator(field, op, value):
+    """Helper to return the query for a given geo query."""
+    if op == "max_distance":
+        value = {"$maxDistance": value}
+    elif op == "min_distance":
+        value = {"$minDistance": value}
+    elif field._geo_index == pymongo.GEO2D:
+        if op == "within_distance":
+            value = {"$within": {"$center": value}}
+        elif op == "within_spherical_distance":
+            value = {"$within": {"$centerSphere": value}}
+        elif op == "within_polygon":
+            value = {"$within": {"$polygon": value}}
+        elif op == "near":
+            value = {"$near": value}
+        elif op == "near_sphere":
+            value = {"$nearSphere": value}
+        elif op == "within_box":
+            value = {"$within": {"$box": value}}
+        else:
+            raise NotImplementedError(
+                'Geo method "%s" has not been ' "implemented for a GeoPointField" % op
+            )
+    else:
+        if op == "geo_within":
+            value = {"$geoWithin": _infer_geometry(value)}
+        elif op == "geo_within_box":
+            value = {"$geoWithin": {"$box": value}}
+        elif op == "geo_within_polygon":
+            value = {"$geoWithin": {"$polygon": value}}
+        elif op == "geo_within_center":
+            value = {"$geoWithin": {"$center": value}}
+        elif op == "geo_within_sphere":
+            value = {"$geoWithin": {"$centerSphere": value}}
+        elif op == "geo_intersects":
+            value = {"$geoIntersects": _infer_geometry(value)}
+        elif op == "near":
+            value = {"$near": _infer_geometry(value)}
+        else:
+            raise NotImplementedError(
+                'Geo method "{}" has not been implemented for a {} '.format(
+                    op, field._name
+                )
+            )
+    return value
+
+
+def _infer_geometry(value):
+    """Helper method that tries to infer the $geometry shape for a
+    given value.
+    """
+    if isinstance(value, dict):
+        if "$geometry" in value:
+            return value
+        elif "coordinates" in value and "type" in value:
+            return {"$geometry": value}
+        raise InvalidQueryError(
+            "Invalid $geometry dictionary should have type and coordinates keys"
+        )
+    elif isinstance(value, (list, set)):
+        # TODO: shouldn't we test value[0][0][0][0] to see if it is MultiPolygon?
+
+        try:
+            value[0][0][0]
+            return {"$geometry": {"type": "Polygon", "coordinates": value}}
+        except (TypeError, IndexError):
+            pass
+
+        try:
+            value[0][0]
+            return {"$geometry": {"type": "LineString", "coordinates": value}}
+        except (TypeError, IndexError):
+            pass
+
+        try:
+            value[0]
+            return {"$geometry": {"type": "Point", "coordinates": value}}
+        except (TypeError, IndexError):
+            pass
+
+    raise InvalidQueryError(
+        "Invalid $geometry data. Can be either a "
+        "dictionary or (nested) lists of coordinate(s)"
+    )
+
+
+def _prepare_query_for_iterable(field, op, value):
+    # We need a special check for BaseDocument, because - although it's iterable - using
+    # it as such in the context of this method is most definitely a mistake.
+    BaseDocument = _import_class("BaseDocument")
+
+    if isinstance(value, BaseDocument):
+        raise TypeError(
+            "When using the `in`, `nin`, `all`, or "
+            "`near`-operators you can't use a "
+            "`Document`, you must wrap your object "
+            "in a list (object -> [object])."
+        )
+
+    if not hasattr(value, "__iter__"):
+        raise TypeError(
+            "The `in`, `nin`, `all`, or "
+            "`near`-operators must be applied to an "
+            "iterable (e.g. a list)."
+        )
+
+    return [field.prepare_query_value(op, v) for v in value]

mongoengine/queryset/visitor.py

@@ -0,0 +1,189 @@
+import copy
+import warnings
+
+from mongoengine.errors import InvalidQueryError
+from mongoengine.queryset import transform
+
+__all__ = ("Q", "QNode")
+
+
+class QNodeVisitor:
+    """Base visitor class for visiting Q-object nodes in a query tree.
+    """
+
+    def visit_combination(self, combination):
+        """Called by QCombination objects.
+        """
+        return combination
+
+    def visit_query(self, query):
+        """Called by (New)Q objects.
+        """
+        return query
+
+
+class DuplicateQueryConditionsError(InvalidQueryError):
+    pass
+
+
+class SimplificationVisitor(QNodeVisitor):
+    """Simplifies query trees by combining unnecessary 'and' connection nodes
+    into a single Q-object.
+    """
+
+    def visit_combination(self, combination):
+        if combination.operation == combination.AND:
+            # The simplification only applies to 'simple' queries
+            if all(isinstance(node, Q) for node in combination.children):
+                queries = [n.query for n in combination.children]
+                try:
+                    return Q(**self._query_conjunction(queries))
+                except DuplicateQueryConditionsError:
+                    # Cannot be simplified
+                    pass
+        return combination
+
+    def _query_conjunction(self, queries):
+        """Merges query dicts - effectively &ing them together.
+        """
+        query_ops = set()
+        combined_query = {}
+        for query in queries:
+            ops = set(query.keys())
+            # Make sure that the same operation isn't applied more than once
+            # to a single field
+            intersection = ops.intersection(query_ops)
+            if intersection:
+                raise DuplicateQueryConditionsError()
+
+            query_ops.update(ops)
+            combined_query.update(copy.deepcopy(query))
+        return combined_query
+
+
+class QueryCompilerVisitor(QNodeVisitor):
+    """Compiles the nodes in a query tree to a PyMongo-compatible query
+    dictionary.
+    """
+
+    def __init__(self, document):
+        self.document = document
+
+    def visit_combination(self, combination):
+        operator = "$and"
+        if combination.operation == combination.OR:
+            operator = "$or"
+        return {operator: combination.children}
+
+    def visit_query(self, query):
+        return transform.query(self.document, **query.query)
+
+
+class QNode:
+    """Base class for nodes in query trees."""
+
+    AND = 0
+    OR = 1
+
+    def to_query(self, document):
+        query = self.accept(SimplificationVisitor())
+        query = query.accept(QueryCompilerVisitor(document))
+        return query
+
+    def accept(self, visitor):
+        raise NotImplementedError
+
+    def _combine(self, other, operation):
+        """Combine this node with another node into a QCombination
+        object.
+        """
+        # If the other Q() is empty, ignore it and just use `self`.
+        if getattr(other, "empty", True):
+            return self
+
+        # Or if this Q is empty, ignore it and just use `other`.
+        if self.empty:
+            return other
+
+        return QCombination(operation, [self, other])
+
+    @property
+    def empty(self):
+        msg = "'empty' property is deprecated in favour of using 'not bool(filter)'"
+        warnings.warn(msg, DeprecationWarning)
+        return False
+
+    def __or__(self, other):
+        return self._combine(other, self.OR)
+
+    def __and__(self, other):
+        return self._combine(other, self.AND)
+
+
+class QCombination(QNode):
+    """Represents the combination of several conditions by a given
+    logical operator.
+    """
+
+    def __init__(self, operation, children):
+        self.operation = operation
+        self.children = []
+        for node in children:
+            # If the child is a combination of the same type, we can merge its
+            # children directly into this combinations children
+            if isinstance(node, QCombination) and node.operation == operation:
+                self.children += node.children
+            else:
+                self.children.append(node)
+
+    def __repr__(self):
+        op = " & " if self.operation is self.AND else " | "
+        return "(%s)" % op.join([repr(node) for node in self.children])
+
+    def __bool__(self):
+        return bool(self.children)
+
+    def accept(self, visitor):
+        for i in range(len(self.children)):
+            if isinstance(self.children[i], QNode):
+                self.children[i] = self.children[i].accept(visitor)
+
+        return visitor.visit_combination(self)
+
+    @property
+    def empty(self):
+        msg = "'empty' property is deprecated in favour of using 'not bool(filter)'"
+        warnings.warn(msg, DeprecationWarning)
+        return not bool(self.children)
+
+    def __eq__(self, other):
+        return (
+            self.__class__ == other.__class__
+            and self.operation == other.operation
+            and self.children == other.children
+        )
+
+
+class Q(QNode):
+    """A simple query object, used in a query tree to build up more complex
+    query structures.
+    """
+
+    def __init__(self, **query):
+        self.query = query
+
+    def __repr__(self):
+        return "Q(**%s)" % repr(self.query)
+
+    def __bool__(self):
+        return bool(self.query)
+
+    def __eq__(self, other):
+        return self.__class__ == other.__class__ and self.query == other.query
+
+    def accept(self, visitor):
+        return visitor.visit_query(self)
+
+    @property
+    def empty(self):
+        return not bool(self.query)

mongoengine/signals.py

@@ -0,0 +1,59 @@
+__all__ = (
+    "pre_init",
+    "post_init",
+    "pre_save",
+    "pre_save_post_validation",
+    "post_save",
+    "pre_delete",
+    "post_delete",
+)
+
+signals_available = False
+try:
+    from blinker import Namespace
+
+    signals_available = True
+except ImportError:
+
+    class Namespace:
+        def signal(self, name, doc=None):
+            return _FakeSignal(name, doc)
+
+    class _FakeSignal:
+        """If blinker is unavailable, create a fake class with the same
+        interface that allows sending of signals but will fail with an
+        error on anything else.  Instead of doing anything on send, it
+        will just ignore the arguments and do nothing instead.
+        """
+
+        def __init__(self, name, doc=None):
+            self.name = name
+            self.__doc__ = doc
+
+        def _fail(self, *args, **kwargs):
+            raise RuntimeError(
+                "signalling support is unavailable "
+                "because the blinker library is "
+                "not installed."
+            )
+
+        send = lambda *a, **kw: None  # noqa
+        connect = (
+            disconnect
+        ) = has_receivers_for = receivers_for = temporarily_connected_to = _fail
+        del _fail
+
+
+# the namespace for code signals.  If you are not mongoengine code, do
+# not put signals in here.  Create your own namespace instead.
+_signals = Namespace()
+
+pre_init = _signals.signal("pre_init")
+post_init = _signals.signal("post_init")
+pre_save = _signals.signal("pre_save")
+pre_save_post_validation = _signals.signal("pre_save_post_validation")
+post_save = _signals.signal("post_save")
+pre_delete = _signals.signal("pre_delete")
+post_delete = _signals.signal("post_delete")
+pre_bulk_insert = _signals.signal("pre_bulk_insert")
+post_bulk_insert = _signals.signal("post_bulk_insert")

python-mongoengine.spec

@@ -0,0 +1,54 @@
+# sitelib for noarch packages, sitearch for others (remove the unneeded one)
+%{!?python_sitelib: %global python_sitelib %(%{__python} -c "from distutils.sysconfig import get_python_lib; print(get_python_lib())")}
+%{!?python_sitearch: %global python_sitearch %(%{__python} -c "from distutils.sysconfig import get_python_lib; print(get_python_lib(1))")}
+
+%define srcname mongoengine
+
+Name:           python-%{srcname}
+Version:        0.8.7
+Release:        1%{?dist}
+Summary:        A Python Document-Object Mapper for working with MongoDB
+
+Group:          Development/Libraries
+License:        MIT
+URL:            https://github.com/MongoEngine/mongoengine
+Source0:        %{srcname}-%{version}.tar.bz2
+
+BuildRequires:  python-devel
+BuildRequires:  python-setuptools
+
+Requires:       mongodb
+Requires:       pymongo
+Requires:       python-blinker
+Requires:       python-imaging
+
+
+%description
+MongoEngine is an ORM-like layer on top of PyMongo.
+
+%prep
+%setup -q -n %{srcname}-%{version}
+
+
+%build
+# Remove CFLAGS=... for noarch packages (unneeded)
+CFLAGS="$RPM_OPT_FLAGS" %{__python} setup.py build
+
+
+%install
+rm -rf $RPM_BUILD_ROOT
+%{__python} setup.py install -O1 --skip-build --root $RPM_BUILD_ROOT
+
+%clean
+rm -rf $RPM_BUILD_ROOT
+
+%files
+%defattr(-,root,root,-)
+%doc docs AUTHORS LICENSE README.rst
+# For noarch packages: sitelib
+ %{python_sitelib}/*
+# For arch-specific packages: sitearch
+# %{python_sitearch}/*
+
+%changelog
+* See: http://docs.mongoengine.org/en/latest/changelog.html

requirements-dev.txt

@@ -0,0 +1,8 @@
+black
+flake8
+flake8-import-order
+pre-commit
+pytest
+ipdb
+ipython
+tox

requirements.txt

@@ -0,0 +1,3 @@
+pymongo>=3.4
+Sphinx==1.5.5
+sphinx-rtd-theme==0.2.4

setup.cfg

@@ -0,0 +1,10 @@
+[flake8]
+ignore=E501,F401,F403,F405,I201,I202,W504, W605, W503
+exclude=build,dist,docs,venv,venv3,.tox,.eggs,tests
+max-complexity=47
+application-import-names=mongoengine,tests
+
+[tool:pytest]
+# Limits the discovery to tests directory
+# avoids that it runs for instance the benchmark
+testpaths = tests

setup.py

@@ -1,50 +1,147 @@
-from setuptools import setup, find_packages
 import os
+import sys
 
-DESCRIPTION = "A Python Document-Object Mapper for working with MongoDB"
+from pkg_resources import normalize_path
+from setuptools import find_packages, setup
+from setuptools.command.test import test as TestCommand
 
-LONG_DESCRIPTION = None
+# Hack to silence atexit traceback in newer python versions
 try:
-    LONG_DESCRIPTION = open('README.rst').read()
-except:
+    import multiprocessing
+except ImportError:
     pass
 
+DESCRIPTION = "MongoEngine is a Python Object-Document Mapper for working with MongoDB."
+
+try:
+    with open("README.rst") as fin:
+        LONG_DESCRIPTION = fin.read()
+except Exception:
+    LONG_DESCRIPTION = None
+
+
 def get_version(version_tuple):
-    version = '%s.%s' % (version_tuple[0], version_tuple[1])
-    if version_tuple[2]:
-        version = '%s.%s' % (version, version_tuple[2])
-    return version
+    """Return the version tuple as a string, e.g. for (0, 10, 7),
+    return '0.10.7'.
+    """
+    return ".".join(map(str, version_tuple))
+
+
+class PyTest(TestCommand):
+    """Will force pytest to search for tests inside the build directory
+    for 2to3 converted code (used by tox), instead of the current directory.
+    Required as long as we need 2to3
+
+    Known Limitation: https://tox.readthedocs.io/en/latest/example/pytest.html#known-issues-and-limitations
+    Source: https://www.hackzine.org/python-testing-with-pytest-and-2to3-plus-tox-and-travis-ci.html
+    """
+
+    # https://pytest.readthedocs.io/en/2.7.3/goodpractises.html#integration-with-setuptools-test-commands
+    # Allows to provide pytest command argument through the test runner command `python setup.py test`
+    # e.g: `python setup.py test -a "-k=test"`
+    # This only works for 1 argument though
+    user_options = [("pytest-args=", "a", "Arguments to pass to py.test")]
+
+    def initialize_options(self):
+        TestCommand.initialize_options(self)
+        self.pytest_args = ""
+
+    def finalize_options(self):
+        TestCommand.finalize_options(self)
+        self.test_args = ["tests"]
+        self.test_suite = True
+
+    def run_tests(self):
+        # import here, cause outside the eggs aren't loaded
+        from pkg_resources import _namespace_packages
+        import pytest
+
+        # Purge modules under test from sys.modules. The test loader will
+        # re-import them from the build location. Required when 2to3 is used
+        # with namespace packages.
+        if sys.version_info >= (3,) and getattr(self.distribution, "use_2to3", False):
+            module = self.test_args[-1].split(".")[0]
+            if module in _namespace_packages:
+                del_modules = []
+                if module in sys.modules:
+                    del_modules.append(module)
+                module += "."
+                for name in sys.modules:
+                    if name.startswith(module):
+                        del_modules.append(name)
+                map(sys.modules.__delitem__, del_modules)
+
+            # Run on the build directory for 2to3-built code
+            # This will prevent the old 2.x code from being found
+            # by py.test discovery mechanism, that apparently
+            # ignores sys.path..
+            ei_cmd = self.get_finalized_command("egg_info")
+            self.test_args = [normalize_path(ei_cmd.egg_base)]
+
+        cmd_args = self.test_args + ([self.pytest_args] if self.pytest_args else [])
+        errno = pytest.main(cmd_args)
+        sys.exit(errno)
+
 
 # Dirty hack to get version number from monogengine/__init__.py - we can't
 # import it as it depends on PyMongo and PyMongo isn't installed until this
 # file is read
-init = os.path.join(os.path.dirname(__file__), 'mongoengine', '__init__.py')
-version_line = filter(lambda l: l.startswith('VERSION'), open(init))[0]
-VERSION = get_version(eval(version_line.split('=')[-1]))
-print VERSION
+init = os.path.join(os.path.dirname(__file__), "mongoengine", "__init__.py")
+version_line = list(filter(lambda l: l.startswith("VERSION"), open(init)))[0]
+
+VERSION = get_version(eval(version_line.split("=")[-1]))
 
 CLASSIFIERS = [
-    'Development Status :: 4 - Beta',
-    'Intended Audience :: Developers',
-    'License :: OSI Approved :: MIT License',
-    'Operating System :: OS Independent',
-    'Programming Language :: Python',
-    'Topic :: Database',
-    'Topic :: Software Development :: Libraries :: Python Modules',
+    "Development Status :: 5 - Production/Stable",
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.5",
+    "Programming Language :: Python :: 3.6",
+    "Programming Language :: Python :: 3.7",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: Implementation :: CPython",
+    "Programming Language :: Python :: Implementation :: PyPy",
+    "Topic :: Database",
+    "Topic :: Software Development :: Libraries :: Python Modules",
 ]
 
-setup(name='mongoengine',
+extra_opts = {
+    "packages": find_packages(exclude=["tests", "tests.*"]),
+    "tests_require": [
+        "pytest<5.0",
+        "pytest-cov",
+        "coverage<5.0",  # recent coverage switched to sqlite format for the .coverage file which isn't handled properly by coveralls
+        "blinker",
+        "Pillow>=2.0.0, <7.0.0",  # 7.0.0 dropped Python2 support
+    ],
+}
+
+if "test" in sys.argv:
+    extra_opts["packages"] = find_packages()
+    extra_opts["package_data"] = {
+        "tests": ["fields/mongoengine.png", "fields/mongodb_leaf.png"]
+    }
+
+setup(
+    name="mongoengine",
     version=VERSION,
-      packages=find_packages(),
-      author='Harry Marr',
-      author_email='harry.marr@{nospam}gmail.com',
-      url='http://hmarr.com/mongoengine/',
-      license='MIT',
+    author="Harry Marr",
+    author_email="harry.marr@gmail.com",
+    maintainer="Stefan Wojcik",
+    maintainer_email="wojcikstefan@gmail.com",
+    url="http://mongoengine.org/",
+    download_url="https://github.com/MongoEngine/mongoengine/tarball/master",
+    license="MIT",
     include_package_data=True,
     description=DESCRIPTION,
     long_description=LONG_DESCRIPTION,
-      platforms=['any'],
+    platforms=["any"],
     classifiers=CLASSIFIERS,
-      install_requires=['pymongo'],
-      test_suite='tests',
+    python_requires=">=3.5",
+    install_requires=["pymongo>=3.4, <4.0"],
+    cmdclass={"test": PyTest},
+    **extra_opts
 )

tests/all_warnings/test_warnings.py

@@ -0,0 +1,35 @@
+"""
+This test has been put into a module.  This is because it tests warnings that
+only get triggered on first hit.  This way we can ensure its imported into the
+top level and called first by the test suite.
+"""
+import unittest
+import warnings
+
+from mongoengine import *
+
+
+class TestAllWarnings(unittest.TestCase):
+    def setUp(self):
+        connect(db="mongoenginetest")
+        self.warning_list = []
+        self.showwarning_default = warnings.showwarning
+        warnings.showwarning = self.append_to_warning_list
+
+    def append_to_warning_list(self, message, category, *args):
+        self.warning_list.append({"message": message, "category": category})
+
+    def tearDown(self):
+        # restore default handling of warnings
+        warnings.showwarning = self.showwarning_default
+
+    def test_document_collection_syntax_warning(self):
+        class NonAbstractBase(Document):
+            meta = {"allow_inheritance": True}
+
+        class InheritedDocumentFailTest(NonAbstractBase):
+            meta = {"collection": "fail"}
+
+        warning = self.warning_list[0]
+        assert SyntaxWarning == warning["category"]
+        assert "non_abstract_base" == InheritedDocumentFailTest._get_collection_name()

tests/document.py

@@ -1,541 +0,0 @@
-import unittest
-import datetime
-import pymongo
-
-from mongoengine import *
-from mongoengine.connection import _get_db
-
-
-class DocumentTest(unittest.TestCase):
-    
-    def setUp(self):
-        connect(db='mongoenginetest')
-        self.db = _get_db()
-
-        class Person(Document):
-            name = StringField()
-            age = IntField()
-        self.Person = Person
-
-    def test_drop_collection(self):
-        """Ensure that the collection may be dropped from the database.
-        """
-        self.Person(name='Test').save()
-
-        collection = self.Person._meta['collection']
-        self.assertTrue(collection in self.db.collection_names())
-
-        self.Person.drop_collection()
-        self.assertFalse(collection in self.db.collection_names())
-
-    def test_definition(self):
-        """Ensure that document may be defined using fields.
-        """
-        name_field = StringField()
-        age_field = IntField()
-
-        class Person(Document):
-            name = name_field
-            age = age_field
-            non_field = True
-        
-        self.assertEqual(Person._fields['name'], name_field)
-        self.assertEqual(Person._fields['age'], age_field)
-        self.assertFalse('non_field' in Person._fields)
-        self.assertTrue('id' in Person._fields)
-        # Test iteration over fields
-        fields = list(Person())
-        self.assertTrue('name' in fields and 'age' in fields)
-        # Ensure Document isn't treated like an actual document
-        self.assertFalse(hasattr(Document, '_fields'))
-
-    def test_get_superclasses(self):
-        """Ensure that the correct list of superclasses is assembled.
-        """
-        class Animal(Document): pass
-        class Fish(Animal): pass
-        class Mammal(Animal): pass
-        class Human(Mammal): pass
-        class Dog(Mammal): pass
-
-        mammal_superclasses = {'Animal': Animal}
-        self.assertEqual(Mammal._superclasses, mammal_superclasses)
-        
-        dog_superclasses = {
-            'Animal': Animal,
-            'Animal.Mammal': Mammal,
-        }
-        self.assertEqual(Dog._superclasses, dog_superclasses)
-
-    def test_get_subclasses(self):
-        """Ensure that the correct list of subclasses is retrieved by the 
-        _get_subclasses method.
-        """
-        class Animal(Document): pass
-        class Fish(Animal): pass
-        class Mammal(Animal): pass
-        class Human(Mammal): pass
-        class Dog(Mammal): pass
-
-        mammal_subclasses = {
-            'Animal.Mammal.Dog': Dog, 
-            'Animal.Mammal.Human': Human
-        }
-        self.assertEqual(Mammal._get_subclasses(), mammal_subclasses)
-        
-        animal_subclasses = {
-            'Animal.Fish': Fish,
-            'Animal.Mammal': Mammal,
-            'Animal.Mammal.Dog': Dog, 
-            'Animal.Mammal.Human': Human
-        }
-        self.assertEqual(Animal._get_subclasses(), animal_subclasses)
-
-    def test_polymorphic_queries(self):
-        """Ensure that the correct subclasses are returned from a query"""
-        class Animal(Document): pass
-        class Fish(Animal): pass
-        class Mammal(Animal): pass
-        class Human(Mammal): pass
-        class Dog(Mammal): pass
-
-        Animal().save()
-        Fish().save()
-        Mammal().save()
-        Human().save()
-        Dog().save()
-
-        classes = [obj.__class__ for obj in Animal.objects]
-        self.assertEqual(classes, [Animal, Fish, Mammal, Human, Dog])
-
-        classes = [obj.__class__ for obj in Mammal.objects]
-        self.assertEqual(classes, [Mammal, Human, Dog])
-
-        classes = [obj.__class__ for obj in Human.objects]
-        self.assertEqual(classes, [Human])
-
-        Animal.drop_collection()
-
-    def test_inheritance(self):
-        """Ensure that document may inherit fields from a superclass document.
-        """
-        class Employee(self.Person):
-            salary = IntField()
-
-        self.assertTrue('name' in Employee._fields)
-        self.assertTrue('salary' in Employee._fields)
-        self.assertEqual(Employee._meta['collection'], 
-                         self.Person._meta['collection'])
-
-    def test_allow_inheritance(self):
-        """Ensure that inheritance may be disabled on simple classes and that
-        _cls and _types will not be used.
-        """
-        class Animal(Document):
-            meta = {'allow_inheritance': False}
-            name = StringField()
-
-        Animal.drop_collection()
-
-        def create_dog_class():
-            class Dog(Animal):
-                pass
-        self.assertRaises(ValueError, create_dog_class)
-        
-        # Check that _cls etc aren't present on simple documents
-        dog = Animal(name='dog')
-        dog.save()
-        collection = self.db[Animal._meta['collection']]
-        obj = collection.find_one()
-        self.assertFalse('_cls' in obj)
-        self.assertFalse('_types' in obj)
-
-        Animal.drop_collection()
-
-        def create_employee_class():
-            class Employee(self.Person):
-                meta = {'allow_inheritance': False}
-        self.assertRaises(ValueError, create_employee_class)
-
-    def test_collection_name(self):
-        """Ensure that a collection with a specified name may be used.
-        """
-        collection = 'personCollTest'
-        if collection in self.db.collection_names():
-            self.db.drop_collection(collection)
-
-        class Person(Document):
-            name = StringField()
-            meta = {'collection': collection}
-        
-        user = Person(name="Test User")
-        user.save()
-        self.assertTrue(collection in self.db.collection_names())
-
-        user_obj = self.db[collection].find_one()
-        self.assertEqual(user_obj['name'], "Test User")
-
-        user_obj = Person.objects[0]
-        self.assertEqual(user_obj.name, "Test User")
-
-        Person.drop_collection()
-        self.assertFalse(collection in self.db.collection_names())
-
-    def test_capped_collection(self):
-        """Ensure that capped collections work properly.
-        """
-        class Log(Document):
-            date = DateTimeField(default=datetime.datetime.now)
-            meta = {
-                'max_documents': 10,
-                'max_size': 90000,
-            }
-
-        Log.drop_collection()
-
-        # Ensure that the collection handles up to its maximum
-        for i in range(10):
-            Log().save()
-
-        self.assertEqual(len(Log.objects), 10)
-
-        # Check that extra documents don't increase the size
-        Log().save()
-        self.assertEqual(len(Log.objects), 10)
-
-        options = Log.objects._collection.options()
-        self.assertEqual(options['capped'], True)
-        self.assertEqual(options['max'], 10)
-        self.assertEqual(options['size'], 90000)
-
-        # Check that the document cannot be redefined with different options
-        def recreate_log_document():
-            class Log(Document):
-                date = DateTimeField(default=datetime.datetime.now)
-                meta = {
-                    'max_documents': 11,
-                }
-            # Create the collection by accessing Document.objects
-            Log.objects
-        self.assertRaises(InvalidCollectionError, recreate_log_document)
-
-        Log.drop_collection()
-
-    def test_indexes(self):
-        """Ensure that indexes are used when meta[indexes] is specified.
-        """
-        class BlogPost(Document):
-            date = DateTimeField(name='addDate', default=datetime.datetime.now)
-            category = StringField()
-            tags = ListField(StringField())
-            meta = {
-                'indexes': [
-                    '-date', 
-                    'tags',
-                    ('category', '-date')
-                ],
-            }
-
-        BlogPost.drop_collection()
-
-        info = BlogPost.objects._collection.index_information()
-        # _id, types, '-date', 'tags', ('cat', 'date')
-        self.assertEqual(len(info), 5) 
-
-        # Indexes are lazy so use list() to perform query
-        list(BlogPost.objects)
-        info = BlogPost.objects._collection.index_information()
-        self.assertTrue([('_types', 1), ('category', 1), ('addDate', -1)] 
-                        in info.values())
-        self.assertTrue([('_types', 1), ('addDate', -1)] in info.values())
-        # tags is a list field so it shouldn't have _types in the index
-        self.assertTrue([('tags', 1)] in info.values())
-        
-        class ExtendedBlogPost(BlogPost):
-            title = StringField()
-            meta = {'indexes': ['title']}
-
-        BlogPost.drop_collection()
-
-        list(ExtendedBlogPost.objects)
-        info = ExtendedBlogPost.objects._collection.index_information()
-        self.assertTrue([('_types', 1), ('category', 1), ('addDate', -1)] 
-                        in info.values())
-        self.assertTrue([('_types', 1), ('addDate', -1)] in info.values())
-        self.assertTrue([('_types', 1), ('title', 1)] in info.values())
-
-        BlogPost.drop_collection()
-
-    def test_unique(self):
-        """Ensure that uniqueness constraints are applied to fields.
-        """
-        class BlogPost(Document):
-            title = StringField()
-            slug = StringField(unique=True)
-
-        BlogPost.drop_collection()
-
-        post1 = BlogPost(title='test1', slug='test')
-        post1.save()
-
-        # Two posts with the same slug is not allowed
-        post2 = BlogPost(title='test2', slug='test')
-        self.assertRaises(OperationError, post2.save)
-
-        class Date(EmbeddedDocument):
-            year = IntField(name='yr')
-
-        class BlogPost(Document):
-            title = StringField()
-            date = EmbeddedDocumentField(Date)
-            slug = StringField(unique_with='date.year')
-
-        BlogPost.drop_collection()
-
-        post1 = BlogPost(title='test1', date=Date(year=2009), slug='test')
-        post1.save()
-
-        # day is different so won't raise exception
-        post2 = BlogPost(title='test2', date=Date(year=2010), slug='test')
-        post2.save()
-
-        # Now there will be two docs with the same slug and the same day: fail
-        post3 = BlogPost(title='test3', date=Date(year=2010), slug='test')
-        self.assertRaises(OperationError, post3.save)
-
-        BlogPost.drop_collection()
-
-    def test_custom_id_field(self):
-        """Ensure that documents may be created with custom primary keys.
-        """
-        class User(Document):
-            username = StringField(primary_key=True)
-            name = StringField()
-
-        User.drop_collection()
-
-        self.assertEqual(User._fields['username'].name, '_id')
-        self.assertEqual(User._meta['id_field'], 'username')
-
-        def create_invalid_user():
-            User(name='test').save() # no primary key field
-        self.assertRaises(ValidationError, create_invalid_user)
-
-        def define_invalid_user():
-            class EmailUser(User):
-                email = StringField(primary_key=True)
-        self.assertRaises(ValueError, define_invalid_user)
-        
-        user = User(username='test', name='test user')
-        user.save()
-
-        user_obj = User.objects.first()
-        self.assertEqual(user_obj.id, 'test')
-
-        user_son = User.objects._collection.find_one()
-        self.assertEqual(user_son['_id'], 'test')
-        self.assertTrue('username' not in user_son['_id'])
-        
-        User.drop_collection()
-
-    def test_creation(self):
-        """Ensure that document may be created using keyword arguments.
-        """
-        person = self.Person(name="Test User", age=30)
-        self.assertEqual(person.name, "Test User")
-        self.assertEqual(person.age, 30)
-
-    def test_reload(self):
-        """Ensure that attributes may be reloaded.
-        """
-        person = self.Person(name="Test User", age=20)
-        person.save()
-
-        person_obj = self.Person.objects.first()
-        person_obj.name = "Mr Test User"
-        person_obj.age = 21
-        person_obj.save()
-
-        self.assertEqual(person.name, "Test User")
-        self.assertEqual(person.age, 20)
-
-        person.reload()
-        self.assertEqual(person.name, "Mr Test User")
-        self.assertEqual(person.age, 21)
-
-    def test_dictionary_access(self):
-        """Ensure that dictionary-style field access works properly.
-        """
-        person = self.Person(name='Test User', age=30)
-        self.assertEquals(person['name'], 'Test User')
-
-        self.assertRaises(KeyError, person.__getitem__, 'salary')
-        self.assertRaises(KeyError, person.__setitem__, 'salary', 50)
-
-        person['name'] = 'Another User'
-        self.assertEquals(person['name'], 'Another User')
-
-        # Length = length(assigned fields + id)
-        self.assertEquals(len(person), 3)
-
-        self.assertTrue('age' in person)
-        person.age = None
-        self.assertFalse('age' in person)
-        self.assertFalse('nationality' in person)
-
-    def test_embedded_document(self):
-        """Ensure that embedded documents are set up correctly.
-        """
-        class Comment(EmbeddedDocument):
-            content = StringField()
-        
-        self.assertTrue('content' in Comment._fields)
-        self.assertFalse('id' in Comment._fields)
-        self.assertFalse(hasattr(Comment, '_meta'))
-    
-    def test_embedded_document_validation(self):
-        """Ensure that embedded documents may be validated.
-        """
-        class Comment(EmbeddedDocument):
-            date = DateTimeField()
-            content = StringField(required=True)
-        
-        comment = Comment()
-        self.assertRaises(ValidationError, comment.validate)
-
-        comment.content = 'test'
-        comment.validate()
-
-        comment.date = 4
-        self.assertRaises(ValidationError, comment.validate)
-
-        comment.date = datetime.datetime.now()
-        comment.validate()
-
-    def test_save(self):
-        """Ensure that a document may be saved in the database.
-        """
-        # Create person object and save it to the database
-        person = self.Person(name='Test User', age=30)
-        person.save()
-        # Ensure that the object is in the database
-        collection = self.db[self.Person._meta['collection']]
-        person_obj = collection.find_one({'name': 'Test User'})
-        self.assertEqual(person_obj['name'], 'Test User')
-        self.assertEqual(person_obj['age'], 30)
-        self.assertEqual(str(person_obj['_id']), person.id)
-
-    def test_delete(self):
-        """Ensure that document may be deleted using the delete method.
-        """
-        person = self.Person(name="Test User", age=30)
-        person.save()
-        self.assertEqual(len(self.Person.objects), 1)
-        person.delete()
-        self.assertEqual(len(self.Person.objects), 0)
-
-    def test_save_custom_id(self):
-        """Ensure that a document may be saved with a custom _id.
-        """
-        # Create person object and save it to the database
-        person = self.Person(name='Test User', age=30, 
-                             id='497ce96f395f2f052a494fd4')
-        person.save()
-        # Ensure that the object is in the database with the correct _id
-        collection = self.db[self.Person._meta['collection']]
-        person_obj = collection.find_one({'name': 'Test User'})
-        self.assertEqual(str(person_obj['_id']), '497ce96f395f2f052a494fd4')
-
-    def test_save_list(self):
-        """Ensure that a list field may be properly saved.
-        """
-        class Comment(EmbeddedDocument):
-            content = StringField()
-
-        class BlogPost(Document):
-            content = StringField()
-            comments = ListField(EmbeddedDocumentField(Comment))
-            tags = ListField(StringField())
-
-        BlogPost.drop_collection()
-
-        post = BlogPost(content='Went for a walk today...')
-        post.tags = tags = ['fun', 'leisure']
-        comments = [Comment(content='Good for you'), Comment(content='Yay.')]
-        post.comments = comments
-        post.save()
-
-        collection = self.db[BlogPost._meta['collection']]
-        post_obj = collection.find_one()
-        self.assertEqual(post_obj['tags'], tags)
-        for comment_obj, comment in zip(post_obj['comments'], comments):
-            self.assertEqual(comment_obj['content'], comment['content'])
-
-        BlogPost.drop_collection()
-
-    def test_save_embedded_document(self):
-        """Ensure that a document with an embedded document field may be 
-        saved in the database.
-        """
-        class EmployeeDetails(EmbeddedDocument):
-            position = StringField()
-
-        class Employee(self.Person):
-            salary = IntField()
-            details = EmbeddedDocumentField(EmployeeDetails)
-
-        # Create employee object and save it to the database
-        employee = Employee(name='Test Employee', age=50, salary=20000)
-        employee.details = EmployeeDetails(position='Developer')
-        employee.save()
-
-        # Ensure that the object is in the database
-        collection = self.db[self.Person._meta['collection']]
-        employee_obj = collection.find_one({'name': 'Test Employee'})
-        self.assertEqual(employee_obj['name'], 'Test Employee')
-        self.assertEqual(employee_obj['age'], 50)
-        # Ensure that the 'details' embedded object saved correctly
-        self.assertEqual(employee_obj['details']['position'], 'Developer')
-
-    def test_save_reference(self):
-        """Ensure that a document reference field may be saved in the database.
-        """
-        
-        class BlogPost(Document):
-            meta = {'collection': 'blogpost_1'}
-            content = StringField()
-            author = ReferenceField(self.Person)
-
-        BlogPost.drop_collection()
-
-        author = self.Person(name='Test User')
-        author.save()
-
-        post = BlogPost(content='Watched some TV today... how exciting.')
-        # Should only reference author when saving
-        post.author = author
-        post.save()
-
-        post_obj = BlogPost.objects.first()
-
-        # Test laziness
-        self.assertTrue(isinstance(post_obj._data['author'], 
-                                   pymongo.dbref.DBRef))
-        self.assertTrue(isinstance(post_obj.author, self.Person))
-        self.assertEqual(post_obj.author.name, 'Test User')
-
-        # Ensure that the dereferenced object may be changed and saved
-        post_obj.author.age = 25
-        post_obj.author.save()
-
-        author = list(self.Person.objects(name='Test User'))[-1]
-        self.assertEqual(author.age, 25)
-
-        BlogPost.drop_collection()
-
-    def tearDown(self):
-        self.Person.drop_collection()
-
-
-if __name__ == '__main__':
-    unittest.main()

tests/document/test_class_methods.py

@@ -0,0 +1,357 @@
+# -*- coding: utf-8 -*-
+import unittest
+
+from mongoengine import *
+from mongoengine.connection import get_db
+from mongoengine.pymongo_support import list_collection_names
+from mongoengine.queryset import NULLIFY, PULL
+
+
+class TestClassMethods(unittest.TestCase):
+    def setUp(self):
+        connect(db="mongoenginetest")
+        self.db = get_db()
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+
+            non_field = True
+
+            meta = {"allow_inheritance": True}
+
+        self.Person = Person
+
+    def tearDown(self):
+        for collection in list_collection_names(self.db):
+            self.db.drop_collection(collection)
+
+    def test_definition(self):
+        """Ensure that document may be defined using fields.
+        """
+        assert ["_cls", "age", "id", "name"] == sorted(self.Person._fields.keys())
+        assert ["IntField", "ObjectIdField", "StringField", "StringField"] == sorted(
+            [x.__class__.__name__ for x in self.Person._fields.values()]
+        )
+
+    def test_get_db(self):
+        """Ensure that get_db returns the expected db.
+        """
+        db = self.Person._get_db()
+        assert self.db == db
+
+    def test_get_collection_name(self):
+        """Ensure that get_collection_name returns the expected collection
+        name.
+        """
+        collection_name = "person"
+        assert collection_name == self.Person._get_collection_name()
+
+    def test_get_collection(self):
+        """Ensure that get_collection returns the expected collection.
+        """
+        collection_name = "person"
+        collection = self.Person._get_collection()
+        assert self.db[collection_name] == collection
+
+    def test_drop_collection(self):
+        """Ensure that the collection may be dropped from the database.
+        """
+        collection_name = "person"
+        self.Person(name="Test").save()
+        assert collection_name in list_collection_names(self.db)
+
+        self.Person.drop_collection()
+        assert collection_name not in list_collection_names(self.db)
+
+    def test_register_delete_rule(self):
+        """Ensure that register delete rule adds a delete rule to the document
+        meta.
+        """
+
+        class Job(Document):
+            employee = ReferenceField(self.Person)
+
+        assert self.Person._meta.get("delete_rules") is None
+
+        self.Person.register_delete_rule(Job, "employee", NULLIFY)
+        assert self.Person._meta["delete_rules"] == {(Job, "employee"): NULLIFY}
+
+    def test_compare_indexes(self):
+        """ Ensure that the indexes are properly created and that
+        compare_indexes identifies the missing/extra indexes
+        """
+
+        class BlogPost(Document):
+            author = StringField()
+            title = StringField()
+            description = StringField()
+            tags = StringField()
+
+            meta = {"indexes": [("author", "title")]}
+
+        BlogPost.drop_collection()
+
+        BlogPost.ensure_indexes()
+        assert BlogPost.compare_indexes() == {"missing": [], "extra": []}
+
+        BlogPost.ensure_index(["author", "description"])
+        assert BlogPost.compare_indexes() == {
+            "missing": [],
+            "extra": [[("author", 1), ("description", 1)]],
+        }
+
+        BlogPost._get_collection().drop_index("author_1_description_1")
+        assert BlogPost.compare_indexes() == {"missing": [], "extra": []}
+
+        BlogPost._get_collection().drop_index("author_1_title_1")
+        assert BlogPost.compare_indexes() == {
+            "missing": [[("author", 1), ("title", 1)]],
+            "extra": [],
+        }
+
+    def test_compare_indexes_inheritance(self):
+        """ Ensure that the indexes are properly created and that
+        compare_indexes identifies the missing/extra indexes for subclassed
+        documents (_cls included)
+        """
+
+        class BlogPost(Document):
+            author = StringField()
+            title = StringField()
+            description = StringField()
+
+            meta = {"allow_inheritance": True}
+
+        class BlogPostWithTags(BlogPost):
+            tags = StringField()
+            tag_list = ListField(StringField())
+
+            meta = {"indexes": [("author", "tags")]}
+
+        BlogPost.drop_collection()
+
+        BlogPost.ensure_indexes()
+        BlogPostWithTags.ensure_indexes()
+        assert BlogPost.compare_indexes() == {"missing": [], "extra": []}
+
+        BlogPostWithTags.ensure_index(["author", "tag_list"])
+        assert BlogPost.compare_indexes() == {
+            "missing": [],
+            "extra": [[("_cls", 1), ("author", 1), ("tag_list", 1)]],
+        }
+
+        BlogPostWithTags._get_collection().drop_index("_cls_1_author_1_tag_list_1")
+        assert BlogPost.compare_indexes() == {"missing": [], "extra": []}
+
+        BlogPostWithTags._get_collection().drop_index("_cls_1_author_1_tags_1")
+        assert BlogPost.compare_indexes() == {
+            "missing": [[("_cls", 1), ("author", 1), ("tags", 1)]],
+            "extra": [],
+        }
+
+    def test_compare_indexes_multiple_subclasses(self):
+        """ Ensure that compare_indexes behaves correctly if called from a
+        class, which base class has multiple subclasses
+        """
+
+        class BlogPost(Document):
+            author = StringField()
+            title = StringField()
+            description = StringField()
+
+            meta = {"allow_inheritance": True}
+
+        class BlogPostWithTags(BlogPost):
+            tags = StringField()
+            tag_list = ListField(StringField())
+
+            meta = {"indexes": [("author", "tags")]}
+
+        class BlogPostWithCustomField(BlogPost):
+            custom = DictField()
+
+            meta = {"indexes": [("author", "custom")]}
+
+        BlogPost.ensure_indexes()
+        BlogPostWithTags.ensure_indexes()
+        BlogPostWithCustomField.ensure_indexes()
+
+        assert BlogPost.compare_indexes() == {"missing": [], "extra": []}
+        assert BlogPostWithTags.compare_indexes() == {"missing": [], "extra": []}
+        assert BlogPostWithCustomField.compare_indexes() == {"missing": [], "extra": []}
+
+    def test_compare_indexes_for_text_indexes(self):
+        """ Ensure that compare_indexes behaves correctly for text indexes """
+
+        class Doc(Document):
+            a = StringField()
+            b = StringField()
+            meta = {
+                "indexes": [
+                    {
+                        "fields": ["$a", "$b"],
+                        "default_language": "english",
+                        "weights": {"a": 10, "b": 2},
+                    }
+                ]
+            }
+
+        Doc.drop_collection()
+        Doc.ensure_indexes()
+        actual = Doc.compare_indexes()
+        expected = {"missing": [], "extra": []}
+        assert actual == expected
+
+    def test_list_indexes_inheritance(self):
+        """ ensure that all of the indexes are listed regardless of the super-
+        or sub-class that we call it from
+        """
+
+        class BlogPost(Document):
+            author = StringField()
+            title = StringField()
+            description = StringField()
+
+            meta = {"allow_inheritance": True}
+
+        class BlogPostWithTags(BlogPost):
+            tags = StringField()
+
+            meta = {"indexes": [("author", "tags")]}
+
+        class BlogPostWithTagsAndExtraText(BlogPostWithTags):
+            extra_text = StringField()
+
+            meta = {"indexes": [("author", "tags", "extra_text")]}
+
+        BlogPost.drop_collection()
+
+        BlogPost.ensure_indexes()
+        BlogPostWithTags.ensure_indexes()
+        BlogPostWithTagsAndExtraText.ensure_indexes()
+
+        assert BlogPost.list_indexes() == BlogPostWithTags.list_indexes()
+        assert BlogPost.list_indexes() == BlogPostWithTagsAndExtraText.list_indexes()
+        assert BlogPost.list_indexes() == [
+            [("_cls", 1), ("author", 1), ("tags", 1)],
+            [("_cls", 1), ("author", 1), ("tags", 1), ("extra_text", 1)],
+            [(u"_id", 1)],
+            [("_cls", 1)],
+        ]
+
+    def test_register_delete_rule_inherited(self):
+        class Vaccine(Document):
+            name = StringField(required=True)
+
+            meta = {"indexes": ["name"]}
+
+        class Animal(Document):
+            family = StringField(required=True)
+            vaccine_made = ListField(
+                ReferenceField("Vaccine", reverse_delete_rule=PULL)
+            )
+
+            meta = {"allow_inheritance": True, "indexes": ["family"]}
+
+        class Cat(Animal):
+            name = StringField(required=True)
+
+        assert Vaccine._meta["delete_rules"][(Animal, "vaccine_made")] == PULL
+        assert Vaccine._meta["delete_rules"][(Cat, "vaccine_made")] == PULL
+
+    def test_collection_naming(self):
+        """Ensure that a collection with a specified name may be used.
+        """
+
+        class DefaultNamingTest(Document):
+            pass
+
+        assert "default_naming_test" == DefaultNamingTest._get_collection_name()
+
+        class CustomNamingTest(Document):
+            meta = {"collection": "pimp_my_collection"}
+
+        assert "pimp_my_collection" == CustomNamingTest._get_collection_name()
+
+        class DynamicNamingTest(Document):
+            meta = {"collection": lambda c: "DYNAMO"}
+
+        assert "DYNAMO" == DynamicNamingTest._get_collection_name()
+
+        # Use Abstract class to handle backwards compatibility
+        class BaseDocument(Document):
+            meta = {"abstract": True, "collection": lambda c: c.__name__.lower()}
+
+        class OldNamingConvention(BaseDocument):
+            pass
+
+        assert "oldnamingconvention" == OldNamingConvention._get_collection_name()
+
+        class InheritedAbstractNamingTest(BaseDocument):
+            meta = {"collection": "wibble"}
+
+        assert "wibble" == InheritedAbstractNamingTest._get_collection_name()
+
+        # Mixin tests
+        class BaseMixin(object):
+            meta = {"collection": lambda c: c.__name__.lower()}
+
+        class OldMixinNamingConvention(Document, BaseMixin):
+            pass
+
+        assert (
+            "oldmixinnamingconvention"
+            == OldMixinNamingConvention._get_collection_name()
+        )
+
+        class BaseMixin(object):
+            meta = {"collection": lambda c: c.__name__.lower()}
+
+        class BaseDocument(Document, BaseMixin):
+            meta = {"allow_inheritance": True}
+
+        class MyDocument(BaseDocument):
+            pass
+
+        assert "basedocument" == MyDocument._get_collection_name()
+
+    def test_custom_collection_name_operations(self):
+        """Ensure that a collection with a specified name is used as expected.
+        """
+        collection_name = "personCollTest"
+
+        class Person(Document):
+            name = StringField()
+            meta = {"collection": collection_name}
+
+        Person(name="Test User").save()
+        assert collection_name in list_collection_names(self.db)
+
+        user_obj = self.db[collection_name].find_one()
+        assert user_obj["name"] == "Test User"
+
+        user_obj = Person.objects[0]
+        assert user_obj.name == "Test User"
+
+        Person.drop_collection()
+        assert collection_name not in list_collection_names(self.db)
+
+    def test_collection_name_and_primary(self):
+        """Ensure that a collection with a specified name may be used.
+        """
+
+        class Person(Document):
+            name = StringField(primary_key=True)
+            meta = {"collection": "app"}
+
+        Person(name="Test User").save()
+
+        user_obj = Person.objects.first()
+        assert user_obj.name == "Test User"
+
+        Person.drop_collection()
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/document/test_delta.py

@@ -0,0 +1,916 @@
+# -*- coding: utf-8 -*-
+import unittest
+
+from bson import SON
+from mongoengine import *
+from mongoengine.pymongo_support import list_collection_names
+from tests.utils import MongoDBTestCase
+
+
+class TestDelta(MongoDBTestCase):
+    def setUp(self):
+        super(TestDelta, self).setUp()
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+
+            non_field = True
+
+            meta = {"allow_inheritance": True}
+
+        self.Person = Person
+
+    def tearDown(self):
+        for collection in list_collection_names(self.db):
+            self.db.drop_collection(collection)
+
+    def test_delta(self):
+        self.delta(Document)
+        self.delta(DynamicDocument)
+
+    def delta(self, DocClass):
+        class Doc(DocClass):
+            string_field = StringField()
+            int_field = IntField()
+            dict_field = DictField()
+            list_field = ListField()
+
+        Doc.drop_collection()
+        doc = Doc()
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc._get_changed_fields() == []
+        assert doc._delta() == ({}, {})
+
+        doc.string_field = "hello"
+        assert doc._get_changed_fields() == ["string_field"]
+        assert doc._delta() == ({"string_field": "hello"}, {})
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        assert doc._get_changed_fields() == ["int_field"]
+        assert doc._delta() == ({"int_field": 1}, {})
+
+        doc._changed_fields = []
+        dict_value = {"hello": "world", "ping": "pong"}
+        doc.dict_field = dict_value
+        assert doc._get_changed_fields() == ["dict_field"]
+        assert doc._delta() == ({"dict_field": dict_value}, {})
+
+        doc._changed_fields = []
+        list_value = ["1", 2, {"hello": "world"}]
+        doc.list_field = list_value
+        assert doc._get_changed_fields() == ["list_field"]
+        assert doc._delta() == ({"list_field": list_value}, {})
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        assert doc._get_changed_fields() == ["dict_field"]
+        assert doc._delta() == ({}, {"dict_field": 1})
+
+        doc._changed_fields = []
+        doc.list_field = []
+        assert doc._get_changed_fields() == ["list_field"]
+        assert doc._delta() == ({}, {"list_field": 1})
+
+    def test_delta_recursive(self):
+        self.delta_recursive(Document, EmbeddedDocument)
+        self.delta_recursive(DynamicDocument, EmbeddedDocument)
+        self.delta_recursive(Document, DynamicEmbeddedDocument)
+        self.delta_recursive(DynamicDocument, DynamicEmbeddedDocument)
+
+    def delta_recursive(self, DocClass, EmbeddedClass):
+        class Embedded(EmbeddedClass):
+            id = StringField()
+            string_field = StringField()
+            int_field = IntField()
+            dict_field = DictField()
+            list_field = ListField()
+
+        class Doc(DocClass):
+            string_field = StringField()
+            int_field = IntField()
+            dict_field = DictField()
+            list_field = ListField()
+            embedded_field = EmbeddedDocumentField(Embedded)
+
+        Doc.drop_collection()
+        doc = Doc()
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc._get_changed_fields() == []
+        assert doc._delta() == ({}, {})
+
+        embedded_1 = Embedded()
+        embedded_1.id = "010101"
+        embedded_1.string_field = "hello"
+        embedded_1.int_field = 1
+        embedded_1.dict_field = {"hello": "world"}
+        embedded_1.list_field = ["1", 2, {"hello": "world"}]
+        doc.embedded_field = embedded_1
+
+        assert doc._get_changed_fields() == ["embedded_field"]
+
+        embedded_delta = {
+            "id": "010101",
+            "string_field": "hello",
+            "int_field": 1,
+            "dict_field": {"hello": "world"},
+            "list_field": ["1", 2, {"hello": "world"}],
+        }
+        assert doc.embedded_field._delta() == (embedded_delta, {})
+        assert doc._delta() == ({"embedded_field": embedded_delta}, {})
+
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.dict_field = {}
+        assert doc._get_changed_fields() == ["embedded_field.dict_field"]
+        assert doc.embedded_field._delta() == ({}, {"dict_field": 1})
+        assert doc._delta() == ({}, {"embedded_field.dict_field": 1})
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.dict_field == {}
+
+        doc.embedded_field.list_field = []
+        assert doc._get_changed_fields() == ["embedded_field.list_field"]
+        assert doc.embedded_field._delta() == ({}, {"list_field": 1})
+        assert doc._delta() == ({}, {"embedded_field.list_field": 1})
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field == []
+
+        embedded_2 = Embedded()
+        embedded_2.string_field = "hello"
+        embedded_2.int_field = 1
+        embedded_2.dict_field = {"hello": "world"}
+        embedded_2.list_field = ["1", 2, {"hello": "world"}]
+
+        doc.embedded_field.list_field = ["1", 2, embedded_2]
+        assert doc._get_changed_fields() == ["embedded_field.list_field"]
+
+        assert doc.embedded_field._delta() == (
+            {
+                "list_field": [
+                    "1",
+                    2,
+                    {
+                        "_cls": "Embedded",
+                        "string_field": "hello",
+                        "dict_field": {"hello": "world"},
+                        "int_field": 1,
+                        "list_field": ["1", 2, {"hello": "world"}],
+                    },
+                ]
+            },
+            {},
+        )
+
+        assert doc._delta() == (
+            {
+                "embedded_field.list_field": [
+                    "1",
+                    2,
+                    {
+                        "_cls": "Embedded",
+                        "string_field": "hello",
+                        "dict_field": {"hello": "world"},
+                        "int_field": 1,
+                        "list_field": ["1", 2, {"hello": "world"}],
+                    },
+                ]
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        assert doc.embedded_field.list_field[0] == "1"
+        assert doc.embedded_field.list_field[1] == 2
+        for k in doc.embedded_field.list_field[2]._fields:
+            assert doc.embedded_field.list_field[2][k] == embedded_2[k]
+
+        doc.embedded_field.list_field[2].string_field = "world"
+        assert doc._get_changed_fields() == ["embedded_field.list_field.2.string_field"]
+        assert doc.embedded_field._delta() == (
+            {"list_field.2.string_field": "world"},
+            {},
+        )
+        assert doc._delta() == (
+            {"embedded_field.list_field.2.string_field": "world"},
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].string_field == "world"
+
+        # Test multiple assignments
+        doc.embedded_field.list_field[2].string_field = "hello world"
+        doc.embedded_field.list_field[2] = doc.embedded_field.list_field[2]
+        assert doc._get_changed_fields() == ["embedded_field.list_field.2"]
+        assert doc.embedded_field._delta() == (
+            {
+                "list_field.2": {
+                    "_cls": "Embedded",
+                    "string_field": "hello world",
+                    "int_field": 1,
+                    "list_field": ["1", 2, {"hello": "world"}],
+                    "dict_field": {"hello": "world"},
+                }
+            },
+            {},
+        )
+        assert doc._delta() == (
+            {
+                "embedded_field.list_field.2": {
+                    "_cls": "Embedded",
+                    "string_field": "hello world",
+                    "int_field": 1,
+                    "list_field": ["1", 2, {"hello": "world"}],
+                    "dict_field": {"hello": "world"},
+                }
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].string_field == "hello world"
+
+        # Test list native methods
+        doc.embedded_field.list_field[2].list_field.pop(0)
+        assert doc._delta() == (
+            {"embedded_field.list_field.2.list_field": [2, {"hello": "world"}]},
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.list_field[2].list_field.append(1)
+        assert doc._delta() == (
+            {"embedded_field.list_field.2.list_field": [2, {"hello": "world"}, 1]},
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].list_field == [2, {"hello": "world"}, 1]
+
+        doc.embedded_field.list_field[2].list_field.sort(key=str)
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].list_field == [1, 2, {"hello": "world"}]
+
+        del doc.embedded_field.list_field[2].list_field[2]["hello"]
+        assert doc._delta() == (
+            {},
+            {"embedded_field.list_field.2.list_field.2.hello": 1},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        del doc.embedded_field.list_field[2].list_field
+        assert doc._delta() == ({}, {"embedded_field.list_field.2.list_field": 1})
+
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.dict_field["Embedded"] = embedded_1
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.dict_field["Embedded"].string_field = "Hello World"
+        assert doc._get_changed_fields() == ["dict_field.Embedded.string_field"]
+        assert doc._delta() == ({"dict_field.Embedded.string_field": "Hello World"}, {})
+
+    def test_circular_reference_deltas(self):
+        self.circular_reference_deltas(Document, Document)
+        self.circular_reference_deltas(Document, DynamicDocument)
+        self.circular_reference_deltas(DynamicDocument, Document)
+        self.circular_reference_deltas(DynamicDocument, DynamicDocument)
+
+    def circular_reference_deltas(self, DocClass1, DocClass2):
+        class Person(DocClass1):
+            name = StringField()
+            owns = ListField(ReferenceField("Organization"))
+
+        class Organization(DocClass2):
+            name = StringField()
+            owner = ReferenceField("Person")
+
+        Person.drop_collection()
+        Organization.drop_collection()
+
+        person = Person(name="owner").save()
+        organization = Organization(name="company").save()
+
+        person.owns.append(organization)
+        organization.owner = person
+
+        person.save()
+        organization.save()
+
+        p = Person.objects[0].select_related()
+        o = Organization.objects.first()
+        assert p.owns[0] == o
+        assert o.owner == p
+
+    def test_circular_reference_deltas_2(self):
+        self.circular_reference_deltas_2(Document, Document)
+        self.circular_reference_deltas_2(Document, DynamicDocument)
+        self.circular_reference_deltas_2(DynamicDocument, Document)
+        self.circular_reference_deltas_2(DynamicDocument, DynamicDocument)
+
+    def circular_reference_deltas_2(self, DocClass1, DocClass2, dbref=True):
+        class Person(DocClass1):
+            name = StringField()
+            owns = ListField(ReferenceField("Organization", dbref=dbref))
+            employer = ReferenceField("Organization", dbref=dbref)
+
+        class Organization(DocClass2):
+            name = StringField()
+            owner = ReferenceField("Person", dbref=dbref)
+            employees = ListField(ReferenceField("Person", dbref=dbref))
+
+        Person.drop_collection()
+        Organization.drop_collection()
+
+        person = Person(name="owner").save()
+        employee = Person(name="employee").save()
+        organization = Organization(name="company").save()
+
+        person.owns.append(organization)
+        organization.owner = person
+
+        organization.employees.append(employee)
+        employee.employer = organization
+
+        person.save()
+        organization.save()
+        employee.save()
+
+        p = Person.objects.get(name="owner")
+        e = Person.objects.get(name="employee")
+        o = Organization.objects.first()
+
+        assert p.owns[0] == o
+        assert o.owner == p
+        assert e.employer == o
+
+        return person, organization, employee
+
+    def test_delta_db_field(self):
+        self.delta_db_field(Document)
+        self.delta_db_field(DynamicDocument)
+
+    def delta_db_field(self, DocClass):
+        class Doc(DocClass):
+            string_field = StringField(db_field="db_string_field")
+            int_field = IntField(db_field="db_int_field")
+            dict_field = DictField(db_field="db_dict_field")
+            list_field = ListField(db_field="db_list_field")
+
+        Doc.drop_collection()
+        doc = Doc()
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc._get_changed_fields() == []
+        assert doc._delta() == ({}, {})
+
+        doc.string_field = "hello"
+        assert doc._get_changed_fields() == ["db_string_field"]
+        assert doc._delta() == ({"db_string_field": "hello"}, {})
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        assert doc._get_changed_fields() == ["db_int_field"]
+        assert doc._delta() == ({"db_int_field": 1}, {})
+
+        doc._changed_fields = []
+        dict_value = {"hello": "world", "ping": "pong"}
+        doc.dict_field = dict_value
+        assert doc._get_changed_fields() == ["db_dict_field"]
+        assert doc._delta() == ({"db_dict_field": dict_value}, {})
+
+        doc._changed_fields = []
+        list_value = ["1", 2, {"hello": "world"}]
+        doc.list_field = list_value
+        assert doc._get_changed_fields() == ["db_list_field"]
+        assert doc._delta() == ({"db_list_field": list_value}, {})
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        assert doc._get_changed_fields() == ["db_dict_field"]
+        assert doc._delta() == ({}, {"db_dict_field": 1})
+
+        doc._changed_fields = []
+        doc.list_field = []
+        assert doc._get_changed_fields() == ["db_list_field"]
+        assert doc._delta() == ({}, {"db_list_field": 1})
+
+        # Test it saves that data
+        doc = Doc()
+        doc.save()
+
+        doc.string_field = "hello"
+        doc.int_field = 1
+        doc.dict_field = {"hello": "world"}
+        doc.list_field = ["1", 2, {"hello": "world"}]
+        doc.save()
+        doc = doc.reload(10)
+
+        assert doc.string_field == "hello"
+        assert doc.int_field == 1
+        assert doc.dict_field == {"hello": "world"}
+        assert doc.list_field == ["1", 2, {"hello": "world"}]
+
+    def test_delta_recursive_db_field(self):
+        self.delta_recursive_db_field(Document, EmbeddedDocument)
+        self.delta_recursive_db_field(Document, DynamicEmbeddedDocument)
+        self.delta_recursive_db_field(DynamicDocument, EmbeddedDocument)
+        self.delta_recursive_db_field(DynamicDocument, DynamicEmbeddedDocument)
+
+    def delta_recursive_db_field(self, DocClass, EmbeddedClass):
+        class Embedded(EmbeddedClass):
+            string_field = StringField(db_field="db_string_field")
+            int_field = IntField(db_field="db_int_field")
+            dict_field = DictField(db_field="db_dict_field")
+            list_field = ListField(db_field="db_list_field")
+
+        class Doc(DocClass):
+            string_field = StringField(db_field="db_string_field")
+            int_field = IntField(db_field="db_int_field")
+            dict_field = DictField(db_field="db_dict_field")
+            list_field = ListField(db_field="db_list_field")
+            embedded_field = EmbeddedDocumentField(
+                Embedded, db_field="db_embedded_field"
+            )
+
+        Doc.drop_collection()
+        doc = Doc()
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc._get_changed_fields() == []
+        assert doc._delta() == ({}, {})
+
+        embedded_1 = Embedded()
+        embedded_1.string_field = "hello"
+        embedded_1.int_field = 1
+        embedded_1.dict_field = {"hello": "world"}
+        embedded_1.list_field = ["1", 2, {"hello": "world"}]
+        doc.embedded_field = embedded_1
+
+        assert doc._get_changed_fields() == ["db_embedded_field"]
+
+        embedded_delta = {
+            "db_string_field": "hello",
+            "db_int_field": 1,
+            "db_dict_field": {"hello": "world"},
+            "db_list_field": ["1", 2, {"hello": "world"}],
+        }
+        assert doc.embedded_field._delta() == (embedded_delta, {})
+        assert doc._delta() == ({"db_embedded_field": embedded_delta}, {})
+
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.dict_field = {}
+        assert doc._get_changed_fields() == ["db_embedded_field.db_dict_field"]
+        assert doc.embedded_field._delta() == ({}, {"db_dict_field": 1})
+        assert doc._delta() == ({}, {"db_embedded_field.db_dict_field": 1})
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.dict_field == {}
+
+        doc.embedded_field.list_field = []
+        assert doc._get_changed_fields() == ["db_embedded_field.db_list_field"]
+        assert doc.embedded_field._delta() == ({}, {"db_list_field": 1})
+        assert doc._delta() == ({}, {"db_embedded_field.db_list_field": 1})
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field == []
+
+        embedded_2 = Embedded()
+        embedded_2.string_field = "hello"
+        embedded_2.int_field = 1
+        embedded_2.dict_field = {"hello": "world"}
+        embedded_2.list_field = ["1", 2, {"hello": "world"}]
+
+        doc.embedded_field.list_field = ["1", 2, embedded_2]
+        assert doc._get_changed_fields() == ["db_embedded_field.db_list_field"]
+        assert doc.embedded_field._delta() == (
+            {
+                "db_list_field": [
+                    "1",
+                    2,
+                    {
+                        "_cls": "Embedded",
+                        "db_string_field": "hello",
+                        "db_dict_field": {"hello": "world"},
+                        "db_int_field": 1,
+                        "db_list_field": ["1", 2, {"hello": "world"}],
+                    },
+                ]
+            },
+            {},
+        )
+
+        assert doc._delta() == (
+            {
+                "db_embedded_field.db_list_field": [
+                    "1",
+                    2,
+                    {
+                        "_cls": "Embedded",
+                        "db_string_field": "hello",
+                        "db_dict_field": {"hello": "world"},
+                        "db_int_field": 1,
+                        "db_list_field": ["1", 2, {"hello": "world"}],
+                    },
+                ]
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        assert doc.embedded_field.list_field[0] == "1"
+        assert doc.embedded_field.list_field[1] == 2
+        for k in doc.embedded_field.list_field[2]._fields:
+            assert doc.embedded_field.list_field[2][k] == embedded_2[k]
+
+        doc.embedded_field.list_field[2].string_field = "world"
+        assert doc._get_changed_fields() == [
+            "db_embedded_field.db_list_field.2.db_string_field"
+        ]
+        assert doc.embedded_field._delta() == (
+            {"db_list_field.2.db_string_field": "world"},
+            {},
+        )
+        assert doc._delta() == (
+            {"db_embedded_field.db_list_field.2.db_string_field": "world"},
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].string_field == "world"
+
+        # Test multiple assignments
+        doc.embedded_field.list_field[2].string_field = "hello world"
+        doc.embedded_field.list_field[2] = doc.embedded_field.list_field[2]
+        assert doc._get_changed_fields() == ["db_embedded_field.db_list_field.2"]
+        assert doc.embedded_field._delta() == (
+            {
+                "db_list_field.2": {
+                    "_cls": "Embedded",
+                    "db_string_field": "hello world",
+                    "db_int_field": 1,
+                    "db_list_field": ["1", 2, {"hello": "world"}],
+                    "db_dict_field": {"hello": "world"},
+                }
+            },
+            {},
+        )
+        assert doc._delta() == (
+            {
+                "db_embedded_field.db_list_field.2": {
+                    "_cls": "Embedded",
+                    "db_string_field": "hello world",
+                    "db_int_field": 1,
+                    "db_list_field": ["1", 2, {"hello": "world"}],
+                    "db_dict_field": {"hello": "world"},
+                }
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].string_field == "hello world"
+
+        # Test list native methods
+        doc.embedded_field.list_field[2].list_field.pop(0)
+        assert doc._delta() == (
+            {
+                "db_embedded_field.db_list_field.2.db_list_field": [
+                    2,
+                    {"hello": "world"},
+                ]
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.list_field[2].list_field.append(1)
+        assert doc._delta() == (
+            {
+                "db_embedded_field.db_list_field.2.db_list_field": [
+                    2,
+                    {"hello": "world"},
+                    1,
+                ]
+            },
+            {},
+        )
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].list_field == [2, {"hello": "world"}, 1]
+
+        doc.embedded_field.list_field[2].list_field.sort(key=str)
+        doc.save()
+        doc = doc.reload(10)
+        assert doc.embedded_field.list_field[2].list_field == [1, 2, {"hello": "world"}]
+
+        del doc.embedded_field.list_field[2].list_field[2]["hello"]
+        assert doc._delta() == (
+            {},
+            {"db_embedded_field.db_list_field.2.db_list_field.2.hello": 1},
+        )
+        doc.save()
+        doc = doc.reload(10)
+
+        del doc.embedded_field.list_field[2].list_field
+        assert doc._delta() == (
+            {},
+            {"db_embedded_field.db_list_field.2.db_list_field": 1},
+        )
+
+    def test_delta_for_dynamic_documents(self):
+        class Person(DynamicDocument):
+            name = StringField()
+            meta = {"allow_inheritance": True}
+
+        Person.drop_collection()
+
+        p = Person(name="James", age=34)
+        assert p._delta() == (
+            SON([("_cls", "Person"), ("name", "James"), ("age", 34)]),
+            {},
+        )
+
+        p.doc = 123
+        del p.doc
+        assert p._delta() == (
+            SON([("_cls", "Person"), ("name", "James"), ("age", 34)]),
+            {},
+        )
+
+        p = Person()
+        p.name = "Dean"
+        p.age = 22
+        p.save()
+
+        p.age = 24
+        assert p.age == 24
+        assert p._get_changed_fields() == ["age"]
+        assert p._delta() == ({"age": 24}, {})
+
+        p = Person.objects(age=22).get()
+        p.age = 24
+        assert p.age == 24
+        assert p._get_changed_fields() == ["age"]
+        assert p._delta() == ({"age": 24}, {})
+
+        p.save()
+        assert 1 == Person.objects(age=24).count()
+
+    def test_dynamic_delta(self):
+        class Doc(DynamicDocument):
+            pass
+
+        Doc.drop_collection()
+        doc = Doc()
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc._get_changed_fields() == []
+        assert doc._delta() == ({}, {})
+
+        doc.string_field = "hello"
+        assert doc._get_changed_fields() == ["string_field"]
+        assert doc._delta() == ({"string_field": "hello"}, {})
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        assert doc._get_changed_fields() == ["int_field"]
+        assert doc._delta() == ({"int_field": 1}, {})
+
+        doc._changed_fields = []
+        dict_value = {"hello": "world", "ping": "pong"}
+        doc.dict_field = dict_value
+        assert doc._get_changed_fields() == ["dict_field"]
+        assert doc._delta() == ({"dict_field": dict_value}, {})
+
+        doc._changed_fields = []
+        list_value = ["1", 2, {"hello": "world"}]
+        doc.list_field = list_value
+        assert doc._get_changed_fields() == ["list_field"]
+        assert doc._delta() == ({"list_field": list_value}, {})
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        assert doc._get_changed_fields() == ["dict_field"]
+        assert doc._delta() == ({}, {"dict_field": 1})
+
+        doc._changed_fields = []
+        doc.list_field = []
+        assert doc._get_changed_fields() == ["list_field"]
+        assert doc._delta() == ({}, {"list_field": 1})
+
+    def test_delta_with_dbref_true(self):
+        person, organization, employee = self.circular_reference_deltas_2(
+            Document, Document, True
+        )
+        employee.name = "test"
+
+        assert organization._get_changed_fields() == []
+
+        updates, removals = organization._delta()
+        assert {} == removals
+        assert {} == updates
+
+        organization.employees.append(person)
+        updates, removals = organization._delta()
+        assert {} == removals
+        assert "employees" in updates
+
+    def test_delta_with_dbref_false(self):
+        person, organization, employee = self.circular_reference_deltas_2(
+            Document, Document, False
+        )
+        employee.name = "test"
+
+        assert organization._get_changed_fields() == []
+
+        updates, removals = organization._delta()
+        assert {} == removals
+        assert {} == updates
+
+        organization.employees.append(person)
+        updates, removals = organization._delta()
+        assert {} == removals
+        assert "employees" in updates
+
+    def test_nested_nested_fields_mark_as_changed(self):
+        class EmbeddedDoc(EmbeddedDocument):
+            name = StringField()
+
+        class MyDoc(Document):
+            subs = MapField(MapField(EmbeddedDocumentField(EmbeddedDoc)))
+            name = StringField()
+
+        MyDoc.drop_collection()
+
+        mydoc = MyDoc(
+            name="testcase1", subs={"a": {"b": EmbeddedDoc(name="foo")}}
+        ).save()
+
+        mydoc = MyDoc.objects.first()
+        subdoc = mydoc.subs["a"]["b"]
+        subdoc.name = "bar"
+
+        assert ["name"] == subdoc._get_changed_fields()
+        assert ["subs.a.b.name"] == mydoc._get_changed_fields()
+
+        mydoc._clear_changed_fields()
+        assert [] == mydoc._get_changed_fields()
+
+    def test_lower_level_mark_as_changed(self):
+        class EmbeddedDoc(EmbeddedDocument):
+            name = StringField()
+
+        class MyDoc(Document):
+            subs = MapField(EmbeddedDocumentField(EmbeddedDoc))
+
+        MyDoc.drop_collection()
+
+        MyDoc().save()
+
+        mydoc = MyDoc.objects.first()
+        mydoc.subs["a"] = EmbeddedDoc()
+        assert ["subs.a"] == mydoc._get_changed_fields()
+
+        subdoc = mydoc.subs["a"]
+        subdoc.name = "bar"
+
+        assert ["name"] == subdoc._get_changed_fields()
+        assert ["subs.a"] == mydoc._get_changed_fields()
+        mydoc.save()
+
+        mydoc._clear_changed_fields()
+        assert [] == mydoc._get_changed_fields()
+
+    def test_upper_level_mark_as_changed(self):
+        class EmbeddedDoc(EmbeddedDocument):
+            name = StringField()
+
+        class MyDoc(Document):
+            subs = MapField(EmbeddedDocumentField(EmbeddedDoc))
+
+        MyDoc.drop_collection()
+
+        MyDoc(subs={"a": EmbeddedDoc(name="foo")}).save()
+
+        mydoc = MyDoc.objects.first()
+        subdoc = mydoc.subs["a"]
+        subdoc.name = "bar"
+
+        assert ["name"] == subdoc._get_changed_fields()
+        assert ["subs.a.name"] == mydoc._get_changed_fields()
+
+        mydoc.subs["a"] = EmbeddedDoc()
+        assert ["subs.a"] == mydoc._get_changed_fields()
+        mydoc.save()
+
+        mydoc._clear_changed_fields()
+        assert [] == mydoc._get_changed_fields()
+
+    def test_referenced_object_changed_attributes(self):
+        """Ensures that when you save a new reference to a field, the referenced object isn't altered"""
+
+        class Organization(Document):
+            name = StringField()
+
+        class User(Document):
+            name = StringField()
+            org = ReferenceField("Organization", required=True)
+
+        Organization.drop_collection()
+        User.drop_collection()
+
+        org1 = Organization(name="Org 1")
+        org1.save()
+
+        org2 = Organization(name="Org 2")
+        org2.save()
+
+        user = User(name="Fred", org=org1)
+        user.save()
+
+        org1.reload()
+        org2.reload()
+        user.reload()
+        assert org1.name == "Org 1"
+        assert org2.name == "Org 2"
+        assert user.name == "Fred"
+
+        user.name = "Harold"
+        user.org = org2
+
+        org2.name = "New Org 2"
+        assert org2.name == "New Org 2"
+
+        user.save()
+        org2.save()
+
+        assert org2.name == "New Org 2"
+        org2.reload()
+        assert org2.name == "New Org 2"
+
+    def test_delta_for_nested_map_fields(self):
+        class UInfoDocument(Document):
+            phone = StringField()
+
+        class EmbeddedRole(EmbeddedDocument):
+            type = StringField()
+
+        class EmbeddedUser(EmbeddedDocument):
+            name = StringField()
+            roles = MapField(field=EmbeddedDocumentField(EmbeddedRole))
+            rolist = ListField(field=EmbeddedDocumentField(EmbeddedRole))
+            info = ReferenceField(UInfoDocument)
+
+        class Doc(Document):
+            users = MapField(field=EmbeddedDocumentField(EmbeddedUser))
+            num = IntField(default=-1)
+
+        Doc.drop_collection()
+
+        doc = Doc(num=1)
+        doc.users["007"] = EmbeddedUser(name="Agent007")
+        doc.save()
+
+        uinfo = UInfoDocument(phone="79089269066")
+        uinfo.save()
+
+        d = Doc.objects(num=1).first()
+        d.users["007"]["roles"]["666"] = EmbeddedRole(type="superadmin")
+        d.users["007"]["rolist"].append(EmbeddedRole(type="oops"))
+        d.users["007"]["info"] = uinfo
+        delta = d._delta()
+        assert True == ("users.007.roles.666" in delta[0])
+        assert True == ("users.007.rolist" in delta[0])
+        assert True == ("users.007.info" in delta[0])
+        assert "superadmin" == delta[0]["users.007.roles.666"]["type"]
+        assert "oops" == delta[0]["users.007.rolist"][0]["type"]
+        assert uinfo.id == delta[0]["users.007.info"]
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/document/test_dynamic.py

@@ -0,0 +1,426 @@
+import unittest
+
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+__all__ = ("TestDynamicDocument",)
+
+
+class TestDynamicDocument(MongoDBTestCase):
+    def setUp(self):
+        super(TestDynamicDocument, self).setUp()
+
+        class Person(DynamicDocument):
+            name = StringField()
+            meta = {"allow_inheritance": True}
+
+        Person.drop_collection()
+
+        self.Person = Person
+
+    def test_simple_dynamic_document(self):
+        """Ensures simple dynamic documents are saved correctly"""
+
+        p = self.Person()
+        p.name = "James"
+        p.age = 34
+
+        assert p.to_mongo() == {"_cls": "Person", "name": "James", "age": 34}
+        assert p.to_mongo().keys() == ["_cls", "name", "age"]
+        p.save()
+        assert p.to_mongo().keys() == ["_id", "_cls", "name", "age"]
+
+        assert self.Person.objects.first().age == 34
+
+        # Confirm no changes to self.Person
+        assert not hasattr(self.Person, "age")
+
+    def test_change_scope_of_variable(self):
+        """Test changing the scope of a dynamic field has no adverse effects"""
+        p = self.Person()
+        p.name = "Dean"
+        p.misc = 22
+        p.save()
+
+        p = self.Person.objects.get()
+        p.misc = {"hello": "world"}
+        p.save()
+
+        p = self.Person.objects.get()
+        assert p.misc == {"hello": "world"}
+
+    def test_delete_dynamic_field(self):
+        """Test deleting a dynamic field works"""
+        self.Person.drop_collection()
+        p = self.Person()
+        p.name = "Dean"
+        p.misc = 22
+        p.save()
+
+        p = self.Person.objects.get()
+        p.misc = {"hello": "world"}
+        p.save()
+
+        p = self.Person.objects.get()
+        assert p.misc == {"hello": "world"}
+        collection = self.db[self.Person._get_collection_name()]
+        obj = collection.find_one()
+        assert sorted(obj.keys()) == ["_cls", "_id", "misc", "name"]
+
+        del p.misc
+        p.save()
+
+        p = self.Person.objects.get()
+        assert not hasattr(p, "misc")
+
+        obj = collection.find_one()
+        assert sorted(obj.keys()) == ["_cls", "_id", "name"]
+
+    def test_reload_after_unsetting(self):
+        p = self.Person()
+        p.misc = 22
+        p.save()
+        p.update(unset__misc=1)
+        p.reload()
+
+    def test_reload_dynamic_field(self):
+        self.Person.objects.delete()
+        p = self.Person.objects.create()
+        p.update(age=1)
+
+        assert len(p._data) == 3
+        assert sorted(p._data.keys()) == ["_cls", "id", "name"]
+
+        p.reload()
+        assert len(p._data) == 4
+        assert sorted(p._data.keys()) == ["_cls", "age", "id", "name"]
+
+    def test_fields_without_underscore(self):
+        """Ensure we can query dynamic fields"""
+        Person = self.Person
+
+        p = self.Person(name="Dean")
+        p.save()
+
+        raw_p = Person.objects.as_pymongo().get(id=p.id)
+        assert raw_p == {"_cls": u"Person", "_id": p.id, "name": u"Dean"}
+
+        p.name = "OldDean"
+        p.newattr = "garbage"
+        p.save()
+        raw_p = Person.objects.as_pymongo().get(id=p.id)
+        assert raw_p == {
+            "_cls": u"Person",
+            "_id": p.id,
+            "name": "OldDean",
+            "newattr": u"garbage",
+        }
+
+    def test_fields_containing_underscore(self):
+        """Ensure we can query dynamic fields"""
+
+        class WeirdPerson(DynamicDocument):
+            name = StringField()
+            _name = StringField()
+
+        WeirdPerson.drop_collection()
+
+        p = WeirdPerson(name="Dean", _name="Dean")
+        p.save()
+
+        raw_p = WeirdPerson.objects.as_pymongo().get(id=p.id)
+        assert raw_p == {"_id": p.id, "_name": u"Dean", "name": u"Dean"}
+
+        p.name = "OldDean"
+        p._name = "NewDean"
+        p._newattr1 = "garbage"  # Unknown fields won't be added
+        p.save()
+        raw_p = WeirdPerson.objects.as_pymongo().get(id=p.id)
+        assert raw_p == {"_id": p.id, "_name": u"NewDean", "name": u"OldDean"}
+
+    def test_dynamic_document_queries(self):
+        """Ensure we can query dynamic fields"""
+        p = self.Person()
+        p.name = "Dean"
+        p.age = 22
+        p.save()
+
+        assert 1 == self.Person.objects(age=22).count()
+        p = self.Person.objects(age=22)
+        p = p.get()
+        assert 22 == p.age
+
+    def test_complex_dynamic_document_queries(self):
+        class Person(DynamicDocument):
+            name = StringField()
+
+        Person.drop_collection()
+
+        p = Person(name="test")
+        p.age = "ten"
+        p.save()
+
+        p1 = Person(name="test1")
+        p1.age = "less then ten and a half"
+        p1.save()
+
+        p2 = Person(name="test2")
+        p2.age = 10
+        p2.save()
+
+        assert Person.objects(age__icontains="ten").count() == 2
+        assert Person.objects(age__gte=10).count() == 1
+
+    def test_complex_data_lookups(self):
+        """Ensure you can query dynamic document dynamic fields"""
+        p = self.Person()
+        p.misc = {"hello": "world"}
+        p.save()
+
+        assert 1 == self.Person.objects(misc__hello="world").count()
+
+    def test_three_level_complex_data_lookups(self):
+        """Ensure you can query three level document dynamic fields"""
+        self.Person.objects.create(misc={"hello": {"hello2": "world"}})
+        assert 1 == self.Person.objects(misc__hello__hello2="world").count()
+
+    def test_complex_embedded_document_validation(self):
+        """Ensure embedded dynamic documents may be validated"""
+
+        class Embedded(DynamicEmbeddedDocument):
+            content = URLField()
+
+        class Doc(DynamicDocument):
+            pass
+
+        Doc.drop_collection()
+        doc = Doc()
+
+        embedded_doc_1 = Embedded(content="http://mongoengine.org")
+        embedded_doc_1.validate()
+
+        embedded_doc_2 = Embedded(content="this is not a url")
+        with pytest.raises(ValidationError):
+            embedded_doc_2.validate()
+
+        doc.embedded_field_1 = embedded_doc_1
+        doc.embedded_field_2 = embedded_doc_2
+        with pytest.raises(ValidationError):
+            doc.validate()
+
+    def test_inheritance(self):
+        """Ensure that dynamic document plays nice with inheritance"""
+
+        class Employee(self.Person):
+            salary = IntField()
+
+        Employee.drop_collection()
+
+        assert "name" in Employee._fields
+        assert "salary" in Employee._fields
+        assert Employee._get_collection_name() == self.Person._get_collection_name()
+
+        joe_bloggs = Employee()
+        joe_bloggs.name = "Joe Bloggs"
+        joe_bloggs.salary = 10
+        joe_bloggs.age = 20
+        joe_bloggs.save()
+
+        assert 1 == self.Person.objects(age=20).count()
+        assert 1 == Employee.objects(age=20).count()
+
+        joe_bloggs = self.Person.objects.first()
+        assert isinstance(joe_bloggs, Employee)
+
+    def test_embedded_dynamic_document(self):
+        """Test dynamic embedded documents"""
+
+        class Embedded(DynamicEmbeddedDocument):
+            pass
+
+        class Doc(DynamicDocument):
+            pass
+
+        Doc.drop_collection()
+        doc = Doc()
+
+        embedded_1 = Embedded()
+        embedded_1.string_field = "hello"
+        embedded_1.int_field = 1
+        embedded_1.dict_field = {"hello": "world"}
+        embedded_1.list_field = ["1", 2, {"hello": "world"}]
+        doc.embedded_field = embedded_1
+
+        assert doc.to_mongo() == {
+            "embedded_field": {
+                "_cls": "Embedded",
+                "string_field": "hello",
+                "int_field": 1,
+                "dict_field": {"hello": "world"},
+                "list_field": ["1", 2, {"hello": "world"}],
+            }
+        }
+        doc.save()
+
+        doc = Doc.objects.first()
+        assert doc.embedded_field.__class__ == Embedded
+        assert doc.embedded_field.string_field == "hello"
+        assert doc.embedded_field.int_field == 1
+        assert doc.embedded_field.dict_field == {"hello": "world"}
+        assert doc.embedded_field.list_field == ["1", 2, {"hello": "world"}]
+
+    def test_complex_embedded_documents(self):
+        """Test complex dynamic embedded documents setups"""
+
+        class Embedded(DynamicEmbeddedDocument):
+            pass
+
+        class Doc(DynamicDocument):
+            pass
+
+        Doc.drop_collection()
+        doc = Doc()
+
+        embedded_1 = Embedded()
+        embedded_1.string_field = "hello"
+        embedded_1.int_field = 1
+        embedded_1.dict_field = {"hello": "world"}
+
+        embedded_2 = Embedded()
+        embedded_2.string_field = "hello"
+        embedded_2.int_field = 1
+        embedded_2.dict_field = {"hello": "world"}
+        embedded_2.list_field = ["1", 2, {"hello": "world"}]
+
+        embedded_1.list_field = ["1", 2, embedded_2]
+        doc.embedded_field = embedded_1
+
+        assert doc.to_mongo() == {
+            "embedded_field": {
+                "_cls": "Embedded",
+                "string_field": "hello",
+                "int_field": 1,
+                "dict_field": {"hello": "world"},
+                "list_field": [
+                    "1",
+                    2,
+                    {
+                        "_cls": "Embedded",
+                        "string_field": "hello",
+                        "int_field": 1,
+                        "dict_field": {"hello": "world"},
+                        "list_field": ["1", 2, {"hello": "world"}],
+                    },
+                ],
+            }
+        }
+        doc.save()
+        doc = Doc.objects.first()
+        assert doc.embedded_field.__class__ == Embedded
+        assert doc.embedded_field.string_field == "hello"
+        assert doc.embedded_field.int_field == 1
+        assert doc.embedded_field.dict_field == {"hello": "world"}
+        assert doc.embedded_field.list_field[0] == "1"
+        assert doc.embedded_field.list_field[1] == 2
+
+        embedded_field = doc.embedded_field.list_field[2]
+
+        assert embedded_field.__class__ == Embedded
+        assert embedded_field.string_field == "hello"
+        assert embedded_field.int_field == 1
+        assert embedded_field.dict_field == {"hello": "world"}
+        assert embedded_field.list_field == ["1", 2, {"hello": "world"}]
+
+    def test_dynamic_and_embedded(self):
+        """Ensure embedded documents play nicely"""
+
+        class Address(EmbeddedDocument):
+            city = StringField()
+
+        class Person(DynamicDocument):
+            name = StringField()
+
+        Person.drop_collection()
+
+        Person(name="Ross", address=Address(city="London")).save()
+
+        person = Person.objects.first()
+        person.address.city = "Lundenne"
+        person.save()
+
+        assert Person.objects.first().address.city == "Lundenne"
+
+        person = Person.objects.first()
+        person.address = Address(city="Londinium")
+        person.save()
+
+        assert Person.objects.first().address.city == "Londinium"
+
+        person = Person.objects.first()
+        person.age = 35
+        person.save()
+        assert Person.objects.first().age == 35
+
+    def test_dynamic_embedded_works_with_only(self):
+        """Ensure custom fieldnames on a dynamic embedded document are found by qs.only()"""
+
+        class Address(DynamicEmbeddedDocument):
+            city = StringField()
+
+        class Person(DynamicDocument):
+            address = EmbeddedDocumentField(Address)
+
+        Person.drop_collection()
+
+        Person(
+            name="Eric", address=Address(city="San Francisco", street_number="1337")
+        ).save()
+
+        assert Person.objects.first().address.street_number == "1337"
+        assert (
+            Person.objects.only("address__street_number").first().address.street_number
+            == "1337"
+        )
+
+    def test_dynamic_and_embedded_dict_access(self):
+        """Ensure embedded dynamic documents work with dict[] style access"""
+
+        class Address(EmbeddedDocument):
+            city = StringField()
+
+        class Person(DynamicDocument):
+            name = StringField()
+
+        Person.drop_collection()
+
+        Person(name="Ross", address=Address(city="London")).save()
+
+        person = Person.objects.first()
+        person.attrval = "This works"
+
+        person["phone"] = "555-1212"  # but this should too
+
+        # Same thing two levels deep
+        person["address"]["city"] = "Lundenne"
+        person.save()
+
+        assert Person.objects.first().address.city == "Lundenne"
+
+        assert Person.objects.first().phone == "555-1212"
+
+        person = Person.objects.first()
+        person.address = Address(city="Londinium")
+        person.save()
+
+        assert Person.objects.first().address.city == "Londinium"
+
+        person = Person.objects.first()
+        person["age"] = 35
+        person.save()
+        assert Person.objects.first().age == 35
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/document/test_inheritance.py

@@ -0,0 +1,631 @@
+# -*- coding: utf-8 -*-
+import unittest
+import warnings
+
+import pytest
+
+from mongoengine import (
+    BooleanField,
+    Document,
+    EmbeddedDocument,
+    EmbeddedDocumentField,
+    GenericReferenceField,
+    IntField,
+    ReferenceField,
+    StringField,
+)
+from mongoengine.pymongo_support import list_collection_names
+from tests.fixtures import Base
+from tests.utils import MongoDBTestCase
+
+
+class TestInheritance(MongoDBTestCase):
+    def tearDown(self):
+        for collection in list_collection_names(self.db):
+            self.db.drop_collection(collection)
+
+    def test_constructor_cls(self):
+        # Ensures _cls is properly set during construction
+        # and when object gets reloaded (prevent regression of #1950)
+        class EmbedData(EmbeddedDocument):
+            data = StringField()
+            meta = {"allow_inheritance": True}
+
+        class DataDoc(Document):
+            name = StringField()
+            embed = EmbeddedDocumentField(EmbedData)
+            meta = {"allow_inheritance": True}
+
+        test_doc = DataDoc(name="test", embed=EmbedData(data="data"))
+        assert test_doc._cls == "DataDoc"
+        assert test_doc.embed._cls == "EmbedData"
+        test_doc.save()
+        saved_doc = DataDoc.objects.with_id(test_doc.id)
+        assert test_doc._cls == saved_doc._cls
+        assert test_doc.embed._cls == saved_doc.embed._cls
+        test_doc.delete()
+
+    def test_superclasses(self):
+        """Ensure that the correct list of superclasses is assembled.
+        """
+
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+        class Fish(Animal):
+            pass
+
+        class Guppy(Fish):
+            pass
+
+        class Mammal(Animal):
+            pass
+
+        class Dog(Mammal):
+            pass
+
+        class Human(Mammal):
+            pass
+
+        assert Animal._superclasses == ()
+        assert Fish._superclasses == ("Animal",)
+        assert Guppy._superclasses == ("Animal", "Animal.Fish")
+        assert Mammal._superclasses == ("Animal",)
+        assert Dog._superclasses == ("Animal", "Animal.Mammal")
+        assert Human._superclasses == ("Animal", "Animal.Mammal")
+
+    def test_external_superclasses(self):
+        """Ensure that the correct list of super classes is assembled when
+        importing part of the model.
+        """
+
+        class Animal(Base):
+            pass
+
+        class Fish(Animal):
+            pass
+
+        class Guppy(Fish):
+            pass
+
+        class Mammal(Animal):
+            pass
+
+        class Dog(Mammal):
+            pass
+
+        class Human(Mammal):
+            pass
+
+        assert Animal._superclasses == ("Base",)
+        assert Fish._superclasses == ("Base", "Base.Animal")
+        assert Guppy._superclasses == ("Base", "Base.Animal", "Base.Animal.Fish")
+        assert Mammal._superclasses == ("Base", "Base.Animal")
+        assert Dog._superclasses == ("Base", "Base.Animal", "Base.Animal.Mammal")
+        assert Human._superclasses == ("Base", "Base.Animal", "Base.Animal.Mammal")
+
+    def test_subclasses(self):
+        """Ensure that the correct list of _subclasses (subclasses) is
+        assembled.
+        """
+
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+        class Fish(Animal):
+            pass
+
+        class Guppy(Fish):
+            pass
+
+        class Mammal(Animal):
+            pass
+
+        class Dog(Mammal):
+            pass
+
+        class Human(Mammal):
+            pass
+
+        assert Animal._subclasses == (
+            "Animal",
+            "Animal.Fish",
+            "Animal.Fish.Guppy",
+            "Animal.Mammal",
+            "Animal.Mammal.Dog",
+            "Animal.Mammal.Human",
+        )
+        assert Fish._subclasses == ("Animal.Fish", "Animal.Fish.Guppy")
+        assert Guppy._subclasses == ("Animal.Fish.Guppy",)
+        assert Mammal._subclasses == (
+            "Animal.Mammal",
+            "Animal.Mammal.Dog",
+            "Animal.Mammal.Human",
+        )
+        assert Human._subclasses == ("Animal.Mammal.Human",)
+
+    def test_external_subclasses(self):
+        """Ensure that the correct list of _subclasses (subclasses) is
+        assembled when importing part of the model.
+        """
+
+        class Animal(Base):
+            pass
+
+        class Fish(Animal):
+            pass
+
+        class Guppy(Fish):
+            pass
+
+        class Mammal(Animal):
+            pass
+
+        class Dog(Mammal):
+            pass
+
+        class Human(Mammal):
+            pass
+
+        assert Animal._subclasses == (
+            "Base.Animal",
+            "Base.Animal.Fish",
+            "Base.Animal.Fish.Guppy",
+            "Base.Animal.Mammal",
+            "Base.Animal.Mammal.Dog",
+            "Base.Animal.Mammal.Human",
+        )
+        assert Fish._subclasses == ("Base.Animal.Fish", "Base.Animal.Fish.Guppy")
+        assert Guppy._subclasses == ("Base.Animal.Fish.Guppy",)
+        assert Mammal._subclasses == (
+            "Base.Animal.Mammal",
+            "Base.Animal.Mammal.Dog",
+            "Base.Animal.Mammal.Human",
+        )
+        assert Human._subclasses == ("Base.Animal.Mammal.Human",)
+
+    def test_dynamic_declarations(self):
+        """Test that declaring an extra class updates meta data"""
+
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+        assert Animal._superclasses == ()
+        assert Animal._subclasses == ("Animal",)
+
+        # Test dynamically adding a class changes the meta data
+        class Fish(Animal):
+            pass
+
+        assert Animal._superclasses == ()
+        assert Animal._subclasses == ("Animal", "Animal.Fish")
+
+        assert Fish._superclasses == ("Animal",)
+        assert Fish._subclasses == ("Animal.Fish",)
+
+        # Test dynamically adding an inherited class changes the meta data
+        class Pike(Fish):
+            pass
+
+        assert Animal._superclasses == ()
+        assert Animal._subclasses == ("Animal", "Animal.Fish", "Animal.Fish.Pike")
+
+        assert Fish._superclasses == ("Animal",)
+        assert Fish._subclasses == ("Animal.Fish", "Animal.Fish.Pike")
+
+        assert Pike._superclasses == ("Animal", "Animal.Fish")
+        assert Pike._subclasses == ("Animal.Fish.Pike",)
+
+    def test_inheritance_meta_data(self):
+        """Ensure that document may inherit fields from a superclass document.
+        """
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+
+            meta = {"allow_inheritance": True}
+
+        class Employee(Person):
+            salary = IntField()
+
+        assert ["_cls", "age", "id", "name", "salary"] == sorted(
+            Employee._fields.keys()
+        )
+        assert Employee._get_collection_name() == Person._get_collection_name()
+
+    def test_inheritance_to_mongo_keys(self):
+        """Ensure that document may inherit fields from a superclass document.
+        """
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+
+            meta = {"allow_inheritance": True}
+
+        class Employee(Person):
+            salary = IntField()
+
+        assert ["_cls", "age", "id", "name", "salary"] == sorted(
+            Employee._fields.keys()
+        )
+        assert Person(name="Bob", age=35).to_mongo().keys() == ["_cls", "name", "age"]
+        assert Employee(name="Bob", age=35, salary=0).to_mongo().keys() == [
+            "_cls",
+            "name",
+            "age",
+            "salary",
+        ]
+        assert Employee._get_collection_name() == Person._get_collection_name()
+
+    def test_indexes_and_multiple_inheritance(self):
+        """ Ensure that all of the indexes are created for a document with
+        multiple inheritance.
+        """
+
+        class A(Document):
+            a = StringField()
+
+            meta = {"allow_inheritance": True, "indexes": ["a"]}
+
+        class B(Document):
+            b = StringField()
+
+            meta = {"allow_inheritance": True, "indexes": ["b"]}
+
+        class C(A, B):
+            pass
+
+        A.drop_collection()
+        B.drop_collection()
+        C.drop_collection()
+
+        C.ensure_indexes()
+
+        assert sorted(
+            [idx["key"] for idx in C._get_collection().index_information().values()]
+        ) == sorted(
+            [[(u"_cls", 1), (u"b", 1)], [(u"_id", 1)], [(u"_cls", 1), (u"a", 1)]]
+        )
+
+    def test_polymorphic_queries(self):
+        """Ensure that the correct subclasses are returned from a query
+        """
+
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+        class Fish(Animal):
+            pass
+
+        class Mammal(Animal):
+            pass
+
+        class Dog(Mammal):
+            pass
+
+        class Human(Mammal):
+            pass
+
+        Animal.drop_collection()
+
+        Animal().save()
+        Fish().save()
+        Mammal().save()
+        Dog().save()
+        Human().save()
+
+        classes = [obj.__class__ for obj in Animal.objects]
+        assert classes == [Animal, Fish, Mammal, Dog, Human]
+
+        classes = [obj.__class__ for obj in Mammal.objects]
+        assert classes == [Mammal, Dog, Human]
+
+        classes = [obj.__class__ for obj in Human.objects]
+        assert classes == [Human]
+
+    def test_allow_inheritance(self):
+        """Ensure that inheritance is disabled by default on simple
+        classes and that _cls will not be used.
+        """
+
+        class Animal(Document):
+            name = StringField()
+
+        # can't inherit because Animal didn't explicitly allow inheritance
+        with pytest.raises(ValueError, match="Document Animal may not be subclassed"):
+
+            class Dog(Animal):
+                pass
+
+        # Check that _cls etc aren't present on simple documents
+        dog = Animal(name="dog").save()
+        assert dog.to_mongo().keys() == ["_id", "name"]
+
+        collection = self.db[Animal._get_collection_name()]
+        obj = collection.find_one()
+        assert "_cls" not in obj
+
+    def test_cant_turn_off_inheritance_on_subclass(self):
+        """Ensure if inheritance is on in a subclass you cant turn it off.
+        """
+
+        class Animal(Document):
+            name = StringField()
+            meta = {"allow_inheritance": True}
+
+        with pytest.raises(ValueError) as exc_info:
+
+            class Mammal(Animal):
+                meta = {"allow_inheritance": False}
+
+        assert (
+            str(exc_info.value)
+            == 'Only direct subclasses of Document may set "allow_inheritance" to False'
+        )
+
+    def test_allow_inheritance_abstract_document(self):
+        """Ensure that abstract documents can set inheritance rules and that
+        _cls will not be used.
+        """
+
+        class FinalDocument(Document):
+            meta = {"abstract": True, "allow_inheritance": False}
+
+        class Animal(FinalDocument):
+            name = StringField()
+
+        with pytest.raises(ValueError):
+
+            class Mammal(Animal):
+                pass
+
+        # Check that _cls isn't present in simple documents
+        doc = Animal(name="dog")
+        assert "_cls" not in doc.to_mongo()
+
+    def test_using_abstract_class_in_reference_field(self):
+        # Ensures no regression of #1920
+        class AbstractHuman(Document):
+            meta = {"abstract": True}
+
+        class Dad(AbstractHuman):
+            name = StringField()
+
+        class Home(Document):
+            dad = ReferenceField(AbstractHuman)  # Referencing the abstract class
+            address = StringField()
+
+        dad = Dad(name="5").save()
+        Home(dad=dad, address="street").save()
+
+        home = Home.objects.first()
+        home.address = "garbage"
+        home.save()  # Was failing with ValidationError
+
+    def test_abstract_class_referencing_self(self):
+        # Ensures no regression of #1920
+        class Human(Document):
+            meta = {"abstract": True}
+            creator = ReferenceField("self", dbref=True)
+
+        class User(Human):
+            name = StringField()
+
+        user = User(name="John").save()
+        user2 = User(name="Foo", creator=user).save()
+
+        user2 = User.objects.with_id(user2.id)
+        user2.name = "Bar"
+        user2.save()  # Was failing with ValidationError
+
+    def test_abstract_handle_ids_in_metaclass_properly(self):
+        class City(Document):
+            continent = StringField()
+            meta = {"abstract": True, "allow_inheritance": False}
+
+        class EuropeanCity(City):
+            name = StringField()
+
+        berlin = EuropeanCity(name="Berlin", continent="Europe")
+        assert len(berlin._db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._reverse_db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._fields_ordered) == 3
+        assert berlin._fields_ordered[0] == "id"
+
+    def test_auto_id_not_set_if_specific_in_parent_class(self):
+        class City(Document):
+            continent = StringField()
+            city_id = IntField(primary_key=True)
+            meta = {"abstract": True, "allow_inheritance": False}
+
+        class EuropeanCity(City):
+            name = StringField()
+
+        berlin = EuropeanCity(name="Berlin", continent="Europe")
+        assert len(berlin._db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._reverse_db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._fields_ordered) == 3
+        assert berlin._fields_ordered[0] == "city_id"
+
+    def test_auto_id_vs_non_pk_id_field(self):
+        class City(Document):
+            continent = StringField()
+            id = IntField()
+            meta = {"abstract": True, "allow_inheritance": False}
+
+        class EuropeanCity(City):
+            name = StringField()
+
+        berlin = EuropeanCity(name="Berlin", continent="Europe")
+        assert len(berlin._db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._reverse_db_field_map) == len(berlin._fields_ordered)
+        assert len(berlin._fields_ordered) == 4
+        assert berlin._fields_ordered[0] == "auto_id_0"
+        berlin.save()
+        assert berlin.pk == berlin.auto_id_0
+
+    def test_abstract_document_creation_does_not_fail(self):
+        class City(Document):
+            continent = StringField()
+            meta = {"abstract": True, "allow_inheritance": False}
+
+        city = City(continent="asia")
+        assert city.pk is None
+        # TODO: expected error? Shouldn't we create a new error type?
+        with pytest.raises(KeyError):
+            setattr(city, "pk", 1)
+
+    def test_allow_inheritance_embedded_document(self):
+        """Ensure embedded documents respect inheritance."""
+
+        class Comment(EmbeddedDocument):
+            content = StringField()
+
+        with pytest.raises(ValueError):
+
+            class SpecialComment(Comment):
+                pass
+
+        doc = Comment(content="test")
+        assert "_cls" not in doc.to_mongo()
+
+        class Comment(EmbeddedDocument):
+            content = StringField()
+            meta = {"allow_inheritance": True}
+
+        doc = Comment(content="test")
+        assert "_cls" in doc.to_mongo()
+
+    def test_document_inheritance(self):
+        """Ensure mutliple inheritance of abstract documents
+        """
+
+        class DateCreatedDocument(Document):
+            meta = {"allow_inheritance": True, "abstract": True}
+
+        class DateUpdatedDocument(Document):
+            meta = {"allow_inheritance": True, "abstract": True}
+
+        try:
+
+            class MyDocument(DateCreatedDocument, DateUpdatedDocument):
+                pass
+
+        except Exception:
+            assert False, "Couldn't create MyDocument class"
+
+    def test_abstract_documents(self):
+        """Ensure that a document superclass can be marked as abstract
+        thereby not using it as the name for the collection."""
+
+        defaults = {
+            "index_background": True,
+            "index_opts": {"hello": "world"},
+            "allow_inheritance": True,
+            "queryset_class": "QuerySet",
+            "db_alias": "myDB",
+            "shard_key": ("hello", "world"),
+        }
+
+        meta_settings = {"abstract": True}
+        meta_settings.update(defaults)
+
+        class Animal(Document):
+            name = StringField()
+            meta = meta_settings
+
+        class Fish(Animal):
+            pass
+
+        class Guppy(Fish):
+            pass
+
+        class Mammal(Animal):
+            meta = {"abstract": True}
+
+        class Human(Mammal):
+            pass
+
+        for k, v in defaults.items():
+            for cls in [Animal, Fish, Guppy]:
+                assert cls._meta[k] == v
+
+        assert "collection" not in Animal._meta
+        assert "collection" not in Mammal._meta
+
+        assert Animal._get_collection_name() is None
+        assert Mammal._get_collection_name() is None
+
+        assert Fish._get_collection_name() == "fish"
+        assert Guppy._get_collection_name() == "fish"
+        assert Human._get_collection_name() == "human"
+
+        # ensure that a subclass of a non-abstract class can't be abstract
+        with pytest.raises(ValueError):
+
+            class EvilHuman(Human):
+                evil = BooleanField(default=True)
+                meta = {"abstract": True}
+
+    def test_abstract_embedded_documents(self):
+        # 789: EmbeddedDocument shouldn't inherit abstract
+        class A(EmbeddedDocument):
+            meta = {"abstract": True}
+
+        class B(A):
+            pass
+
+        assert not B._meta["abstract"]
+
+    def test_inherited_collections(self):
+        """Ensure that subclassed documents don't override parents'
+        collections
+        """
+
+        class Drink(Document):
+            name = StringField()
+            meta = {"allow_inheritance": True}
+
+        class Drinker(Document):
+            drink = GenericReferenceField()
+
+        try:
+            warnings.simplefilter("error")
+
+            class AcloholicDrink(Drink):
+                meta = {"collection": "booze"}
+
+        except SyntaxWarning:
+            warnings.simplefilter("ignore")
+
+            class AlcoholicDrink(Drink):
+                meta = {"collection": "booze"}
+
+        else:
+            raise AssertionError("SyntaxWarning should be triggered")
+
+        warnings.resetwarnings()
+
+        Drink.drop_collection()
+        AlcoholicDrink.drop_collection()
+        Drinker.drop_collection()
+
+        red_bull = Drink(name="Red Bull")
+        red_bull.save()
+
+        programmer = Drinker(drink=red_bull)
+        programmer.save()
+
+        beer = AlcoholicDrink(name="Beer")
+        beer.save()
+        real_person = Drinker(drink=beer)
+        real_person.save()
+
+        assert Drinker.objects[0].drink.name == red_bull.name
+        assert Drinker.objects[1].drink.name == beer.name
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/document/test_json_serialisation.py

@@ -0,0 +1,106 @@
+import unittest
+import uuid
+
+from datetime import datetime
+from bson import ObjectId
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestJson(MongoDBTestCase):
+    def test_json_names(self):
+        """
+        Going to test reported issue:
+            https://github.com/MongoEngine/mongoengine/issues/654
+        where the reporter asks for the availability to perform
+        a to_json with the original class names and not the abreviated
+        mongodb document keys
+        """
+
+        class Embedded(EmbeddedDocument):
+            string = StringField(db_field="s")
+
+        class Doc(Document):
+            string = StringField(db_field="s")
+            embedded = EmbeddedDocumentField(Embedded, db_field="e")
+
+        doc = Doc(string="Hello", embedded=Embedded(string="Inner Hello"))
+        doc_json = doc.to_json(
+            sort_keys=True, use_db_field=False, separators=(",", ":")
+        )
+
+        expected_json = """{"embedded":{"string":"Inner Hello"},"string":"Hello"}"""
+
+        assert doc_json == expected_json
+
+    def test_json_simple(self):
+        class Embedded(EmbeddedDocument):
+            string = StringField()
+
+        class Doc(Document):
+            string = StringField()
+            embedded_field = EmbeddedDocumentField(Embedded)
+
+            def __eq__(self, other):
+                return (
+                    self.string == other.string
+                    and self.embedded_field == other.embedded_field
+                )
+
+        doc = Doc(string="Hi", embedded_field=Embedded(string="Hi"))
+
+        doc_json = doc.to_json(sort_keys=True, separators=(",", ":"))
+        expected_json = """{"embedded_field":{"string":"Hi"},"string":"Hi"}"""
+        assert doc_json == expected_json
+
+        assert doc == Doc.from_json(doc.to_json())
+
+    def test_json_complex(self):
+        class EmbeddedDoc(EmbeddedDocument):
+            pass
+
+        class Simple(Document):
+            pass
+
+        class Doc(Document):
+            string_field = StringField(default="1")
+            int_field = IntField(default=1)
+            float_field = FloatField(default=1.1)
+            boolean_field = BooleanField(default=True)
+            datetime_field = DateTimeField(default=datetime.now)
+            embedded_document_field = EmbeddedDocumentField(
+                EmbeddedDoc, default=lambda: EmbeddedDoc()
+            )
+            list_field = ListField(default=lambda: [1, 2, 3])
+            dict_field = DictField(default=lambda: {"hello": "world"})
+            objectid_field = ObjectIdField(default=ObjectId)
+            reference_field = ReferenceField(Simple, default=lambda: Simple().save())
+            map_field = MapField(IntField(), default=lambda: {"simple": 1})
+            decimal_field = DecimalField(default=1.0)
+            complex_datetime_field = ComplexDateTimeField(default=datetime.now)
+            url_field = URLField(default="http://mongoengine.org")
+            dynamic_field = DynamicField(default=1)
+            generic_reference_field = GenericReferenceField(
+                default=lambda: Simple().save()
+            )
+            sorted_list_field = SortedListField(IntField(), default=lambda: [1, 2, 3])
+            email_field = EmailField(default="ross@example.com")
+            geo_point_field = GeoPointField(default=lambda: [1, 2])
+            sequence_field = SequenceField()
+            uuid_field = UUIDField(default=uuid.uuid4)
+            generic_embedded_document_field = GenericEmbeddedDocumentField(
+                default=lambda: EmbeddedDoc()
+            )
+
+            def __eq__(self, other):
+                import json
+
+                return json.loads(self.to_json()) == json.loads(other.to_json())
+
+        doc = Doc()
+        assert doc == Doc.from_json(doc.to_json())
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/document/test_validation.py

@@ -0,0 +1,223 @@
+# -*- coding: utf-8 -*-
+import unittest
+from datetime import datetime
+
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestValidatorError(MongoDBTestCase):
+    def test_to_dict(self):
+        """Ensure a ValidationError handles error to_dict correctly.
+        """
+        error = ValidationError("root")
+        assert error.to_dict() == {}
+
+        # 1st level error schema
+        error.errors = {"1st": ValidationError("bad 1st")}
+        assert "1st" in error.to_dict()
+        assert error.to_dict()["1st"] == "bad 1st"
+
+        # 2nd level error schema
+        error.errors = {
+            "1st": ValidationError(
+                "bad 1st", errors={"2nd": ValidationError("bad 2nd")}
+            )
+        }
+        assert "1st" in error.to_dict()
+        assert isinstance(error.to_dict()["1st"], dict)
+        assert "2nd" in error.to_dict()["1st"]
+        assert error.to_dict()["1st"]["2nd"] == "bad 2nd"
+
+        # moar levels
+        error.errors = {
+            "1st": ValidationError(
+                "bad 1st",
+                errors={
+                    "2nd": ValidationError(
+                        "bad 2nd",
+                        errors={
+                            "3rd": ValidationError(
+                                "bad 3rd", errors={"4th": ValidationError("Inception")}
+                            )
+                        },
+                    )
+                },
+            )
+        }
+        assert "1st" in error.to_dict()
+        assert "2nd" in error.to_dict()["1st"]
+        assert "3rd" in error.to_dict()["1st"]["2nd"]
+        assert "4th" in error.to_dict()["1st"]["2nd"]["3rd"]
+        assert error.to_dict()["1st"]["2nd"]["3rd"]["4th"] == "Inception"
+
+        assert error.message == "root(2nd.3rd.4th.Inception: ['1st'])"
+
+    def test_model_validation(self):
+        class User(Document):
+            username = StringField(primary_key=True)
+            name = StringField(required=True)
+
+        try:
+            User().validate()
+        except ValidationError as e:
+            assert "User:None" in e.message
+            assert e.to_dict() == {
+                "username": "Field is required",
+                "name": "Field is required",
+            }
+
+        user = User(username="RossC0", name="Ross").save()
+        user.name = None
+        try:
+            user.save()
+        except ValidationError as e:
+            assert "User:RossC0" in e.message
+            assert e.to_dict() == {"name": "Field is required"}
+
+    def test_fields_rewrite(self):
+        class BasePerson(Document):
+            name = StringField()
+            age = IntField()
+            meta = {"abstract": True}
+
+        class Person(BasePerson):
+            name = StringField(required=True)
+
+        p = Person(age=15)
+        with pytest.raises(ValidationError):
+            p.validate()
+
+    def test_embedded_document_validation(self):
+        """Ensure that embedded documents may be validated.
+        """
+
+        class Comment(EmbeddedDocument):
+            date = DateTimeField()
+            content = StringField(required=True)
+
+        comment = Comment()
+        with pytest.raises(ValidationError):
+            comment.validate()
+
+        comment.content = "test"
+        comment.validate()
+
+        comment.date = 4
+        with pytest.raises(ValidationError):
+            comment.validate()
+
+        comment.date = datetime.now()
+        comment.validate()
+        assert comment._instance is None
+
+    def test_embedded_db_field_validate(self):
+        class SubDoc(EmbeddedDocument):
+            val = IntField(required=True)
+
+        class Doc(Document):
+            id = StringField(primary_key=True)
+            e = EmbeddedDocumentField(SubDoc, db_field="eb")
+
+        try:
+            Doc(id="bad").validate()
+        except ValidationError as e:
+            assert "SubDoc:None" in e.message
+            assert e.to_dict() == {"e": {"val": "OK could not be converted to int"}}
+
+        Doc.drop_collection()
+
+        Doc(id="test", e=SubDoc(val=15)).save()
+
+        doc = Doc.objects.first()
+        keys = doc._data.keys()
+        assert 2 == len(keys)
+        assert "e" in keys
+        assert "id" in keys
+
+        doc.e.val = "OK"
+        try:
+            doc.save()
+        except ValidationError as e:
+            assert "Doc:test" in e.message
+            assert e.to_dict() == {"e": {"val": "OK could not be converted to int"}}
+
+    def test_embedded_weakref(self):
+        class SubDoc(EmbeddedDocument):
+            val = IntField(required=True)
+
+        class Doc(Document):
+            e = EmbeddedDocumentField(SubDoc, db_field="eb")
+
+        Doc.drop_collection()
+
+        d1 = Doc()
+        d2 = Doc()
+
+        s = SubDoc()
+
+        with pytest.raises(ValidationError):
+            s.validate()
+
+        d1.e = s
+        d2.e = s
+
+        del d1
+
+        with pytest.raises(ValidationError):
+            d2.validate()
+
+    def test_parent_reference_in_child_document(self):
+        """
+        Test to ensure a ReferenceField can store a reference to a parent
+        class when inherited. Issue #954.
+        """
+
+        class Parent(Document):
+            meta = {"allow_inheritance": True}
+            reference = ReferenceField("self")
+
+        class Child(Parent):
+            pass
+
+        parent = Parent()
+        parent.save()
+
+        child = Child(reference=parent)
+
+        # Saving child should not raise a ValidationError
+        try:
+            child.save()
+        except ValidationError as e:
+            self.fail("ValidationError raised: %s" % e.message)
+
+    def test_parent_reference_set_as_attribute_in_child_document(self):
+        """
+        Test to ensure a ReferenceField can store a reference to a parent
+        class when inherited and when set via attribute. Issue #954.
+        """
+
+        class Parent(Document):
+            meta = {"allow_inheritance": True}
+            reference = ReferenceField("self")
+
+        class Child(Parent):
+            pass
+
+        parent = Parent()
+        parent.save()
+
+        child = Child()
+        child.reference = parent
+
+        # Saving the child should not raise a ValidationError
+        try:
+            child.save()
+        except ValidationError as e:
+            self.fail("ValidationError raised: %s" % e.message)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/fields.py

@@ -1,264 +0,0 @@
-import unittest
-import datetime
-
-from mongoengine import *
-from mongoengine.connection import _get_db
-
-
-class FieldTest(unittest.TestCase):
-
-    def setUp(self):
-        connect(db='mongoenginetest')
-        self.db = _get_db()
-
-    def test_default_values(self):
-        """Ensure that default field values are used when creating a document.
-        """
-        class Person(Document):
-            name = StringField()
-            age = IntField(default=30)
-            userid = StringField(default=lambda: 'test')
-
-        person = Person(name='Test Person')
-        self.assertEqual(person._data['age'], 30)
-        self.assertEqual(person._data['userid'], 'test')
-
-    def test_required_values(self):
-        """Ensure that required field constraints are enforced.
-        """
-        class Person(Document):
-            name = StringField(required=True)
-            age = IntField(required=True)
-            userid = StringField()
-
-        person = Person(name="Test User")
-        self.assertRaises(ValidationError, person.validate)
-        person = Person(age=30)
-        self.assertRaises(ValidationError, person.validate)
-
-    def test_object_id_validation(self):
-        """Ensure that invalid values cannot be assigned to string fields.
-        """
-        class Person(Document):
-            name = StringField()
-        
-        person = Person(name='Test User')
-        self.assertEqual(person.id, None)
-
-        person.id = 47
-        self.assertRaises(ValidationError, person.validate)
-
-        person.id = 'abc'
-        self.assertRaises(ValidationError, person.validate)
-
-        person.id = '497ce96f395f2f052a494fd4'
-        person.validate()
-
-    def test_string_validation(self):
-        """Ensure that invalid values cannot be assigned to string fields.
-        """
-        class Person(Document):
-            name = StringField(max_length=20)
-            userid = StringField(r'[0-9a-z_]+$')
-
-        person = Person(name=34)
-        self.assertRaises(ValidationError, person.validate)
-
-        # Test regex validation on userid
-        person = Person(userid='test.User')
-        self.assertRaises(ValidationError, person.validate)
-
-        person.userid = 'test_user'
-        self.assertEqual(person.userid, 'test_user')
-        person.validate()
-
-        # Test max length validation on name
-        person = Person(name='Name that is more than twenty characters')
-        self.assertRaises(ValidationError, person.validate)
-
-        person.name = 'Shorter name'
-        person.validate()
-
-    def test_int_validation(self):
-        """Ensure that invalid values cannot be assigned to int fields.
-        """
-        class Person(Document):
-            age = IntField(min_value=0, max_value=110)
-
-        person = Person()
-        person.age = 50
-        person.validate()
-
-        person.age = -1
-        self.assertRaises(ValidationError, person.validate)
-        person.age = 120
-        self.assertRaises(ValidationError, person.validate)
-        person.age = 'ten'
-        self.assertRaises(ValidationError, person.validate)
-
-    def test_float_validation(self):
-        """Ensure that invalid values cannot be assigned to float fields.
-        """
-        class Person(Document):
-            height = FloatField(min_value=0.1, max_value=3.5)
-
-        person = Person()
-        person.height = 1.89
-        person.validate()
-
-        person.height = 2
-        self.assertRaises(ValidationError, person.validate)
-        person.height = 0.01
-        self.assertRaises(ValidationError, person.validate)
-        person.height = 4.0
-        self.assertRaises(ValidationError, person.validate)
-
-    def test_boolean_validation(self):
-        """Ensure that invalid values cannot be assigned to boolean fields.
-        """
-        class Person(Document):
-            admin = BooleanField()
-
-        person = Person()
-        person.admin = True
-        person.validate()
-
-        person.admin = 2
-        self.assertRaises(ValidationError, person.validate)
-        person.admin = 'Yes'
-        self.assertRaises(ValidationError, person.validate)
-
-    def test_datetime_validation(self):
-        """Ensure that invalid values cannot be assigned to datetime fields.
-        """
-        class LogEntry(Document):
-            time = DateTimeField()
-
-        log = LogEntry()
-        log.time = datetime.datetime.now()
-        log.validate()
-
-        log.time = -1
-        self.assertRaises(ValidationError, log.validate)
-        log.time = '1pm'
-        self.assertRaises(ValidationError, log.validate)
-
-    def test_list_validation(self):
-        """Ensure that a list field only accepts lists with valid elements.
-        """
-        class Comment(EmbeddedDocument):
-            content = StringField()
-
-        class BlogPost(Document):
-            content = StringField()
-            comments = ListField(EmbeddedDocumentField(Comment))
-            tags = ListField(StringField())
-
-        post = BlogPost(content='Went for a walk today...')
-        post.validate()
-
-        post.tags = 'fun'
-        self.assertRaises(ValidationError, post.validate)
-        post.tags = [1, 2]
-        self.assertRaises(ValidationError, post.validate)
-
-        post.tags = ['fun', 'leisure']
-        post.validate()
-        post.tags = ('fun', 'leisure')
-        post.validate()
-
-        comments = [Comment(content='Good for you'), Comment(content='Yay.')]
-        post.comments = comments
-        post.validate()
-
-        post.comments = ['a']
-        self.assertRaises(ValidationError, post.validate)
-        post.comments = 'yay'
-        self.assertRaises(ValidationError, post.validate)
-
-    def test_embedded_document_validation(self):
-        """Ensure that invalid embedded documents cannot be assigned to
-        embedded document fields.
-        """
-        class Comment(EmbeddedDocument):
-            content = StringField()
-
-        class PersonPreferences(EmbeddedDocument):
-            food = StringField()
-            number = IntField()
-
-        class Person(Document):
-            name = StringField()
-            preferences = EmbeddedDocumentField(PersonPreferences)
-
-        person = Person(name='Test User')
-        person.preferences = 'My Preferences'
-        self.assertRaises(ValidationError, person.validate)
-
-        person.preferences = Comment(content='Nice blog post...')
-        self.assertRaises(ValidationError, person.validate)
-
-        person.preferences = PersonPreferences(food='Cheese', number=47)
-        self.assertEqual(person.preferences.food, 'Cheese')
-        person.validate()
-
-    def test_embedded_document_inheritance(self):
-        """Ensure that subclasses of embedded documents may be provided to 
-        EmbeddedDocumentFields of the superclass' type.
-        """
-        class User(EmbeddedDocument):
-            name = StringField()
-
-        class PowerUser(User):
-            power = IntField()
-
-        class BlogPost(Document):
-            content = StringField()
-            author = EmbeddedDocumentField(User)
-        
-        post = BlogPost(content='What I did today...')
-        post.author = User(name='Test User')
-        post.author = PowerUser(name='Test User', power=47)
-
-    def test_reference_validation(self):
-        """Ensure that invalid docment objects cannot be assigned to reference
-        fields.
-        """
-        class User(Document):
-            name = StringField()
-
-        class BlogPost(Document):
-            content = StringField()
-            author = ReferenceField(User)
-
-        User.drop_collection()
-        BlogPost.drop_collection()
-
-        self.assertRaises(ValidationError, ReferenceField, EmbeddedDocument)
-
-        user = User(name='Test User')
-
-        # Ensure that the referenced object must have been saved
-        post1 = BlogPost(content='Chips and gravy taste good.')
-        post1.author = user
-        self.assertRaises(ValidationError, post1.save)
-
-        # Check that an invalid object type cannot be used
-        post2 = BlogPost(content='Chips and chilli taste good.')
-        post1.author = post2
-        self.assertRaises(ValidationError, post1.validate)
-
-        user.save()
-        post1.author = user
-        post1.save()
-
-        post2.save()
-        post1.author = post2
-        self.assertRaises(ValidationError, post1.validate)
-
-        User.drop_collection()
-        BlogPost.drop_collection()
-
-
-if __name__ == '__main__':
-    unittest.main()

tests/fields/test_binary_field.py

@@ -0,0 +1,143 @@
+# -*- coding: utf-8 -*-
+import uuid
+
+from bson import Binary
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+BIN_VALUE = "\xa9\xf3\x8d(\xd7\x03\x84\xb4k[\x0f\xe3\xa2\x19\x85p[J\xa3\xd2>\xde\xe6\x87\xb1\x7f\xc6\xe6\xd9r\x18\xf5".encode(
+    "latin-1"
+)
+
+
+class TestBinaryField(MongoDBTestCase):
+    def test_binary_fields(self):
+        """Ensure that binary fields can be stored and retrieved.
+        """
+
+        class Attachment(Document):
+            content_type = StringField()
+            blob = BinaryField()
+
+        BLOB = "\xe6\x00\xc4\xff\x07".encode("latin-1")
+        MIME_TYPE = "application/octet-stream"
+
+        Attachment.drop_collection()
+
+        attachment = Attachment(content_type=MIME_TYPE, blob=BLOB)
+        attachment.save()
+
+        attachment_1 = Attachment.objects().first()
+        assert MIME_TYPE == attachment_1.content_type
+        assert BLOB == bytes(attachment_1.blob)
+
+    def test_validation_succeeds(self):
+        """Ensure that valid values can be assigned to binary fields.
+        """
+
+        class AttachmentRequired(Document):
+            blob = BinaryField(required=True)
+
+        class AttachmentSizeLimit(Document):
+            blob = BinaryField(max_bytes=4)
+
+        attachment_required = AttachmentRequired()
+        with pytest.raises(ValidationError):
+            attachment_required.validate()
+        attachment_required.blob = Binary("\xe6\x00\xc4\xff\x07".encode("latin-1"))
+        attachment_required.validate()
+
+        _5_BYTES = "\xe6\x00\xc4\xff\x07".encode("latin-1")
+        _4_BYTES = "\xe6\x00\xc4\xff".encode("latin-1")
+        with pytest.raises(ValidationError):
+            AttachmentSizeLimit(blob=_5_BYTES).validate()
+        AttachmentSizeLimit(blob=_4_BYTES).validate()
+
+    def test_validation_fails(self):
+        """Ensure that invalid values cannot be assigned to binary fields."""
+
+        class Attachment(Document):
+            blob = BinaryField()
+
+        for invalid_data in (2, u"Im_a_unicode", ["some_str"]):
+            with pytest.raises(ValidationError):
+                Attachment(blob=invalid_data).validate()
+
+    def test__primary(self):
+        class Attachment(Document):
+            id = BinaryField(primary_key=True)
+
+        Attachment.drop_collection()
+        binary_id = uuid.uuid4().bytes
+        att = Attachment(id=binary_id).save()
+        assert 1 == Attachment.objects.count()
+        assert 1 == Attachment.objects.filter(id=att.id).count()
+        att.delete()
+        assert 0 == Attachment.objects.count()
+
+    def test_primary_filter_by_binary_pk_as_str(self):
+        class Attachment(Document):
+            id = BinaryField(primary_key=True)
+
+        Attachment.drop_collection()
+        binary_id = uuid.uuid4().bytes
+        att = Attachment(id=binary_id).save()
+        assert 1 == Attachment.objects.filter(id=binary_id).count()
+        att.delete()
+        assert 0 == Attachment.objects.count()
+
+    def test_match_querying_with_bytes(self):
+        class MyDocument(Document):
+            bin_field = BinaryField()
+
+        MyDocument.drop_collection()
+
+        doc = MyDocument(bin_field=BIN_VALUE).save()
+        matched_doc = MyDocument.objects(bin_field=BIN_VALUE).first()
+        assert matched_doc.id == doc.id
+
+    def test_match_querying_with_binary(self):
+        class MyDocument(Document):
+            bin_field = BinaryField()
+
+        MyDocument.drop_collection()
+
+        doc = MyDocument(bin_field=BIN_VALUE).save()
+
+        matched_doc = MyDocument.objects(bin_field=Binary(BIN_VALUE)).first()
+        assert matched_doc.id == doc.id
+
+    def test_modify_operation__set(self):
+        """Ensures no regression of bug #1127"""
+
+        class MyDocument(Document):
+            some_field = StringField()
+            bin_field = BinaryField()
+
+        MyDocument.drop_collection()
+
+        doc = MyDocument.objects(some_field="test").modify(
+            upsert=True, new=True, set__bin_field=BIN_VALUE
+        )
+        assert doc.some_field == "test"
+        assert doc.bin_field == BIN_VALUE
+
+    def test_update_one(self):
+        """Ensures no regression of bug #1127"""
+
+        class MyDocument(Document):
+            bin_field = BinaryField()
+
+        MyDocument.drop_collection()
+
+        bin_data = "\xe6\x00\xc4\xff\x07".encode("latin-1")
+        doc = MyDocument(bin_field=bin_data).save()
+
+        n_updated = MyDocument.objects(bin_field=bin_data).update_one(
+            bin_field=BIN_VALUE
+        )
+        assert n_updated == 1
+        fetched = MyDocument.objects.with_id(doc.id)
+        assert fetched.bin_field == BIN_VALUE

tests/fields/test_boolean_field.py

@@ -0,0 +1,52 @@
+# -*- coding: utf-8 -*-
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase, get_as_pymongo
+
+
+class TestBooleanField(MongoDBTestCase):
+    def test_storage(self):
+        class Person(Document):
+            admin = BooleanField()
+
+        person = Person(admin=True)
+        person.save()
+        assert get_as_pymongo(person) == {"_id": person.id, "admin": True}
+
+    def test_validation(self):
+        """Ensure that invalid values cannot be assigned to boolean
+        fields.
+        """
+
+        class Person(Document):
+            admin = BooleanField()
+
+        person = Person()
+        person.admin = True
+        person.validate()
+
+        person.admin = 2
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.admin = "Yes"
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.admin = "False"
+        with pytest.raises(ValidationError):
+            person.validate()
+
+    def test_weirdness_constructor(self):
+        """When attribute is set in contructor, it gets cast into a bool
+        which causes some weird behavior. We dont necessarily want to maintain this behavior
+        but its a known issue
+        """
+
+        class Person(Document):
+            admin = BooleanField()
+
+        new_person = Person(admin="False")
+        assert new_person.admin
+
+        new_person = Person(admin="0")
+        assert new_person.admin

tests/fields/test_cached_reference_field.py

@@ -0,0 +1,377 @@
+# -*- coding: utf-8 -*-
+from decimal import Decimal
+
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestCachedReferenceField(MongoDBTestCase):
+    def test_get_and_save(self):
+        """
+        Tests #1047: CachedReferenceField creates DBRefs on to_python,
+        but can't save them on to_mongo.
+        """
+
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocorrence(Document):
+            person = StringField()
+            animal = CachedReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocorrence.drop_collection()
+
+        Ocorrence(
+            person="testte", animal=Animal(name="Leopard", tag="heavy").save()
+        ).save()
+        p = Ocorrence.objects.get()
+        p.person = "new_testte"
+        p.save()
+
+    def test_general_things(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocorrence(Document):
+            person = StringField()
+            animal = CachedReferenceField(Animal, fields=["tag"])
+
+        Animal.drop_collection()
+        Ocorrence.drop_collection()
+
+        a = Animal(name="Leopard", tag="heavy")
+        a.save()
+
+        assert Animal._cached_reference_fields == [Ocorrence.animal]
+        o = Ocorrence(person="teste", animal=a)
+        o.save()
+
+        p = Ocorrence(person="Wilson")
+        p.save()
+
+        assert Ocorrence.objects(animal=None).count() == 1
+
+        assert a.to_mongo(fields=["tag"]) == {"tag": "heavy", "_id": a.pk}
+
+        assert o.to_mongo()["animal"]["tag"] == "heavy"
+
+        # counts
+        Ocorrence(person="teste 2").save()
+        Ocorrence(person="teste 3").save()
+
+        count = Ocorrence.objects(animal__tag="heavy").count()
+        assert count == 1
+
+        ocorrence = Ocorrence.objects(animal__tag="heavy").first()
+        assert ocorrence.person == "teste"
+        assert isinstance(ocorrence.animal, Animal)
+
+    def test_with_decimal(self):
+        class PersonAuto(Document):
+            name = StringField()
+            salary = DecimalField()
+
+        class SocialTest(Document):
+            group = StringField()
+            person = CachedReferenceField(PersonAuto, fields=("salary",))
+
+        PersonAuto.drop_collection()
+        SocialTest.drop_collection()
+
+        p = PersonAuto(name="Alberto", salary=Decimal("7000.00"))
+        p.save()
+
+        s = SocialTest(group="dev", person=p)
+        s.save()
+
+        assert SocialTest.objects._collection.find_one({"person.salary": 7000.00}) == {
+            "_id": s.pk,
+            "group": s.group,
+            "person": {"_id": p.pk, "salary": 7000.00},
+        }
+
+    def test_cached_reference_field_reference(self):
+        class Group(Document):
+            name = StringField()
+
+        class Person(Document):
+            name = StringField()
+            group = ReferenceField(Group)
+
+        class SocialData(Document):
+            obs = StringField()
+            tags = ListField(StringField())
+            person = CachedReferenceField(Person, fields=("group",))
+
+        Group.drop_collection()
+        Person.drop_collection()
+        SocialData.drop_collection()
+
+        g1 = Group(name="dev")
+        g1.save()
+
+        g2 = Group(name="designers")
+        g2.save()
+
+        p1 = Person(name="Alberto", group=g1)
+        p1.save()
+
+        p2 = Person(name="Andre", group=g1)
+        p2.save()
+
+        p3 = Person(name="Afro design", group=g2)
+        p3.save()
+
+        s1 = SocialData(obs="testing 123", person=p1, tags=["tag1", "tag2"])
+        s1.save()
+
+        s2 = SocialData(obs="testing 321", person=p3, tags=["tag3", "tag4"])
+        s2.save()
+
+        assert SocialData.objects._collection.find_one({"tags": "tag2"}) == {
+            "_id": s1.pk,
+            "obs": "testing 123",
+            "tags": ["tag1", "tag2"],
+            "person": {"_id": p1.pk, "group": g1.pk},
+        }
+
+        assert SocialData.objects(person__group=g2).count() == 1
+        assert SocialData.objects(person__group=g2).first() == s2
+
+    def test_cached_reference_field_push_with_fields(self):
+        class Product(Document):
+            name = StringField()
+
+        Product.drop_collection()
+
+        class Basket(Document):
+            products = ListField(CachedReferenceField(Product, fields=["name"]))
+
+        Basket.drop_collection()
+        product1 = Product(name="abc").save()
+        product2 = Product(name="def").save()
+        basket = Basket(products=[product1]).save()
+        assert Basket.objects._collection.find_one() == {
+            "_id": basket.pk,
+            "products": [{"_id": product1.pk, "name": product1.name}],
+        }
+        # push to list
+        basket.update(push__products=product2)
+        basket.reload()
+        assert Basket.objects._collection.find_one() == {
+            "_id": basket.pk,
+            "products": [
+                {"_id": product1.pk, "name": product1.name},
+                {"_id": product2.pk, "name": product2.name},
+            ],
+        }
+
+    def test_cached_reference_field_update_all(self):
+        class Person(Document):
+            TYPES = (("pf", "PF"), ("pj", "PJ"))
+            name = StringField()
+            tp = StringField(choices=TYPES)
+            father = CachedReferenceField("self", fields=("tp",))
+
+        Person.drop_collection()
+
+        a1 = Person(name="Wilson Father", tp="pj")
+        a1.save()
+
+        a2 = Person(name="Wilson Junior", tp="pf", father=a1)
+        a2.save()
+
+        a2 = Person.objects.with_id(a2.id)
+        assert a2.father.tp == a1.tp
+
+        assert dict(a2.to_mongo()) == {
+            "_id": a2.pk,
+            "name": u"Wilson Junior",
+            "tp": u"pf",
+            "father": {"_id": a1.pk, "tp": u"pj"},
+        }
+
+        assert Person.objects(father=a1)._query == {"father._id": a1.pk}
+        assert Person.objects(father=a1).count() == 1
+
+        Person.objects.update(set__tp="pf")
+        Person.father.sync_all()
+
+        a2.reload()
+        assert dict(a2.to_mongo()) == {
+            "_id": a2.pk,
+            "name": u"Wilson Junior",
+            "tp": u"pf",
+            "father": {"_id": a1.pk, "tp": u"pf"},
+        }
+
+    def test_cached_reference_fields_on_embedded_documents(self):
+        with pytest.raises(InvalidDocumentError):
+
+            class Test(Document):
+                name = StringField()
+
+            type(
+                "WrongEmbeddedDocument",
+                (EmbeddedDocument,),
+                {"test": CachedReferenceField(Test)},
+            )
+
+    def test_cached_reference_auto_sync(self):
+        class Person(Document):
+            TYPES = (("pf", "PF"), ("pj", "PJ"))
+            name = StringField()
+            tp = StringField(choices=TYPES)
+
+            father = CachedReferenceField("self", fields=("tp",))
+
+        Person.drop_collection()
+
+        a1 = Person(name="Wilson Father", tp="pj")
+        a1.save()
+
+        a2 = Person(name="Wilson Junior", tp="pf", father=a1)
+        a2.save()
+
+        a1.tp = "pf"
+        a1.save()
+
+        a2.reload()
+        assert dict(a2.to_mongo()) == {
+            "_id": a2.pk,
+            "name": "Wilson Junior",
+            "tp": "pf",
+            "father": {"_id": a1.pk, "tp": "pf"},
+        }
+
+    def test_cached_reference_auto_sync_disabled(self):
+        class Persone(Document):
+            TYPES = (("pf", "PF"), ("pj", "PJ"))
+            name = StringField()
+            tp = StringField(choices=TYPES)
+
+            father = CachedReferenceField("self", fields=("tp",), auto_sync=False)
+
+        Persone.drop_collection()
+
+        a1 = Persone(name="Wilson Father", tp="pj")
+        a1.save()
+
+        a2 = Persone(name="Wilson Junior", tp="pf", father=a1)
+        a2.save()
+
+        a1.tp = "pf"
+        a1.save()
+
+        assert Persone.objects._collection.find_one({"_id": a2.pk}) == {
+            "_id": a2.pk,
+            "name": "Wilson Junior",
+            "tp": "pf",
+            "father": {"_id": a1.pk, "tp": "pj"},
+        }
+
+    def test_cached_reference_embedded_fields(self):
+        class Owner(EmbeddedDocument):
+            TPS = (("n", "Normal"), ("u", "Urgent"))
+            name = StringField()
+            tp = StringField(verbose_name="Type", db_field="t", choices=TPS)
+
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+            owner = EmbeddedDocumentField(Owner)
+
+        class Ocorrence(Document):
+            person = StringField()
+            animal = CachedReferenceField(Animal, fields=["tag", "owner.tp"])
+
+        Animal.drop_collection()
+        Ocorrence.drop_collection()
+
+        a = Animal(
+            name="Leopard", tag="heavy", owner=Owner(tp="u", name="Wilson Júnior")
+        )
+        a.save()
+
+        o = Ocorrence(person="teste", animal=a)
+        o.save()
+        assert dict(a.to_mongo(fields=["tag", "owner.tp"])) == {
+            "_id": a.pk,
+            "tag": "heavy",
+            "owner": {"t": "u"},
+        }
+        assert o.to_mongo()["animal"]["tag"] == "heavy"
+        assert o.to_mongo()["animal"]["owner"]["t"] == "u"
+
+        # Check to_mongo with fields
+        assert "animal" not in o.to_mongo(fields=["person"])
+
+        # counts
+        Ocorrence(person="teste 2").save()
+        Ocorrence(person="teste 3").save()
+
+        count = Ocorrence.objects(animal__tag="heavy", animal__owner__tp="u").count()
+        assert count == 1
+
+        ocorrence = Ocorrence.objects(
+            animal__tag="heavy", animal__owner__tp="u"
+        ).first()
+        assert ocorrence.person == "teste"
+        assert isinstance(ocorrence.animal, Animal)
+
+    def test_cached_reference_embedded_list_fields(self):
+        class Owner(EmbeddedDocument):
+            name = StringField()
+            tags = ListField(StringField())
+
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+            owner = EmbeddedDocumentField(Owner)
+
+        class Ocorrence(Document):
+            person = StringField()
+            animal = CachedReferenceField(Animal, fields=["tag", "owner.tags"])
+
+        Animal.drop_collection()
+        Ocorrence.drop_collection()
+
+        a = Animal(
+            name="Leopard",
+            tag="heavy",
+            owner=Owner(tags=["cool", "funny"], name="Wilson Júnior"),
+        )
+        a.save()
+
+        o = Ocorrence(person="teste 2", animal=a)
+        o.save()
+        assert dict(a.to_mongo(fields=["tag", "owner.tags"])) == {
+            "_id": a.pk,
+            "tag": "heavy",
+            "owner": {"tags": ["cool", "funny"]},
+        }
+
+        assert o.to_mongo()["animal"]["tag"] == "heavy"
+        assert o.to_mongo()["animal"]["owner"]["tags"] == ["cool", "funny"]
+
+        # counts
+        Ocorrence(person="teste 2").save()
+        Ocorrence(person="teste 3").save()
+
+        query = Ocorrence.objects(
+            animal__tag="heavy", animal__owner__tags="cool"
+        )._query
+        assert query == {"animal.owner.tags": "cool", "animal.tag": "heavy"}
+
+        ocorrence = Ocorrence.objects(
+            animal__tag="heavy", animal__owner__tags="cool"
+        ).first()
+        assert ocorrence.person == "teste 2"
+        assert isinstance(ocorrence.animal, Animal)

tests/fields/test_complex_datetime_field.py

@@ -0,0 +1,210 @@
+# -*- coding: utf-8 -*-
+import datetime
+import itertools
+import math
+import re
+
+import pytest
+
+from mongoengine import *
+
+from tests.utils import MongoDBTestCase
+
+
+class ComplexDateTimeFieldTest(MongoDBTestCase):
+    def test_complexdatetime_storage(self):
+        """Tests for complex datetime fields - which can handle
+        microseconds without rounding.
+        """
+
+        class LogEntry(Document):
+            date = ComplexDateTimeField()
+            date_with_dots = ComplexDateTimeField(separator=".")
+
+        LogEntry.drop_collection()
+
+        # Post UTC - microseconds are rounded (down) nearest millisecond and
+        # dropped - with default datetimefields
+        d1 = datetime.datetime(1970, 1, 1, 0, 0, 1, 999)
+        log = LogEntry()
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date == d1
+
+        # Post UTC - microseconds are rounded (down) nearest millisecond - with
+        # default datetimefields
+        d1 = datetime.datetime(1970, 1, 1, 0, 0, 1, 9999)
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date == d1
+
+        # Pre UTC dates microseconds below 1000 are dropped - with default
+        # datetimefields
+        d1 = datetime.datetime(1969, 12, 31, 23, 59, 59, 999)
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date == d1
+
+        # Pre UTC microseconds above 1000 is wonky - with default datetimefields
+        # log.date has an invalid microsecond value so I can't construct
+        # a date to compare.
+        for i in range(1001, 3113, 33):
+            d1 = datetime.datetime(1969, 12, 31, 23, 59, 59, i)
+            log.date = d1
+            log.save()
+            log.reload()
+            assert log.date == d1
+            log1 = LogEntry.objects.get(date=d1)
+            assert log == log1
+
+        # Test string padding
+        microsecond = map(int, [math.pow(10, x) for x in range(6)])
+        mm = dd = hh = ii = ss = [1, 10]
+
+        for values in itertools.product([2014], mm, dd, hh, ii, ss, microsecond):
+            stored = LogEntry(date=datetime.datetime(*values)).to_mongo()["date"]
+            assert (
+                re.match(r"^\d{4},\d{2},\d{2},\d{2},\d{2},\d{2},\d{6}$", stored)
+                is not None
+            )
+
+        # Test separator
+        stored = LogEntry(date_with_dots=datetime.datetime(2014, 1, 1)).to_mongo()[
+            "date_with_dots"
+        ]
+        assert (
+            re.match(r"^\d{4}.\d{2}.\d{2}.\d{2}.\d{2}.\d{2}.\d{6}$", stored) is not None
+        )
+
+    def test_complexdatetime_usage(self):
+        """Tests for complex datetime fields - which can handle
+        microseconds without rounding.
+        """
+
+        class LogEntry(Document):
+            date = ComplexDateTimeField()
+
+        LogEntry.drop_collection()
+
+        d1 = datetime.datetime(1950, 1, 1, 0, 0, 1, 999)
+        log = LogEntry()
+        log.date = d1
+        log.save()
+
+        log1 = LogEntry.objects.get(date=d1)
+        assert log == log1
+
+        # create extra 59 log entries for a total of 60
+        for i in range(1951, 2010):
+            d = datetime.datetime(i, 1, 1, 0, 0, 1, 999)
+            LogEntry(date=d).save()
+
+        assert LogEntry.objects.count() == 60
+
+        # Test ordering
+        logs = LogEntry.objects.order_by("date")
+        i = 0
+        while i < 59:
+            assert logs[i].date <= logs[i + 1].date
+            i += 1
+
+        logs = LogEntry.objects.order_by("-date")
+        i = 0
+        while i < 59:
+            assert logs[i].date >= logs[i + 1].date
+            i += 1
+
+        # Test searching
+        logs = LogEntry.objects.filter(date__gte=datetime.datetime(1980, 1, 1))
+        assert logs.count() == 30
+
+        logs = LogEntry.objects.filter(date__lte=datetime.datetime(1980, 1, 1))
+        assert logs.count() == 30
+
+        logs = LogEntry.objects.filter(
+            date__lte=datetime.datetime(2011, 1, 1),
+            date__gte=datetime.datetime(2000, 1, 1),
+        )
+        assert logs.count() == 10
+
+        LogEntry.drop_collection()
+
+        # Test microsecond-level ordering/filtering
+        for microsecond in (99, 999, 9999, 10000):
+            LogEntry(date=datetime.datetime(2015, 1, 1, 0, 0, 0, microsecond)).save()
+
+        logs = list(LogEntry.objects.order_by("date"))
+        for next_idx, log in enumerate(logs[:-1], start=1):
+            next_log = logs[next_idx]
+            assert log.date < next_log.date
+
+        logs = list(LogEntry.objects.order_by("-date"))
+        for next_idx, log in enumerate(logs[:-1], start=1):
+            next_log = logs[next_idx]
+            assert log.date > next_log.date
+
+        logs = LogEntry.objects.filter(
+            date__lte=datetime.datetime(2015, 1, 1, 0, 0, 0, 10000)
+        )
+        assert logs.count() == 4
+
+    def test_no_default_value(self):
+        class Log(Document):
+            timestamp = ComplexDateTimeField()
+
+        Log.drop_collection()
+
+        log = Log()
+        assert log.timestamp is None
+        log.save()
+
+        fetched_log = Log.objects.with_id(log.id)
+        assert fetched_log.timestamp is None
+
+    def test_default_static_value(self):
+        NOW = datetime.datetime.utcnow()
+
+        class Log(Document):
+            timestamp = ComplexDateTimeField(default=NOW)
+
+        Log.drop_collection()
+
+        log = Log()
+        assert log.timestamp == NOW
+        log.save()
+
+        fetched_log = Log.objects.with_id(log.id)
+        assert fetched_log.timestamp == NOW
+
+    def test_default_callable(self):
+        NOW = datetime.datetime.utcnow()
+
+        class Log(Document):
+            timestamp = ComplexDateTimeField(default=datetime.datetime.utcnow)
+
+        Log.drop_collection()
+
+        log = Log()
+        assert log.timestamp >= NOW
+        log.save()
+
+        fetched_log = Log.objects.with_id(log.id)
+        assert fetched_log.timestamp >= NOW
+
+    def test_setting_bad_value_does_not_raise_unless_validate_is_called(self):
+        # test regression of #2253
+
+        class Log(Document):
+            timestamp = ComplexDateTimeField()
+
+        Log.drop_collection()
+
+        log = Log(timestamp="garbage")
+        with pytest.raises(ValidationError):
+            log.validate()
+
+        with pytest.raises(ValidationError):
+            log.save()

tests/fields/test_date_field.py

@@ -0,0 +1,164 @@
+# -*- coding: utf-8 -*-
+import datetime
+
+import pytest
+
+try:
+    import dateutil
+except ImportError:
+    dateutil = None
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestDateField(MongoDBTestCase):
+    def test_date_from_empty_string(self):
+        """
+        Ensure an exception is raised when trying to
+        cast an empty string to datetime.
+        """
+
+        class MyDoc(Document):
+            dt = DateField()
+
+        md = MyDoc(dt="")
+        with pytest.raises(ValidationError):
+            md.save()
+
+    def test_date_from_whitespace_string(self):
+        """
+        Ensure an exception is raised when trying to
+        cast a whitespace-only string to datetime.
+        """
+
+        class MyDoc(Document):
+            dt = DateField()
+
+        md = MyDoc(dt="   ")
+        with pytest.raises(ValidationError):
+            md.save()
+
+    def test_default_values_today(self):
+        """Ensure that default field values are used when creating
+        a document.
+        """
+
+        class Person(Document):
+            day = DateField(default=datetime.date.today)
+
+        person = Person()
+        person.validate()
+        assert person.day == person.day
+        assert person.day == datetime.date.today()
+        assert person._data["day"] == person.day
+
+    def test_date(self):
+        """Tests showing pymongo date fields
+
+        See: http://api.mongodb.org/python/current/api/bson/son.html#dt
+        """
+
+        class LogEntry(Document):
+            date = DateField()
+
+        LogEntry.drop_collection()
+
+        # Test can save dates
+        log = LogEntry()
+        log.date = datetime.date.today()
+        log.save()
+        log.reload()
+        assert log.date == datetime.date.today()
+
+        d1 = datetime.datetime(1970, 1, 1, 0, 0, 1, 999)
+        d2 = datetime.datetime(1970, 1, 1, 0, 0, 1)
+        log = LogEntry()
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date == d1.date()
+        assert log.date == d2.date()
+
+        d1 = datetime.datetime(1970, 1, 1, 0, 0, 1, 9999)
+        d2 = datetime.datetime(1970, 1, 1, 0, 0, 1, 9000)
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date == d1.date()
+        assert log.date == d2.date()
+
+    def test_regular_usage(self):
+        """Tests for regular datetime fields"""
+
+        class LogEntry(Document):
+            date = DateField()
+
+        LogEntry.drop_collection()
+
+        d1 = datetime.datetime(1970, 1, 1, 0, 0, 1)
+        log = LogEntry()
+        log.date = d1
+        log.validate()
+        log.save()
+
+        for query in (d1, d1.isoformat(" ")):
+            log1 = LogEntry.objects.get(date=query)
+            assert log == log1
+
+        if dateutil:
+            log1 = LogEntry.objects.get(date=d1.isoformat("T"))
+            assert log == log1
+
+        # create additional 19 log entries for a total of 20
+        for i in range(1971, 1990):
+            d = datetime.datetime(i, 1, 1, 0, 0, 1)
+            LogEntry(date=d).save()
+
+        assert LogEntry.objects.count() == 20
+
+        # Test ordering
+        logs = LogEntry.objects.order_by("date")
+        i = 0
+        while i < 19:
+            assert logs[i].date <= logs[i + 1].date
+            i += 1
+
+        logs = LogEntry.objects.order_by("-date")
+        i = 0
+        while i < 19:
+            assert logs[i].date >= logs[i + 1].date
+            i += 1
+
+        # Test searching
+        logs = LogEntry.objects.filter(date__gte=datetime.datetime(1980, 1, 1))
+        assert logs.count() == 10
+
+    def test_validation(self):
+        """Ensure that invalid values cannot be assigned to datetime
+        fields.
+        """
+
+        class LogEntry(Document):
+            time = DateField()
+
+        log = LogEntry()
+        log.time = datetime.datetime.now()
+        log.validate()
+
+        log.time = datetime.date.today()
+        log.validate()
+
+        log.time = datetime.datetime.now().isoformat(" ")
+        log.validate()
+
+        if dateutil:
+            log.time = datetime.datetime.now().isoformat("T")
+            log.validate()
+
+        log.time = -1
+        with pytest.raises(ValidationError):
+            log.validate()
+        log.time = "ABC"
+        with pytest.raises(ValidationError):
+            log.validate()

tests/fields/test_datetime_field.py

@@ -0,0 +1,234 @@
+# -*- coding: utf-8 -*-
+import datetime as dt
+
+import pytest
+
+try:
+    import dateutil
+except ImportError:
+    dateutil = None
+
+from mongoengine import *
+from mongoengine import connection
+
+from tests.utils import MongoDBTestCase
+
+
+class TestDateTimeField(MongoDBTestCase):
+    def test_datetime_from_empty_string(self):
+        """
+        Ensure an exception is raised when trying to
+        cast an empty string to datetime.
+        """
+
+        class MyDoc(Document):
+            dt = DateTimeField()
+
+        md = MyDoc(dt="")
+        with pytest.raises(ValidationError):
+            md.save()
+
+    def test_datetime_from_whitespace_string(self):
+        """
+        Ensure an exception is raised when trying to
+        cast a whitespace-only string to datetime.
+        """
+
+        class MyDoc(Document):
+            dt = DateTimeField()
+
+        md = MyDoc(dt="   ")
+        with pytest.raises(ValidationError):
+            md.save()
+
+    def test_default_value_utcnow(self):
+        """Ensure that default field values are used when creating
+        a document.
+        """
+
+        class Person(Document):
+            created = DateTimeField(default=dt.datetime.utcnow)
+
+        utcnow = dt.datetime.utcnow()
+        person = Person()
+        person.validate()
+        person_created_t0 = person.created
+        assert person.created - utcnow < dt.timedelta(seconds=1)
+        assert person_created_t0 == person.created  # make sure it does not change
+        assert person._data["created"] == person.created
+
+    def test_handling_microseconds(self):
+        """Tests showing pymongo datetime fields handling of microseconds.
+        Microseconds are rounded to the nearest millisecond and pre UTC
+        handling is wonky.
+
+        See: http://api.mongodb.org/python/current/api/bson/son.html#dt
+        """
+
+        class LogEntry(Document):
+            date = DateTimeField()
+
+        LogEntry.drop_collection()
+
+        # Test can save dates
+        log = LogEntry()
+        log.date = dt.date.today()
+        log.save()
+        log.reload()
+        assert log.date.date() == dt.date.today()
+
+        # Post UTC - microseconds are rounded (down) nearest millisecond and
+        # dropped
+        d1 = dt.datetime(1970, 1, 1, 0, 0, 1, 999)
+        d2 = dt.datetime(1970, 1, 1, 0, 0, 1)
+        log = LogEntry()
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date != d1
+        assert log.date == d2
+
+        # Post UTC - microseconds are rounded (down) nearest millisecond
+        d1 = dt.datetime(1970, 1, 1, 0, 0, 1, 9999)
+        d2 = dt.datetime(1970, 1, 1, 0, 0, 1, 9000)
+        log.date = d1
+        log.save()
+        log.reload()
+        assert log.date != d1
+        assert log.date == d2
+
+    def test_regular_usage(self):
+        """Tests for regular datetime fields"""
+
+        class LogEntry(Document):
+            date = DateTimeField()
+
+        LogEntry.drop_collection()
+
+        d1 = dt.datetime(1970, 1, 1, 0, 0, 1)
+        log = LogEntry()
+        log.date = d1
+        log.validate()
+        log.save()
+
+        for query in (d1, d1.isoformat(" ")):
+            log1 = LogEntry.objects.get(date=query)
+            assert log == log1
+
+        if dateutil:
+            log1 = LogEntry.objects.get(date=d1.isoformat("T"))
+            assert log == log1
+
+        # create additional 19 log entries for a total of 20
+        for i in range(1971, 1990):
+            d = dt.datetime(i, 1, 1, 0, 0, 1)
+            LogEntry(date=d).save()
+
+        assert LogEntry.objects.count() == 20
+
+        # Test ordering
+        logs = LogEntry.objects.order_by("date")
+        i = 0
+        while i < 19:
+            assert logs[i].date <= logs[i + 1].date
+            i += 1
+
+        logs = LogEntry.objects.order_by("-date")
+        i = 0
+        while i < 19:
+            assert logs[i].date >= logs[i + 1].date
+            i += 1
+
+        # Test searching
+        logs = LogEntry.objects.filter(date__gte=dt.datetime(1980, 1, 1))
+        assert logs.count() == 10
+
+        logs = LogEntry.objects.filter(date__lte=dt.datetime(1980, 1, 1))
+        assert logs.count() == 10
+
+        logs = LogEntry.objects.filter(
+            date__lte=dt.datetime(1980, 1, 1), date__gte=dt.datetime(1975, 1, 1)
+        )
+        assert logs.count() == 5
+
+    def test_datetime_validation(self):
+        """Ensure that invalid values cannot be assigned to datetime
+        fields.
+        """
+
+        class LogEntry(Document):
+            time = DateTimeField()
+
+        log = LogEntry()
+        log.time = dt.datetime.now()
+        log.validate()
+
+        log.time = dt.date.today()
+        log.validate()
+
+        log.time = dt.datetime.now().isoformat(" ")
+        log.validate()
+
+        log.time = "2019-05-16 21:42:57.897847"
+        log.validate()
+
+        if dateutil:
+            log.time = dt.datetime.now().isoformat("T")
+            log.validate()
+
+        log.time = -1
+        with pytest.raises(ValidationError):
+            log.validate()
+        log.time = "ABC"
+        with pytest.raises(ValidationError):
+            log.validate()
+        log.time = "2019-05-16 21:GARBAGE:12"
+        with pytest.raises(ValidationError):
+            log.validate()
+        log.time = "2019-05-16 21:42:57.GARBAGE"
+        with pytest.raises(ValidationError):
+            log.validate()
+        log.time = "2019-05-16 21:42:57.123.456"
+        with pytest.raises(ValidationError):
+            log.validate()
+
+    def test_parse_datetime_as_str(self):
+        class DTDoc(Document):
+            date = DateTimeField()
+
+        date_str = "2019-03-02 22:26:01"
+
+        # make sure that passing a parsable datetime works
+        dtd = DTDoc()
+        dtd.date = date_str
+        assert isinstance(dtd.date, str)
+        dtd.save()
+        dtd.reload()
+
+        assert isinstance(dtd.date, dt.datetime)
+        assert str(dtd.date) == date_str
+
+        dtd.date = "January 1st, 9999999999"
+        with pytest.raises(ValidationError):
+            dtd.validate()
+
+
+class TestDateTimeTzAware(MongoDBTestCase):
+    def test_datetime_tz_aware_mark_as_changed(self):
+        # Reset the connections
+        connection._connection_settings = {}
+        connection._connections = {}
+        connection._dbs = {}
+
+        connect(db="mongoenginetest", tz_aware=True)
+
+        class LogEntry(Document):
+            time = DateTimeField()
+
+        LogEntry.drop_collection()
+
+        LogEntry(time=dt.datetime(2013, 1, 1, 0, 0, 0)).save()
+
+        log = LogEntry.objects.first()
+        log.time = dt.datetime(2013, 1, 1, 0, 0, 0)
+        assert ["time"] == log._changed_fields

tests/fields/test_decimal_field.py

@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+from decimal import Decimal
+
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestDecimalField(MongoDBTestCase):
+    def test_validation(self):
+        """Ensure that invalid values cannot be assigned to decimal fields.
+        """
+
+        class Person(Document):
+            height = DecimalField(min_value=Decimal("0.1"), max_value=Decimal("3.5"))
+
+        Person.drop_collection()
+
+        Person(height=Decimal("1.89")).save()
+        person = Person.objects.first()
+        assert person.height == Decimal("1.89")
+
+        person.height = "2.0"
+        person.save()
+        person.height = 0.01
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.height = Decimal("0.01")
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.height = Decimal("4.0")
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.height = "something invalid"
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person_2 = Person(height="something invalid")
+        with pytest.raises(ValidationError):
+            person_2.validate()
+
+    def test_comparison(self):
+        class Person(Document):
+            money = DecimalField()
+
+        Person.drop_collection()
+
+        Person(money=6).save()
+        Person(money=7).save()
+        Person(money=8).save()
+        Person(money=10).save()
+
+        assert 2 == Person.objects(money__gt=Decimal("7")).count()
+        assert 2 == Person.objects(money__gt=7).count()
+        assert 2 == Person.objects(money__gt="7").count()
+
+        assert 3 == Person.objects(money__gte="7").count()
+
+    def test_storage(self):
+        class Person(Document):
+            float_value = DecimalField(precision=4)
+            string_value = DecimalField(precision=4, force_string=True)
+
+        Person.drop_collection()
+        values_to_store = [
+            10,
+            10.1,
+            10.11,
+            "10.111",
+            Decimal("10.1111"),
+            Decimal("10.11111"),
+        ]
+        for store_at_creation in [True, False]:
+            for value in values_to_store:
+                # to_python is called explicitly if values were sent in the kwargs of __init__
+                if store_at_creation:
+                    Person(float_value=value, string_value=value).save()
+                else:
+                    person = Person.objects.create()
+                    person.float_value = value
+                    person.string_value = value
+                    person.save()
+
+        # How its stored
+        expected = [
+            {"float_value": 10.0, "string_value": "10.0000"},
+            {"float_value": 10.1, "string_value": "10.1000"},
+            {"float_value": 10.11, "string_value": "10.1100"},
+            {"float_value": 10.111, "string_value": "10.1110"},
+            {"float_value": 10.1111, "string_value": "10.1111"},
+            {"float_value": 10.1111, "string_value": "10.1111"},
+        ]
+        expected.extend(expected)
+        actual = list(Person.objects.exclude("id").as_pymongo())
+        assert expected == actual
+
+        # How it comes out locally
+        expected = [
+            Decimal("10.0000"),
+            Decimal("10.1000"),
+            Decimal("10.1100"),
+            Decimal("10.1110"),
+            Decimal("10.1111"),
+            Decimal("10.1111"),
+        ]
+        expected.extend(expected)
+        for field_name in ["float_value", "string_value"]:
+            actual = list(Person.objects().scalar(field_name))
+            assert expected == actual

tests/fields/test_dict_field.py

@@ -0,0 +1,368 @@
+# -*- coding: utf-8 -*-
+from bson import InvalidDocument
+import pytest
+
+from mongoengine import *
+from mongoengine.base import BaseDict
+from mongoengine.mongodb_support import MONGODB_36, get_mongodb_version
+
+from tests.utils import MongoDBTestCase, get_as_pymongo
+
+
+class TestDictField(MongoDBTestCase):
+    def test_storage(self):
+        class BlogPost(Document):
+            info = DictField()
+
+        BlogPost.drop_collection()
+
+        info = {"testkey": "testvalue"}
+        post = BlogPost(info=info).save()
+        assert get_as_pymongo(post) == {"_id": post.id, "info": info}
+
+    def test_validate_invalid_type(self):
+        class BlogPost(Document):
+            info = DictField()
+
+        BlogPost.drop_collection()
+
+        invalid_infos = ["my post", ["test", "test"], {1: "test"}]
+        for invalid_info in invalid_infos:
+            with pytest.raises(ValidationError):
+                BlogPost(info=invalid_info).validate()
+
+    def test_keys_with_dots_or_dollars(self):
+        class BlogPost(Document):
+            info = DictField()
+
+        BlogPost.drop_collection()
+
+        post = BlogPost()
+
+        post.info = {"$title": "test"}
+        with pytest.raises(ValidationError):
+            post.validate()
+
+        post.info = {"nested": {"$title": "test"}}
+        with pytest.raises(ValidationError):
+            post.validate()
+
+        post.info = {"$title.test": "test"}
+        with pytest.raises(ValidationError):
+            post.validate()
+
+        post.info = {"nested": {"the.title": "test"}}
+        if get_mongodb_version() < MONGODB_36:
+            # MongoDB < 3.6 rejects dots
+            # To avoid checking the mongodb version from the DictField class
+            # we rely on MongoDB to reject the data during the save
+            post.validate()
+            with pytest.raises(InvalidDocument):
+                post.save()
+        else:
+            post.validate()
+
+        post.info = {"dollar_and_dot": {"te$st.test": "test"}}
+        if get_mongodb_version() < MONGODB_36:
+            post.validate()
+            with pytest.raises(InvalidDocument):
+                post.save()
+        else:
+            post.validate()
+
+    def test_general_things(self):
+        """Ensure that dict types work as expected."""
+
+        class BlogPost(Document):
+            info = DictField()
+
+        BlogPost.drop_collection()
+
+        post = BlogPost(info={"title": "test"})
+        post.save()
+
+        post = BlogPost()
+        post.info = {"title": "dollar_sign", "details": {"te$t": "test"}}
+        post.save()
+
+        post = BlogPost()
+        post.info = {"details": {"test": "test"}}
+        post.save()
+
+        post = BlogPost()
+        post.info = {"details": {"test": 3}}
+        post.save()
+
+        assert BlogPost.objects.count() == 4
+        assert BlogPost.objects.filter(info__title__exact="test").count() == 1
+        assert BlogPost.objects.filter(info__details__test__exact="test").count() == 1
+
+        post = BlogPost.objects.filter(info__title__exact="dollar_sign").first()
+        assert "te$t" in post["info"]["details"]
+
+        # Confirm handles non strings or non existing keys
+        assert BlogPost.objects.filter(info__details__test__exact=5).count() == 0
+        assert BlogPost.objects.filter(info__made_up__test__exact="test").count() == 0
+
+        post = BlogPost.objects.create(info={"title": "original"})
+        post.info.update({"title": "updated"})
+        post.save()
+        post.reload()
+        assert "updated" == post.info["title"]
+
+        post.info.setdefault("authors", [])
+        post.save()
+        post.reload()
+        assert [] == post.info["authors"]
+
+    def test_dictfield_dump_document(self):
+        """Ensure a DictField can handle another document's dump."""
+
+        class Doc(Document):
+            field = DictField()
+
+        class ToEmbed(Document):
+            id = IntField(primary_key=True, default=1)
+            recursive = DictField()
+
+        class ToEmbedParent(Document):
+            id = IntField(primary_key=True, default=1)
+            recursive = DictField()
+
+            meta = {"allow_inheritance": True}
+
+        class ToEmbedChild(ToEmbedParent):
+            pass
+
+        to_embed_recursive = ToEmbed(id=1).save()
+        to_embed = ToEmbed(
+            id=2, recursive=to_embed_recursive.to_mongo().to_dict()
+        ).save()
+        doc = Doc(field=to_embed.to_mongo().to_dict())
+        doc.save()
+        assert isinstance(doc.field, dict)
+        assert doc.field == {"_id": 2, "recursive": {"_id": 1, "recursive": {}}}
+        # Same thing with a Document with a _cls field
+        to_embed_recursive = ToEmbedChild(id=1).save()
+        to_embed_child = ToEmbedChild(
+            id=2, recursive=to_embed_recursive.to_mongo().to_dict()
+        ).save()
+        doc = Doc(field=to_embed_child.to_mongo().to_dict())
+        doc.save()
+        assert isinstance(doc.field, dict)
+        expected = {
+            "_id": 2,
+            "_cls": "ToEmbedParent.ToEmbedChild",
+            "recursive": {
+                "_id": 1,
+                "_cls": "ToEmbedParent.ToEmbedChild",
+                "recursive": {},
+            },
+        }
+        assert doc.field == expected
+
+    def test_dictfield_strict(self):
+        """Ensure that dict field handles validation if provided a strict field type."""
+
+        class Simple(Document):
+            mapping = DictField(field=IntField())
+
+        Simple.drop_collection()
+
+        e = Simple()
+        e.mapping["someint"] = 1
+        e.save()
+
+        # try creating an invalid mapping
+        with pytest.raises(ValidationError):
+            e.mapping["somestring"] = "abc"
+            e.save()
+
+    def test_dictfield_complex(self):
+        """Ensure that the dict field can handle the complex types."""
+
+        class SettingBase(EmbeddedDocument):
+            meta = {"allow_inheritance": True}
+
+        class StringSetting(SettingBase):
+            value = StringField()
+
+        class IntegerSetting(SettingBase):
+            value = IntField()
+
+        class Simple(Document):
+            mapping = DictField()
+
+        Simple.drop_collection()
+
+        e = Simple()
+        e.mapping["somestring"] = StringSetting(value="foo")
+        e.mapping["someint"] = IntegerSetting(value=42)
+        e.mapping["nested_dict"] = {
+            "number": 1,
+            "string": "Hi!",
+            "float": 1.001,
+            "complex": IntegerSetting(value=42),
+            "list": [IntegerSetting(value=42), StringSetting(value="foo")],
+        }
+        e.save()
+
+        e2 = Simple.objects.get(id=e.id)
+        assert isinstance(e2.mapping["somestring"], StringSetting)
+        assert isinstance(e2.mapping["someint"], IntegerSetting)
+
+        # Test querying
+        assert Simple.objects.filter(mapping__someint__value=42).count() == 1
+        assert Simple.objects.filter(mapping__nested_dict__number=1).count() == 1
+        assert (
+            Simple.objects.filter(mapping__nested_dict__complex__value=42).count() == 1
+        )
+        assert (
+            Simple.objects.filter(mapping__nested_dict__list__0__value=42).count() == 1
+        )
+        assert (
+            Simple.objects.filter(mapping__nested_dict__list__1__value="foo").count()
+            == 1
+        )
+
+        # Confirm can update
+        Simple.objects().update(set__mapping={"someint": IntegerSetting(value=10)})
+        Simple.objects().update(
+            set__mapping__nested_dict__list__1=StringSetting(value="Boo")
+        )
+        assert (
+            Simple.objects.filter(mapping__nested_dict__list__1__value="foo").count()
+            == 0
+        )
+        assert (
+            Simple.objects.filter(mapping__nested_dict__list__1__value="Boo").count()
+            == 1
+        )
+
+    def test_push_dict(self):
+        class MyModel(Document):
+            events = ListField(DictField())
+
+        doc = MyModel(events=[{"a": 1}]).save()
+        raw_doc = get_as_pymongo(doc)
+        expected_raw_doc = {"_id": doc.id, "events": [{"a": 1}]}
+        assert raw_doc == expected_raw_doc
+
+        MyModel.objects(id=doc.id).update(push__events={})
+        raw_doc = get_as_pymongo(doc)
+        expected_raw_doc = {"_id": doc.id, "events": [{"a": 1}, {}]}
+        assert raw_doc == expected_raw_doc
+
+    def test_ensure_unique_default_instances(self):
+        """Ensure that every field has it's own unique default instance."""
+
+        class D(Document):
+            data = DictField()
+            data2 = DictField(default=lambda: {})
+
+        d1 = D()
+        d1.data["foo"] = "bar"
+        d1.data2["foo"] = "bar"
+        d2 = D()
+        assert d2.data == {}
+        assert d2.data2 == {}
+
+    def test_dict_field_invalid_dict_value(self):
+        class DictFieldTest(Document):
+            dictionary = DictField(required=True)
+
+        DictFieldTest.drop_collection()
+
+        test = DictFieldTest(dictionary=None)
+        test.dictionary  # Just access to test getter
+        with pytest.raises(ValidationError):
+            test.validate()
+
+        test = DictFieldTest(dictionary=False)
+        test.dictionary  # Just access to test getter
+        with pytest.raises(ValidationError):
+            test.validate()
+
+    def test_dict_field_raises_validation_error_if_wrongly_assign_embedded_doc(self):
+        class DictFieldTest(Document):
+            dictionary = DictField(required=True)
+
+        DictFieldTest.drop_collection()
+
+        class Embedded(EmbeddedDocument):
+            name = StringField()
+
+        embed = Embedded(name="garbage")
+        doc = DictFieldTest(dictionary=embed)
+        with pytest.raises(ValidationError) as exc_info:
+            doc.validate()
+
+        error_msg = str(exc_info.value)
+        assert "'dictionary'" in error_msg
+        assert "Only dictionaries may be used in a DictField" in error_msg
+
+    def test_atomic_update_dict_field(self):
+        """Ensure that the entire DictField can be atomically updated."""
+
+        class Simple(Document):
+            mapping = DictField(field=ListField(IntField(required=True)))
+
+        Simple.drop_collection()
+
+        e = Simple()
+        e.mapping["someints"] = [1, 2]
+        e.save()
+        e.update(set__mapping={"ints": [3, 4]})
+        e.reload()
+        assert isinstance(e.mapping, BaseDict)
+        assert {"ints": [3, 4]} == e.mapping
+
+        # try creating an invalid mapping
+        with pytest.raises(ValueError):
+            e.update(set__mapping={"somestrings": ["foo", "bar"]})
+
+    def test_dictfield_with_referencefield_complex_nesting_cases(self):
+        """Ensure complex nesting inside DictField handles dereferencing of ReferenceField(dbref=True | False)"""
+        # Relates to Issue #1453
+        class Doc(Document):
+            s = StringField()
+
+        class Simple(Document):
+            mapping0 = DictField(ReferenceField(Doc, dbref=True))
+            mapping1 = DictField(ReferenceField(Doc, dbref=False))
+            mapping2 = DictField(ListField(ReferenceField(Doc, dbref=True)))
+            mapping3 = DictField(ListField(ReferenceField(Doc, dbref=False)))
+            mapping4 = DictField(DictField(field=ReferenceField(Doc, dbref=True)))
+            mapping5 = DictField(DictField(field=ReferenceField(Doc, dbref=False)))
+            mapping6 = DictField(ListField(DictField(ReferenceField(Doc, dbref=True))))
+            mapping7 = DictField(ListField(DictField(ReferenceField(Doc, dbref=False))))
+            mapping8 = DictField(
+                ListField(DictField(ListField(ReferenceField(Doc, dbref=True))))
+            )
+            mapping9 = DictField(
+                ListField(DictField(ListField(ReferenceField(Doc, dbref=False))))
+            )
+
+        Doc.drop_collection()
+        Simple.drop_collection()
+
+        d = Doc(s="aa").save()
+        e = Simple()
+        e.mapping0["someint"] = e.mapping1["someint"] = d
+        e.mapping2["someint"] = e.mapping3["someint"] = [d]
+        e.mapping4["someint"] = e.mapping5["someint"] = {"d": d}
+        e.mapping6["someint"] = e.mapping7["someint"] = [{"d": d}]
+        e.mapping8["someint"] = e.mapping9["someint"] = [{"d": [d]}]
+        e.save()
+
+        s = Simple.objects.first()
+        assert isinstance(s.mapping0["someint"], Doc)
+        assert isinstance(s.mapping1["someint"], Doc)
+        assert isinstance(s.mapping2["someint"][0], Doc)
+        assert isinstance(s.mapping3["someint"][0], Doc)
+        assert isinstance(s.mapping4["someint"]["d"], Doc)
+        assert isinstance(s.mapping5["someint"]["d"], Doc)
+        assert isinstance(s.mapping6["someint"][0]["d"], Doc)
+        assert isinstance(s.mapping7["someint"][0]["d"], Doc)
+        assert isinstance(s.mapping8["someint"][0]["d"][0], Doc)
+        assert isinstance(s.mapping9["someint"][0]["d"][0], Doc)

tests/fields/test_email_field.py

@@ -0,0 +1,137 @@
+# -*- coding: utf-8 -*-
+import sys
+
+import pytest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestEmailField(MongoDBTestCase):
+    def test_generic_behavior(self):
+        class User(Document):
+            email = EmailField()
+
+        user = User(email="ross@example.com")
+        user.validate()
+
+        user = User(email="ross@example.co.uk")
+        user.validate()
+
+        user = User(
+            email=("Kofq@rhom0e4klgauOhpbpNdogawnyIKvQS0wk2mjqrgGQ5SaJIazqqWkm7.net")
+        )
+        user.validate()
+
+        user = User(email="new-tld@example.technology")
+        user.validate()
+
+        user = User(email="ross@example.com.")
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # unicode domain
+        user = User(email=u"user@пример.рф")
+        user.validate()
+
+        # invalid unicode domain
+        user = User(email=u"user@пример")
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # invalid data type
+        user = User(email=123)
+        with pytest.raises(ValidationError):
+            user.validate()
+
+    def test_email_field_unicode_user(self):
+        class User(Document):
+            email = EmailField()
+
+        # unicode user shouldn't validate by default...
+        user = User(email=u"Dörte@Sörensen.example.com")
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # ...but it should be fine with allow_utf8_user set to True
+        class User(Document):
+            email = EmailField(allow_utf8_user=True)
+
+        user = User(email=u"Dörte@Sörensen.example.com")
+        user.validate()
+
+    def test_email_field_domain_whitelist(self):
+        class User(Document):
+            email = EmailField()
+
+        # localhost domain shouldn't validate by default...
+        user = User(email="me@localhost")
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # ...but it should be fine if it's whitelisted
+        class User(Document):
+            email = EmailField(domain_whitelist=["localhost"])
+
+        user = User(email="me@localhost")
+        user.validate()
+
+    def test_email_domain_validation_fails_if_invalid_idn(self):
+        class User(Document):
+            email = EmailField()
+
+        invalid_idn = ".google.com"
+        user = User(email="me@%s" % invalid_idn)
+
+        with pytest.raises(ValidationError) as exc_info:
+            user.validate()
+        assert "domain failed IDN encoding" in str(exc_info.value)
+
+    def test_email_field_ip_domain(self):
+        class User(Document):
+            email = EmailField()
+
+        valid_ipv4 = "email@[127.0.0.1]"
+        valid_ipv6 = "email@[2001:dB8::1]"
+        invalid_ip = "email@[324.0.0.1]"
+
+        # IP address as a domain shouldn't validate by default...
+        user = User(email=valid_ipv4)
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        user = User(email=valid_ipv6)
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        user = User(email=invalid_ip)
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # ...but it should be fine with allow_ip_domain set to True
+        class User(Document):
+            email = EmailField(allow_ip_domain=True)
+
+        user = User(email=valid_ipv4)
+        user.validate()
+
+        user = User(email=valid_ipv6)
+        user.validate()
+
+        # invalid IP should still fail validation
+        user = User(email=invalid_ip)
+        with pytest.raises(ValidationError):
+            user.validate()
+
+    def test_email_field_honors_regex(self):
+        class User(Document):
+            email = EmailField(regex=r"\w+@example.com")
+
+        # Fails regex validation
+        user = User(email="me@foo.com")
+        with pytest.raises(ValidationError):
+            user.validate()
+
+        # Passes regex validation
+        user = User(email="me@example.com")
+        assert user.validate() is None

tests/fields/test_embedded_document_field.py

@@ -0,0 +1,354 @@
+# -*- coding: utf-8 -*-
+import pytest
+
+from mongoengine import (
+    Document,
+    EmbeddedDocument,
+    EmbeddedDocumentField,
+    GenericEmbeddedDocumentField,
+    IntField,
+    InvalidQueryError,
+    ListField,
+    LookUpError,
+    StringField,
+    ValidationError,
+)
+
+from tests.utils import MongoDBTestCase
+
+
+class TestEmbeddedDocumentField(MongoDBTestCase):
+    def test___init___(self):
+        class MyDoc(EmbeddedDocument):
+            name = StringField()
+
+        field = EmbeddedDocumentField(MyDoc)
+        assert field.document_type_obj == MyDoc
+
+        field2 = EmbeddedDocumentField("MyDoc")
+        assert field2.document_type_obj == "MyDoc"
+
+    def test___init___throw_error_if_document_type_is_not_EmbeddedDocument(self):
+        with pytest.raises(ValidationError):
+            EmbeddedDocumentField(dict)
+
+    def test_document_type_throw_error_if_not_EmbeddedDocument_subclass(self):
+        class MyDoc(Document):
+            name = StringField()
+
+        emb = EmbeddedDocumentField("MyDoc")
+        with pytest.raises(ValidationError) as exc_info:
+            emb.document_type
+        assert (
+            "Invalid embedded document class provided to an EmbeddedDocumentField"
+            in str(exc_info.value)
+        )
+
+    def test_embedded_document_field_only_allow_subclasses_of_embedded_document(self):
+        # Relates to #1661
+        class MyDoc(Document):
+            name = StringField()
+
+        with pytest.raises(ValidationError):
+
+            class MyFailingDoc(Document):
+                emb = EmbeddedDocumentField(MyDoc)
+
+        with pytest.raises(ValidationError):
+
+            class MyFailingdoc2(Document):
+                emb = EmbeddedDocumentField("MyDoc")
+
+    def test_query_embedded_document_attribute(self):
+        class AdminSettings(EmbeddedDocument):
+            foo1 = StringField()
+            foo2 = StringField()
+
+        class Person(Document):
+            settings = EmbeddedDocumentField(AdminSettings)
+            name = StringField()
+
+        Person.drop_collection()
+
+        p = Person(settings=AdminSettings(foo1="bar1", foo2="bar2"), name="John").save()
+
+        # Test non exiting attribute
+        with pytest.raises(InvalidQueryError) as exc_info:
+            Person.objects(settings__notexist="bar").first()
+        assert str(exc_info.value) == u'Cannot resolve field "notexist"'
+
+        with pytest.raises(LookUpError):
+            Person.objects.only("settings.notexist")
+
+        # Test existing attribute
+        assert Person.objects(settings__foo1="bar1").first().id == p.id
+        only_p = Person.objects.only("settings.foo1").first()
+        assert only_p.settings.foo1 == p.settings.foo1
+        assert only_p.settings.foo2 is None
+        assert only_p.name is None
+
+        exclude_p = Person.objects.exclude("settings.foo1").first()
+        assert exclude_p.settings.foo1 is None
+        assert exclude_p.settings.foo2 == p.settings.foo2
+        assert exclude_p.name == p.name
+
+    def test_query_embedded_document_attribute_with_inheritance(self):
+        class BaseSettings(EmbeddedDocument):
+            meta = {"allow_inheritance": True}
+            base_foo = StringField()
+
+        class AdminSettings(BaseSettings):
+            sub_foo = StringField()
+
+        class Person(Document):
+            settings = EmbeddedDocumentField(BaseSettings)
+
+        Person.drop_collection()
+
+        p = Person(settings=AdminSettings(base_foo="basefoo", sub_foo="subfoo"))
+        p.save()
+
+        # Test non exiting attribute
+        with pytest.raises(InvalidQueryError) as exc_info:
+            assert Person.objects(settings__notexist="bar").first().id == p.id
+        assert str(exc_info.value) == u'Cannot resolve field "notexist"'
+
+        # Test existing attribute
+        assert Person.objects(settings__base_foo="basefoo").first().id == p.id
+        assert Person.objects(settings__sub_foo="subfoo").first().id == p.id
+
+        only_p = Person.objects.only("settings.base_foo", "settings._cls").first()
+        assert only_p.settings.base_foo == "basefoo"
+        assert only_p.settings.sub_foo is None
+
+    def test_query_list_embedded_document_with_inheritance(self):
+        class Post(EmbeddedDocument):
+            title = StringField(max_length=120, required=True)
+            meta = {"allow_inheritance": True}
+
+        class TextPost(Post):
+            content = StringField()
+
+        class MoviePost(Post):
+            author = StringField()
+
+        class Record(Document):
+            posts = ListField(EmbeddedDocumentField(Post))
+
+        record_movie = Record(posts=[MoviePost(author="John", title="foo")]).save()
+        record_text = Record(posts=[TextPost(content="a", title="foo")]).save()
+
+        records = list(Record.objects(posts__author=record_movie.posts[0].author))
+        assert len(records) == 1
+        assert records[0].id == record_movie.id
+
+        records = list(Record.objects(posts__content=record_text.posts[0].content))
+        assert len(records) == 1
+        assert records[0].id == record_text.id
+
+        assert Record.objects(posts__title="foo").count() == 2
+
+
+class TestGenericEmbeddedDocumentField(MongoDBTestCase):
+    def test_generic_embedded_document(self):
+        class Car(EmbeddedDocument):
+            name = StringField()
+
+        class Dish(EmbeddedDocument):
+            food = StringField(required=True)
+            number = IntField()
+
+        class Person(Document):
+            name = StringField()
+            like = GenericEmbeddedDocumentField()
+
+        Person.drop_collection()
+
+        person = Person(name="Test User")
+        person.like = Car(name="Fiat")
+        person.save()
+
+        person = Person.objects.first()
+        assert isinstance(person.like, Car)
+
+        person.like = Dish(food="arroz", number=15)
+        person.save()
+
+        person = Person.objects.first()
+        assert isinstance(person.like, Dish)
+
+    def test_generic_embedded_document_choices(self):
+        """Ensure you can limit GenericEmbeddedDocument choices."""
+
+        class Car(EmbeddedDocument):
+            name = StringField()
+
+        class Dish(EmbeddedDocument):
+            food = StringField(required=True)
+            number = IntField()
+
+        class Person(Document):
+            name = StringField()
+            like = GenericEmbeddedDocumentField(choices=(Dish,))
+
+        Person.drop_collection()
+
+        person = Person(name="Test User")
+        person.like = Car(name="Fiat")
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person.like = Dish(food="arroz", number=15)
+        person.save()
+
+        person = Person.objects.first()
+        assert isinstance(person.like, Dish)
+
+    def test_generic_list_embedded_document_choices(self):
+        """Ensure you can limit GenericEmbeddedDocument choices inside
+        a list field.
+        """
+
+        class Car(EmbeddedDocument):
+            name = StringField()
+
+        class Dish(EmbeddedDocument):
+            food = StringField(required=True)
+            number = IntField()
+
+        class Person(Document):
+            name = StringField()
+            likes = ListField(GenericEmbeddedDocumentField(choices=(Dish,)))
+
+        Person.drop_collection()
+
+        person = Person(name="Test User")
+        person.likes = [Car(name="Fiat")]
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person.likes = [Dish(food="arroz", number=15)]
+        person.save()
+
+        person = Person.objects.first()
+        assert isinstance(person.likes[0], Dish)
+
+    def test_choices_validation_documents(self):
+        """
+        Ensure fields with document choices validate given a valid choice.
+        """
+
+        class UserComments(EmbeddedDocument):
+            author = StringField()
+            message = StringField()
+
+        class BlogPost(Document):
+            comments = ListField(GenericEmbeddedDocumentField(choices=(UserComments,)))
+
+        # Ensure Validation Passes
+        BlogPost(comments=[UserComments(author="user2", message="message2")]).save()
+
+    def test_choices_validation_documents_invalid(self):
+        """
+        Ensure fields with document choices validate given an invalid choice.
+        This should throw a ValidationError exception.
+        """
+
+        class UserComments(EmbeddedDocument):
+            author = StringField()
+            message = StringField()
+
+        class ModeratorComments(EmbeddedDocument):
+            author = StringField()
+            message = StringField()
+
+        class BlogPost(Document):
+            comments = ListField(GenericEmbeddedDocumentField(choices=(UserComments,)))
+
+        # Single Entry Failure
+        post = BlogPost(comments=[ModeratorComments(author="mod1", message="message1")])
+        with pytest.raises(ValidationError):
+            post.save()
+
+        # Mixed Entry Failure
+        post = BlogPost(
+            comments=[
+                ModeratorComments(author="mod1", message="message1"),
+                UserComments(author="user2", message="message2"),
+            ]
+        )
+        with pytest.raises(ValidationError):
+            post.save()
+
+    def test_choices_validation_documents_inheritance(self):
+        """
+        Ensure fields with document choices validate given subclass of choice.
+        """
+
+        class Comments(EmbeddedDocument):
+            meta = {"abstract": True}
+            author = StringField()
+            message = StringField()
+
+        class UserComments(Comments):
+            pass
+
+        class BlogPost(Document):
+            comments = ListField(GenericEmbeddedDocumentField(choices=(Comments,)))
+
+        # Save Valid EmbeddedDocument Type
+        BlogPost(comments=[UserComments(author="user2", message="message2")]).save()
+
+    def test_query_generic_embedded_document_attribute(self):
+        class AdminSettings(EmbeddedDocument):
+            foo1 = StringField()
+
+        class NonAdminSettings(EmbeddedDocument):
+            foo2 = StringField()
+
+        class Person(Document):
+            settings = GenericEmbeddedDocumentField(
+                choices=(AdminSettings, NonAdminSettings)
+            )
+
+        Person.drop_collection()
+
+        p1 = Person(settings=AdminSettings(foo1="bar1")).save()
+        p2 = Person(settings=NonAdminSettings(foo2="bar2")).save()
+
+        # Test non exiting attribute
+        with pytest.raises(InvalidQueryError) as exc_info:
+            Person.objects(settings__notexist="bar").first()
+        assert str(exc_info.value) == u'Cannot resolve field "notexist"'
+
+        with pytest.raises(LookUpError):
+            Person.objects.only("settings.notexist")
+
+        # Test existing attribute
+        assert Person.objects(settings__foo1="bar1").first().id == p1.id
+        assert Person.objects(settings__foo2="bar2").first().id == p2.id
+
+    def test_query_generic_embedded_document_attribute_with_inheritance(self):
+        class BaseSettings(EmbeddedDocument):
+            meta = {"allow_inheritance": True}
+            base_foo = StringField()
+
+        class AdminSettings(BaseSettings):
+            sub_foo = StringField()
+
+        class Person(Document):
+            settings = GenericEmbeddedDocumentField(choices=[BaseSettings])
+
+        Person.drop_collection()
+
+        p = Person(settings=AdminSettings(base_foo="basefoo", sub_foo="subfoo"))
+        p.save()
+
+        # Test non exiting attribute
+        with pytest.raises(InvalidQueryError) as exc_info:
+            assert Person.objects(settings__notexist="bar").first().id == p.id
+        assert str(exc_info.value) == u'Cannot resolve field "notexist"'
+
+        # Test existing attribute
+        assert Person.objects(settings__base_foo="basefoo").first().id == p.id
+        assert Person.objects(settings__sub_foo="subfoo").first().id == p.id

tests/fields/test_file_field.py

@@ -0,0 +1,587 @@
+# -*- coding: utf-8 -*-
+import copy
+import os
+import tempfile
+import unittest
+from io import BytesIO
+
+import gridfs
+import pytest
+
+from mongoengine import *
+from mongoengine.connection import get_db
+
+try:
+    from PIL import Image
+
+    HAS_PIL = True
+except ImportError:
+    HAS_PIL = False
+
+from tests.utils import MongoDBTestCase
+
+require_pil = pytest.mark.skipif(not HAS_PIL, reason="PIL not installed")
+
+TEST_IMAGE_PATH = os.path.join(os.path.dirname(__file__), "mongoengine.png")
+TEST_IMAGE2_PATH = os.path.join(os.path.dirname(__file__), "mongodb_leaf.png")
+
+
+def get_file(path):
+    """Use a BytesIO instead of a file to allow
+    to have a one-liner and avoid that the file remains opened"""
+    bytes_io = BytesIO()
+    with open(path, "rb") as f:
+        bytes_io.write(f.read())
+    bytes_io.seek(0)
+    return bytes_io
+
+
+class TestFileField(MongoDBTestCase):
+    def tearDown(self):
+        self.db.drop_collection("fs.files")
+        self.db.drop_collection("fs.chunks")
+
+    def test_file_field_optional(self):
+        # Make sure FileField is optional and not required
+        class DemoFile(Document):
+            the_file = FileField()
+
+        DemoFile.objects.create()
+
+    def test_file_fields(self):
+        """Ensure that file fields can be written to and their data retrieved
+        """
+
+        class PutFile(Document):
+            the_file = FileField()
+
+        PutFile.drop_collection()
+
+        text = "Hello, World!".encode("latin-1")
+        content_type = "text/plain"
+
+        putfile = PutFile()
+        putfile.the_file.put(text, content_type=content_type, filename="hello")
+        putfile.save()
+
+        result = PutFile.objects.first()
+        assert putfile == result
+        assert (
+            "%s" % result.the_file
+            == "<GridFSProxy: hello (%s)>" % result.the_file.grid_id
+        )
+        assert result.the_file.read() == text
+        assert result.the_file.content_type == content_type
+        result.the_file.delete()  # Remove file from GridFS
+        PutFile.objects.delete()
+
+        # Ensure file-like objects are stored
+        PutFile.drop_collection()
+
+        putfile = PutFile()
+        putstring = BytesIO()
+        putstring.write(text)
+        putstring.seek(0)
+        putfile.the_file.put(putstring, content_type=content_type)
+        putfile.save()
+
+        result = PutFile.objects.first()
+        assert putfile == result
+        assert result.the_file.read() == text
+        assert result.the_file.content_type == content_type
+        result.the_file.delete()
+
+    def test_file_fields_stream(self):
+        """Ensure that file fields can be written to and their data retrieved
+        """
+
+        class StreamFile(Document):
+            the_file = FileField()
+
+        StreamFile.drop_collection()
+
+        text = "Hello, World!".encode("latin-1")
+        more_text = "Foo Bar".encode("latin-1")
+        content_type = "text/plain"
+
+        streamfile = StreamFile()
+        streamfile.the_file.new_file(content_type=content_type)
+        streamfile.the_file.write(text)
+        streamfile.the_file.write(more_text)
+        streamfile.the_file.close()
+        streamfile.save()
+
+        result = StreamFile.objects.first()
+        assert streamfile == result
+        assert result.the_file.read() == text + more_text
+        assert result.the_file.content_type == content_type
+        result.the_file.seek(0)
+        assert result.the_file.tell() == 0
+        assert result.the_file.read(len(text)) == text
+        assert result.the_file.tell() == len(text)
+        assert result.the_file.read(len(more_text)) == more_text
+        assert result.the_file.tell() == len(text + more_text)
+        result.the_file.delete()
+
+        # Ensure deleted file returns None
+        assert result.the_file.read() is None
+
+    def test_file_fields_stream_after_none(self):
+        """Ensure that a file field can be written to after it has been saved as
+        None
+        """
+
+        class StreamFile(Document):
+            the_file = FileField()
+
+        StreamFile.drop_collection()
+
+        text = "Hello, World!".encode("latin-1")
+        more_text = "Foo Bar".encode("latin-1")
+
+        streamfile = StreamFile()
+        streamfile.save()
+        streamfile.the_file.new_file()
+        streamfile.the_file.write(text)
+        streamfile.the_file.write(more_text)
+        streamfile.the_file.close()
+        streamfile.save()
+
+        result = StreamFile.objects.first()
+        assert streamfile == result
+        assert result.the_file.read() == text + more_text
+        # assert result.the_file.content_type == content_type
+        result.the_file.seek(0)
+        assert result.the_file.tell() == 0
+        assert result.the_file.read(len(text)) == text
+        assert result.the_file.tell() == len(text)
+        assert result.the_file.read(len(more_text)) == more_text
+        assert result.the_file.tell() == len(text + more_text)
+        result.the_file.delete()
+
+        # Ensure deleted file returns None
+        assert result.the_file.read() is None
+
+    def test_file_fields_set(self):
+        class SetFile(Document):
+            the_file = FileField()
+
+        text = "Hello, World!".encode("latin-1")
+        more_text = "Foo Bar".encode("latin-1")
+
+        SetFile.drop_collection()
+
+        setfile = SetFile()
+        setfile.the_file = text
+        setfile.save()
+
+        result = SetFile.objects.first()
+        assert setfile == result
+        assert result.the_file.read() == text
+
+        # Try replacing file with new one
+        result.the_file.replace(more_text)
+        result.save()
+
+        result = SetFile.objects.first()
+        assert setfile == result
+        assert result.the_file.read() == more_text
+        result.the_file.delete()
+
+    def test_file_field_no_default(self):
+        class GridDocument(Document):
+            the_file = FileField()
+
+        GridDocument.drop_collection()
+
+        with tempfile.TemporaryFile() as f:
+            f.write("Hello World!".encode("latin-1"))
+            f.flush()
+
+            # Test without default
+            doc_a = GridDocument()
+            doc_a.save()
+
+            doc_b = GridDocument.objects.with_id(doc_a.id)
+            doc_b.the_file.replace(f, filename="doc_b")
+            doc_b.save()
+            assert doc_b.the_file.grid_id is not None
+
+            # Test it matches
+            doc_c = GridDocument.objects.with_id(doc_b.id)
+            assert doc_b.the_file.grid_id == doc_c.the_file.grid_id
+
+            # Test with default
+            doc_d = GridDocument(the_file="".encode("latin-1"))
+            doc_d.save()
+
+            doc_e = GridDocument.objects.with_id(doc_d.id)
+            assert doc_d.the_file.grid_id == doc_e.the_file.grid_id
+
+            doc_e.the_file.replace(f, filename="doc_e")
+            doc_e.save()
+
+            doc_f = GridDocument.objects.with_id(doc_e.id)
+            assert doc_e.the_file.grid_id == doc_f.the_file.grid_id
+
+        db = GridDocument._get_db()
+        grid_fs = gridfs.GridFS(db)
+        assert ["doc_b", "doc_e"] == grid_fs.list()
+
+    def test_file_uniqueness(self):
+        """Ensure that each instance of a FileField is unique
+        """
+
+        class TestFile(Document):
+            name = StringField()
+            the_file = FileField()
+
+        # First instance
+        test_file = TestFile()
+        test_file.name = "Hello, World!"
+        test_file.the_file.put("Hello, World!".encode("latin-1"))
+        test_file.save()
+
+        # Second instance
+        test_file_dupe = TestFile()
+        data = test_file_dupe.the_file.read()  # Should be None
+
+        assert test_file.name != test_file_dupe.name
+        assert test_file.the_file.read() != data
+
+        TestFile.drop_collection()
+
+    def test_file_saving(self):
+        """Ensure you can add meta data to file"""
+
+        class Animal(Document):
+            genus = StringField()
+            family = StringField()
+            photo = FileField()
+
+        Animal.drop_collection()
+        marmot = Animal(genus="Marmota", family="Sciuridae")
+
+        marmot_photo_content = get_file(TEST_IMAGE_PATH)  # Retrieve a photo from disk
+        marmot.photo.put(marmot_photo_content, content_type="image/jpeg", foo="bar")
+        marmot.photo.close()
+        marmot.save()
+
+        marmot = Animal.objects.get()
+        assert marmot.photo.content_type == "image/jpeg"
+        assert marmot.photo.foo == "bar"
+
+    def test_file_reassigning(self):
+        class TestFile(Document):
+            the_file = FileField()
+
+        TestFile.drop_collection()
+
+        test_file = TestFile(the_file=get_file(TEST_IMAGE_PATH)).save()
+        assert test_file.the_file.get().length == 8313
+
+        test_file = TestFile.objects.first()
+        test_file.the_file = get_file(TEST_IMAGE2_PATH)
+        test_file.save()
+        assert test_file.the_file.get().length == 4971
+
+    def test_file_boolean(self):
+        """Ensure that a boolean test of a FileField indicates its presence
+        """
+
+        class TestFile(Document):
+            the_file = FileField()
+
+        TestFile.drop_collection()
+
+        test_file = TestFile()
+        assert not bool(test_file.the_file)
+        test_file.the_file.put(
+            "Hello, World!".encode("latin-1"), content_type="text/plain"
+        )
+        test_file.save()
+        assert bool(test_file.the_file)
+
+        test_file = TestFile.objects.first()
+        assert test_file.the_file.content_type == "text/plain"
+
+    def test_file_cmp(self):
+        """Test comparing against other types"""
+
+        class TestFile(Document):
+            the_file = FileField()
+
+        test_file = TestFile()
+        assert test_file.the_file not in [{"test": 1}]
+
+    def test_file_disk_space(self):
+        """ Test disk space usage when we delete/replace a file """
+
+        class TestFile(Document):
+            the_file = FileField()
+
+        text = "Hello, World!".encode("latin-1")
+        content_type = "text/plain"
+
+        testfile = TestFile()
+        testfile.the_file.put(text, content_type=content_type, filename="hello")
+        testfile.save()
+
+        # Now check fs.files and fs.chunks
+        db = TestFile._get_db()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 1
+        assert len(list(chunks)) == 1
+
+        # Deleting the docoument should delete the files
+        testfile.delete()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 0
+        assert len(list(chunks)) == 0
+
+        # Test case where we don't store a file in the first place
+        testfile = TestFile()
+        testfile.save()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 0
+        assert len(list(chunks)) == 0
+
+        testfile.delete()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 0
+        assert len(list(chunks)) == 0
+
+        # Test case where we overwrite the file
+        testfile = TestFile()
+        testfile.the_file.put(text, content_type=content_type, filename="hello")
+        testfile.save()
+
+        text = "Bonjour, World!".encode("latin-1")
+        testfile.the_file.replace(text, content_type=content_type, filename="hello")
+        testfile.save()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 1
+        assert len(list(chunks)) == 1
+
+        testfile.delete()
+
+        files = db.fs.files.find()
+        chunks = db.fs.chunks.find()
+        assert len(list(files)) == 0
+        assert len(list(chunks)) == 0
+
+    @require_pil
+    def test_image_field(self):
+        class TestImage(Document):
+            image = ImageField()
+
+        TestImage.drop_collection()
+
+        with tempfile.TemporaryFile() as f:
+            f.write("Hello World!".encode("latin-1"))
+            f.flush()
+
+            t = TestImage()
+            try:
+                t.image.put(f)
+                self.fail("Should have raised an invalidation error")
+            except ValidationError as e:
+                assert "%s" % e == "Invalid image: cannot identify image file %s" % f
+
+        t = TestImage()
+        t.image.put(get_file(TEST_IMAGE_PATH))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        assert t.image.format == "PNG"
+
+        w, h = t.image.size
+        assert w == 371
+        assert h == 76
+
+        t.image.delete()
+
+    @require_pil
+    def test_image_field_reassigning(self):
+        class TestFile(Document):
+            the_file = ImageField()
+
+        TestFile.drop_collection()
+
+        test_file = TestFile(the_file=get_file(TEST_IMAGE_PATH)).save()
+        assert test_file.the_file.size == (371, 76)
+
+        test_file = TestFile.objects.first()
+        test_file.the_file = get_file(TEST_IMAGE2_PATH)
+        test_file.save()
+        assert test_file.the_file.size == (45, 101)
+
+    @require_pil
+    def test_image_field_resize(self):
+        class TestImage(Document):
+            image = ImageField(size=(185, 37))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(get_file(TEST_IMAGE_PATH))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        assert t.image.format == "PNG"
+        w, h = t.image.size
+
+        assert w == 185
+        assert h == 37
+
+        t.image.delete()
+
+    @require_pil
+    def test_image_field_resize_force(self):
+        class TestImage(Document):
+            image = ImageField(size=(185, 37, True))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(get_file(TEST_IMAGE_PATH))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        assert t.image.format == "PNG"
+        w, h = t.image.size
+
+        assert w == 185
+        assert h == 37
+
+        t.image.delete()
+
+    @require_pil
+    def test_image_field_thumbnail(self):
+        class TestImage(Document):
+            image = ImageField(thumbnail_size=(92, 18))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(get_file(TEST_IMAGE_PATH))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        assert t.image.thumbnail.format == "PNG"
+        assert t.image.thumbnail.width == 92
+        assert t.image.thumbnail.height == 18
+
+        t.image.delete()
+
+    def test_file_multidb(self):
+        register_connection("test_files", "test_files")
+
+        class TestFile(Document):
+            name = StringField()
+            the_file = FileField(db_alias="test_files", collection_name="macumba")
+
+        TestFile.drop_collection()
+
+        # delete old filesystem
+        get_db("test_files").macumba.files.drop()
+        get_db("test_files").macumba.chunks.drop()
+
+        # First instance
+        test_file = TestFile()
+        test_file.name = "Hello, World!"
+        test_file.the_file.put("Hello, World!".encode("latin-1"), name="hello.txt")
+        test_file.save()
+
+        data = get_db("test_files").macumba.files.find_one()
+        assert data.get("name") == "hello.txt"
+
+        test_file = TestFile.objects.first()
+        assert test_file.the_file.read() == "Hello, World!".encode("latin-1")
+
+        test_file = TestFile.objects.first()
+        test_file.the_file = "Hello, World!".encode("latin-1")
+        test_file.save()
+
+        test_file = TestFile.objects.first()
+        assert test_file.the_file.read() == "Hello, World!".encode("latin-1")
+
+    def test_copyable(self):
+        class PutFile(Document):
+            the_file = FileField()
+
+        PutFile.drop_collection()
+
+        text = "Hello, World!".encode("latin-1")
+        content_type = "text/plain"
+
+        putfile = PutFile()
+        putfile.the_file.put(text, content_type=content_type)
+        putfile.save()
+
+        class TestFile(Document):
+            name = StringField()
+
+        assert putfile == copy.copy(putfile)
+        assert putfile == copy.deepcopy(putfile)
+
+    @require_pil
+    def test_get_image_by_grid_id(self):
+        class TestImage(Document):
+
+            image1 = ImageField()
+            image2 = ImageField()
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image1.put(get_file(TEST_IMAGE_PATH))
+        t.image2.put(get_file(TEST_IMAGE2_PATH))
+        t.save()
+
+        test = TestImage.objects.first()
+        grid_id = test.image1.grid_id
+
+        assert 1 == TestImage.objects(Q(image1=grid_id) or Q(image2=grid_id)).count()
+
+    def test_complex_field_filefield(self):
+        """Ensure you can add meta data to file"""
+
+        class Animal(Document):
+            genus = StringField()
+            family = StringField()
+            photos = ListField(FileField())
+
+        Animal.drop_collection()
+        marmot = Animal(genus="Marmota", family="Sciuridae")
+
+        with open(TEST_IMAGE_PATH, "rb") as marmot_photo:  # Retrieve a photo from disk
+            photos_field = marmot._fields["photos"].field
+            new_proxy = photos_field.get_proxy_obj("photos", marmot)
+            new_proxy.put(marmot_photo, content_type="image/jpeg", foo="bar")
+
+        marmot.photos.append(new_proxy)
+        marmot.save()
+
+        marmot = Animal.objects.get()
+        assert marmot.photos[0].content_type == "image/jpeg"
+        assert marmot.photos[0].foo == "bar"
+        assert marmot.photos[0].get().length == 8313
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/fields/test_float_field.py

@@ -0,0 +1,62 @@
+# -*- coding: utf-8 -*-
+import pytest
+
+from mongoengine import *
+
+from tests.utils import MongoDBTestCase
+
+
+class TestFloatField(MongoDBTestCase):
+    def test_float_ne_operator(self):
+        class TestDocument(Document):
+            float_fld = FloatField()
+
+        TestDocument.drop_collection()
+
+        TestDocument(float_fld=None).save()
+        TestDocument(float_fld=1).save()
+
+        assert 1 == TestDocument.objects(float_fld__ne=None).count()
+        assert 1 == TestDocument.objects(float_fld__ne=1).count()
+
+    def test_validation(self):
+        """Ensure that invalid values cannot be assigned to float fields.
+        """
+
+        class Person(Document):
+            height = FloatField(min_value=0.1, max_value=3.5)
+
+        class BigPerson(Document):
+            height = FloatField()
+
+        person = Person()
+        person.height = 1.89
+        person.validate()
+
+        person.height = "2.0"
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person.height = 0.01
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person.height = 4.0
+        with pytest.raises(ValidationError):
+            person.validate()
+
+        person_2 = Person(height="something invalid")
+        with pytest.raises(ValidationError):
+            person_2.validate()
+
+        big_person = BigPerson()
+
+        big_person.height = int(0)
+        big_person.validate()
+
+        big_person.height = 2 ** 500
+        big_person.validate()
+
+        big_person.height = 2 ** 100000  # Too big for a float value
+        with pytest.raises(ValidationError):
+            big_person.validate()

tests/fields/test_geo_fields.py

@@ -0,0 +1,417 @@
+# -*- coding: utf-8 -*-
+import unittest
+
+from mongoengine import *
+from tests.utils import MongoDBTestCase
+
+
+class TestGeoField(MongoDBTestCase):
+    def _test_for_expected_error(self, Cls, loc, expected):
+        try:
+            Cls(loc=loc).validate()
+            self.fail("Should not validate the location {0}".format(loc))
+        except ValidationError as e:
+            assert expected == e.to_dict()["loc"]
+
+    def test_geopoint_validation(self):
+        class Location(Document):
+            loc = GeoPointField()
+
+        invalid_coords = [{"x": 1, "y": 2}, 5, "a"]
+        expected = "GeoPointField can only accept tuples or lists of (x, y)"
+
+        for coord in invalid_coords:
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [[], [1], [1, 2, 3]]
+        for coord in invalid_coords:
+            expected = "Value (%s) must be a two-dimensional point" % repr(coord)
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [[{}, {}], ("a", "b")]
+        for coord in invalid_coords:
+            expected = "Both values (%s) in point must be float or int" % repr(coord)
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [21, 4, "a"]
+        for coord in invalid_coords:
+            expected = "GeoPointField can only accept tuples or lists of (x, y)"
+            self._test_for_expected_error(Location, coord, expected)
+
+    def test_point_validation(self):
+        class Location(Document):
+            loc = PointField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = (
+            "PointField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": []}
+        expected = 'PointField type must be "Point"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "Point", "coordinates": [1, 2, 3]}
+        expected = "Value ([1, 2, 3]) must be a two-dimensional point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [5, "a"]
+        expected = "PointField can only accept lists of [x, y]"
+        for coord in invalid_coords:
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [[], [1], [1, 2, 3]]
+        for coord in invalid_coords:
+            expected = "Value (%s) must be a two-dimensional point" % repr(coord)
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [[{}, {}], ("a", "b")]
+        for coord in invalid_coords:
+            expected = "Both values (%s) in point must be float or int" % repr(coord)
+            self._test_for_expected_error(Location, coord, expected)
+
+        Location(loc=[1, 2]).validate()
+        Location(
+            loc={"type": "Point", "coordinates": [81.4471435546875, 23.61432859499169]}
+        ).validate()
+
+    def test_linestring_validation(self):
+        class Location(Document):
+            loc = LineStringField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = "LineStringField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": [[]]}
+        expected = 'LineStringField type must be "LineString"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "LineString", "coordinates": [[1, 2, 3]]}
+        expected = (
+            "Invalid LineString:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [5, "a"]
+        expected = "Invalid LineString must contain at least one valid point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[1]]
+        expected = (
+            "Invalid LineString:\nValue (%s) must be a two-dimensional point"
+            % repr(invalid_coords[0])
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[1, 2, 3]]
+        expected = (
+            "Invalid LineString:\nValue (%s) must be a two-dimensional point"
+            % repr(invalid_coords[0])
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[{}, {}]], [("a", "b")]]
+        for coord in invalid_coords:
+            expected = (
+                "Invalid LineString:\nBoth values (%s) in point must be float or int"
+                % repr(coord[0])
+            )
+            self._test_for_expected_error(Location, coord, expected)
+
+        Location(loc=[[1, 2], [3, 4], [5, 6], [1, 2]]).validate()
+
+    def test_polygon_validation(self):
+        class Location(Document):
+            loc = PolygonField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = (
+            "PolygonField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": [[]]}
+        expected = 'PolygonField type must be "Polygon"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "Polygon", "coordinates": [[[1, 2, 3]]]}
+        expected = "Invalid Polygon:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[5, "a"]]]
+        expected = (
+            "Invalid Polygon:\nBoth values ([5, 'a']) in point must be float or int"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[]]]
+        expected = "Invalid Polygon must contain at least one valid linestring"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[1, 2, 3]]]
+        expected = "Invalid Polygon:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[{}, {}]], [("a", "b")]]
+        expected = "Invalid Polygon:\nBoth values ([{}, {}]) in point must be float or int, Both values (('a', 'b')) in point must be float or int"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[1, 2], [3, 4]]]
+        expected = "Invalid Polygon:\nLineStrings must start and end at the same point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        Location(loc=[[[1, 2], [3, 4], [5, 6], [1, 2]]]).validate()
+
+    def test_multipoint_validation(self):
+        class Location(Document):
+            loc = MultiPointField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = "MultiPointField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": [[]]}
+        expected = 'MultiPointField type must be "MultiPoint"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MultiPoint", "coordinates": [[1, 2, 3]]}
+        expected = "Value ([1, 2, 3]) must be a two-dimensional point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[]]
+        expected = "Invalid MultiPoint must contain at least one valid point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[1]], [[1, 2, 3]]]
+        for coord in invalid_coords:
+            expected = "Value (%s) must be a two-dimensional point" % repr(coord[0])
+            self._test_for_expected_error(Location, coord, expected)
+
+        invalid_coords = [[[{}, {}]], [("a", "b")]]
+        for coord in invalid_coords:
+            expected = "Both values (%s) in point must be float or int" % repr(coord[0])
+            self._test_for_expected_error(Location, coord, expected)
+
+        Location(loc=[[1, 2]]).validate()
+        Location(
+            loc={
+                "type": "MultiPoint",
+                "coordinates": [[1, 2], [81.4471435546875, 23.61432859499169]],
+            }
+        ).validate()
+
+    def test_multilinestring_validation(self):
+        class Location(Document):
+            loc = MultiLineStringField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = "MultiLineStringField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": [[]]}
+        expected = 'MultiLineStringField type must be "MultiLineString"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MultiLineString", "coordinates": [[[1, 2, 3]]]}
+        expected = "Invalid MultiLineString:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [5, "a"]
+        expected = "Invalid MultiLineString must contain at least one valid linestring"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[1]]]
+        expected = (
+            "Invalid MultiLineString:\nValue (%s) must be a two-dimensional point"
+            % repr(invalid_coords[0][0])
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[1, 2, 3]]]
+        expected = (
+            "Invalid MultiLineString:\nValue (%s) must be a two-dimensional point"
+            % repr(invalid_coords[0][0])
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[{}, {}]]], [[("a", "b")]]]
+        for coord in invalid_coords:
+            expected = (
+                "Invalid MultiLineString:\nBoth values (%s) in point must be float or int"
+                % repr(coord[0][0])
+            )
+            self._test_for_expected_error(Location, coord, expected)
+
+        Location(loc=[[[1, 2], [3, 4], [5, 6], [1, 2]]]).validate()
+
+    def test_multipolygon_validation(self):
+        class Location(Document):
+            loc = MultiPolygonField()
+
+        invalid_coords = {"x": 1, "y": 2}
+        expected = "MultiPolygonField can only accept a valid GeoJson dictionary or lists of (x, y)"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MadeUp", "coordinates": [[]]}
+        expected = 'MultiPolygonField type must be "MultiPolygon"'
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = {"type": "MultiPolygon", "coordinates": [[[[1, 2, 3]]]]}
+        expected = (
+            "Invalid MultiPolygon:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[5, "a"]]]]
+        expected = "Invalid MultiPolygon:\nBoth values ([5, 'a']) in point must be float or int"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[]]]]
+        expected = "Invalid MultiPolygon must contain at least one valid Polygon"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[1, 2, 3]]]]
+        expected = (
+            "Invalid MultiPolygon:\nValue ([1, 2, 3]) must be a two-dimensional point"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[{}, {}]]], [[("a", "b")]]]
+        expected = "Invalid MultiPolygon:\nBoth values ([{}, {}]) in point must be float or int, Both values (('a', 'b')) in point must be float or int"
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        invalid_coords = [[[[1, 2], [3, 4]]]]
+        expected = (
+            "Invalid MultiPolygon:\nLineStrings must start and end at the same point"
+        )
+        self._test_for_expected_error(Location, invalid_coords, expected)
+
+        Location(loc=[[[[1, 2], [3, 4], [5, 6], [1, 2]]]]).validate()
+
+    def test_indexes_geopoint(self):
+        """Ensure that indexes are created automatically for GeoPointFields.
+        """
+
+        class Event(Document):
+            title = StringField()
+            location = GeoPointField()
+
+        geo_indicies = Event._geo_indices()
+        assert geo_indicies == [{"fields": [("location", "2d")]}]
+
+    def test_geopoint_embedded_indexes(self):
+        """Ensure that indexes are created automatically for GeoPointFields on
+        embedded documents.
+        """
+
+        class Venue(EmbeddedDocument):
+            location = GeoPointField()
+            name = StringField()
+
+        class Event(Document):
+            title = StringField()
+            venue = EmbeddedDocumentField(Venue)
+
+        geo_indicies = Event._geo_indices()
+        assert geo_indicies == [{"fields": [("venue.location", "2d")]}]
+
+    def test_indexes_2dsphere(self):
+        """Ensure that indexes are created automatically for GeoPointFields.
+        """
+
+        class Event(Document):
+            title = StringField()
+            point = PointField()
+            line = LineStringField()
+            polygon = PolygonField()
+
+        geo_indicies = Event._geo_indices()
+        assert {"fields": [("line", "2dsphere")]} in geo_indicies
+        assert {"fields": [("polygon", "2dsphere")]} in geo_indicies
+        assert {"fields": [("point", "2dsphere")]} in geo_indicies
+
+    def test_indexes_2dsphere_embedded(self):
+        """Ensure that indexes are created automatically for GeoPointFields.
+        """
+
+        class Venue(EmbeddedDocument):
+            name = StringField()
+            point = PointField()
+            line = LineStringField()
+            polygon = PolygonField()
+
+        class Event(Document):
+            title = StringField()
+            venue = EmbeddedDocumentField(Venue)
+
+        geo_indicies = Event._geo_indices()
+        assert {"fields": [("venue.line", "2dsphere")]} in geo_indicies
+        assert {"fields": [("venue.polygon", "2dsphere")]} in geo_indicies
+        assert {"fields": [("venue.point", "2dsphere")]} in geo_indicies
+
+    def test_geo_indexes_recursion(self):
+        class Location(Document):
+            name = StringField()
+            location = GeoPointField()
+
+        class Parent(Document):
+            name = StringField()
+            location = ReferenceField(Location)
+
+        Location.drop_collection()
+        Parent.drop_collection()
+
+        Parent(name="Berlin").save()
+        info = Parent._get_collection().index_information()
+        assert "location_2d" not in info
+        info = Location._get_collection().index_information()
+        assert "location_2d" in info
+
+        assert len(Parent._geo_indices()) == 0
+        assert len(Location._geo_indices()) == 1
+
+    def test_geo_indexes_auto_index(self):
+
+        # Test just listing the fields
+        class Log(Document):
+            location = PointField(auto_index=False)
+            datetime = DateTimeField()
+
+            meta = {"indexes": [[("location", "2dsphere"), ("datetime", 1)]]}
+
+        assert [] == Log._geo_indices()
+
+        Log.drop_collection()
+        Log.ensure_indexes()
+
+        info = Log._get_collection().index_information()
+        assert info["location_2dsphere_datetime_1"]["key"] == [
+            ("location", "2dsphere"),
+            ("datetime", 1),
+        ]
+
+        # Test listing explicitly
+        class Log(Document):
+            location = PointField(auto_index=False)
+            datetime = DateTimeField()
+
+            meta = {
+                "indexes": [{"fields": [("location", "2dsphere"), ("datetime", 1)]}]
+            }
+
+        assert [] == Log._geo_indices()
+
+        Log.drop_collection()
+        Log.ensure_indexes()
+
+        info = Log._get_collection().index_information()
+        assert info["location_2dsphere_datetime_1"]["key"] == [
+            ("location", "2dsphere"),
+            ("datetime", 1),
+        ]
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/fields/test_int_field.py

@@ -0,0 +1,47 @@
+# -*- coding: utf-8 -*-
+import pytest
+
+from mongoengine import *
+
+from tests.utils import MongoDBTestCase
+
+
+class TestIntField(MongoDBTestCase):
+    def test_int_validation(self):
+        """Ensure that invalid values cannot be assigned to int fields.
+        """
+
+        class Person(Document):
+            age = IntField(min_value=0, max_value=110)
+
+        person = Person()
+        person.age = 0
+        person.validate()
+
+        person.age = 50
+        person.validate()
+
+        person.age = 110
+        person.validate()
+
+        person.age = -1
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.age = 120
+        with pytest.raises(ValidationError):
+            person.validate()
+        person.age = "ten"
+        with pytest.raises(ValidationError):
+            person.validate()
+
+    def test_ne_operator(self):
+        class TestDocument(Document):
+            int_fld = IntField()
+
+        TestDocument.drop_collection()
+
+        TestDocument(int_fld=None).save()
+        TestDocument(int_fld=1).save()
+
+        assert 1 == TestDocument.objects(int_fld__ne=None).count()
+        assert 1 == TestDocument.objects(int_fld__ne=1).count()

tests/fields/test_lazy_reference_field.py

@@ -0,0 +1,576 @@
+# -*- coding: utf-8 -*-
+from bson import DBRef, ObjectId
+import pytest
+
+from mongoengine import *
+from mongoengine.base import LazyReference
+
+from tests.utils import MongoDBTestCase
+
+
+class TestLazyReferenceField(MongoDBTestCase):
+    def test_lazy_reference_config(self):
+        # Make sure ReferenceField only accepts a document class or a string
+        # with a document class name.
+        with pytest.raises(ValidationError):
+            LazyReferenceField(EmbeddedDocument)
+
+    def test___repr__(self):
+        class Animal(Document):
+            pass
+
+        class Ocurrence(Document):
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal()
+        oc = Ocurrence(animal=animal)
+        assert "LazyReference" in repr(oc.animal)
+
+    def test___getattr___unknown_attr_raises_attribute_error(self):
+        class Animal(Document):
+            pass
+
+        class Ocurrence(Document):
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal().save()
+        oc = Ocurrence(animal=animal)
+        with pytest.raises(AttributeError):
+            oc.animal.not_exist
+
+    def test_lazy_reference_simple(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        Ocurrence(person="test", animal=animal).save()
+        p = Ocurrence.objects.get()
+        assert isinstance(p.animal, LazyReference)
+        fetched_animal = p.animal.fetch()
+        assert fetched_animal == animal
+        # `fetch` keep cache on referenced document by default...
+        animal.tag = "not so heavy"
+        animal.save()
+        double_fetch = p.animal.fetch()
+        assert fetched_animal is double_fetch
+        assert double_fetch.tag == "heavy"
+        # ...unless specified otherwise
+        fetch_force = p.animal.fetch(force=True)
+        assert fetch_force is not fetched_animal
+        assert fetch_force.tag == "not so heavy"
+
+    def test_lazy_reference_fetch_invalid_ref(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        Ocurrence(person="test", animal=animal).save()
+        animal.delete()
+        p = Ocurrence.objects.get()
+        assert isinstance(p.animal, LazyReference)
+        with pytest.raises(DoesNotExist):
+            p.animal.fetch()
+
+    def test_lazy_reference_set(self):
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        class SubAnimal(Animal):
+            nick = StringField()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        sub_animal = SubAnimal(nick="doggo", name="dog").save()
+        for ref in (
+            animal,
+            animal.pk,
+            DBRef(animal._get_collection_name(), animal.pk),
+            LazyReference(Animal, animal.pk),
+            sub_animal,
+            sub_animal.pk,
+            DBRef(sub_animal._get_collection_name(), sub_animal.pk),
+            LazyReference(SubAnimal, sub_animal.pk),
+        ):
+            p = Ocurrence(person="test", animal=ref).save()
+            p.reload()
+            assert isinstance(p.animal, LazyReference)
+            p.animal.fetch()
+
+    def test_lazy_reference_bad_set(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        class BadDoc(Document):
+            pass
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        baddoc = BadDoc().save()
+        for bad in (
+            42,
+            "foo",
+            baddoc,
+            DBRef(baddoc._get_collection_name(), animal.pk),
+            LazyReference(BadDoc, animal.pk),
+        ):
+            with pytest.raises(ValidationError):
+                Ocurrence(person="test", animal=bad).save()
+
+    def test_lazy_reference_query_conversion(self):
+        """Ensure that LazyReferenceFields can be queried using objects and values
+        of the type of the primary key of the referenced object.
+        """
+
+        class Member(Document):
+            user_num = IntField(primary_key=True)
+
+        class BlogPost(Document):
+            title = StringField()
+            author = LazyReferenceField(Member, dbref=False)
+
+        Member.drop_collection()
+        BlogPost.drop_collection()
+
+        m1 = Member(user_num=1)
+        m1.save()
+        m2 = Member(user_num=2)
+        m2.save()
+
+        post1 = BlogPost(title="post 1", author=m1)
+        post1.save()
+
+        post2 = BlogPost(title="post 2", author=m2)
+        post2.save()
+
+        post = BlogPost.objects(author=m1).first()
+        assert post.id == post1.id
+
+        post = BlogPost.objects(author=m2).first()
+        assert post.id == post2.id
+
+        # Same thing by passing a LazyReference instance
+        post = BlogPost.objects(author=LazyReference(Member, m2.pk)).first()
+        assert post.id == post2.id
+
+    def test_lazy_reference_query_conversion_dbref(self):
+        """Ensure that LazyReferenceFields can be queried using objects and values
+        of the type of the primary key of the referenced object.
+        """
+
+        class Member(Document):
+            user_num = IntField(primary_key=True)
+
+        class BlogPost(Document):
+            title = StringField()
+            author = LazyReferenceField(Member, dbref=True)
+
+        Member.drop_collection()
+        BlogPost.drop_collection()
+
+        m1 = Member(user_num=1)
+        m1.save()
+        m2 = Member(user_num=2)
+        m2.save()
+
+        post1 = BlogPost(title="post 1", author=m1)
+        post1.save()
+
+        post2 = BlogPost(title="post 2", author=m2)
+        post2.save()
+
+        post = BlogPost.objects(author=m1).first()
+        assert post.id == post1.id
+
+        post = BlogPost.objects(author=m2).first()
+        assert post.id == post2.id
+
+        # Same thing by passing a LazyReference instance
+        post = BlogPost.objects(author=LazyReference(Member, m2.pk)).first()
+        assert post.id == post2.id
+
+    def test_lazy_reference_passthrough(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            animal = LazyReferenceField(Animal, passthrough=False)
+            animal_passthrough = LazyReferenceField(Animal, passthrough=True)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        Ocurrence(animal=animal, animal_passthrough=animal).save()
+        p = Ocurrence.objects.get()
+        assert isinstance(p.animal, LazyReference)
+        with pytest.raises(KeyError):
+            p.animal["name"]
+        with pytest.raises(AttributeError):
+            p.animal.name
+        assert p.animal.pk == animal.pk
+
+        assert p.animal_passthrough.name == "Leopard"
+        assert p.animal_passthrough["name"] == "Leopard"
+
+        # Should not be able to access referenced document's methods
+        with pytest.raises(AttributeError):
+            p.animal.save
+        with pytest.raises(KeyError):
+            p.animal["save"]
+
+    def test_lazy_reference_not_set(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        Ocurrence(person="foo").save()
+        p = Ocurrence.objects.get()
+        assert p.animal is None
+
+    def test_lazy_reference_equality(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        Animal.drop_collection()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        animalref = LazyReference(Animal, animal.pk)
+        assert animal == animalref
+        assert animalref == animal
+
+        other_animalref = LazyReference(Animal, ObjectId("54495ad94c934721ede76f90"))
+        assert animal != other_animalref
+        assert other_animalref != animal
+
+    def test_lazy_reference_embedded(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class EmbeddedOcurrence(EmbeddedDocument):
+            in_list = ListField(LazyReferenceField(Animal))
+            direct = LazyReferenceField(Animal)
+
+        class Ocurrence(Document):
+            in_list = ListField(LazyReferenceField(Animal))
+            in_embedded = EmbeddedDocumentField(EmbeddedOcurrence)
+            direct = LazyReferenceField(Animal)
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal1 = Animal(name="doggo").save()
+        animal2 = Animal(name="cheeta").save()
+
+        def check_fields_type(occ):
+            assert isinstance(occ.direct, LazyReference)
+            for elem in occ.in_list:
+                assert isinstance(elem, LazyReference)
+            assert isinstance(occ.in_embedded.direct, LazyReference)
+            for elem in occ.in_embedded.in_list:
+                assert isinstance(elem, LazyReference)
+
+        occ = Ocurrence(
+            in_list=[animal1, animal2],
+            in_embedded={"in_list": [animal1, animal2], "direct": animal1},
+            direct=animal1,
+        ).save()
+        check_fields_type(occ)
+        occ.reload()
+        check_fields_type(occ)
+        occ.direct = animal1.id
+        occ.in_list = [animal1.id, animal2.id]
+        occ.in_embedded.direct = animal1.id
+        occ.in_embedded.in_list = [animal1.id, animal2.id]
+        check_fields_type(occ)
+
+
+class TestGenericLazyReferenceField(MongoDBTestCase):
+    def test_generic_lazy_reference_simple(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = GenericLazyReferenceField()
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        Ocurrence(person="test", animal=animal).save()
+        p = Ocurrence.objects.get()
+        assert isinstance(p.animal, LazyReference)
+        fetched_animal = p.animal.fetch()
+        assert fetched_animal == animal
+        # `fetch` keep cache on referenced document by default...
+        animal.tag = "not so heavy"
+        animal.save()
+        double_fetch = p.animal.fetch()
+        assert fetched_animal is double_fetch
+        assert double_fetch.tag == "heavy"
+        # ...unless specified otherwise
+        fetch_force = p.animal.fetch(force=True)
+        assert fetch_force is not fetched_animal
+        assert fetch_force.tag == "not so heavy"
+
+    def test_generic_lazy_reference_choices(self):
+        class Animal(Document):
+            name = StringField()
+
+        class Vegetal(Document):
+            name = StringField()
+
+        class Mineral(Document):
+            name = StringField()
+
+        class Ocurrence(Document):
+            living_thing = GenericLazyReferenceField(choices=[Animal, Vegetal])
+            thing = GenericLazyReferenceField()
+
+        Animal.drop_collection()
+        Vegetal.drop_collection()
+        Mineral.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal(name="Leopard").save()
+        vegetal = Vegetal(name="Oak").save()
+        mineral = Mineral(name="Granite").save()
+
+        occ_animal = Ocurrence(living_thing=animal, thing=animal).save()
+        _ = Ocurrence(living_thing=vegetal, thing=vegetal).save()
+        with pytest.raises(ValidationError):
+            Ocurrence(living_thing=mineral).save()
+
+        occ = Ocurrence.objects.get(living_thing=animal)
+        assert occ == occ_animal
+        assert isinstance(occ.thing, LazyReference)
+        assert isinstance(occ.living_thing, LazyReference)
+
+        occ.thing = vegetal
+        occ.living_thing = vegetal
+        occ.save()
+
+        occ.thing = mineral
+        occ.living_thing = mineral
+        with pytest.raises(ValidationError):
+            occ.save()
+
+    def test_generic_lazy_reference_set(self):
+        class Animal(Document):
+            meta = {"allow_inheritance": True}
+
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = GenericLazyReferenceField()
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        class SubAnimal(Animal):
+            nick = StringField()
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        sub_animal = SubAnimal(nick="doggo", name="dog").save()
+        for ref in (
+            animal,
+            LazyReference(Animal, animal.pk),
+            {"_cls": "Animal", "_ref": DBRef(animal._get_collection_name(), animal.pk)},
+            sub_animal,
+            LazyReference(SubAnimal, sub_animal.pk),
+            {
+                "_cls": "SubAnimal",
+                "_ref": DBRef(sub_animal._get_collection_name(), sub_animal.pk),
+            },
+        ):
+            p = Ocurrence(person="test", animal=ref).save()
+            p.reload()
+            assert isinstance(p.animal, (LazyReference, Document))
+            p.animal.fetch()
+
+    def test_generic_lazy_reference_bad_set(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = GenericLazyReferenceField(choices=["Animal"])
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        class BadDoc(Document):
+            pass
+
+        animal = Animal(name="Leopard", tag="heavy").save()
+        baddoc = BadDoc().save()
+        for bad in (42, "foo", baddoc, LazyReference(BadDoc, animal.pk)):
+            with pytest.raises(ValidationError):
+                Ocurrence(person="test", animal=bad).save()
+
+    def test_generic_lazy_reference_query_conversion(self):
+        class Member(Document):
+            user_num = IntField(primary_key=True)
+
+        class BlogPost(Document):
+            title = StringField()
+            author = GenericLazyReferenceField()
+
+        Member.drop_collection()
+        BlogPost.drop_collection()
+
+        m1 = Member(user_num=1)
+        m1.save()
+        m2 = Member(user_num=2)
+        m2.save()
+
+        post1 = BlogPost(title="post 1", author=m1)
+        post1.save()
+
+        post2 = BlogPost(title="post 2", author=m2)
+        post2.save()
+
+        post = BlogPost.objects(author=m1).first()
+        assert post.id == post1.id
+
+        post = BlogPost.objects(author=m2).first()
+        assert post.id == post2.id
+
+        # Same thing by passing a LazyReference instance
+        post = BlogPost.objects(author=LazyReference(Member, m2.pk)).first()
+        assert post.id == post2.id
+
+    def test_generic_lazy_reference_not_set(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = GenericLazyReferenceField()
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        Ocurrence(person="foo").save()
+        p = Ocurrence.objects.get()
+        assert p.animal is None
+
+    def test_generic_lazy_reference_accepts_string_instead_of_class(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class Ocurrence(Document):
+            person = StringField()
+            animal = GenericLazyReferenceField("Animal")
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal = Animal().save()
+        Ocurrence(animal=animal).save()
+        p = Ocurrence.objects.get()
+        assert p.animal == animal
+
+    def test_generic_lazy_reference_embedded(self):
+        class Animal(Document):
+            name = StringField()
+            tag = StringField()
+
+        class EmbeddedOcurrence(EmbeddedDocument):
+            in_list = ListField(GenericLazyReferenceField())
+            direct = GenericLazyReferenceField()
+
+        class Ocurrence(Document):
+            in_list = ListField(GenericLazyReferenceField())
+            in_embedded = EmbeddedDocumentField(EmbeddedOcurrence)
+            direct = GenericLazyReferenceField()
+
+        Animal.drop_collection()
+        Ocurrence.drop_collection()
+
+        animal1 = Animal(name="doggo").save()
+        animal2 = Animal(name="cheeta").save()
+
+        def check_fields_type(occ):
+            assert isinstance(occ.direct, LazyReference)
+            for elem in occ.in_list:
+                assert isinstance(elem, LazyReference)
+            assert isinstance(occ.in_embedded.direct, LazyReference)
+            for elem in occ.in_embedded.in_list:
+                assert isinstance(elem, LazyReference)
+
+        occ = Ocurrence(
+            in_list=[animal1, animal2],
+            in_embedded={"in_list": [animal1, animal2], "direct": animal1},
+            direct=animal1,
+        ).save()
+        check_fields_type(occ)
+        occ.reload()
+        check_fields_type(occ)
+        animal1_ref = {
+            "_cls": "Animal",
+            "_ref": DBRef(animal1._get_collection_name(), animal1.pk),
+        }
+        animal2_ref = {
+            "_cls": "Animal",
+            "_ref": DBRef(animal2._get_collection_name(), animal2.pk),
+        }
+        occ.direct = animal1_ref
+        occ.in_list = [animal1_ref, animal2_ref]
+        occ.in_embedded.direct = animal1_ref
+        occ.in_embedded.in_list = [animal1_ref, animal2_ref]
+        check_fields_type(occ)