Merge branch 'master' into 0.8

Do not fail on delete() when blinker is not available

.gitignore

@@ -1,7 +1,17 @@
-*.pyc
+.*
-.*.swp
+!.gitignore
+*~
+*.py[co]
+.*.sw[po]
+*.egg
 docs/.build
 docs/_build
 build/
 dist/
-mongoengine.egg-info/
+mongoengine.egg-info/
+env/
+.settings
+.project
+.pydevproject
+tests/test_bugfix.py
+htmlcov/

.travis.yml

@@ -0,0 +1,26 @@
+# http://travis-ci.org/#!/MongoEngine/mongoengine
+language: python
+services: mongodb
+python:
+    - "2.6"
+    - "2.7"
+    - "3.2"
+    - "3.3"
+env:
+  - PYMONGO=dev
+  - PYMONGO=2.5
+  - PYMONGO=2.4.2
+install:
+    - if [[ $TRAVIS_PYTHON_VERSION == '2.'* ]]; then cp /usr/lib/*/libz.so $VIRTUAL_ENV/lib/; fi
+    - if [[ $TRAVIS_PYTHON_VERSION == '2.'* ]]; then pip install pil --use-mirrors ; true; fi
+    - if [[ $PYMONGO == 'dev' ]]; then pip install https://github.com/mongodb/mongo-python-driver/tarball/master; true; fi
+    - if [[ $PYMONGO != 'dev' ]]; then pip install pymongo==$PYMONGO --use-mirrors; true; fi
+    - python setup.py install
+script:
+    - python setup.py test
+notifications:
+  irc: "irc.freenode.org#mongoengine"
+branches:
+  only:
+    - master
+    - "0.8"

AUTHORS

@@ -1,4 +1,160 @@
+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
+
+Dervived 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:
+
+ * Harry Marr
+ * Ross Lawley
+ * blackbrrr
+ * Florian Schlachter
+ * Vincent Driessen
+ * Steve Challis
+ * flosch
+ * Deepak Thukral
+ * Colin Howe
+ * Wilson Júnior
+ * 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
+ * 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
+ * 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
+ * lraucy
+ * hellysmile
+ * Jaepil Jeong
+ * Daniil Sharou
+ * Stefan Wójcik
+ * Pete Campton
+ * Martyn Smith
+ * Marcelo Anton
+ * Aleksey Porfirov
+ * 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
+

CONTRIBUTING.rst

@@ -0,0 +1,61 @@
+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
+----------------------
+
+PyMongo supports CPython 2.5 and newer. Language
+features not supported by all interpreters can not be used.
+Please also ensure that your code is properly converted by
+`2to3 <http://docs.python.org/library/2to3.html>`_ for Python 3 support.
+
+Style Guide
+-----------
+
+MongoEngine aims to follow `PEP8 <http://www.python.org/dev/peps/pep-0008/>`_
+including 4 space indents and 79 character line limits.
+
+Testing
+-------
+
+All tests are run on `Travis <http://travis-ci.org/MongoEngine/mongoengine>`_
+and any pull requests are automatically tested by Travis. Any pull requests
+without tests will take longer to be integrated and might be refused.
+
+General Guidelines
+------------------
+
+- Avoid backward breaking changes if at all possible.
+- 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).
+- Add yourself to AUTHORS.rst :)
+
+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.

LICENSE

@@ -1,5 +1,5 @@
-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
 files (the "Software"), to deal in the Software without
@@ -8,10 +8,10 @@ copy, modify, merge, publish, distribute, sublicense, and/or sell
 copies of the Software, and to permit persons to whom the
 Software is furnished to do so, subject to the following
 conditions:
- 
+
 The above copyright notice and this permission notice shall be
 included in all copies or substantial portions of the Software.
- 
+
 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
 OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND

README.rst

@@ -2,26 +2,31 @@
 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: Ross Lawley (http://github.com/rozza)
+
+.. image:: https://secure.travis-ci.org/MongoEngine/mongoengine.png?branch=master
+  :target: http://travis-ci.org/MongoEngine/mongoengine
 
 About
 =====
-MongoEngine is a Python Object-Document Mapper for working with MongoDB. 
+MongoEngine is a Python Object-Document Mapper for working with MongoDB.
-Documentation available at http://hmarr.com/mongoengine/ - there is currently 
+Documentation available at http://mongoengine-odm.rtfd.org - there is currently
-a `tutorial <http://hmarr.com/mongoengine/tutorial.html>`_, a `user guide 
+a `tutorial <http://readthedocs.org/docs/mongoengine-odm/en/latest/tutorial.html>`_, a `user guide
-<http://hmarr.com/mongoengine/userguide.html>`_ and an `API reference
+<https://mongoengine-odm.readthedocs.org/en/latest/guide/index.html>`_ and an `API reference
-<http://hmarr.com/mongoengine/apireference.html>`_.
+<http://readthedocs.org/docs/mongoengine-odm/en/latest/apireference.html>`_.
 
 Installation
 ============
 If you have `setuptools <http://peak.telecommunity.com/DevCenter/setuptools>`_
 you can use ``easy_install -U mongoengine``. Otherwise, you can download the
-source from `GitHub <http://github.com/hmarr/mongoengine>`_ and run ``python
+source from `GitHub <http://github.com/MongoEngine/mongoengine>`_ and run ``python
 setup.py install``.
 
 Dependencies
 ============
-- pymongo 1.1+
+- pymongo 2.1.1+
 - sphinx (optional - for documentation generation)
 
 Examples
@@ -58,11 +63,6 @@ Some simple examples of what MongoEngine code looks like::
     ...         print 'Link:', post.url
     ...     print
     ...
-    === Using MongoEngine ===
-    See the tutorial
-
-    === MongoEngine Docs ===
-    Link: hmarr.com/mongoengine
 
     >>> len(BlogPost.objects)
     2
@@ -80,18 +80,16 @@ Some simple examples of what MongoEngine code looks like::
 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 run: ``python setup.py test``.
 
 Community
 =========
-- `MongoEngine Users mailing list 
+- `MongoEngine Users mailing list
   <http://groups.google.com/group/mongoengine-users>`_
-- `MongoEngine Developers mailing list 
+- `MongoEngine Developers mailing list
   <http://groups.google.com/group/mongoengine-dev>`_
-- `#mongoengine IRC channel <irc://irc.freenode.net/mongoengine>`_
+- `#mongoengine IRC channel <http://webchat.freenode.net/?channels=mongoengine>`_
 
 Contributing
 ============
-The source is available on `GitHub <http://github.com/hmarr/mongoengine>`_ - to
+We welcome contributions! see  the`Contribution guidelines <https://github.com/MongoEngine/mongoengine/blob/master/CONTRIBUTING.rst>`_
-contribute to the project, fork it on GitHub and send a pull request, all
-contributions and suggestions are welcome!

benchmark.py

@@ -0,0 +1,199 @@
+#!/usr/bin/env python
+
+import timeit
+
+
+def cprofile_main():
+    from pymongo import Connection
+    connection = Connection()
+    connection.drop_database('timeit_test')
+    connection.disconnect()
+
+    from mongoengine import Document, DictField, connect
+    connect("timeit_test")
+
+    class Noddy(Document):
+        fields = DictField()
+
+    for i in xrange(1):
+        noddy = Noddy()
+        for j in range(20):
+            noddy.fields["key" + str(j)] = "value " + str(j)
+        noddy.save()
+
+
+def main():
+    """
+    0.4 Performance Figures ...
+
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - Pymongo
+    3.86744189262
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine
+    6.23374891281
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False
+    5.33027005196
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False, cascade=False
+    pass - No Cascade
+
+    0.5.X
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - Pymongo
+    3.89597702026
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine
+    21.7735359669
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False
+    19.8670389652
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False, cascade=False
+    pass - No Cascade
+
+    0.6.X
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - Pymongo
+    3.81559205055
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine
+    10.0446798801
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False
+    9.51354718208
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False, cascade=False
+    9.02567505836
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, force=True
+    8.44933390617
+
+    0.7.X
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - Pymongo
+    3.78801012039
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine
+    9.73050498962
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False
+    8.33456707001
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, safe=False, validate=False, cascade=False
+    8.37778115273
+    ----------------------------------------------------------------------------------------------------
+    Creating 10000 dictionaries - MongoEngine, force=True
+    8.36906409264
+    """
+
+    setup = """
+from pymongo import Connection
+connection = Connection()
+connection.drop_database('timeit_test')
+"""
+
+    stmt = """
+from pymongo import Connection
+connection = Connection()
+
+db = connection.timeit_test
+noddy = db.noddy
+
+for i in xrange(10000):
+    example = {'fields': {}}
+    for j in range(20):
+        example['fields']["key"+str(j)] = "value "+str(j)
+
+    noddy.insert(example)
+
+myNoddys = noddy.find()
+[n for n in myNoddys] # iterate
+"""
+
+    print "-" * 100
+    print """Creating 10000 dictionaries - Pymongo"""
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print t.timeit(1)
+
+    setup = """
+from pymongo import Connection
+connection = Connection()
+connection.drop_database('timeit_test')
+connection.disconnect()
+
+from mongoengine import Document, DictField, connect
+connect("timeit_test")
+
+class Noddy(Document):
+    fields = DictField()
+"""
+
+    stmt = """
+for i in xrange(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 """Creating 10000 dictionaries - MongoEngine"""
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print t.timeit(1)
+
+    stmt = """
+for i in xrange(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(safe=False, validate=False)
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print "-" * 100
+    print """Creating 10000 dictionaries - MongoEngine, safe=False, validate=False"""
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print t.timeit(1)
+
+
+    stmt = """
+for i in xrange(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(safe=False, validate=False, cascade=False)
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print "-" * 100
+    print """Creating 10000 dictionaries - MongoEngine, safe=False, validate=False, cascade=False"""
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print t.timeit(1)
+
+    stmt = """
+for i in xrange(10000):
+    noddy = Noddy()
+    for j in range(20):
+        noddy.fields["key"+str(j)] = "value "+str(j)
+    noddy.save(force_insert=True, safe=False, validate=False, cascade=False)
+
+myNoddys = Noddy.objects()
+[n for n in myNoddys] # iterate
+"""
+
+    print "-" * 100
+    print """Creating 10000 dictionaries - MongoEngine, force=True"""
+    t = timeit.Timer(stmt=stmt, setup=setup)
+    print t.timeit(1)
+
+if __name__ == "__main__":
+    main()

docs/apireference.rst

@@ -6,6 +6,7 @@ Connecting
 ==========
 
 .. autofunction:: mongoengine.connect
+.. autofunction:: mongoengine.register_connection
 
 Documents
 =========
@@ -15,15 +16,31 @@ Documents
 
    .. attribute:: objects
 
-      A :class:`~mongoengine.queryset.QuerySet` object that is created lazily 
+      A :class:`~mongoengine.queryset.QuerySet` object that is created lazily
       on access.
 
 .. autoclass:: mongoengine.EmbeddedDocument
    :members:
-   
+
+.. autoclass:: mongoengine.DynamicDocument
+   :members:
+
+.. autoclass:: mongoengine.DynamicEmbeddedDocument
+   :members:
+
 .. autoclass:: mongoengine.document.MapReduceDocument
   :members:
 
+.. autoclass:: mongoengine.ValidationError
+  :members:
+
+Context Managers
+================
+
+.. autoclass:: mongoengine.context_managers.switch_db
+.. autoclass:: mongoengine.context_managers.no_dereference
+.. autoclass:: mongoengine.context_managers.query_counter
+
 Querying
 ========
 
@@ -31,36 +48,34 @@ Querying
    :members:
 
    .. automethod:: mongoengine.queryset.QuerySet.__call__
-   
+
 .. autofunction:: mongoengine.queryset.queryset_manager
 
 Fields
 ======
 
-.. autoclass:: mongoengine.StringField
-
-.. autoclass:: mongoengine.URLField
-
-.. autoclass:: mongoengine.IntField
-
-.. autoclass:: mongoengine.FloatField
-
-.. autoclass:: mongoengine.DecimalField
-
-.. autoclass:: mongoengine.BooleanField
-
-.. autoclass:: mongoengine.DateTimeField
-
-.. autoclass:: mongoengine.EmbeddedDocumentField
-
-.. autoclass:: mongoengine.DictField
-
-.. autoclass:: mongoengine.ListField
-
 .. autoclass:: mongoengine.BinaryField
-
+.. autoclass:: mongoengine.BooleanField
-.. autoclass:: mongoengine.ObjectIdField
+.. autoclass:: mongoengine.ComplexDateTimeField
-
+.. autoclass:: mongoengine.DateTimeField
-.. autoclass:: mongoengine.ReferenceField
+.. autoclass:: mongoengine.DecimalField
-
+.. autoclass:: mongoengine.DictField
+.. autoclass:: mongoengine.DynamicField
+.. autoclass:: mongoengine.EmailField
+.. autoclass:: mongoengine.EmbeddedDocumentField
+.. autoclass:: mongoengine.FileField
+.. autoclass:: mongoengine.FloatField
+.. autoclass:: mongoengine.GenericEmbeddedDocumentField
 .. autoclass:: mongoengine.GenericReferenceField
+.. autoclass:: mongoengine.GeoPointField
+.. autoclass:: mongoengine.ImageField
+.. autoclass:: mongoengine.IntField
+.. autoclass:: mongoengine.ListField
+.. autoclass:: mongoengine.MapField
+.. autoclass:: mongoengine.ObjectIdField
+.. autoclass:: mongoengine.ReferenceField
+.. autoclass:: mongoengine.SequenceField
+.. autoclass:: mongoengine.SortedListField
+.. autoclass:: mongoengine.StringField
+.. autoclass:: mongoengine.URLField
+.. autoclass:: mongoengine.UUIDField

docs/changelog.rst

@@ -2,11 +2,450 @@
 Changelog
 =========
 
+Changes in 0.8.X
+================
+- Fixed db_alias and inherited Documents (#143)
+- Documentation update for document errors (#124)
+- Deprecated `get_or_create` (#35)
+- Updated inheritable objects created by upsert now contain _cls (#118)
+- Added support for creating documents with embedded documents in a single operation (#6)
+- Added to_json and from_json to Document (#1)
+- Added to_json and from_json to QuerySet (#131)
+- Updated index creation now tied to Document class (#102)
+- Added none() to queryset (#127)
+- Updated SequenceFields to allow post processing of the calculated counter value (#141)
+- Added clean method to documents for pre validation data cleaning (#60)
+- Added support setting for read prefrence at a query level (#157)
+- Added _instance to EmbeddedDocuments pointing to the parent (#139)
+- Inheritance is off by default (#122)
+- Remove _types and just use _cls for inheritance (#148)
+- Only allow QNode instances to be passed as query objects (#199)
+- Dynamic fields are now validated on save (#153) (#154)
+- Added support for multiple slices and made slicing chainable. (#170) (#190) (#191)
+- Fixed GridFSProxy __getattr__ behaviour (#196)
+- Fix Django timezone support (#151)
+- Simplified Q objects, removed QueryTreeTransformerVisitor (#98) (#171)
+- FileFields now copyable (#198)
+- Querysets now return clones and are no longer edit in place (#56)
+- Added support for $maxDistance (#179)
+- Uses getlasterror to test created on updated saves (#163)
+- Fixed inheritance and unique index creation (#140)
+- Fixed reverse delete rule with inheritance (#197)
+- Fixed validation for GenericReferences which havent been dereferenced
+- Added switch_db context manager (#106)
+- Added switch_db method to document instances (#106)
+- Added no_dereference context manager (#82) (#61)
+- Added switch_collection context manager (#220)
+- Added switch_collection method to document instances (#220)
+- Added support for compound primary keys (#149) (#121)
+- Fixed overriding objects with custom manager (#58)
+- Added no_dereference method for querysets (#82) (#61)
+- Undefined data should not override instance methods (#49)
+- Added Django Group and Permission (#142)
+- Added Doc class and pk to Validation messages (#69)
+- Fixed Documents deleted via a queryset don't call any signals (#105)
+- Added the "get_decoded" method to the MongoSession class (#216)
+- Fixed invalid choices error bubbling (#214)
+- Updated Save so it calls $set and $unset in a single operation (#211)
+- Fixed inner queryset looping (#204)
+
+Changes in 0.7.10
+=================
+- Fix UnicodeEncodeError for dbref (#278)
+- Allow construction using positional parameters (#268)
+- Updated EmailField length to support long domains (#243)
+- Added 64-bit integer support (#251)
+- Added Django sessions TTL support (#224)
+- Fixed issue with numerical keys in MapField(EmbeddedDocumentField()) (#240)
+- Fixed clearing _changed_fields for complex nested embedded documents (#237, #239, #242)
+- Added "id" back to _data dictionary (#255)
+- Only mark a field as changed if the value has changed (#258)
+- Explicitly check for Document instances when dereferencing (#261)
+- Fixed order_by chaining issue (#265)
+- Added dereference support for tuples (#250)
+- Resolve field name to db field name when using distinct(#260, #264, #269)
+- Added kwargs to doc.save to help interop with django (#223, #270)
+- Fixed cloning querysets in PY3
+- Int fields no longer unset in save when changed to 0 (#272)
+- Fixed ReferenceField query chaining bug fixed (#254)
+
+Changes in 0.7.9
+================
+- Better fix handling for old style _types
+- Embedded SequenceFields follow collection naming convention
+
+Changes in 0.7.8
+================
+- Fix sequence fields in embedded documents (#166)
+- Fix query chaining with .order_by() (#176)
+- Added optional encoding and collection config for Django sessions (#180, #181, #183)
+- Fixed EmailField so can add extra validation (#173, #174, #187)
+- Fixed bulk inserts can now handle custom pk's (#192)
+- Added as_pymongo method to return raw or cast results from pymongo (#193)
+
+Changes in 0.7.7
+================
+- Fix handling for old style _types
+
+Changes in 0.7.6
+================
+- Unicode fix for repr (#133)
+- Allow updates with match operators (#144)
+- Updated URLField - now can have a override the regex (#136)
+- Allow Django AuthenticationBackends to work with Django user (hmarr/mongoengine#573)
+- Fixed reload issue with ReferenceField where dbref=False (#138)
+
+Changes in 0.7.5
+================
+- ReferenceFields with dbref=False use ObjectId instead of strings (#134)
+  See ticket for upgrade notes (#134)
+
+Changes in 0.7.4
+================
+- Fixed index inheritance issues - firmed up testcases (#123) (#125)
+
+Changes in 0.7.3
+================
+- Reverted EmbeddedDocuments meta handling - now can turn off inheritance (#119)
+
+Changes in 0.7.2
+================
+- Update index spec generation so its not destructive (#113)
+
+Changes in 0.7.1
+=================
+- Fixed index spec inheritance (#111)
+
+Changes in 0.7.0
+=================
+- Updated queryset.delete so you can use with skip / limit (#107)
+- Updated index creation allows kwargs to be passed through refs (#104)
+- Fixed Q object merge edge case (#109)
+- Fixed reloading on sharded documents (hmarr/mongoengine#569)
+- Added NotUniqueError for duplicate keys (#62)
+- Added custom collection / sequence naming for SequenceFields (#92)
+- Fixed UnboundLocalError in composite index with pk field (#88)
+- Updated ReferenceField's to optionally store ObjectId strings
+  this will become the default in 0.8 (#89)
+- Added FutureWarning - save will default to `cascade=False` in 0.8
+- Added example of indexing embedded document fields (#75)
+- Fixed ImageField resizing when forcing size (#80)
+- Add flexibility for fields handling bad data (#78)
+- Embedded Documents no longer handle meta definitions
+- Use weakref proxies in base lists / dicts (#74)
+- Improved queryset filtering (hmarr/mongoengine#554)
+- Fixed Dynamic Documents and Embedded Documents (hmarr/mongoengine#561)
+- Fixed abstract classes and shard keys (#64)
+- Fixed Python 2.5 support
+- Added Python 3 support (thanks to Laine Heron)
+
+Changes in 0.6.20
+=================
+- Added support for distinct and db_alias (#59)
+- Improved support for chained querysets when constraining the same fields (hmarr/mongoengine#554)
+- Fixed BinaryField lookup re (#48)
+
+Changes in 0.6.19
+=================
+
+- Added Binary support to UUID (#47)
+- Fixed MapField lookup for fields without declared lookups (#46)
+- Fixed BinaryField python value issue (#48)
+- Fixed SequenceField non numeric value lookup (#41)
+- Fixed queryset manager issue (#52)
+- Fixed FileField comparision (hmarr/mongoengine#547)
+
+Changes in 0.6.18
+=================
+- Fixed recursion loading bug in _get_changed_fields
+
+Changes in 0.6.17
+=================
+- Fixed issue with custom queryset manager expecting explict variable names
+
+Changes in 0.6.16
+=================
+- Fixed issue where db_alias wasn't inherited
+
+Changes in 0.6.15
+=================
+- Updated validation error messages
+- Added support for null / zero / false values in item_frequencies
+- Fixed cascade save edge case
+- Fixed geo index creation through reference fields
+- Added support for args / kwargs when using @queryset_manager
+- Deref list custom id fix
+
+Changes in 0.6.14
+=================
+- Fixed error dict with nested validation
+- Fixed Int/Float fields and not equals None
+- Exclude tests from installation
+- Allow tuples for index meta
+- Fixed use of str in instance checks
+- Fixed unicode support in transform update
+- Added support for add_to_set and each
+
+Changes in 0.6.13
+=================
+- Fixed EmbeddedDocument db_field validation issue
+- Fixed StringField unicode issue
+- Fixes __repr__ modifying the cursor
+
+Changes in 0.6.12
+=================
+- Fixes scalar lookups for primary_key
+- Fixes error with _delta handling DBRefs
+
+Changes in 0.6.11
+==================
+- Fixed inconsistency handling None values field attrs
+- Fixed map_field embedded db_field issue
+- Fixed .save() _delta issue with DbRefs
+- Fixed Django TestCase
+- Added cmp to Embedded Document
+- Added PULL reverse_delete_rule
+- Fixed CASCADE delete bug
+- Fixed db_field data load error
+- Fixed recursive save with FileField
+
+Changes in 0.6.10
+=================
+- Fixed basedict / baselist to return super(..)
+- Promoted BaseDynamicField to DynamicField
+
+Changes in 0.6.9
+================
+- Fixed sparse indexes on inherited docs
+- Removed FileField auto deletion, needs more work maybe 0.7
+
+Changes in 0.6.8
+================
+- Fixed FileField losing reference when no default set
+- Removed possible race condition from FileField (grid_file)
+- Added assignment to save, can now do: `b = MyDoc(**kwargs).save()`
+- Added support for pull operations on nested EmbeddedDocuments
+- Added support for choices with GenericReferenceFields
+- Added support for choices with GenericEmbeddedDocumentFields
+- Fixed Django 1.4 sessions first save data loss
+- FileField now automatically delete files on .delete()
+- Fix for GenericReference to_mongo method
+- Fixed connection regression
+- Updated Django User document, now allows inheritance
+
+Changes in 0.6.7
+================
+- Fixed indexing on '_id' or 'pk' or 'id'
+- Invalid data from the DB now raises a InvalidDocumentError
+- Cleaned up the Validation Error - docs and code
+- Added meta `auto_create_index` so you can disable index creation
+- Added write concern options to inserts
+- Fixed typo in meta for index options
+- Bug fix Read preference now passed correctly
+- Added support for File like objects for GridFS
+- Fix for #473 - Dereferencing abstracts
+
+Changes in 0.6.6
+================
+- Django 1.4 fixed (finally)
+- Added tests for Django
+
+Changes in 0.6.5
+================
+- More Django updates
+
+Changes in 0.6.4
+================
+
+- Refactored connection / fixed replicasetconnection
+- Bug fix for unknown connection alias error message
+- Sessions support Django 1.3 and Django 1.4
+- Minor fix for ReferenceField
+
+Changes in 0.6.3
+================
+- Updated sessions for Django 1.4
+- Bug fix for updates where listfields contain embedded documents
+- Bug fix for collection naming and mixins
+
+Changes in 0.6.2
+================
+- Updated documentation for ReplicaSet connections
+- Hack round _types issue with SERVER-5247 - querying other arrays may also cause problems.
+
+Changes in 0.6.1
+================
+- Fix for replicaSet connections
+
+Changes in 0.6
+================
+
+- Added FutureWarning to inherited classes not declaring 'allow_inheritance' as the default will change in 0.7
+- Added support for covered indexes when inheritance is off
+- No longer always upsert on save for items with a '_id'
+- Error raised if update doesn't have an operation
+- DeReferencing is now thread safe
+- Errors raised if trying to perform a join in a query
+- Updates can now take __raw__ queries
+- Added custom 2D index declarations
+- Added replicaSet connection support
+- Updated deprecated imports from pymongo (safe for pymongo 2.2)
+- Added uri support for connections
+- Added scalar for efficiently returning partial data values (aliased to values_list)
+- Fixed limit skip bug
+- Improved Inheritance / Mixin
+- Added sharding support
+- Added pymongo 2.1 support
+- Fixed Abstract documents can now declare indexes
+- Added db_alias support to individual documents
+- Fixed GridFS documents can now be pickled
+- Added Now raises an InvalidDocumentError when declaring multiple fields with the same db_field
+- Added InvalidQueryError when calling with_id with a filter
+- Added support for DBRefs in distinct()
+- Fixed issue saving False booleans
+- Fixed issue with dynamic documents deltas
+- Added Reverse Delete Rule support to ListFields - MapFields aren't supported
+- Added customisable cascade kwarg options
+- Fixed Handle None values for non-required fields
+- Removed Document._get_subclasses() - no longer required
+- Fixed bug requiring subclasses when not actually needed
+- Fixed deletion of dynamic data
+- Added support for the $elementMatch operator
+- Added reverse option to SortedListFields
+- Fixed dereferencing - multi directional list dereferencing
+- Fixed issue creating indexes with recursive embedded documents
+- Fixed recursive lookup in _unique_with_indexes
+- Fixed passing ComplexField defaults to constructor for ReferenceFields
+- Fixed validation of DictField Int keys
+- Added optional cascade saving
+- Fixed dereferencing - max_depth now taken into account
+- Fixed document mutation saving issue
+- Fixed positional operator when replacing embedded documents
+- Added Non-Django Style choices back (you can have either)
+- Fixed __repr__ of a sliced queryset
+- Added recursive validation error of documents / complex fields
+- Fixed breaking during queryset iteration
+- Added pre and post bulk-insert signals
+- Added ImageField - requires PIL
+- Fixed Reference Fields can be None in get_or_create / queries
+- Fixed accessing pk on an embedded document
+- Fixed calling a queryset after drop_collection now recreates the collection
+- Add field name to validation exception messages
+- Added UUID field
+- Improved efficiency of .get()
+- Updated ComplexFields so if required they won't accept empty lists / dicts
+- Added spec file for rpm-based distributions
+- Fixed ListField so it doesnt accept strings
+- Added DynamicDocument and EmbeddedDynamicDocument classes for expando schemas
+
+Changes in v0.5.2
+=================
+
+- A Robust Circular reference bugfix
+
+
+Changes in v0.5.1
+=================
+
+- Fixed simple circular reference bug
+
+Changes in v0.5
+===============
+
+- Added InvalidDocumentError - so Document core methods can't be overwritten
+- Added GenericEmbeddedDocument - so you can embed any type of embeddable document
+- Added within_polygon support - for those with mongodb 1.9
+- Updated sum / average to use map_reduce as db.eval doesn't work in sharded environments
+- Added where() - filter to allowing users to specify query expressions as Javascript
+- Added SequenceField - for creating sequential counters
+- Added update() convenience method to a document
+- Added cascading saves - so changes to Referenced documents are saved on .save()
+- Added select_related() support
+- Added support for the positional operator
+- Updated geo index checking to be recursive and check in embedded documents
+- Updated default collection naming convention
+- Added Document Mixin support
+- Fixed queryet __repr__ mid iteration
+- Added hint() support, so cantell Mongo the proper index to use for the query
+- Fixed issue with inconsitent setting of _cls breaking inherited referencing
+- Added help_text and verbose_name to fields to help with some form libs
+- Updated item_frequencies to handle embedded document lookups
+- Added delta tracking now only sets / unsets explicitly changed fields
+- Fixed saving so sets updated values rather than overwrites
+- Added ComplexDateTimeField - Handles datetimes correctly with microseconds
+- Added ComplexBaseField - for improved flexibility and performance
+- Added get_FIELD_display() method for easy choice field displaying
+- Added queryset.slave_okay(enabled) method
+- Updated queryset.timeout(enabled) and queryset.snapshot(enabled) to be chainable
+- Added insert method for bulk inserts
+- Added blinker signal support
+- Added query_counter context manager for tests
+- Added map_reduce method item_frequencies and set as default (as db.eval doesn't work in sharded environments)
+- Added inline_map_reduce option to map_reduce
+- Updated connection exception so it provides more info on the cause.
+- Added searching multiple levels deep in ``DictField``
+- Added ``DictField`` entries containing strings to use matching operators
+- Added ``MapField``, similar to ``DictField``
+- Added Abstract Base Classes
+- Added Custom Objects Managers
+- Added sliced subfields updating
+- Added ``NotRegistered`` exception if dereferencing ``Document`` not in the registry
+- Added a write concern for ``save``, ``update``, ``update_one`` and ``get_or_create``
+- Added slicing / subarray fetching controls
+- Fixed various unique index and other index issues
+- Fixed threaded connection issues
+- Added spherical geospatial query operators
+- Updated queryset to handle latest version of pymongo
+  map_reduce now requires an output.
+- Added ``Document`` __hash__, __ne__ for pickling
+- Added ``FileField`` optional size arg for read method
+- Fixed ``FileField`` seek and tell methods for reading files
+- Added ``QuerySet.clone`` to support copying querysets
+- Fixed item_frequencies when using name thats the same as a native js function
+- Added reverse delete rules
+- Fixed issue with unset operation
+- Fixed Q-object bug
+- Added ``QuerySet.all_fields`` resets previous .only() and .exclude()
+- Added ``QuerySet.exclude``
+- Added django style choices
+- Fixed order and filter issue
+- Added ``QuerySet.only`` subfield support
+- Added creation_counter to ``BaseField`` allowing fields to be sorted in the
+  way the user has specified them
+- Fixed various errors
+- Added many tests
+
+Changes in v0.4
+===============
+- Added ``GridFSStorage`` Django storage backend
+- Added ``FileField`` for GridFS support
+- New Q-object implementation, which is no longer based on Javascript
+- Added ``SortedListField``
+- Added ``EmailField``
+- Added ``GeoPointField``
+- Added ``exact`` and ``iexact`` match operators to ``QuerySet``
+- Added ``get_document_or_404`` and ``get_list_or_404`` Django shortcuts
+- Added new query operators for Geo queries
+- Added ``not`` query operator
+- Added new update operators: ``pop`` and ``add_to_set``
+- Added ``__raw__`` query parameter
+- Added support for custom querysets
+- Fixed document inheritance primary key issue
+- Added support for querying by array element position
+- Base class can now be defined for ``DictField``
+- Fixed MRO error that occured on document inheritance
+- Added ``QuerySet.distinct``, ``QuerySet.create``, ``QuerySet.snapshot``,
+  ``QuerySet.timeout`` and ``QuerySet.all``
+- Subsequent calls to ``connect()`` now work
+- Introduced ``min_length`` for ``StringField``
+- Fixed multi-process connection issue
+- Other minor fixes
+
 Changes in v0.3
 ===============
 - Added MapReduce support
 - Added ``contains``, ``startswith`` and ``endswith`` query operators (and
-  case-insensitive versions that are prefixed with 'i') 
+  case-insensitive versions that are prefixed with 'i')
 - Deprecated fields' ``name`` parameter, replaced with ``db_field``
 - Added ``QuerySet.only`` for only retrieving specific fields
 - Added ``QuerySet.in_bulk()`` for bulk querying using ids
@@ -53,7 +492,7 @@ Changes in v0.2
 ===============
 - Added ``Q`` class for building advanced queries
 - Added ``QuerySet`` methods for atomic updates to documents
-- Fields may now specify ``unique=True`` to enforce uniqueness across a 
+- Fields may now specify ``unique=True`` to enforce uniqueness across a
   collection
 - Added option for default document ordering
 - Fixed bug in index definitions
@@ -61,7 +500,7 @@ Changes in v0.2
 Changes in v0.1.3
 =================
 - Added Django authentication backend
-- Added ``Document.meta`` support for indexes, which are ensured just before 
+- Added ``Document.meta`` support for indexes, which are ensured just before
   querying takes place
 - A few minor bugfixes
 

docs/conf.py

@@ -16,7 +16,7 @@ import sys, os
 # 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 -----------------------------------------------------
 
@@ -38,7 +38,7 @@ master_doc = 'index'
 
 # General information about the project.
 project = u'MongoEngine'
-copyright = u'2009-2010, Harry Marr'
+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
@@ -121,7 +121,7 @@ html_theme_path = ['_themes']
 # 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.

docs/django.rst

@@ -2,29 +2,41 @@
 Using MongoEngine with Django
 =============================
 
+.. note:: Updated to support Django 1.4
+
 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 
+also plan to use the ORM in your project), and instead call
 :func:`~mongoengine.connect` somewhere in the settings module.
 
+.. note ::
+   If you are not using another Database backend you may need to add a dummy
+   database backend to ``settings.py`` eg::
+
+        DATABASES = {
+            'default': {
+                'ENGINE': 'django.db.backends.dummy'
+            }
+        }
+
 Authentication
 ==============
 MongoEngine includes a Django authentication backend, which uses MongoDB. The
-:class:`~mongoengine.django.auth.User` model is a MongoEngine 
+:class:`~mongoengine.django.auth.User` model is a MongoEngine
-:class:`~mongoengine.Document`, but implements most of the methods and 
+: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 
+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::
-    
+
     AUTHENTICATION_BACKENDS = (
         'mongoengine.django.auth.MongoEngineBackend',
     )
 
-The :mod:`~mongoengine.django.auth` module also contains a 
+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.
 
@@ -43,4 +55,49 @@ into you settings module::
 
     SESSION_ENGINE = 'mongoengine.django.sessions'
 
+Django provides session cookie, which expires after ```SESSION_COOKIE_AGE``` seconds, but doesnt delete cookie at sessions backend, so ``'mongoengine.django.sessions'`` supports  `mongodb TTL
+<http://docs.mongodb.org/manual/tutorial/expire-data/>`_.
+
 .. versionadded:: 0.2.1
+
+Storage
+=======
+With MongoEngine's support for GridFS via the :class:`~mongoengine.FileField`,
+it is useful to have a Django file storage backend that wraps this. The new
+storage module is called :class:`~mongoengine.django.storage.GridFSStorage`.
+Using it is very similar to using the default FileSystemStorage.::
+
+    from mongoengine.django.storage import GridFSStorage
+    fs = GridFSStorage()
+
+    filename = fs.save('hello.txt', 'Hello, World!')
+
+All of the `Django Storage API methods
+<http://docs.djangoproject.com/en/dev/ref/files/storage/>`_ have been
+implemented except :func:`path`. If the filename provided already exists, an
+underscore and a number (before # the file extension, if one exists) will be
+appended to the filename until the generated filename doesn't exist. The
+:func:`save` method will return the new filename.::
+
+    >>> fs.exists('hello.txt')
+    True
+    >>> fs.open('hello.txt').read()
+    'Hello, World!'
+    >>> fs.size('hello.txt')
+    13
+    >>> fs.url('hello.txt')
+    'http://your_media_url/hello.txt'
+    >>> fs.open('hello.txt').name
+    'hello.txt'
+    >>> fs.listdir()
+    ([], [u'hello.txt'])
+
+All files will be saved and retrieved in GridFS via the :class::`FileDocument`
+document, allowing easy access to the files without the GridFSStorage
+backend.::
+
+    >>> from mongoengine.django.storage import FileDocument
+    >>> FileDocument.objects()
+    [<FileDocument: FileDocument object>]
+
+.. versionadded:: 0.4

docs/guide/connecting.rst

@@ -3,6 +3,7 @@
 =====================
 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
@@ -18,3 +19,71 @@ provide :attr:`host` and :attr:`port` arguments to
 :func:`~mongoengine.connect`::
 
     connect('project1', host='192.168.1.35', port=12345)
+
+Uri style connections are also supported as long as you include the database
+name - just supply the uri as the :attr:`host` to
+:func:`~mongoengine.connect`::
+
+    connect('project1', host='mongodb://localhost/database_name')
+
+ReplicaSets
+===========
+
+MongoEngine now supports :func:`~pymongo.replica_set_connection.ReplicaSetConnection`
+to use them please use a URI style connection and provide the `replicaSet` name in the
+connection kwargs.
+
+Read preferences are supported throught 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
+==================
+
+Multiple database support was added in MongoEngine 0.6. 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.
+
+Individual documents can also support multiple 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::
+
+        class User(Document):
+            name = StringField()
+
+            meta = {"db_alias": "user-db"}
+
+        class Book(Document):
+            name = StringField()
+
+            meta = {"db_alias": "book-db"}
+
+        class AuthorBooks(Document):
+            author = ReferenceField(User)
+            book = ReferenceField(Book)
+
+            meta = {"db_alias": "users-books-db"}
+
+
+Switch Database Context Manager
+===============================
+
+Sometimes you might want to switch the database to query against for a class.
+The :class:`~mongoengine.context_managers.switch_db` context manager allows
+you to change the database alias for a class eg ::
+
+        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'

docs/guide/defining-documents.rst

@@ -4,14 +4,14 @@ 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 principle difference is that no schema
-is enforced at a database level. 
+is enforced at a database level.
 
 Defining a document's schema
 ============================
 MongoEngine allows you to define schemata for documents as this helps to reduce
 coding errors, and allows for utility methods to be defined on fields which may
-be present. 
+be present.
 
 To define a schema for a document, create a class that inherits from
 :class:`~mongoengine.Document`. Fields are specified by adding **field
@@ -19,10 +19,38 @@ objects** as class attributes to the document class::
 
     from mongoengine import *
     import datetime
-    
+
     class Page(Document):
         title = StringField(max_length=200, required=True)
-        date_modified = DateTimeField(default=datetime.now)
+        date_modified = DateTimeField(default=datetime.datetime.now)
+
+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 `_`
+
 
 Fields
 ======
@@ -31,21 +59,115 @@ By default, fields are not required. To make a field mandatory, set the
 validation constraints available (such as :attr:`max_length` in the example
 above). Fields may also take default values, which will be used if a value is
 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 
+to retrieve the value (such as in the above example). The field types available
 are as follows:
 
+* :class:`~mongoengine.BinaryField`
+* :class:`~mongoengine.BooleanField`
+* :class:`~mongoengine.ComplexDateTimeField`
+* :class:`~mongoengine.DateTimeField`
+* :class:`~mongoengine.DecimalField`
+* :class:`~mongoengine.DictField`
+* :class:`~mongoengine.DynamicField`
+* :class:`~mongoengine.EmailField`
+* :class:`~mongoengine.EmbeddedDocumentField`
+* :class:`~mongoengine.FileField`
+* :class:`~mongoengine.FloatField`
+* :class:`~mongoengine.GenericEmbeddedDocumentField`
+* :class:`~mongoengine.GenericReferenceField`
+* :class:`~mongoengine.GeoPointField`
+* :class:`~mongoengine.ImageField`
+* :class:`~mongoengine.IntField`
+* :class:`~mongoengine.ListField`
+* :class:`~mongoengine.MapField`
+* :class:`~mongoengine.ObjectIdField`
+* :class:`~mongoengine.ReferenceField`
+* :class:`~mongoengine.SequenceField`
+* :class:`~mongoengine.SortedListField`
 * :class:`~mongoengine.StringField`
 * :class:`~mongoengine.URLField`
-* :class:`~mongoengine.IntField`
+* :class:`~mongoengine.UUIDField`
-* :class:`~mongoengine.FloatField`
+
-* :class:`~mongoengine.DecimalField`
+Field arguments
-* :class:`~mongoengine.DateTimeField`
+---------------
-* :class:`~mongoengine.ListField`
+Each field type can be customized by keyword arguments.  The following keyword
-* :class:`~mongoengine.DictField`
+arguments can be set on all fields:
-* :class:`~mongoengine.ObjectIdField`
+
-* :class:`~mongoengine.EmbeddedDocumentField`
+:attr:`db_field` (Default: None)
-* :class:`~mongoengine.ReferenceField`
+    The MongoDB field name.
-* :class:`~mongoengine.GenericReferenceField`
+
+:attr:`name` (Default: None)
+    The mongoengine 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 definion 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.ListField` or :class:`~mongoengine.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])
+
+
+: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.
+
+:attr:`choices` (Default: None)
+    An iterable (e.g. a list or tuple) of choices to which the value of this
+    field should be limited.
+
+    Can be either be a 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:`help_text` (Default: None)
+    Optional help text to output with the field - used by form libraries
+
+:attr:`verbose_name` (Default: None)
+    Optional human-readable name for the field - used by form libraries
+
 
 List fields
 -----------
@@ -62,7 +184,7 @@ 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
-inherit from :class:`~mongoengine.EmbeddedDocument` rather than 
+inherit from :class:`~mongoengine.EmbeddedDocument` rather than
 :class:`~mongoengine.Document`::
 
     class Comment(EmbeddedDocument):
@@ -75,8 +197,8 @@ document class as the first argument::
     class Page(Document):
         comments = ListField(EmbeddedDocumentField(Comment))
 
-    comment1 = Comment('Good work!')
+    comment1 = Comment(content='Good work!')
-    comment2 = Comment('Nice article!')
+    comment2 = Comment(content='Nice article!')
     page = Page(comments=[comment1, comment2])
 
 Dictionary Fields
@@ -85,7 +207,7 @@ Often, an embedded document may be used instead of a dictionary -- generally
 this is 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.DictField` is appropriate::
-    
+
     class SurveyResponse(Document):
         date = DateTimeField()
         user = ReferenceField(User)
@@ -93,16 +215,19 @@ store; in this situation a :class:`~mongoengine.DictField` is appropriate::
 
     survey_response = SurveyResponse(date=datetime.now(), user=request.user)
     response_form = ResponseForm(request.POST)
-    survey_response.answers = response_form.cleaned_data()   
+    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
 first argument to the constructor, then simply assign document objects to the
 field::
-    
+
     class User(Document):
         name = StringField()
 
@@ -134,19 +259,109 @@ as the constructor's argument::
     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])
+
+
+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 Employee(Document):
+        ...
+        profile_page = ReferenceField('ProfilePage', reverse_delete_rule=mongoengine.NULLIFY)
+
+The declaration in this example means that when an :class:`Employee` object is
+removed, the :class:`ProfilePage` that belongs to 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 removed
+  (using MongoDB's "unset" operation), effectively nullifying the relationship.
+:const:`mongoengine.CASCADE`
+  Any object containing fields that are refererring 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.ListField` (:class:`~mongoengine.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.
+
+
+.. warning::
+   Signals are not triggered when doing cascading updates / deletes - if this
+   is required you must manually handle the update / delete.
+
 Generic reference fields
 ''''''''''''''''''''''''
 A second kind of reference field also exists,
 :class:`~mongoengine.GenericReferenceField`. This allows you to reference any
-kind of :class:`~mongoengine.Document`, and hence doesn't take a 
+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()
 
@@ -160,9 +375,10 @@ kind of :class:`~mongoengine.Document`, and hence doesn't take a
     Bookmark(bookmark_object=post).save()
 
 .. note::
+
    Using :class:`~mongoengine.GenericReferenceField`\ s is slightly less
    efficient than the standard :class:`~mongoengine.ReferenceField`\ s, so if
-   you will only be referencing one document type, prefer the standard 
+   you will only be referencing one document type, prefer the standard
    :class:`~mongoengine.ReferenceField`.
 
 Uniqueness constraints
@@ -170,7 +386,7 @@ 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
 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 
+field as a document that is already in the database, a
 :class:`~mongoengine.OperationError` 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::
@@ -178,7 +394,21 @@ 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 caling 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
 ====================
@@ -203,7 +433,7 @@ A :class:`~mongoengine.Document` may use a **Capped Collection** by specifying
 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).
-The following example shows a :class:`Log` document that will be limited to 
+The following example shows a :class:`Log` document that will be limited to
 1000 entries and 2MB of disk space::
 
     class Log(Document):
@@ -212,12 +442,14 @@ The following example shows a :class:`Log` document that will be limited to
 
 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
+either be a single field name, a tuple containing multiple field names, or a
-direction may be specified on fields by prefixing the field name with a **+**
+dictionary containing a full index definition. A direction may be specified on
-or a **-** sign. Note that direction only matters on multi-field indexes. ::
+fields by prefixing the field name with a **+** or a **-** sign. Note that
+direction only matters on multi-field indexes. ::
 
     class Page(Document):
         title = StringField()
@@ -225,7 +457,57 @@ or a **-** sign. Note that direction only matters on multi-field indexes. ::
         meta = {
             'indexes': ['title', ('title', '-rating')]
         }
-        
+
+If a dictionary is passed then the following options are available:
+
+: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.
+
+.. note::
+
+    Inheritance adds extra fields indices see: :ref:`document-inheritance`.
+
+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 Embeedded / dictionary fields,
+in this case use 'dot' notation to identify the value to index eg: `rank.title`
+
+Geospatial indexes
+------------------
+
+Geospatial indexes will be automatically created for all
+:class:`~mongoengine.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.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',
+            ],
+        }
+
 Ordering
 ========
 A default ordering can be specified for your
@@ -247,7 +529,7 @@ 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_2 = BlogPost(title="Blog Post #2") 
+    blog_post_2 = BlogPost(title="Blog Post #2")
     blog_post_2.published_date = datetime(2010, 1, 6, 0, 0 ,0)
 
     blog_post_3 = BlogPost(title="Blog Post #3")
@@ -259,46 +541,77 @@ 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() 
+    latest_post = BlogPost.objects.first()
     assert latest_post.title == "Blog Post #3"
 
     # override default ordering, order BlogPosts by "published_date"
     first_post = BlogPost.objects.order_by("+published_date").first()
     assert first_post.title == "Blog Post #1"
 
+Shard keys
+==========
+
+If your collection is sharded, then you need to specify the shard key as a tuple,
+using the :attr:`shard_key` attribute of :attr:`-mongoengine.Document.meta`.
+This ensures that the shard key is sent with the query when calling the
+:meth:`~mongoengine.document.Document.save` or
+:meth:`~mongoengine.document.Document.update` method on an existing
+:class:`-mongoengine.Document` instance::
+
+    class LogEntry(Document):
+        machine = StringField()
+        app = StringField()
+        timestamp = DateTimeField()
+        data = StringField()
+
+        meta = {
+            'shard_key': ('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
 :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 you must declare :attr:`allow_inheritance` defaults
+          to False, meaning you must set it to True to use inheritance.
+
+
 Working with existing data
 --------------------------
-To enable correct retrieval of documents involved in this kind of heirarchy,
+As MongoEngine no longer defaults to needing :attr:`_cls` you can quickly and
-two extra attributes are stored on each document in the database: :attr:`_cls`
+easily get working with existing data.  Just define the document to match
-and :attr:`_types`. These are hidden from the user through the MongoEngine
+the expected schema in your database ::
-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::
 
     # Will work with data in an existing collection named 'cmsPage'
     class Page(Document):
         title = StringField(max_length=200, required=True)
         meta = {
-            'collection': 'cmsPage',
+            'collection': 'cmsPage'
-            'allow_inheritance': False,
         }
+
+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.

docs/guide/document-instances.rst

@@ -4,12 +4,12 @@ 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.
 You may provide values for any of the fields on the document::
-    
+
     >>> page = Page(title="Test Page")
     >>> page.title
     'Test Page'
 
-You may also assign values to the document's fields using standard object 
+You may also assign values to the document's fields using standard object
 attribute syntax::
 
     >>> page.title = "Example Page"
@@ -18,18 +18,68 @@ attribute syntax::
 
 Saving and deleting documents
 =============================
-To save the document to the database, call the
+MongoEngine tracks changes to documents to provide efficient saving.  To save
-:meth:`~mongoengine.Document.save` method. If the document does not exist in
+the document to the database, call the :meth:`~mongoengine.Document.save` method.
-the database, it will be created. If it does already exist, it will be
+If the document does not exist in the database, it will be created. If it does
-updated.
+already exist, then any changes will be updated atomically.  For example::
 
-To delete a document, call the :meth:`~mongoengine.Document.delete` method.
+    >>> page = Page(title="Test Page")
-Note that this will only work if the document exists in the database and has a
+    >>> page.save()  # Performs an insert
-valide :attr:`id`.
+    >>> 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.pop(0)`` - *sets* the resulting list
+    * ``del(list_field)``   - *unsets* whole list
 
 .. 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 the pub_date if published and 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.ReferenceField` or
+:class:`~mongoengine.GenericReferenceField` objects, then by default the
+:meth:`~mongoengine.Document.save` method will automatically save any changes to
+those objects as well.  If this is not desired passing :attr:`cascade` as False
+to the save method turns this feature off.
+
+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
@@ -59,7 +109,15 @@ 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; in
+is an alias to :attr:`id`::
+
+    >>> page = Page(title="Another Test Page")
+    >>> page.save()
+    >>> page.id == page.pk
+
 .. 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
+   required, so a :class:`~mongoengine.ValidationError` will be thrown if
-   it.
+   you don't provide it.

docs/guide/gridfs.rst

@@ -0,0 +1,74 @@
+======
+GridFS
+======
+
+.. versionadded:: 0.4
+
+Writing
+-------
+
+GridFS support comes in the form of the :class:`~mongoengine.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. 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')
+
+    marmot_photo = open('marmot.jpg', 'r')
+    marmot.photo.put(marmot_photo, content_type = 'image/jpeg')
+    marmot.save()
+
+Retrieval
+---------
+
+So using the :class:`~mongoengine.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
+
+Streaming
+---------
+
+Streaming data into a :class:`~mongoengine.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.photo.save()
+
+Deletion
+--------
+
+Deleting stored files is achieved with the :func:`delete` method::
+
+    marmot.photo.delete()
+
+.. 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', 'r')
+    marmot.photo.replace(another_marmot, content_type='image/png')

docs/guide/index.rst

@@ -10,3 +10,5 @@ User Guide
    defining-documents
    document-instances
    querying
+   gridfs
+   signals

docs/guide/installing.rst

@@ -1,31 +1,31 @@
 ======================
 Installing MongoEngine
 ======================
+
 To use MongoEngine, you will need to download `MongoDB <http://mongodb.org/>`_
 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 
+MongoEngine is available on PyPI, so to use it you can use :program:`pip`:
-:program:`easy_install`:
+
-    
 .. code-block:: console
 
-    # easy_install mongoengine
+    $ pip install mongoengine
 
-Alternatively, if you don't have setuptools installed, `download it from PyPi 
+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:
-    
+
 .. code-block:: console
 
-    # git clone git://github.com/hmarr/mongoengine
+    $ git clone git://github.com/hmarr/mongoengine
-    # cd mongoengine
+    $ cd mongoengine
-    # python setup.py install
+    $ python setup.py install

docs/guide/querying.rst

@@ -5,8 +5,8 @@ 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 on access. The
-:class:`~mongoengine.queryset.QuerySet` object may may be iterated over to
+: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
@@ -14,6 +14,7 @@ fetch documents from the database::
         print user.name
 
 .. note::
+
    Once the iteration finishes (when :class:`StopIteration` is raised),
    :meth:`~mongoengine.queryset.QuerySet.rewind` will be called so that the
    :class:`~mongoengine.queryset.QuerySet` may be iterated over again. The
@@ -23,7 +24,7 @@ fetch documents from the database::
 Filtering queries
 =================
 The query may be filtered by calling the
-:class:`~mongoengine.queryset.QuerySet` object with field lookup keyword 
+:class:`~mongoengine.queryset.QuerySet` object with field lookup keyword
 arguments. The keys in the keyword arguments correspond to fields on the
 :class:`~mongoengine.Document` you are querying::
 
@@ -34,11 +35,67 @@ arguments. The keys in the keyword arguments correspond to fields on the
 Fields on embedded documents may also be referred to using field lookup syntax
 by using a double-underscore in place of the dot in object attribute access
 syntax::
-    
+
     # This will return a QuerySet that will only iterate over pages that have
     # been written by a user whose 'country' field is set to 'uk'
     uk_pages = Page.objects(author__country='uk')
 
+
+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)``)
+* ``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
+
+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
+
+There are a few special operators for performing geographical queries, that
+may used with :class:`~mongoengine.GeoPointField`\ s:
+
+* ``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.
+
+
 Querying lists
 --------------
 On most fields, this syntax will look up documents where the field specified
@@ -53,39 +110,39 @@ lists that contain that item will be matched::
     # 'tags' list
     Page.objects(tags='coding')
 
-Query operators
+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``,
-Operators other than equality may also be used in queries; just attach the
+you could use the following query::
-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:
+    Page.objects(tags__0='db')
 
-* ``neq`` -- not equal to
+If you only want to fetch part of a list eg: you want to paginate a list, then
-* ``lt`` -- less than
+the `slice` operator is required::
-* ``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
 
-The following operators are available as shortcuts to querying with regular
+    # comments - skip 5, limit 10
-expressions:
+    Page.objects.fields(slice__comments=[5, 10])
 
-* ``contains`` -- string field contains value
+For updating documents, if you don't know the position in a list, you can use
-* ``icontains`` -- string field contains value (case insensitive)
+the $ positional operator ::
-* ``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)
 
-.. versionadded:: 0.3
+    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 Mongo currently the $ operator only applies to the first matched item in the query.
+
+
+Raw queries
+-----------
+It is possible to provide a raw 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
 
 Limiting and skipping results
 =============================
@@ -109,7 +166,7 @@ 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]
@@ -124,19 +181,21 @@ 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,
+:class:`~mongoengine.queryset.DoesNotExist` if
-and :class:`~mongoengine.queryset.MultipleObjectsReturned` if more than one
+no document matches the query, and
-document matched the query.
+:class:`~mongoengine.queryset.MultipleObjectsReturned`
+if more than one document matched the query.  These exceptions are merged into
+your document defintions eg: `MyDoc.DoesNotExist`
 
-A variation of this method exists, 
+A variation of this method exists,
 :meth:`~mongoengine.queryset.Queryset.get_or_create`, that will create a new
-document with the query arguments if no documents match the query. An 
+document with the query arguments if no documents match the query. An
 additional keyword argument, :attr:`defaults` may be provided, which will be
 used as default values for the new document, in the case that it should need
 to be created::
 
-    >>> a = User.objects.get_or_create(name='User A', defaults={'age': 30})
+    >>> a, created = User.objects.get_or_create(name='User A', defaults={'age': 30})
-    >>> b = User.objects.get_or_create(name='User A', defaults={'age': 40})
+    >>> b, created = User.objects.get_or_create(name='User A', defaults={'age': 40})
     >>> a.name == b.name and a.age == b.age
     True
 
@@ -172,12 +231,33 @@ custom manager methods as you like::
 
         @queryset_manager
         def live_posts(doc_cls, queryset):
-            return queryset.order_by('-date')
+            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
+    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
 ===========
@@ -189,7 +269,7 @@ 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 -- 
+:class:`~mongoengine.queryset.QuerySet` objects --
 :meth:`~mongoengine.queryset.QuerySet.count`, but there is also a more Pythonic
 way of achieving this::
 
@@ -203,6 +283,7 @@ You may sum over the values of a specific field on documents using
     yearly_expense = Employee.objects.sum('salary')
 
 .. note::
+
    If the field isn't present on a document, that document will be ignored from
    the sum.
 
@@ -225,8 +306,16 @@ would be generating "tag-clouds"::
     from operator import itemgetter
     top_tags = sorted(tag_freqs.items(), key=itemgetter(1), reverse=True)[:10]
 
+
+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
@@ -251,18 +340,65 @@ will be given::
     >>> 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.ListField` or
+:class:`~mongoengine.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.ReferenceField` objects or
+:class:`~mongoengine.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, ObjectId))
+
+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, ObjectId))
+
+    # 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 
+need to combine a number of constraints using *and* and *or*. This is made
 possible in MongoEngine through the :class:`~mongoengine.queryset.Q` class.
 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 
+documents. To build a complex query, you may combine
 :class:`~mongoengine.queryset.Q` objects using the ``&`` (and) and ``|`` (or)
 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
@@ -274,11 +410,66 @@ calling it with keyword arguments::
     # Get top posts
     Post.objects((Q(featured=True) & Q(hits__gte=1000)) | Q(hits__gte=5000))
 
-.. warning::
+.. _guide-atomic-updates:
-   Only use these advanced queries if absolutely necessary as they will execute
+
-   significantly slower than regular queries. This is because they are not
+Atomic updates
-   natively supported by MongoDB -- they are compiled to Javascript and sent
+==============
-   to the server for execution.
+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:
+
+* ``set`` -- set a particular value
+* ``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
+* ``pop`` -- remove the last item from a list
+* ``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
+* ``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
+
+The syntax for atomic updates is similar to the querying syntax, but the
+modifier comes before the field, not after it::
+
+    >>> post = BlogPost(title='Test', page_views=0, tags=['database'])
+    >>> post.save()
+    >>> BlogPost.objects(id=post.id).update_one(inc__page_views=1)
+    >>> post.reload()  # the document has been changed, so we need to reload it
+    >>> post.page_views
+    1
+    >>> BlogPost.objects(id=post.id).update_one(set__title='Example Post')
+    >>> post.reload()
+    >>> post.title
+    'Example Post'
+    >>> BlogPost.objects(id=post.id).update_one(push__tags='nosql')
+    >>> post.reload()
+    >>> post.tags
+    ['database', 'nosql']
+
+.. 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']
+
+.. 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
 ================================
@@ -382,40 +573,3 @@ following example shows how the substitutions are made::
         return comments;
     }
     """)
-
-.. _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:
-
-* ``set`` -- set a particular value
-* ``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
-* ``pull`` -- remove a value from a list
-* ``pull_all`` -- remove several values from a list
-
-The syntax for atomic updates is similar to the querying syntax, but the 
-modifier comes before the field, not after it::
-    
-    >>> post = BlogPost(title='Test', page_views=0, tags=['database'])
-    >>> post.save()
-    >>> BlogPost.objects(id=post.id).update_one(inc__page_views=1)
-    >>> post.reload()  # the document has been changed, so we need to reload it
-    >>> post.page_views
-    1
-    >>> BlogPost.objects(id=post.id).update_one(set__title='Example Post')
-    >>> post.reload()
-    >>> post.title
-    'Example Post'
-    >>> BlogPost.objects(id=post.id).update_one(push__tags='nosql')
-    >>> post.reload()
-    >>> post.tags
-    ['database', 'nosql']

docs/guide/signals.rst

@@ -0,0 +1,60 @@
+.. _signals:
+
+Signals
+=======
+
+.. versionadded:: 0.5
+
+.. note::
+
+  Signal support is provided by the excellent `blinker`_ library and
+  will gracefully fall back if it is not available.
+
+
+The following document signals exist in MongoEngine and are pretty self-explanatory:
+
+  * `mongoengine.signals.pre_init`
+  * `mongoengine.signals.post_init`
+  * `mongoengine.signals.pre_save`
+  * `mongoengine.signals.post_save`
+  * `mongoengine.signals.pre_delete`
+  * `mongoengine.signals.post_delete`
+  * `mongoengine.signals.pre_bulk_insert`
+  * `mongoengine.signals.post_bulk_insert`
+
+Example usage::
+
+    from mongoengine import *
+    from mongoengine import signals
+
+    class Author(Document):
+        name = StringField()
+
+        def __unicode__(self):
+            return self.name
+
+        @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)
+
+
+ReferenceFields and signals
+---------------------------
+
+Currently `reverse_delete_rules` do not trigger signals on the other part of
+the relationship.  If this is required you must manually handled the
+reverse deletion.
+
+.. _blinker: http://pypi.python.org/pypi/blinker

docs/index.rst

@@ -2,34 +2,66 @@
 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 -U mongoengine
+    # pip install -U mongoengine
 
-The source is available on `GitHub <http://github.com/hmarr/mongoengine>`_.
+:doc:`tutorial`
+  Start here for a quick overview.
+
+:doc:`guide/index`
+  The Full guide to MongoEngine
+
+:doc:`apireference`
+  The complete API documentation.
+
+:doc:`upgrade`
+  How to upgrade MongoEngine.
+
+:doc:`django`
+  Using MongoEngine and Django
+
+Community
+---------
 
 To get help with using MongoEngine, use the `MongoEngine Users mailing list
 <http://groups.google.com/group/mongoengine-users>`_ or come chat on the
 `#mongoengine IRC channel <irc://irc.freenode.net/mongoengine>`_.
 
-If you are interested in contributing, join the developers' `mailing list 
+Contributing
+------------
+
+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. To contribute, fork the project on
+`GitHub <http://github.com/MongoEngine/mongoengine>`_ and send a
+pull request.
+
+Also, you can join the developers' `mailing list
 <http://groups.google.com/group/mongoengine-dev>`_.
 
+Changes
+-------
+See the :doc:`changelog` for a full list of changes to MongoEngine and
+:doc:`upgrade` for upgrade information.
+
 .. toctree::
-   :maxdepth: 2
+   :hidden:
 
    tutorial
    guide/index
    apireference
    django
    changelog
+   upgrade
 
 Indices and tables
-==================
+------------------
 
 * :ref:`genindex`
+* :ref:`modindex`
 * :ref:`search`
 

docs/tutorial.rst

@@ -22,7 +22,7 @@ function. 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`.
@@ -84,12 +84,15 @@ 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 modelling 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()
 
@@ -112,7 +115,7 @@ 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 
+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
 :class:`Post` class::
 
@@ -152,6 +155,26 @@ 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.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.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
@@ -250,5 +273,5 @@ 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 %d posts with tag "mongodb"' % num_posts
-    
+

docs/upgrade.rst

@@ -0,0 +1,302 @@
+=========
+Upgrading
+=========
+
+0.7 to 0.8
+==========
+
+Inheritance
+-----------
+
+Data Model
+~~~~~~~~~~
+
+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.iteritems()
+                       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 - its 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 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'
+
+Querysets
+~~~~~~~~~
+
+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 fitler 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
+
+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.
+
+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, its now much more pythonic and readable as
+its 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 +
+===================
+
+Its 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

@@ -6,19 +6,19 @@ import connection
 from connection import *
 import queryset
 from queryset import *
+import signals
+from signals import *
+import django
 
-__all__ = (document.__all__ + fields.__all__ + connection.__all__ +
+__all__ = (list(document.__all__) + fields.__all__ + connection.__all__ +
-           queryset.__all__)
+           list(queryset.__all__) + signals.__all__)
 
-__author__ = 'Harry Marr'
+VERSION = (0, 8, 0, '+')
 
-VERSION = (0, 3, 0)
 
 def get_version():
-    version = '%s.%s' % (VERSION[0], VERSION[1])
+    if isinstance(VERSION[-1], basestring):
-    if VERSION[2]:
+        return '.'.join(map(str, VERSION[:-1])) + VERSION[-1]
-        version = '%s.%s' % (version, VERSION[2])
+    return '.'.join(map(str, VERSION))
-    return version
 
 __version__ = get_version()
-

mongoengine/base.py

@@ -1,419 +0,0 @@
-from queryset import QuerySet, QuerySetManager
-
-import pymongo
-
-
-_document_registry = {}
-
-def get_document(name):
-    return _document_registry[name]
-
-
-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, db_field=None, name=None, required=False, default=None, 
-                 unique=False, unique_with=None, primary_key=False):
-        self.db_field = (db_field or name) if not primary_key else '_id'
-        if name:
-            import warnings
-            msg = "Fields' 'name' attribute deprecated in favour of 'db_field'"
-            warnings.warn(msg, DeprecationWarning)
-        self.name = None
-        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, op, 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 value
-        # return unicode(value)
-
-    def to_mongo(self, value):
-        if not isinstance(value, pymongo.objectid.ObjectId):
-            try:
-                return pymongo.objectid.ObjectId(str(value))
-            except Exception, e:
-                #e.message attribute has been deprecated since Python 2.6
-                raise ValidationError(str(e))
-        return value
-
-    def prepare_query_value(self, op, 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 = {}
-        simple_class = True
-        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)
-                
-            if hasattr(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
-
-        meta = attrs.get('_meta', attrs.get('meta', {}))
-
-        if 'allow_inheritance' not in meta:
-            meta['allow_inheritance'] = True
-
-        # 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
-
-        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):
-                attr_value.name = attr_name
-                if not attr_value.db_field:
-                    attr_value.db_field = attr_name
-                doc_fields[attr_name] = attr_value
-        attrs['_fields'] = doc_fields
-
-        new_class = super_new(cls, name, bases, attrs)
-        for field in new_class._fields.values():
-            field.owner_document = new_class
-
-        return new_class
-
-
-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):
-        global _document_registry
-
-        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()
-        
-        id_field = None
-        base_indexes = []
-
-        # Subclassed documents inherit collection from superclass
-        for base in bases:
-            if hasattr(base, '_meta') and 'collection' in base._meta:
-                collection = base._meta['collection']
-
-                id_field = id_field or base._meta.get('id_field')
-                base_indexes += base._meta.get('indexes', [])
-
-        meta = {
-            'collection': collection,
-            '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', {}))
-        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.db_field 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._fields['id'] = ObjectIdField(db_field='_id')
-            new_class.id = new_class._fields['id']
-
-        _document_registry[name] = new_class
-
-        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.db_field] = 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 SON.
-        """
-        # 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]
-
-        present_fields = data.keys()
-
-        for field_name, field in cls._fields.items():
-            if field.db_field in data:
-                data[field_name] = field.to_python(data[field.db_field])
-
-        obj = cls(**data)
-        obj._present_fields = present_fields
-        return obj
-    
-    def __eq__(self, other):
-        if isinstance(other, self.__class__) and hasattr(other, 'id'):
-            if self.id == other.id:
-                return True
-        return False

mongoengine/base/__init__.py

@@ -0,0 +1,5 @@
+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 *

mongoengine/base/common.py

@@ -0,0 +1,26 @@
+from mongoengine.errors import NotRegistered
+
+__all__ = ('ALLOW_INHERITANCE', 'get_document', '_document_registry')
+
+ALLOW_INHERITANCE = False
+
+_document_registry = {}
+
+
+def get_document(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.keys()
+                          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

mongoengine/base/datastructures.py

@@ -0,0 +1,142 @@
+import weakref
+from mongoengine.common import _import_class
+
+__all__ = ("BaseDict", "BaseList")
+
+
+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):
+        self._instance = weakref.proxy(instance)
+        self._name = name
+        return super(BaseDict, self).__init__(dict_items)
+
+    def __getitem__(self, *args, **kwargs):
+        value = super(BaseDict, self).__getitem__(*args, **kwargs)
+
+        EmbeddedDocument = _import_class('EmbeddedDocument')
+        if isinstance(value, EmbeddedDocument) and value._instance is None:
+            value._instance = self._instance
+        return value
+
+    def __setitem__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).__setitem__(*args, **kwargs)
+
+    def __delete__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).__delete__(*args, **kwargs)
+
+    def __delitem__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).__delitem__(*args, **kwargs)
+
+    def __delattr__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).__delattr__(*args, **kwargs)
+
+    def __getstate__(self):
+        self.instance = None
+        self._dereferenced = False
+        return self
+
+    def __setstate__(self, state):
+        self = state
+        return self
+
+    def clear(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).clear(*args, **kwargs)
+
+    def pop(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).pop(*args, **kwargs)
+
+    def popitem(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).popitem(*args, **kwargs)
+
+    def update(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseDict, self).update(*args, **kwargs)
+
+    def _mark_as_changed(self):
+        if hasattr(self._instance, '_mark_as_changed'):
+            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):
+        self._instance = weakref.proxy(instance)
+        self._name = name
+        return super(BaseList, self).__init__(list_items)
+
+    def __getitem__(self, *args, **kwargs):
+        value = super(BaseList, self).__getitem__(*args, **kwargs)
+
+        EmbeddedDocument = _import_class('EmbeddedDocument')
+        if isinstance(value, EmbeddedDocument) and value._instance is None:
+            value._instance = self._instance
+        return value
+
+    def __setitem__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).__setitem__(*args, **kwargs)
+
+    def __delitem__(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).__delitem__(*args, **kwargs)
+
+    def __getstate__(self):
+        self.instance = None
+        self._dereferenced = False
+        return self
+
+    def __setstate__(self, state):
+        self = state
+        return self
+
+    def append(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).append(*args, **kwargs)
+
+    def extend(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).extend(*args, **kwargs)
+
+    def insert(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).insert(*args, **kwargs)
+
+    def pop(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).pop(*args, **kwargs)
+
+    def remove(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).remove(*args, **kwargs)
+
+    def reverse(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).reverse(*args, **kwargs)
+
+    def sort(self, *args, **kwargs):
+        self._mark_as_changed()
+        return super(BaseList, self).sort(*args, **kwargs)
+
+    def _mark_as_changed(self):
+        if hasattr(self._instance, '_mark_as_changed'):
+            self._instance._mark_as_changed(self._name)

mongoengine/base/document.py

@@ -0,0 +1,801 @@
+import copy
+import operator
+import numbers
+from functools import partial
+
+import pymongo
+from bson import json_util
+from bson.dbref import DBRef
+
+from mongoengine import signals
+from mongoengine.common import _import_class
+from mongoengine.errors import (ValidationError, InvalidDocumentError,
+                                LookUpError)
+from mongoengine.python_support import (PY3, UNICODE_KWARGS, txt_type,
+                                        to_str_keys_recursive)
+
+from mongoengine.base.common import get_document, ALLOW_INHERITANCE
+from mongoengine.base.datastructures import BaseDict, BaseList
+from mongoengine.base.fields import ComplexBaseField
+
+__all__ = ('BaseDocument', 'NON_FIELD_ERRORS')
+
+NON_FIELD_ERRORS = '__all__'
+
+
+class BaseDocument(object):
+
+    _dynamic = False
+    _created = True
+    _dynamic_lock = True
+    _initialised = False
+
+    def __init__(self, *args, **values):
+        """
+        Initialise a document or embedded document
+
+        :param __auto_convert: Try and will cast python objects to Object types
+        :param values: A dictionary of values for the document
+        """
+        if args:
+            # Combine positional arguments with named arguments.
+            # We only want named arguments.
+            field = iter(self._fields_ordered)
+            for value in args:
+                name = next(field)
+                if name in values:
+                    raise TypeError("Multiple values for keyword argument '" + name + "'")
+                values[name] = value
+        __auto_convert = values.pop("__auto_convert", True)
+        signals.pre_init.send(self.__class__, document=self, values=values)
+
+        self._data = {}
+
+        # Assign default values to instance
+        for key, field in self._fields.iteritems():
+            if self._db_field_map.get(key, key) in values:
+                continue
+            value = getattr(self, key, None)
+            setattr(self, key, value)
+
+        # Set passed values after initialisation
+        if self._dynamic:
+            self._dynamic_fields = {}
+            dynamic_data = {}
+            for key, value in values.iteritems():
+                if key in self._fields or key == '_id':
+                    setattr(self, key, value)
+                elif self._dynamic:
+                    dynamic_data[key] = value
+        else:
+            FileField = _import_class('FileField')
+            for key, value in values.iteritems():
+                if key == '__auto_convert':
+                    continue
+                key = self._reverse_db_field_map.get(key, key)
+                if key in self._fields or key in ('id', 'pk', '_cls'):
+                    if __auto_convert and value is not None:
+                        field = self._fields.get(key)
+                        if field and not isinstance(field, FileField):
+                            value = field.to_python(value)
+                    setattr(self, key, value)
+                else:
+                    self._data[key] = value
+
+        # Set any get_fieldname_display methods
+        self.__set_field_display()
+
+        if self._dynamic:
+            self._dynamic_lock = False
+            for key, value in dynamic_data.iteritems():
+                setattr(self, key, value)
+
+        # Flag initialised
+        self._initialised = True
+        signals.post_init.send(self.__class__, document=self)
+
+    def __delattr__(self, *args, **kwargs):
+        """Handle deletions of fields"""
+        field_name = args[0]
+        if field_name in self._fields:
+            default = self._fields[field_name].default
+            if callable(default):
+                default = default()
+            setattr(self, field_name, default)
+        else:
+            super(BaseDocument, self).__delattr__(*args, **kwargs)
+
+    def __setattr__(self, name, value):
+        # Handle dynamic data only if an initialised dynamic document
+        if self._dynamic and not self._dynamic_lock:
+
+            field = None
+            if not hasattr(self, name) and not name.startswith('_'):
+                DynamicField = _import_class("DynamicField")
+                field = DynamicField(db_field=name)
+                field.name = name
+                self._dynamic_fields[name] = field
+
+            if not name.startswith('_'):
+                value = self.__expand_dynamic_values(name, value)
+
+            # Handle marking data as changed
+            if name in self._dynamic_fields:
+                self._data[name] = value
+                if hasattr(self, '_changed_fields'):
+                    self._mark_as_changed(name)
+
+        if (self._is_document and not self._created and
+           name in self._meta.get('shard_key', tuple()) and
+           self._data.get(name) != value):
+            OperationError = _import_class('OperationError')
+            msg = "Shard Keys are immutable. Tried to update %s" % name
+            raise OperationError(msg)
+
+        # Check if the user has created a new instance of a class
+        if (self._is_document and self._initialised
+           and self._created and name == self._meta['id_field']):
+                super(BaseDocument, self).__setattr__('_created', False)
+
+        super(BaseDocument, self).__setattr__(name, value)
+
+    def __getstate__(self):
+        removals = ("get_%s_display" % k
+                    for k, v in self._fields.items() if v.choices)
+        for k in removals:
+            if hasattr(self, k):
+                delattr(self, k)
+        return self.__dict__
+
+    def __setstate__(self, __dict__):
+        self.__dict__ = __dict__
+        self.__set_field_display()
+
+    def __iter__(self):
+        if 'id' in self._fields and 'id' not in self._fields_ordered:
+            return iter(('id', ) + self._fields_ordered)
+
+        return iter(self._fields_ordered)
+
+    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 = self.__str__()
+        except (UnicodeEncodeError, UnicodeDecodeError):
+            u = '[Bad Unicode data]'
+        repr_type = type(u)
+        return repr_type('<%s: %s>' % (self.__class__.__name__, u))
+
+    def __str__(self):
+        if hasattr(self, '__unicode__'):
+            if PY3:
+                return self.__unicode__()
+            else:
+                return unicode(self).encode('utf-8')
+        return txt_type('%s object' % self.__class__.__name__)
+
+    def __eq__(self, other):
+        if isinstance(other, self.__class__) and hasattr(other, 'id'):
+            if self.id == other.id:
+                return True
+        return False
+
+    def __ne__(self, other):
+        return not self.__eq__(other)
+
+    def __hash__(self):
+        if self.pk is None:
+            # For new object
+            return super(BaseDocument, self).__hash__()
+        else:
+            return hash(self.pk)
+
+    def clean(self):
+        """
+        Hook for doing document level data cleaning before validation is run.
+
+        Any ValidationError raised by this method will not be associated with
+        a particular field; it will have a special-case association with the
+        field defined by NON_FIELD_ERRORS.
+        """
+        pass
+
+    def to_mongo(self):
+        """Return data dictionary ready for use with MongoDB.
+        """
+        data = {}
+        for field_name, field in self._fields.iteritems():
+            value = self._data.get(field_name, None)
+            if value is not None:
+                value = field.to_mongo(value)
+
+            # Handle self generating fields
+            if value is None and field._auto_gen:
+                value = field.generate()
+                self._data[field_name] = value
+
+            if value is not None:
+                data[field.db_field] = value
+
+        # Only add _cls if allow_inheritance is True
+        if (hasattr(self, '_meta') and
+            self._meta.get('allow_inheritance', ALLOW_INHERITANCE) == True):
+            data['_cls'] = self._class_name
+
+        if '_id' in data and data['_id'] is None:
+            del data['_id']
+
+        if not self._dynamic:
+            return data
+
+        for name, field in self._dynamic_fields.items():
+            data[name] = field.to_mongo(self._data.get(name, None))
+        return data
+
+    def validate(self, clean=True):
+        """Ensure that all fields' values are valid and that required fields
+        are present.
+        """
+        # Ensure that each field is matched to a valid value
+        errors = {}
+        if clean:
+            try:
+                self.clean()
+            except ValidationError, error:
+                errors[NON_FIELD_ERRORS] = error
+
+        # Get a list of tuples of field names and their current values
+        fields = [(field, self._data.get(name))
+                  for name, field in self._fields.items()]
+        if self._dynamic:
+            fields += [(field, self._data.get(name))
+                       for name, field in self._dynamic_fields.items()]
+
+        EmbeddedDocumentField = _import_class("EmbeddedDocumentField")
+        GenericEmbeddedDocumentField = _import_class("GenericEmbeddedDocumentField")
+
+        for field, value in fields:
+            if value is not None:
+                try:
+                    if isinstance(field, (EmbeddedDocumentField,
+                                          GenericEmbeddedDocumentField)):
+                        field._validate(value, clean=clean)
+                    else:
+                        field._validate(value)
+                except ValidationError, error:
+                    errors[field.name] = error.errors or error
+                except (ValueError, AttributeError, AssertionError), error:
+                    errors[field.name] = error
+            elif field.required and not getattr(field, '_auto_gen', False):
+                errors[field.name] = ValidationError('Field is required',
+                                                     field_name=field.name)
+
+        if errors:
+            pk = "None"
+            if hasattr(self, 'pk'):
+                pk = self.pk
+            elif self._instance:
+                pk = self._instance.pk
+            message = "ValidationError (%s:%s) " % (self._class_name, pk)
+            raise ValidationError(message, errors=errors)
+
+    def to_json(self):
+        """Converts a document to JSON"""
+        return json_util.dumps(self.to_mongo())
+
+    @classmethod
+    def from_json(cls, json_data):
+        """Converts json data to an unsaved document instance"""
+        return cls._from_son(json_util.loads(json_data))
+
+    def __expand_dynamic_values(self, name, value):
+        """expand any dynamic values to their correct types / values"""
+        if not isinstance(value, (dict, list, tuple)):
+            return value
+
+        is_list = False
+        if not hasattr(value, 'items'):
+            is_list = True
+            value = dict([(k, v) for k, v in enumerate(value)])
+
+        if not is_list and '_cls' in value:
+            cls = get_document(value['_cls'])
+            return cls(**value)
+
+        data = {}
+        for k, v in value.items():
+            key = name if is_list else k
+            data[k] = self.__expand_dynamic_values(key, v)
+
+        if is_list:  # Convert back to a list
+            data_items = sorted(data.items(), key=operator.itemgetter(0))
+            value = [v for k, v in data_items]
+        else:
+            value = data
+
+        # Convert lists / values so we can watch for any changes on them
+        if (isinstance(value, (list, tuple)) and
+           not isinstance(value, BaseList)):
+            value = BaseList(value, self, name)
+        elif isinstance(value, dict) and not isinstance(value, BaseDict):
+            value = BaseDict(value, self, name)
+
+        return value
+
+    def _mark_as_changed(self, key):
+        """Marks a key as explicitly changed by the user
+        """
+        if not key:
+            return
+        key = self._db_field_map.get(key, key)
+        if (hasattr(self, '_changed_fields') and
+           key not in self._changed_fields):
+            self._changed_fields.append(key)
+
+    def _clear_changed_fields(self):
+        self._changed_fields = []
+        EmbeddedDocumentField = _import_class("EmbeddedDocumentField")
+        for field_name, field in self._fields.iteritems():
+            if (isinstance(field, ComplexBaseField) and
+               isinstance(field.field, EmbeddedDocumentField)):
+                field_value = getattr(self, field_name, None)
+                if field_value:
+                    for idx in (field_value if isinstance(field_value, dict)
+                                else xrange(len(field_value))):
+                        field_value[idx]._clear_changed_fields()
+            elif isinstance(field, EmbeddedDocumentField):
+                field_value = getattr(self, field_name, None)
+                if field_value:
+                    field_value._clear_changed_fields()
+
+    def _get_changed_fields(self, key='', inspected=None):
+        """Returns a list of all fields that have explicitly been changed.
+        """
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        DynamicEmbeddedDocument = _import_class("DynamicEmbeddedDocument")
+        _changed_fields = []
+        _changed_fields += getattr(self, '_changed_fields', [])
+
+        inspected = inspected or set()
+        if hasattr(self, 'id'):
+            if self.id in inspected:
+                return _changed_fields
+            inspected.add(self.id)
+
+        field_list = self._fields.copy()
+        if self._dynamic:
+            field_list.update(self._dynamic_fields)
+
+        for field_name in field_list:
+
+            db_field_name = self._db_field_map.get(field_name, field_name)
+            key = '%s.' % db_field_name
+            field = self._data.get(field_name, None)
+            if hasattr(field, 'id'):
+                if field.id in inspected:
+                    continue
+                inspected.add(field.id)
+
+            if (isinstance(field, (EmbeddedDocument, DynamicEmbeddedDocument))
+               and db_field_name not in _changed_fields):
+                 # Find all embedded fields that have been changed
+                changed = field._get_changed_fields(key, inspected)
+                _changed_fields += ["%s%s" % (key, k) for k in changed if k]
+            elif (isinstance(field, (list, tuple, dict)) and
+                    db_field_name not in _changed_fields):
+                # Loop list / dict fields as they contain documents
+                # Determine the iterator to use
+                if not hasattr(field, 'items'):
+                    iterator = enumerate(field)
+                else:
+                    iterator = field.iteritems()
+                for index, value in iterator:
+                    if not hasattr(value, '_get_changed_fields'):
+                        continue
+                    list_key = "%s%s." % (key, index)
+                    changed = value._get_changed_fields(list_key, inspected)
+                    _changed_fields += ["%s%s" % (list_key, k)
+                                        for k in changed if k]
+        return _changed_fields
+
+    def _delta(self):
+        """Returns the delta (set, unset) of the changes for a document.
+        Gets any values that have been explicitly changed.
+        """
+        # Handles cases where not loaded from_son but has _id
+        doc = self.to_mongo()
+
+        set_fields = self._get_changed_fields()
+        set_data = {}
+        unset_data = {}
+        parts = []
+        if hasattr(self, '_changed_fields'):
+            set_data = {}
+            # Fetch each set item from its path
+            for path in set_fields:
+                parts = path.split('.')
+                d = doc
+                new_path = []
+                for p in parts:
+                    if isinstance(d, DBRef):
+                        break
+                    elif isinstance(d, list) and p.isdigit():
+                        d = d[int(p)]
+                    elif hasattr(d, 'get'):
+                        d = d.get(p)
+                    new_path.append(p)
+                path = '.'.join(new_path)
+                set_data[path] = d
+        else:
+            set_data = doc
+            if '_id' in set_data:
+                del(set_data['_id'])
+
+        # Determine if any changed items were actually unset.
+        for path, value in set_data.items():
+            if value or isinstance(value, (numbers.Number, bool)):
+                continue
+
+            # If we've set a value that ain't the default value dont unset it.
+            default = None
+            if (self._dynamic and len(parts) and parts[0] in
+               self._dynamic_fields):
+                del(set_data[path])
+                unset_data[path] = 1
+                continue
+            elif path in self._fields:
+                default = self._fields[path].default
+            else:  # Perform a full lookup for lists / embedded lookups
+                d = self
+                parts = path.split('.')
+                db_field_name = parts.pop()
+                for p in parts:
+                    if isinstance(d, list) and p.isdigit():
+                        d = d[int(p)]
+                    elif (hasattr(d, '__getattribute__') and
+                          not isinstance(d, dict)):
+                        real_path = d._reverse_db_field_map.get(p, p)
+                        d = getattr(d, real_path)
+                    else:
+                        d = d.get(p)
+
+                if hasattr(d, '_fields'):
+                    field_name = d._reverse_db_field_map.get(db_field_name,
+                                                             db_field_name)
+                    if field_name in d._fields:
+                        default = d._fields.get(field_name).default
+                    else:
+                        default = None
+
+            if default is not None:
+                if callable(default):
+                    default = default()
+
+            if default != value:
+                continue
+
+            del(set_data[path])
+            unset_data[path] = 1
+        return set_data, unset_data
+
+    @classmethod
+    def _get_collection_name(cls):
+        """Returns the collection name for this class.
+        """
+        return cls._meta.get('collection', None)
+
+    @classmethod
+    def _from_son(cls, son, _auto_dereference=True):
+        """Create an instance of a Document (subclass) from a PyMongo SON.
+        """
+
+        # get the class name from the document, falling back to the given
+        # class if unavailable
+        class_name = son.get('_cls', cls._class_name)
+        data = dict(("%s" % key, value) for key, value in son.iteritems())
+        if not UNICODE_KWARGS:
+            # python 2.6.4 and lower cannot handle unicode keys
+            # passed to class constructor example: cls(**data)
+            to_str_keys_recursive(data)
+
+        # Return correct subclass for document type
+        if class_name != cls._class_name:
+            cls = get_document(class_name)
+
+        changed_fields = []
+        errors_dict = {}
+
+        fields = cls._fields
+        if not _auto_dereference:
+            fields = copy.copy(fields)
+
+        for field_name, field in fields.iteritems():
+            field._auto_dereference = _auto_dereference
+            if field.db_field in data:
+                value = data[field.db_field]
+                try:
+                    data[field_name] = (value if value is None
+                                        else field.to_python(value))
+                    if field_name != field.db_field:
+                        del data[field.db_field]
+                except (AttributeError, ValueError), e:
+                    errors_dict[field_name] = e
+            elif field.default:
+                default = field.default
+                if callable(default):
+                    default = default()
+                if isinstance(default, BaseDocument):
+                    changed_fields.append(field_name)
+
+        if errors_dict:
+            errors = "\n".join(["%s - %s" % (k, v)
+                     for k, v in errors_dict.items()])
+            msg = ("Invalid data to create a `%s` instance.\n%s"
+                   % (cls._class_name, errors))
+            raise InvalidDocumentError(msg)
+
+        obj = cls(__auto_convert=False, **data)
+        obj._changed_fields = changed_fields
+        obj._created = False
+        if not _auto_dereference:
+            obj._fields = fields
+        return obj
+
+    @classmethod
+    def _build_index_specs(cls, meta_indexes):
+        """Generate and merge the full index specs
+        """
+
+        geo_indices = cls._geo_indices()
+        unique_indices = cls._unique_with_indexes()
+        index_specs = [cls._build_index_spec(spec)
+                       for spec in meta_indexes]
+
+        def merge_index_specs(index_specs, indices):
+            if not indices:
+                return index_specs
+
+            spec_fields = [v['fields']
+                           for k, v in enumerate(index_specs)]
+            # Merge unqiue_indexes with existing specs
+            for k, v in enumerate(indices):
+                if v['fields'] in spec_fields:
+                    index_specs[spec_fields.index(v['fields'])].update(v)
+                else:
+                    index_specs.append(v)
+            return index_specs
+
+        index_specs = merge_index_specs(index_specs, geo_indices)
+        index_specs = merge_index_specs(index_specs, unique_indices)
+        return index_specs
+
+    @classmethod
+    def _build_index_spec(cls, spec):
+        """Build a PyMongo index spec from a MongoEngine index spec.
+        """
+        if isinstance(spec, basestring):
+            spec = {'fields': [spec]}
+        elif isinstance(spec, (list, tuple)):
+            spec = {'fields': list(spec)}
+        elif isinstance(spec, dict):
+            spec = dict(spec)
+
+        index_list = []
+        direction = None
+
+        # Check to see if we need to include _cls
+        allow_inheritance = cls._meta.get('allow_inheritance',
+                                          ALLOW_INHERITANCE)
+        include_cls = allow_inheritance and not spec.get('sparse', False)
+
+        for key in spec['fields']:
+            # If inherited spec continue
+            if isinstance(key, (list, tuple)):
+                continue
+
+            # ASCENDING from +,
+            # DESCENDING from -
+            # GEO2D from *
+            direction = pymongo.ASCENDING
+            if key.startswith("-"):
+                direction = pymongo.DESCENDING
+            elif key.startswith("*"):
+                direction = pymongo.GEO2D
+            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('.')
+            if parts in (['pk'], ['id'], ['_id']):
+                key = '_id'
+                fields = []
+            else:
+                fields = cls._lookup_field(parts)
+                parts = [field if field == '_id' else field.db_field
+                         for field in fields]
+                key = '.'.join(parts)
+            index_list.append((key, direction))
+
+        # Don't add cls to a geo index
+        if include_cls and direction is not pymongo.GEO2D:
+            index_list.insert(0, ('_cls', 1))
+
+        spec['fields'] = index_list
+        if spec.get('sparse', False) and len(spec['fields']) > 1:
+            raise ValueError(
+                'Sparse indexes can only have one field in them. '
+                'See https://jira.mongodb.org/browse/SERVER-2193')
+
+        return spec
+
+    @classmethod
+    def _unique_with_indexes(cls, namespace=""):
+        """
+        Find and set unique indexes
+        """
+        unique_indexes = []
+        for field_name, field in cls._fields.items():
+            sparse = False
+            # Generate a list of indexes needed by uniqueness constraints
+            if field.unique:
+                field.required = True
+                unique_fields = [field.db_field]
+
+                # 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 = cls._lookup_field(parts)
+                        name_parts = [part.db_field for part in parts]
+                        unique_with.append('.'.join(name_parts))
+                        # Unique field should be required
+                        parts[-1].required = True
+                        sparse = (not sparse and
+                                  parts[-1].name not in cls.__dict__)
+                    unique_fields += unique_with
+
+                # Add the new index to the list
+                fields = [("%s%s" % (namespace, f), pymongo.ASCENDING)
+                         for f in unique_fields]
+                index = {'fields': fields, 'unique': True, 'sparse': sparse}
+                unique_indexes.append(index)
+
+            # Grab any embedded document field unique indexes
+            if (field.__class__.__name__ == "EmbeddedDocumentField" and
+                field.document_type != cls):
+                field_namespace = "%s." % field_name
+                doc_cls = field.document_type
+                unique_indexes += doc_cls._unique_with_indexes(field_namespace)
+
+        return unique_indexes
+
+    @classmethod
+    def _geo_indices(cls, inspected=None):
+        inspected = inspected or []
+        geo_indices = []
+        inspected.append(cls)
+
+        EmbeddedDocumentField = _import_class("EmbeddedDocumentField")
+        GeoPointField = _import_class("GeoPointField")
+
+        for field in cls._fields.values():
+            if not isinstance(field, (EmbeddedDocumentField, GeoPointField)):
+                continue
+            if hasattr(field, 'document_type'):
+                field_cls = field.document_type
+                if field_cls in inspected:
+                    continue
+                if hasattr(field_cls, '_geo_indices'):
+                    geo_indices += field_cls._geo_indices(inspected)
+            elif field._geo_index:
+                geo_indices.append({'fields':
+                                   [(field.db_field, pymongo.GEO2D)]})
+        return geo_indices
+
+    @classmethod
+    def _lookup_field(cls, 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:
+            # Handle ListField indexing:
+            if field_name.isdigit():
+                new_field = field.field
+                fields.append(field_name)
+                continue
+
+            if field is None:
+                # Look up first field from the document
+                if field_name == 'pk':
+                    # Deal with "primary key" alias
+                    field_name = cls._meta['id_field']
+                if field_name in cls._fields:
+                    field = cls._fields[field_name]
+                elif cls._dynamic:
+                    DynamicField = _import_class('DynamicField')
+                    field = DynamicField(db_field=field_name)
+                else:
+                    raise LookUpError('Cannot resolve field "%s"'
+                                      % field_name)
+            else:
+                ReferenceField = _import_class('ReferenceField')
+                GenericReferenceField = _import_class('GenericReferenceField')
+                if isinstance(field, (ReferenceField, GenericReferenceField)):
+                    raise LookUpError('Cannot perform join in mongoDB: %s' %
+                                      '__'.join(parts))
+                if hasattr(getattr(field, 'field', None), 'lookup_member'):
+                    new_field = field.field.lookup_member(field_name)
+                else:
+                   # Look up subfield on the previous field
+                    new_field = field.lookup_member(field_name)
+                if not new_field and isinstance(field, ComplexBaseField):
+                    fields.append(field_name)
+                    continue
+                elif not new_field:
+                    raise LookUpError('Cannot resolve field "%s"'
+                                      % field_name)
+                field = new_field  # update field to the new field type
+            fields.append(field)
+        return fields
+
+    @classmethod
+    def _translate_field_name(cls, field, sep='.'):
+        """Translate a field attribute name to a database field name.
+        """
+        parts = field.split(sep)
+        parts = [f.db_field for f in cls._lookup_field(parts)]
+        return '.'.join(parts)
+
+    def __set_field_display(self):
+        """Dynamically set the display value for a field with choices"""
+        for attr_name, field in self._fields.items():
+            if field.choices:
+                setattr(self,
+                        'get_%s_display' % attr_name,
+                        partial(self.__get_field_display, field=field))
+
+    def __get_field_display(self, field):
+        """Returns the display value for a choice field"""
+        value = getattr(self, field.name)
+        if field.choices and isinstance(field.choices[0], (list, tuple)):
+            return dict(field.choices).get(value, value)
+        return value

mongoengine/base/fields.py

@@ -0,0 +1,395 @@
+import operator
+import warnings
+import weakref
+
+from bson import DBRef, ObjectId
+
+from mongoengine.common import _import_class
+from mongoengine.errors import ValidationError
+
+from mongoengine.base.common import ALLOW_INHERITANCE
+from mongoengine.base.datastructures import BaseDict, BaseList
+
+__all__ = ("BaseField", "ComplexBaseField", "ObjectIdField")
+
+
+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.
+
+    .. 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, name=None, required=False, default=None,
+                 unique=False, unique_with=None, primary_key=False,
+                 validation=None, choices=None, verbose_name=None,
+                 help_text=None):
+        self.db_field = (db_field or name) if not primary_key else '_id'
+        if name:
+            msg = "Fields' 'name' attribute deprecated in favour of 'db_field'"
+            warnings.warn(msg, DeprecationWarning)
+        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.verbose_name = verbose_name
+        self.help_text = help_text
+
+        # 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. 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()
+
+        EmbeddedDocument = _import_class('EmbeddedDocument')
+        if isinstance(value, EmbeddedDocument) and value._instance is None:
+            value._instance = weakref.proxy(instance)
+        return value
+
+    def __set__(self, instance, value):
+        """Descriptor for assigning a value to a field in a document.
+        """
+        changed = False
+        if (self.name not in instance._data or
+           instance._data[self.name] != value):
+            changed = True
+            instance._data[self.name] = value
+        if changed and instance._initialised:
+            instance._mark_as_changed(self.name)
+
+    def error(self, message="", errors=None, field_name=None):
+        """Raises 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 prepare_query_value(self, op, value):
+        """Prepare a value that is being used in a query for PyMongo.
+        """
+        return value
+
+    def validate(self, value, clean=True):
+        """Perform validation on a value.
+        """
+        pass
+
+    def _validate(self, value, **kwargs):
+        Document = _import_class('Document')
+        EmbeddedDocument = _import_class('EmbeddedDocument')
+        # check choices
+        if self.choices:
+            is_cls = isinstance(value, (Document, EmbeddedDocument))
+            value_to_check = value.__class__ if is_cls else value
+            err_msg = 'an instance' if is_cls else 'one'
+            if isinstance(self.choices[0], (list, tuple)):
+                option_keys = [k for k, v in self.choices]
+                if value_to_check not in option_keys:
+                    msg = ('Value must be %s of %s' %
+                           (err_msg, unicode(option_keys)))
+                    self.error(msg)
+            elif value_to_check not in self.choices:
+                msg = ('Value must be %s of %s' %
+                       (err_msg, unicode(self.choices)))
+                self.error(msg)
+
+        # check validation argument
+        if self.validation is not None:
+            if callable(self.validation):
+                if not self.validation(value):
+                    self.error('Value does not match custom validation method')
+            else:
+                raise ValueError('validation argument for "%s" must be a '
+                                 'callable.' % self.name)
+
+        self.validate(value, **kwargs)
+
+
+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
+    __dereference = False
+
+    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')
+        dereference = (self._auto_dereference and
+                       (self.field is None or isinstance(self.field,
+                        (GenericReferenceField, ReferenceField))))
+
+        self._auto_dereference = instance._fields[self.name]._auto_dereference
+        if not self.__dereference and instance._initialised and dereference:
+            instance._data[self.name] = self._dereference(
+                instance._data.get(self.name), max_depth=1, instance=instance,
+                name=self.name
+            )
+
+        value = super(ComplexBaseField, self).__get__(instance, owner)
+
+        # Convert lists / values so we can watch for any changes on them
+        if (isinstance(value, (list, tuple)) and
+            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 (self._auto_dereference and instance._initialised and
+            isinstance(value, (BaseList, BaseDict))
+            and not value._dereferenced):
+            value = self._dereference(
+                value, max_depth=1, instance=instance, name=self.name
+            )
+            value._dereferenced = True
+            instance._data[self.name] = value
+
+        return value
+
+    def __set__(self, instance, value):
+        """Descriptor for assigning a value to a field in a document.
+        """
+        instance._data[self.name] = value
+        instance._mark_as_changed(self.name)
+
+    def to_python(self, value):
+        """Convert a MongoDB-compatible type to a Python type.
+        """
+        Document = _import_class('Document')
+
+        if isinstance(value, basestring):
+            return value
+
+        if hasattr(value, 'to_python'):
+            return value.to_python()
+
+        is_list = False
+        if not hasattr(value, 'items'):
+            try:
+                is_list = True
+                value = dict([(k, v) for k, v in enumerate(value)])
+            except TypeError:  # Not iterable return the value
+                return value
+
+        if self.field:
+            value_dict = dict([(key, self.field.to_python(item))
+                                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')
+                    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 k, v in sorted(value_dict.items(),
+                                         key=operator.itemgetter(0))]
+        return value_dict
+
+    def to_mongo(self, value):
+        """Convert a Python type to a MongoDB-compatible type.
+        """
+        Document = _import_class("Document")
+        EmbeddedDocument = _import_class("EmbeddedDocument")
+        GenericReferenceField = _import_class("GenericReferenceField")
+
+        if isinstance(value, basestring):
+            return value
+
+        if hasattr(value, 'to_mongo'):
+            if isinstance(value, Document):
+                return GenericReferenceField().to_mongo(value)
+            cls = value.__class__
+            val = value.to_mongo()
+            # If we its a document thats 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 = dict([(k, v) for k, v in enumerate(value)])
+            except TypeError:  # Not iterable return the value
+                return value
+
+        if self.field:
+            value_dict = dict([(key, self.field.to_mongo(item))
+                                for key, item in value.iteritems()])
+        else:
+            value_dict = {}
+            for k, v in value.iteritems():
+                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', ALLOW_INHERITANCE)
+                        == True)
+                    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()
+                    # If we its a document thats 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)
+
+        if is_list:  # Convert back to a list
+            return [v for k, 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, 'iteritems') or hasattr(value, 'items'):
+                sequence = value.iteritems()
+            else:
+                sequence = enumerate(value)
+            for k, v in sequence:
+                try:
+                    self.field._validate(v)
+                except ValidationError, error:
+                    errors[k] = error.errors or error
+                except (ValueError, AssertionError), error:
+                    errors[k] = error
+
+            if errors:
+                field_class = self.field.__class__.__name__
+                self.error('Invalid %s item (%s)' % (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
+
+    def _get_owner_document(self, owner_document):
+        self._owner_document = owner_document
+
+    owner_document = property(_get_owner_document, _set_owner_document)
+
+    @property
+    def _dereference(self,):
+        if not self.__dereference:
+            DeReference = _import_class("DeReference")
+            self.__dereference = DeReference()  # Cached
+        return self.__dereference
+
+
+class ObjectIdField(BaseField):
+    """A field wrapper around MongoDB's ObjectIds.
+    """
+
+    def to_python(self, value):
+        if not isinstance(value, ObjectId):
+            value = ObjectId(value)
+        return value
+
+    def to_mongo(self, value):
+        if not isinstance(value, ObjectId):
+            try:
+                return ObjectId(unicode(value))
+            except Exception, e:
+                # e.message attribute has been deprecated since Python 2.6
+                self.error(unicode(e))
+        return value
+
+    def prepare_query_value(self, op, value):
+        return self.to_mongo(value)
+
+    def validate(self, value):
+        try:
+            ObjectId(unicode(value))
+        except:
+            self.error('Invalid Object ID')

mongoengine/base/metaclasses.py

@@ -0,0 +1,396 @@
+import warnings
+
+import pymongo
+
+from mongoengine.common import _import_class
+from mongoengine.errors import InvalidDocumentError
+from mongoengine.python_support import PY3
+from mongoengine.queryset import (DO_NOTHING, DoesNotExist,
+                                  MultipleObjectsReturned,
+                                  QuerySet, QuerySetManager)
+
+from mongoengine.base.common import _document_registry, ALLOW_INHERITANCE
+from mongoengine.base.fields import BaseField, ComplexBaseField, ObjectIdField
+
+__all__ = ('DocumentMetaclass', 'TopLevelDocumentMetaclass')
+
+
+class DocumentMetaclass(type):
+    """Metaclass for all documents.
+    """
+
+    def __new__(cls, name, bases, attrs):
+        flattened_bases = cls._get_bases(bases)
+        super_new = super(DocumentMetaclass, cls).__new__
+
+        # If a base class just call super
+        metaclass = attrs.get('my_metaclass')
+        if metaclass and issubclass(metaclass, DocumentMetaclass):
+            return super_new(cls, name, bases, attrs)
+
+        attrs['_is_document'] = attrs.get('_is_document', False)
+
+        # 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
+
+        # 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__.iteritems():
+                    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.iteritems():
+            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'] = dict([(k, getattr(v, 'db_field', k))
+                                      for k, v in doc_fields.iteritems()])
+        attrs['_fields_ordered'] = tuple(i[1] for i in sorted(
+                                         (v.creation_counter, v.name)
+                                         for v in doc_fields.itervalues()))
+        attrs['_reverse_db_field_map'] = dict(
+            (v, k) for k, v in attrs['_db_field_map'].iteritems())
+
+        #
+        # 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 heirarchy 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',
+                                                   ALLOW_INHERITANCE)
+                if (allow_inheritance is not True and
+                   not base._meta.get('abstract')):
+                    raise ValueError('Document %s may not be subclassed' %
+                                     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(cls, 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
+
+        # Handle delete rules
+        Document, EmbeddedDocument, DictField = cls._import_classes()
+        for field in new_class._fields.itervalues():
+            f = field
+            f.owner_document = new_class
+            delete_rule = getattr(f, 'reverse_delete_rule', DO_NOTHING)
+            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)
+
+        if issubclass(new_class, Document):
+            new_class._collection = None
+
+        # Add class to the _document_registry
+        _document_registry[new_class._class_name] = new_class
+
+        # In Python 2, User-defined methods objects have special read-only
+        # attributes 'im_func' and 'im_self' which contain the function obj
+        # and class instance object respectively.  With Python 3 these special
+        # attributes have been replaced by __func__ and __self__.  The Blinker
+        # module continues to use im_func and im_self, so the code below
+        # copies __func__ into im_func and __self__ into im_self for
+        # classmethod objects in Document derived classes.
+        if PY3:
+            for key, val in new_class.__dict__.items():
+                if isinstance(val, classmethod):
+                    f = val.__get__(new_class)
+                    if hasattr(f, '__func__') and not hasattr(f, 'im_func'):
+                        f.__dict__.update({'im_func': getattr(f, '__func__')})
+                    if hasattr(f, '__self__') and not hasattr(f, 'im_self'):
+                        f.__dict__.update({'im_self': getattr(f, '__self__')})
+
+        return new_class
+
+    def add_to_class(self, name, value):
+        setattr(self, name, value)
+
+    @classmethod
+    def _get_bases(cls, bases):
+        if isinstance(bases, BasesTuple):
+            return bases
+        seen = []
+        bases = cls.__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(cls, bases):
+        for base in bases:
+            if base is object:
+                continue
+            yield base
+            for child_base in cls.__get_bases(base.__bases__):
+                yield child_base
+
+    @classmethod
+    def _import_classes(cls):
+        Document = _import_class('Document')
+        EmbeddedDocument = _import_class('EmbeddedDocument')
+        DictField = _import_class('DictField')
+        return (Document, EmbeddedDocument, DictField)
+
+
+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):
+        flattened_bases = cls._get_bases(bases)
+        super_new = super(TopLevelDocumentMetaclass, cls).__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_drop_dups': False,
+                'index_opts': None,
+                'delete_rules': None,
+                'allow_inheritance': None,
+            }
+            attrs['_is_base_cls'] = True
+            attrs['_meta'].update(attrs.get('meta', {}))
+        else:
+            attrs['_meta'] = attrs.get('meta', {})
+            # Explictly 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(cls, 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 (direct subclasses of Document)
+        # may set 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'] == 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(cls, 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 one has been set
+        manager = attrs.get('objects', QuerySetManager())
+        new_class.objects = manager
+
+        # Validate the fields and set primary key if needed
+        for field_name, field in new_class._fields.iteritems():
+            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
+
+        # Set primary key if not defined by the document
+        if not new_class._meta.get('id_field'):
+            new_class._meta['id_field'] = 'id'
+            new_class._fields['id'] = ObjectIdField(db_field='_id')
+            new_class._fields['id'].name = 'id'
+            new_class.id = new_class._fields['id']
+
+        # 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 new exception and set to new_class
+            exception = type(name, parents, {'__module__': module})
+            setattr(new_class, name, exception)
+
+        return new_class
+
+
+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.iteritems():
+            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/common.py

@@ -0,0 +1,36 @@
+_class_registry_cache = {}
+
+
+def _import_class(cls_name):
+    """Cached mechanism for imports"""
+    if cls_name in _class_registry_cache:
+        return _class_registry_cache.get(cls_name)
+
+    doc_classes = ('Document', 'DynamicEmbeddedDocument', 'EmbeddedDocument',
+                   'MapReduceDocument')
+    field_classes = ('DictField', 'DynamicField', 'EmbeddedDocumentField',
+                     'FileField', 'GenericReferenceField',
+                     'GenericEmbeddedDocumentField', 'GeoPointField',
+                     'ReferenceField', 'StringField', 'ComplexBaseField')
+    queryset_classes = ('OperationError',)
+    deref_classes = ('DeReference',)
+
+    if 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 queryset_classes:
+        from mongoengine import queryset as module
+        import_classes = queryset_classes
+    elif cls_name in deref_classes:
+        from mongoengine import dereference as module
+        import_classes = deref_classes
+    else:
+        raise ValueError('No import set for: ' % 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 +1,169 @@
-from pymongo import Connection
+import pymongo
+from pymongo import Connection, ReplicaSetConnection, uri_parser
 
 
-__all__ = ['ConnectionError', 'connect']
+__all__ = ['ConnectionError', 'connect', 'register_connection',
+           'DEFAULT_CONNECTION_NAME']
 
 
-_connection_settings = {
+DEFAULT_CONNECTION_NAME = 'default'
-    'host': 'localhost',
-    'port': 27017,
-}
-_connection = None
-
-_db_name = None
-_db_username = None
-_db_password = None
-_db = None
 
 
 class ConnectionError(Exception):
     pass
 
 
-def _get_connection():
+_connection_settings = {}
-    global _connection
+_connections = {}
-    # Connect to the database if not already connected
+_dbs = {}
-    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:
+def register_connection(alias, name, host='localhost', port=27017,
-        # _db_name will be None if the user hasn't called connect()
+                        is_slave=False, read_preference=False, slaves=None,
-        if _db_name is None:
+                        username=None, password=None, **kwargs):
-            raise ConnectionError('Not connected to the database')
+    """Add a connection.
 
-        # Get DB from current connection and authenticate if necessary
+    :param alias: the name that will be used to refer to this connection
-        _db = _connection[_db_name]
+        throughout MongoEngine
-        if _db_username and _db_password:
+    :param name: the name of the specific database to use
-            _db.authenticate(_db_username, _db_password)
+    :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 is_slave: whether the connection can act as a slave
+      ** Depreciated pymongo 2.0.1+
+    :param read_preference: The read preference for the collection
+       ** Added pymongo 2.1
+    :param slaves: a list of aliases of slave connections; each of these must
+        be a registered connection that has :attr:`is_slave` set to ``True``
+    :param username: username to authenticate with
+    :param password: password to authenticate with
+    :param kwargs: allow ad-hoc parameters to be passed into the pymongo driver
 
-    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.
     """
-    global _connection_settings, _db_name, _db_username, _db_password
+    global _connection_settings
-    _connection_settings.update(kwargs)
+
-    _db_name = db
+    conn_settings = {
-    _db_username = username
+        'name': name,
-    _db_password = password
+        'host': host,
-    return _get_db()
+        'port': port,
+        'is_slave': is_slave,
+        'slaves': slaves or [],
+        'username': username,
+        'password': password,
+        'read_preference': read_preference
+    }
+
+    # Handle uri style connections
+    if "://" in host:
+        uri_dict = uri_parser.parse_uri(host)
+        if uri_dict.get('database') is None:
+            raise ConnectionError("If using URI style connection include "\
+                                  "database name in string")
+        conn_settings.update({
+            'host': host,
+            'name': uri_dict.get('database'),
+            'username': uri_dict.get('username'),
+            'password': uri_dict.get('password'),
+            'read_preference': read_preference,
+        })
+        if "replicaSet" in host:
+            conn_settings['replicaSet'] = True
+
+    conn_settings.update(kwargs)
+    _connection_settings[alias] = conn_settings
+
+
+def disconnect(alias=DEFAULT_CONNECTION_NAME):
+    global _connections
+    global _dbs
+
+    if alias in _connections:
+        get_connection(alias=alias).disconnect()
+        del _connections[alias]
+    if alias in _dbs:
+        del _dbs[alias]
+
+
+def get_connection(alias=DEFAULT_CONNECTION_NAME, reconnect=False):
+    global _connections
+    # Connect to the database if not already connected
+    if reconnect:
+        disconnect(alias)
+
+    if alias not in _connections:
+        if alias not in _connection_settings:
+            msg = 'Connection with alias "%s" has not been defined' % alias
+            if alias == DEFAULT_CONNECTION_NAME:
+                msg = 'You have not defined a default connection'
+            raise ConnectionError(msg)
+        conn_settings = _connection_settings[alias].copy()
+
+        if hasattr(pymongo, 'version_tuple'):  # Support for 2.1+
+            conn_settings.pop('name', None)
+            conn_settings.pop('slaves', None)
+            conn_settings.pop('is_slave', None)
+            conn_settings.pop('username', None)
+            conn_settings.pop('password', None)
+        else:
+            # Get all the slave connections
+            if 'slaves' in conn_settings:
+                slaves = []
+                for slave_alias in conn_settings['slaves']:
+                    slaves.append(get_connection(slave_alias))
+                conn_settings['slaves'] = slaves
+                conn_settings.pop('read_preference', None)
+
+        connection_class = Connection
+        if 'replicaSet' in conn_settings:
+            conn_settings['hosts_or_uri'] = conn_settings.pop('host', None)
+            # Discard port since it can't be used on ReplicaSetConnection
+            conn_settings.pop('port', None)
+            # Discard replicaSet if not base string
+            if not isinstance(conn_settings['replicaSet'], basestring):
+                conn_settings.pop('replicaSet', None)
+            connection_class = ReplicaSetConnection
+
+        try:
+            _connections[alias] = connection_class(**conn_settings)
+        except Exception, e:
+            raise ConnectionError("Cannot connect to database %s :\n%s" % (alias, e))
+    return _connections[alias]
+
+
+def get_db(alias=DEFAULT_CONNECTION_NAME, reconnect=False):
+    global _dbs
+    if reconnect:
+        disconnect(alias)
+
+    if alias not in _dbs:
+        conn = get_connection(alias)
+        conn_settings = _connection_settings[alias]
+        _dbs[alias] = conn[conn_settings['name']]
+        # Authenticate if necessary
+        if conn_settings['username'] and conn_settings['password']:
+            _dbs[alias].authenticate(conn_settings['username'],
+                                     conn_settings['password'])
+    return _dbs[alias]
+
+
+def connect(db, 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`.
+
+    .. versionchanged:: 0.6 - added multiple database support.
+    """
+    global _connections
+    if alias not in _connections:
+        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 +1,194 @@
+from mongoengine.common import _import_class
+from mongoengine.connection import DEFAULT_CONNECTION_NAME, get_db
+from mongoengine.queryset import OperationError, QuerySet
+
+__all__ = ("switch_db", "switch_collection", "no_dereference", "query_counter")
+
+
+class switch_db(object):
+    """ 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(object):
+    """ 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(object):
+    """ 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.iteritems()
+                             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 QuerySetNoDeRef(QuerySet):
+    """Special no_dereference QuerySet"""
+    def __dereference(items, max_depth=1, instance=None, name=None):
+            return items
+
+
+class query_counter(object):
+    """ Query_counter context manager to get the number of queries. """
+
+    def __init__(self):
+        """ Construct the query_counter. """
+        self.counter = 0
+        self.db = get_db()
+
+    def __enter__(self):
+        """ On every with block we need to drop the profile collection. """
+        self.db.set_profiling_level(0)
+        self.db.system.profile.drop()
+        self.db.set_profiling_level(2)
+        return self
+
+    def __exit__(self, t, value, traceback):
+        """ Reset the profiling level. """
+        self.db.set_profiling_level(0)
+
+    def __eq__(self, value):
+        """ == Compare querycounter. """
+        return value == self._get_count()
+
+    def __ne__(self, value):
+        """ != Compare querycounter. """
+        return not self.__eq__(value)
+
+    def __lt__(self, value):
+        """ < Compare querycounter. """
+        return self._get_count() < value
+
+    def __le__(self, value):
+        """ <= Compare querycounter. """
+        return self._get_count() <= value
+
+    def __gt__(self, value):
+        """ > Compare querycounter. """
+        return self._get_count() > value
+
+    def __ge__(self, value):
+        """ >= Compare querycounter. """
+        return self._get_count() >= value
+
+    def __int__(self):
+        """ int representation. """
+        return self._get_count()
+
+    def __repr__(self):
+        """ repr query_counter as the number of queries. """
+        return u"%s" % self._get_count()
+
+    def _get_count(self):
+        """ Get the number of queries. """
+        count = self.db.system.profile.find().count() - self.counter
+        self.counter += 1
+        return count

mongoengine/dereference.py

@@ -0,0 +1,215 @@
+from bson import DBRef, SON
+
+from base import (BaseDict, BaseList, TopLevelDocumentMetaclass, get_document)
+from fields import (ReferenceField, ListField, DictField, MapField)
+from connection import get_db
+from queryset import QuerySet
+from document import Document
+
+
+class DeReference(object):
+
+    def __call__(self, items, max_depth=1, instance=None, name=None):
+        """
+        Cheaply dereferences the items to a set depth.
+        Also handles the convertion 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, basestring):
+            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, TopLevelDocumentMetaclass)):
+            doc_type = instance._fields.get(name)
+            if 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:
+                    if not hasattr(items, 'items'):
+                        items = [field.to_python(v)
+                             if not isinstance(v, (DBRef, Document)) else v
+                             for v in items]
+                    else:
+                        items = dict([
+                            (k, field.to_python(v))
+                            if not isinstance(v, (DBRef, Document)) else (k, v)
+                            for k, v in items.iteritems()]
+                        )
+
+        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 not hasattr(items, 'items'):
+            iterator = enumerate(items)
+        else:
+            iterator = items.iteritems()
+
+        # Recursively find dbreferences
+        depth += 1
+        for k, item in iterator:
+            if isinstance(item, Document):
+                for field_name, field in item._fields.iteritems():
+                    v = item._data.get(field_name, None)
+                    if isinstance(v, (DBRef)):
+                        reference_map.setdefault(field.document_type, []).append(v.id)
+                    elif isinstance(v, (dict, SON)) and '_ref' in v:
+                        reference_map.setdefault(get_document(v['_cls']), []).append(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.iteritems():
+                            if isinstance(field_cls, (Document, TopLevelDocumentMetaclass)):
+                                key = field_cls
+                            reference_map.setdefault(key, []).extend(refs)
+            elif isinstance(item, (DBRef)):
+                reference_map.setdefault(item.collection, []).append(item.id)
+            elif isinstance(item, (dict, SON)) and '_ref' in item:
+                reference_map.setdefault(get_document(item['_cls']), []).append(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.iteritems():
+                    reference_map.setdefault(key, []).extend(refs)
+
+        return reference_map
+
+    def _fetch_objects(self, doc_type=None):
+        """Fetch all references and convert to their document objects
+        """
+        object_map = {}
+        for col, dbrefs in self.reference_map.iteritems():
+            keys = object_map.keys()
+            refs = list(set([dbref for dbref in dbrefs if unicode(dbref).encode('utf-8') not in keys]))
+            if hasattr(col, 'objects'):  # We have a document class for the refs
+                references = col.objects.in_bulk(refs)
+                for key, doc in references.iteritems():
+                    object_map[key] = doc
+            else:  # Generic reference: use the refs data to convert to document
+                if isinstance(doc_type, (ListField, DictField, MapField,)):
+                    continue
+
+                if doc_type:
+                    references = doc_type._get_db()[col].find({'_id': {'$in': refs}})
+                    for ref in references:
+                        doc = doc_type._from_son(ref)
+                        object_map[doc.id] = doc
+                else:
+                    references = get_db()[col].find({'_id': {'$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 col.split('_')))._from_son(ref)
+                        else:
+                            doc = doc_type._from_son(ref)
+                        object_map[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'].id, items)
+            elif '_cls' in items:
+                doc = get_document(items['_cls'])._from_son(items)
+                doc._data = self._attach_objects(doc._data, depth, doc, None)
+                return doc
+
+        if not hasattr(items, 'items'):
+            is_list = True
+            as_tuple = isinstance(items, tuple)
+            iterator = enumerate(items)
+            data = []
+        else:
+            is_list = False
+            iterator = items.iteritems()
+            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):
+                for field_name, field in v._fields.iteritems():
+                    v = data[k]._data.get(field_name, None)
+                    if isinstance(v, (DBRef)):
+                        data[k]._data[field_name] = self.object_map.get(v.id, v)
+                    elif isinstance(v, (dict, SON)) and '_ref' in v:
+                        data[k]._data[field_name] = self.object_map.get(v['_ref'].id, v)
+                    elif isinstance(v, dict) and depth <= self.max_depth:
+                        data[k]._data[field_name] = self._attach_objects(v, depth, instance=instance, name=name)
+                    elif isinstance(v, (list, tuple)) and depth <= self.max_depth:
+                        data[k]._data[field_name] = self._attach_objects(v, depth, instance=instance, name=name)
+            elif isinstance(v, (dict, list, tuple)) and depth <= self.max_depth:
+                data[k] = self._attach_objects(v, depth - 1, instance=instance, name=name)
+            elif hasattr(v, 'id'):
+                data[k] = self.object_map.get(v.id, v)
+
+        if instance and name:
+            if is_list:
+                return tuple(data) if as_tuple else BaseList(data, instance, name)
+            return BaseDict(data, instance, name)
+        depth += 1
+        return data

mongoengine/django/auth.py

@@ -1,36 +1,223 @@
 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 _user_get_all_permissions
+from django.contrib.auth.models import _user_has_perm
+from django.db import models
+from django.contrib.contenttypes.models import ContentTypeManager
+from django.contrib import auth
 from django.contrib.auth.models import AnonymousUser
+from django.utils.translation import ugettext_lazy as _
 
-import datetime
+try:
+    from django.contrib.auth.hashers import check_password, make_password
+except ImportError:
+    """Handle older versions of Django"""
+    from django.utils.hashcompat import md5_constructor, sha_constructor
+
+    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')
+
+    def check_password(raw_password, password):
+        algo, salt, hash = password.split('$')
+        return hash == get_hexdigest(algo, salt, raw_password)
+
+    def make_password(raw_password):
+        from random import random
+        algo = 'sha1'
+        salt = get_hexdigest(algo, str(random()), str(random()))[:5]
+        hash = get_hexdigest(algo, salt, raw_password)
+        return '%s$%s$%s' % (algo, salt, hash)
+
+from .utils import datetime_now
 
 REDIRECT_FIELD_NAME = 'next'
 
-def get_hexdigest(algorithm, salt, raw_password):
+class ContentType(Document):
-    raw_password, salt = smart_str(raw_password), smart_str(salt)
+    name = StringField(max_length=100)
-    if algorithm == 'md5':
+    app_label = StringField(max_length=100)
-        return md5_constructor(salt + raw_password).hexdigest()
+    model = StringField(max_length=100, verbose_name=_('python model class name'),
-    elif algorithm == 'sha1':
+        unique_with='app_label')
-        return sha_constructor(salt + raw_password).hexdigest()
+    objects = ContentTypeManager()
-    raise ValueError('Got unknown password algorithm type in password')
+
+    class Meta:
+        verbose_name = _('content type')
+        verbose_name_plural = _('content types')
+        # db_table = 'django_content_type'
+        # ordering = ('name',)
+        # unique_together = (('app_label', 'model'),)
+
+    def __unicode__(self):
+        return self.name
+
+    def model_class(self):
+        "Returns the Python model class for this type of content."
+        from django.db import models
+        return models.get_model(self.app_label, self.model)
+
+    def get_object_for_this_type(self, **kwargs):
+        """
+        Returns an object of this type for the keyword arguments given.
+        Basically, this is a proxy around this object_type's get_object() model
+        method. The ObjectNotExist exception, if thrown, will not be caught,
+        so code that calls this method should catch it.
+        """
+        return self.model_class()._default_manager.using(self._state.db).get(**kwargs)
+
+    def natural_key(self):
+        return (self.app_label, self.model)
+
+class SiteProfileNotAvailable(Exception):
+    pass
+
+class PermissionManager(models.Manager):
+    def get_by_natural_key(self, codename, app_label, model):
+        return self.get(
+            codename=codename,
+            content_type=ContentType.objects.get_by_natural_key(app_label, model)
+        )
+
+class Permission(Document):
+    """The permissions system provides a way to assign permissions to specific users and groups of users.
+
+    The permission system is used by the Django admin site, but may also be useful in your own code. The Django admin site uses permissions as follows:
+
+        - The "add" permission limits the user's ability to view the "add" form and add an object.
+        - The "change" permission limits a user's ability to view the change list, view the "change" form and change an object.
+        - The "delete" permission limits the ability to delete an object.
+
+    Permissions are set globally per type of object, not per specific object instance. It is possible to say "Mary may change news stories," but it's not currently possible to say "Mary may change news stories, but only the ones she created herself" or "Mary may only change news stories that have a certain status or publication date."
+
+    Three basic permissions -- add, change and delete -- are automatically created for each Django model.
+    """
+    name = StringField(max_length=50, verbose_name=_('username'))
+    content_type = ReferenceField(ContentType)
+    codename = StringField(max_length=100, verbose_name=_('codename'))
+        # FIXME: don't access field of the other class
+        # unique_with=['content_type__app_label', 'content_type__model'])
+
+    objects = PermissionManager()
+
+    class Meta:
+        verbose_name = _('permission')
+        verbose_name_plural = _('permissions')
+        # unique_together = (('content_type', 'codename'),)
+        # ordering = ('content_type__app_label', 'content_type__model', 'codename')
+
+    def __unicode__(self):
+        return u"%s | %s | %s" % (
+            unicode(self.content_type.app_label),
+            unicode(self.content_type),
+            unicode(self.name))
+
+    def natural_key(self):
+        return (self.codename,) + self.content_type.natural_key()
+    natural_key.dependencies = ['contenttypes.contenttype']
+
+class Group(Document):
+    """Groups are a generic way of categorizing users to apply permissions, or some other label, to those users. A user can belong to any number of groups.
+
+    A user in a group automatically has all the permissions granted to that group. For example, if the group Site editors has the permission can_edit_home_page, any user in that group will have that permission.
+
+    Beyond permissions, groups are a convenient way to categorize users to apply some label, or extended functionality, to them. For example, you could create a group 'Special users', and you could write code that would do special things to those users -- such as giving them access to a members-only portion of your site, or sending them members-only e-mail messages.
+    """
+    name = StringField(max_length=80, unique=True, verbose_name=_('name'))
+    # permissions = models.ManyToManyField(Permission, verbose_name=_('permissions'), blank=True)
+    permissions = ListField(ReferenceField(Permission, verbose_name=_('permissions'), required=False))
+
+    class Meta:
+        verbose_name = _('group')
+        verbose_name_plural = _('groups')
+
+    def __unicode__(self):
+        return self.name
+
+class UserManager(models.Manager):
+    def create_user(self, username, email, password=None):
+        """
+        Creates and saves a User with the given username, e-mail and password.
+        """
+        now = datetime_now()
+
+        # Normalize the address by lowercasing the domain part of the email
+        # address.
+        try:
+            email_name, domain_part = email.strip().split('@', 1)
+        except ValueError:
+            pass
+        else:
+            email = '@'.join([email_name, domain_part.lower()])
+
+        user = self.model(username=username, email=email, is_staff=False,
+                         is_active=True, is_superuser=False, last_login=now,
+                         date_joined=now)
+
+        user.set_password(password)
+        user.save(using=self._db)
+        return user
+
+    def create_superuser(self, username, email, password):
+        u = self.create_user(username, email, password)
+        u.is_staff = True
+        u.is_active = True
+        u.is_superuser = True
+        u.save(using=self._db)
+        return u
+
+    def make_random_password(self, length=10, allowed_chars='abcdefghjkmnpqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ23456789'):
+        "Generates a random password with the given length and given allowed_chars"
+        # Note that default value of allowed_chars does not have "I" or letters
+        # that look like it -- just to avoid confusion.
+        from random import choice
+        return ''.join([choice(allowed_chars) for i in range(length)])
+
 
 
 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)
+    username = StringField(max_length=30, required=True,
-    first_name = StringField(max_length=30)
+                           verbose_name=_('username'),
-    last_name = StringField(max_length=30)
+                           help_text=_("Required. 30 characters or fewer. Letters, numbers and @/./+/-/_ characters"))
-    email = StringField()
+
-    password = StringField(max_length=128)
+    first_name = StringField(max_length=30,
-    is_staff = BooleanField(default=False)
+                             verbose_name=_('first name'))
-    is_active = BooleanField(default=True)
+
-    is_superuser = BooleanField(default=False)
+    last_name = StringField(max_length=30,
-    last_login = DateTimeField(default=datetime.datetime.now)
+                            verbose_name=_('last name'))
-    date_joined = DateTimeField(default=datetime.datetime.now)
+    email = EmailField(verbose_name=_('e-mail address'))
+    password = StringField(max_length=128,
+                           verbose_name=_('password'),
+                           help_text=_("Use '[algo]$[iterations]$[salt]$[hexdigest]' or use the <a href=\"password/\">change password form</a>."))
+    is_staff = BooleanField(default=False,
+                            verbose_name=_('staff status'),
+                            help_text=_("Designates whether the user can log into this admin site."))
+    is_active = BooleanField(default=True,
+                             verbose_name=_('active'),
+                             help_text=_("Designates whether this user should be treated as active. Unselect this instead of deleting accounts."))
+    is_superuser = BooleanField(default=False,
+                                verbose_name=_('superuser status'),
+                                help_text=_("Designates that this user has all permissions without explicitly assigning them."))
+    last_login = DateTimeField(default=datetime_now,
+                               verbose_name=_('last login'))
+    date_joined = DateTimeField(default=datetime_now,
+                                verbose_name=_('date joined'))
+
+    meta = {
+        'allow_inheritance': True,
+        'indexes': [
+            {'fields': ['username'], 'unique': True, 'sparse': True}
+        ]
+    }
+
+    def __unicode__(self):
+        return self.username
 
     def get_full_name(self):
         """Returns the users first and last names, separated by a space.
@@ -49,11 +236,7 @@ class User(Document):
         assigning to :attr:`~mongoengine.django.auth.User.password` as the
         password is hashed before storage.
         """
-        from random import random
+        self.password = make_password(raw_password)
-        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)
         self.save()
         return self
 
@@ -63,19 +246,36 @@ class User(Document):
         :attr:`~mongoengine.django.auth.User.password` as the password is
         hashed before storage.
         """
-        algo, salt, hash = self.password.split('$')
+        return check_password(raw_password, self.password)
-        return hash == get_hexdigest(algo, salt, raw_password)
+
+    def get_all_permissions(self, obj=None):
+        return _user_get_all_permissions(self, obj)
+
+    def has_perm(self, perm, obj=None):
+        """
+        Returns True if the user has the specified permission. This method
+        queries all available auth backends, but returns immediately if any
+        backend returns True. Thus, a user who has permission from a single
+        auth backend is assumed to have permission in general. If an object is
+        provided, permissions for this specific object are checked.
+        """
+
+        # Active superusers have all permissions.
+        if self.is_active and self.is_superuser:
+            return True
+
+        # Otherwise we need to check the backends.
+        return _user_has_perm(self, perm, obj)
 
     @classmethod
     def create_user(cls, username, password, email=None):
         """Create (and save) a new user with the given username, password and
         email address.
         """
-        now = datetime.datetime.now()
+        now = datetime_now()
-        
+
         # Normalize the address by lowercasing the domain part of the email
         # address.
-        # Not sure why we'r allowing null email when its not allowed in django
         if email is not None:
             try:
                 email_name, domain_part = email.strip().split('@', 1)
@@ -83,24 +283,132 @@ class User(Document):
                 pass
             else:
                 email = '@'.join([email_name, domain_part.lower()])
-            
+
-        user = User(username=username, email=email, date_joined=now)
+        user = cls(username=username, email=email, date_joined=now)
         user.set_password(password)
         user.save()
         return user
-    
+
+    def get_all_permissions(self, obj=None):
+        permissions = set()
+        anon = self.is_anonymous()
+        for backend in auth.get_backends():
+            if not anon or backend.supports_anonymous_user:
+                if hasattr(backend, "get_all_permissions"):
+                    if obj is not None:
+                        if backend.supports_object_permissions:
+                            permissions.update(
+                                backend.get_all_permissions(user, obj)
+                            )
+                    else:
+                        permissions.update(backend.get_all_permissions(self))
+        return permissions
+
     def get_and_delete_messages(self):
         return []
 
+    def has_perm(self, perm, obj=None):
+        anon = self.is_anonymous()
+        active = self.is_active
+        for backend in auth.get_backends():
+            if (not active and not anon and backend.supports_inactive_user) or \
+                        (not anon or backend.supports_anonymous_user):
+                if hasattr(backend, "has_perm"):
+                    if obj is not None:
+                        if (backend.supports_object_permissions and
+                            backend.has_perm(self, perm, obj)):
+                                return True
+                    else:
+                        if backend.has_perm(self, perm):
+                            return True
+        return False
+
+    def has_perms(self, perm_list, obj=None):
+        """
+        Returns True if the user has each of the specified permissions.
+        If object is passed, it checks if the user has all required perms
+        for this object.
+        """
+        for perm in perm_list:
+            if not self.has_perm(perm, obj):
+                return False
+        return True
+
+    def has_module_perms(self, app_label):
+        anon = self.is_anonymous()
+        active = self.is_active
+        for backend in auth.get_backends():
+            if (not active and not anon and backend.supports_inactive_user) or \
+                        (not anon or backend.supports_anonymous_user):
+                if hasattr(backend, "has_module_perms"):
+                    if backend.has_module_perms(self, app_label):
+                        return True
+        return False
+
+    def get_and_delete_messages(self):
+        messages = []
+        for m in self.message_set.all():
+            messages.append(m.message)
+            m.delete()
+        return messages
+
+    def email_user(self, subject, message, from_email=None):
+        "Sends an e-mail to this User."
+        from django.core.mail import send_mail
+        send_mail(subject, message, from_email, [self.email])
+
+    def get_profile(self):
+        """
+        Returns site-specific profile for this user. Raises
+        SiteProfileNotAvailable if this site does not allow profiles.
+        """
+        if not hasattr(self, '_profile_cache'):
+            from django.conf import settings
+            if not getattr(settings, 'AUTH_PROFILE_MODULE', False):
+                raise SiteProfileNotAvailable('You need to set AUTH_PROFILE_MO'
+                                              'DULE in your project settings')
+            try:
+                app_label, model_name = settings.AUTH_PROFILE_MODULE.split('.')
+            except ValueError:
+                raise SiteProfileNotAvailable('app_label and model_name should'
+                        ' be separated by a dot in the AUTH_PROFILE_MODULE set'
+                        'ting')
+
+            try:
+                model = models.get_model(app_label, model_name)
+                if model is None:
+                    raise SiteProfileNotAvailable('Unable to load the profile '
+                        'model, check AUTH_PROFILE_MODULE in your project sett'
+                        'ings')
+                self._profile_cache = model._default_manager.using(self._state.db).get(user__id__exact=self.id)
+                self._profile_cache.user = self
+            except (ImportError, ImproperlyConfigured):
+                raise SiteProfileNotAvailable
+        return self._profile_cache
+
+    def _get_message_set(self):
+        import warnings
+        warnings.warn('The user messaging API is deprecated. Please update'
+                      ' your code to use the new messages framework.',
+                      category=DeprecationWarning)
+        return self._message_set
+    message_set = property(_get_message_set)
+
 
 class MongoEngineBackend(object):
     """Authenticate using MongoEngine and mongoengine.django.auth.User.
     """
 
+    supports_object_permissions = False
+    supports_anonymous_user = False
+    supports_inactive_user = False
+
     def authenticate(self, username=None, password=None):
         user = User.objects(username=username).first()
         if user:
             if password and user.check_password(password):
+                backend = auth.get_backends()[0]
+                user.backend = "%s.%s" % (backend.__module__, backend.__class__.__name__)
                 return user
         return None
 

mongoengine/django/sessions.py

@@ -1,3 +1,4 @@
+from django.conf import settings
 from django.contrib.sessions.backends.base import SessionBase, CreateError
 from django.core.exceptions import SuspiciousOperation
 from django.utils.encoding import force_unicode
@@ -5,16 +6,46 @@ from django.utils.encoding import force_unicode
 from mongoengine.document import Document
 from mongoengine import fields
 from mongoengine.queryset import OperationError
+from mongoengine.connection import DEFAULT_CONNECTION_NAME
 
-from datetime import datetime
+from .utils import datetime_now
+
+
+MONGOENGINE_SESSION_DB_ALIAS = getattr(
+    settings, 'MONGOENGINE_SESSION_DB_ALIAS',
+    DEFAULT_CONNECTION_NAME)
+
+# a setting for the name of the collection used to store sessions
+MONGOENGINE_SESSION_COLLECTION = getattr(
+    settings, 'MONGOENGINE_SESSION_COLLECTION',
+    'django_session')
+
+# a setting for whether session data is stored encoded or not
+MONGOENGINE_SESSION_DATA_ENCODE = getattr(
+    settings, 'MONGOENGINE_SESSION_DATA_ENCODE',
+    True)
 
 
 class MongoSession(Document):
     session_key = fields.StringField(primary_key=True, max_length=40)
-    session_data = fields.StringField()
+    session_data = fields.StringField() if MONGOENGINE_SESSION_DATA_ENCODE \
+                                        else fields.DictField()
     expire_date = fields.DateTimeField()
-    
+
-    meta = {'collection': 'django_session', 'allow_inheritance': False}
+    meta = {
+        'collection': MONGOENGINE_SESSION_COLLECTION,
+        'db_alias': MONGOENGINE_SESSION_DB_ALIAS,
+        'allow_inheritance': False,
+        'indexes': [
+            {
+                'fields': ['expire_date'],
+                'expireAfterSeconds': settings.SESSION_COOKIE_AGE
+            }
+        ]
+    }
+
+    def get_decoded(self):
+        return SessionStore().decode(self.session_data)
 
 
 class SessionStore(SessionBase):
@@ -24,8 +55,11 @@ class SessionStore(SessionBase):
     def load(self):
         try:
             s = MongoSession.objects(session_key=self.session_key,
-                                     expire_date__gt=datetime.now())[0]
+                                     expire_date__gt=datetime_now)[0]
-            return self.decode(force_unicode(s.session_data))
+            if MONGOENGINE_SESSION_DATA_ENCODE:
+                return self.decode(force_unicode(s.session_data))
+            else:
+                return s.session_data
         except (IndexError, SuspiciousOperation):
             self.create()
             return {}
@@ -35,7 +69,7 @@ class SessionStore(SessionBase):
 
     def create(self):
         while True:
-            self.session_key = self._get_new_session_key()
+            self._session_key = self._get_new_session_key()
             try:
                 self.save(must_create=True)
             except CreateError:
@@ -45,8 +79,13 @@ class SessionStore(SessionBase):
             return
 
     def save(self, must_create=False):
+        if self.session_key is None:
+            self._session_key = self._get_new_session_key()
         s = MongoSession(session_key=self.session_key)
-        s.session_data = self.encode(self._get_session(no_load=must_create))
+        if MONGOENGINE_SESSION_DATA_ENCODE:
+            s.session_data = self.encode(self._get_session(no_load=must_create))
+        else:
+            s.session_data = self._get_session(no_load=must_create)
         s.expire_date = self.get_expiry_date()
         try:
             s.save(force_insert=must_create, safe=True)

mongoengine/django/shortcuts.py

@@ -0,0 +1,47 @@
+from mongoengine.queryset import QuerySet
+from mongoengine.base import BaseDocument
+from mongoengine.errors import ValidationError
+
+def _get_queryset(cls):
+    """Inspired by django.shortcuts.*"""
+    if isinstance(cls, QuerySet):
+        return cls
+    else:
+        return cls.objects
+
+def get_document_or_404(cls, *args, **kwargs):
+    """
+    Uses get() to return an document, or raises a Http404 exception if the document
+    does not exist.
+
+    cls may be a Document or QuerySet object. All other passed
+    arguments and keyword arguments are used in the get() query.
+
+    Note: Like with get(), an MultipleObjectsReturned will be raised if more than one
+    object is found.
+
+    Inspired by django.shortcuts.*
+    """
+    queryset = _get_queryset(cls)
+    try:
+        return queryset.get(*args, **kwargs)
+    except (queryset._document.DoesNotExist, ValidationError):
+        from django.http import Http404
+        raise Http404('No %s matches the given query.' % queryset._document._class_name)
+
+def get_list_or_404(cls, *args, **kwargs):
+    """
+    Uses filter() to return a list of documents, or raise a Http404 exception if
+    the list is empty.
+
+    cls may be a Document or QuerySet object. All other passed
+    arguments and keyword arguments are used in the filter() query.
+
+    Inspired by django.shortcuts.*
+    """
+    queryset = _get_queryset(cls)
+    obj_list = list(queryset.filter(*args, **kwargs))
+    if not obj_list:
+        from django.http import Http404
+        raise Http404('No %s matches the given query.' % queryset._document._class_name)
+    return obj_list

mongoengine/django/storage.py

@@ -0,0 +1,112 @@
+import os
+import itertools
+import urlparse
+
+from mongoengine import *
+from django.conf import settings
+from django.core.files.storage import Storage
+from django.core.exceptions import ImproperlyConfigured
+
+
+class FileDocument(Document):
+    """A document used to store a single file in GridFS.
+    """
+    file = FileField()
+
+
+class GridFSStorage(Storage):
+    """A custom storage backend to store files in GridFS
+    """
+
+    def __init__(self, base_url=None):
+
+        if base_url is None:
+            base_url = settings.MEDIA_URL
+        self.base_url = base_url
+        self.document = FileDocument
+        self.field = 'file'
+
+    def delete(self, name):
+        """Deletes the specified file from the storage system.
+        """
+        if self.exists(name):
+            doc = self.document.objects.first()
+            field = getattr(doc, self.field)
+            self._get_doc_with_name(name).delete()  # Delete the FileField
+            field.delete()                          # Delete the FileDocument
+
+    def exists(self, name):
+        """Returns True if a file referened by the given name already exists in the
+        storage system, or False if the name is available for a new file.
+        """
+        doc = self._get_doc_with_name(name)
+        if doc:
+            field = getattr(doc, self.field)
+            return bool(field.name)
+        else:
+            return False
+
+    def listdir(self, path=None):
+        """Lists the contents of the specified path, returning a 2-tuple of lists;
+        the first item being directories, the second item being files.
+        """
+        def name(doc):
+            return getattr(doc, self.field).name
+        docs = self.document.objects
+        return [], [name(d) for d in docs if name(d)]
+
+    def size(self, name):
+        """Returns the total size, in bytes, of the file specified by name.
+        """
+        doc = self._get_doc_with_name(name)
+        if doc:
+            return getattr(doc, self.field).length
+        else:
+            raise ValueError("No such file or directory: '%s'" % name)
+
+    def url(self, name):
+        """Returns an absolute URL where the file's contents can be accessed
+        directly by a web browser.
+        """
+        if self.base_url is None:
+            raise ValueError("This file is not accessible via a URL.")
+        return urlparse.urljoin(self.base_url, name).replace('\\', '/')
+
+    def _get_doc_with_name(self, name):
+        """Find the documents in the store with the given name
+        """
+        docs = self.document.objects
+        doc = [d for d in docs if getattr(d, self.field).name == name]
+        if doc:
+            return doc[0]
+        else:
+            return None
+
+    def _open(self, name, mode='rb'):
+        doc = self._get_doc_with_name(name)
+        if doc:
+            return getattr(doc, self.field)
+        else:
+            raise ValueError("No file found with the name '%s'." % name)
+
+    def get_available_name(self, name):
+        """Returns a filename that's free on the target storage system, and
+        available for new content to be written to.
+        """
+        file_root, file_ext = os.path.splitext(name)
+        # If the filename already exists, add an underscore and a number (before
+        # the file extension, if one exists) to the filename until the generated
+        # filename doesn't exist.
+        count = itertools.count(1)
+        while self.exists(name):
+            # file_ext includes the dot.
+            name = os.path.join("%s_%s%s" % (file_root, count.next(), file_ext))
+
+        return name
+
+    def _save(self, name, content):
+        doc = self.document()
+        getattr(doc, self.field).put(content, filename=name)
+        doc.save()
+
+        return name

mongoengine/django/tests.py

@@ -0,0 +1,39 @@
+#coding: utf-8
+from nose.plugins.skip import SkipTest
+
+from mongoengine.python_support import PY3
+from mongoengine import connect
+
+try:
+    from django.test import TestCase
+    from django.conf import settings
+except Exception as err:
+    if PY3:
+        from unittest import TestCase
+        # Dummy value so no error
+        class settings:
+            MONGO_DATABASE_NAME = 'dummy'
+    else:
+        raise err
+
+
+class MongoTestCase(TestCase):
+
+    def setUp(self):
+        if PY3:
+            raise SkipTest('django does not have Python 3 support')
+
+    """
+    TestCase class that clear the collection between the tests
+    """
+    db_name = 'test_%s' % settings.MONGO_DATABASE_NAME
+    def __init__(self, methodName='runtest'):
+        self.db = connect(self.db_name).get_db()
+        super(MongoTestCase, self).__init__(methodName)
+
+    def _post_teardown(self):
+        super(MongoTestCase, self)._post_teardown()
+        for collection in self.db.collection_names():
+            if collection == 'system.indexes':
+                continue
+            self.db.drop_collection(collection)

mongoengine/django/utils.py

@@ -0,0 +1,6 @@
+try:
+    # django >= 1.4
+    from django.utils.timezone import now as datetime_now
+except ImportError:
+    from datetime import datetime
+    datetime_now = datetime.now

mongoengine/document.py

@@ -1,12 +1,25 @@
-from base import (DocumentMetaclass, TopLevelDocumentMetaclass, BaseDocument,
+from __future__ import with_statement
-                  ValidationError)
+import warnings
-from queryset import OperationError
-from connection import _get_db
 
 import pymongo
+import re
+
+from bson.dbref import DBRef
+from mongoengine import signals
+from mongoengine.base import (DocumentMetaclass, TopLevelDocumentMetaclass,
+                              BaseDocument, BaseDict, BaseList,
+                              ALLOW_INHERITANCE, get_document)
+from mongoengine.queryset import OperationError, NotUniqueError, QuerySet
+from mongoengine.connection import get_db, DEFAULT_CONNECTION_NAME
+from mongoengine.context_managers import switch_db, switch_collection
+
+__all__ = ('Document', 'EmbeddedDocument', 'DynamicDocument',
+           'DynamicEmbeddedDocument', 'OperationError',
+           'InvalidCollectionError', 'NotUniqueError', 'MapReduceDocument')
 
 
-__all__ = ['Document', 'EmbeddedDocument', 'ValidationError', 'OperationError']
+class InvalidCollectionError(Exception):
+    pass
 
 
 class EmbeddedDocument(BaseDocument):
@@ -14,10 +27,32 @@ class EmbeddedDocument(BaseDocument):
     collection.  :class:`~mongoengine.EmbeddedDocument`\ s should be used as
     fields on :class:`~mongoengine.Document`\ s through the
     :class:`~mongoengine.EmbeddedDocumentField` field type.
+
+    A :class:`~mongoengine.EmbeddedDocument` subclass may be itself subclassed,
+    to create a specialised version of the embedded document that will be
+    stored in the same collection. To facilitate this behaviour a `_cls`
+    field is added to documents (hidden though the MongoEngine interface).
+    To disable this behaviour and remove the dependence on the presence of
+    `_cls` set :attr:`allow_inheritance` to ``False`` in the :attr:`meta`
+    dictionary.
     """
-    
+
+    # The __metaclass__ attribute is removed by 2to3 when running with Python3
+    # my_metaclass is defined so that metaclass can be queried in Python 2 & 3
+    my_metaclass  = DocumentMetaclass
     __metaclass__ = DocumentMetaclass
 
+    _instance = None
+
+    def __init__(self, *args, **kwargs):
+        super(EmbeddedDocument, self).__init__(*args, **kwargs)
+        self._changed_fields = []
+
+    def __eq__(self, other):
+        if isinstance(other, self.__class__):
+            return self._data == other._data
+        return False
+
 
 class Document(BaseDocument):
     """The base class used for defining the structure and properties of
@@ -34,55 +69,280 @@ class Document(BaseDocument):
 
     A :class:`~mongoengine.Document` subclass may be itself subclassed, to
     create a specialised version of the document that will be stored in the
-    same collection. To facilitate this behaviour, `_cls` and `_types`
+    same collection. To facilitate this behaviour a `_cls`
-    fields are added to documents (hidden though the MongoEngine interface
+    field is added to documents (hidden though the MongoEngine interface).
-    though). To disable this behaviour and remove the dependence on the
+    To disable this behaviour and remove the dependence on the presence of
-    presence of `_cls` and `_types`, set :attr:`allow_inheritance` to
+    `_cls` set :attr:`allow_inheritance` to ``False`` in the :attr:`meta`
-    ``False`` in the :attr:`meta` dictionary.
+    dictionary.
 
-    A :class:`~mongoengine.Document` may use a **Capped Collection** by 
+    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 
+    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 
+    maximum size of the collection in bytes. If :attr:`max_size` is not
-    specified and :attr:`max_documents` is, :attr:`max_size` defaults to 
+    specified and :attr:`max_documents` is, :attr:`max_size` defaults to
     10000000 bytes (10MB).
 
     Indexes may be created by specifying :attr:`indexes` in the :attr:`meta`
-    dictionary. The value should be a list of field names or tuples of field 
+    dictionary. The value should be a list of field names or tuples of field
     names. Index direction may be specified by prefixing the field names with
     a **+** or **-** sign.
+
+    Automatic index creation can be disabled by specifying
+    attr:`auto_create_index` in the :attr:`meta` dictionary. If this is set to
+    False then indexes will not be created by MongoEngine.  This is useful in
+    production systems where index creation is performed as part of a
+    deployment system.
+
+    By default, _cls will be added to the start of every index (that
+    doesn't contain a list) if allow_inheritance is True. This can be
+    disabled by either setting cls to False on the specific index or
+    by setting index_cls to False on the meta dictionary for the document.
     """
 
+    # The __metaclass__ attribute is removed by 2to3 when running with Python3
+    # my_metaclass is defined so that metaclass can be queried in Python 2 & 3
+    my_metaclass  = TopLevelDocumentMetaclass
     __metaclass__ = TopLevelDocumentMetaclass
 
-    def save(self, safe=True, force_insert=False):
+    def pk():
+        """Primary key alias
+        """
+        def fget(self):
+            return getattr(self, self._meta['id_field'])
+
+        def fset(self, value):
+            return setattr(self, self._meta['id_field'], value)
+        return property(fget, fset)
+    pk = pk()
+
+    @classmethod
+    def _get_db(cls):
+        """Some Model using other db_alias"""
+        return get_db(cls._meta.get("db_alias", DEFAULT_CONNECTION_NAME))
+
+    @classmethod
+    def _get_collection(cls):
+        """Returns the collection for the document."""
+        if not hasattr(cls, '_collection') or cls._collection is None:
+            db = cls._get_db()
+            collection_name = cls._get_collection_name()
+            # Create collection as a capped collection if specified
+            if cls._meta['max_size'] or cls._meta['max_documents']:
+                # Get max document limit and max byte size from meta
+                max_size = cls._meta['max_size'] or 10000000  # 10MB default
+                max_documents = cls._meta['max_documents']
+
+                if collection_name in db.collection_names():
+                    cls._collection = db[collection_name]
+                    # The collection already exists, check if its capped
+                    # options match the specified capped options
+                    options = cls._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')
+                                % cls._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
+                    cls._collection = db.create_collection(
+                        collection_name, **opts
+                    )
+            else:
+                cls._collection = db[collection_name]
+            if cls._meta.get('auto_create_index', True):
+                cls.ensure_indexes()
+        return cls._collection
+
+    def save(self, safe=True, force_insert=False, validate=True, clean=True,
+             write_options=None,  cascade=None, cascade_kwargs=None,
+             _refs=None, **kwargs):
         """Save the :class:`~mongoengine.Document` to the database. If the
         document already exists, it will be updated, otherwise it will be
         created.
 
-        If ``safe=True`` and the operation is unsuccessful, an 
+        If ``safe=True`` and the operation is unsuccessful, an
         :class:`~mongoengine.OperationError` will be raised.
 
         :param safe: check if the operation succeeded before returning
-        :param force_insert: only try to create a new document, don't allow 
+        :param force_insert: only try to create a new document, don't allow
             updates of existing documents
+        :param validate: validates the document; set to ``False`` to skip.
+        :param clean: call the document clean method, requires `validate` to be
+            True.
+        :param write_options: Extra keyword arguments are passed down to
+            :meth:`~pymongo.collection.Collection.save` OR
+            :meth:`~pymongo.collection.Collection.insert`
+            which will be used as options for the resultant
+            ``getLastError`` command.  For example,
+            ``save(..., write_options={w: 2, fsync: True}, ...)`` will
+            wait until at least two servers have recorded the write and
+            will force an fsync on the primary server.
+        :param cascade: Sets the flag for cascading saves.  You can set a
+            default by setting "cascade" in the document __meta__
+        :param cascade_kwargs: optional kwargs dictionary to be passed throw
+            to cascading saves
+        :param _refs: A list of processed references used in cascading saves
+
+        .. versionchanged:: 0.5
+            In existing documents it only saves changed fields using
+            set / unset.  Saves are cascaded and any
+            :class:`~bson.dbref.DBRef` objects that have changes are
+            saved as well.
+        .. versionchanged:: 0.6
+            Cascade saves are optional = defaults to True, if you want
+            fine grain control then you can turn off using document
+            meta['cascade'] = False  Also you can pass different kwargs to
+            the cascade save using cascade_kwargs which overwrites the
+            existing kwargs with custom values
         """
-        self.validate()
+        signals.pre_save.send(self.__class__, document=self)
+
+        if validate:
+            self.validate(clean=clean)
+
+        if not write_options:
+            write_options = {}
+
         doc = self.to_mongo()
+
+        created = ('_id' not in doc or self._created or force_insert)
+
         try:
-            collection = self.__class__.objects._collection
+            collection = self._get_collection()
-            if force_insert:
+            if created:
-                object_id = collection.insert(doc, safe=safe)
+                if force_insert:
+                    object_id = collection.insert(doc, safe=safe,
+                                                  **write_options)
+                else:
+                    object_id = collection.save(doc, safe=safe,
+                                                **write_options)
             else:
-                object_id = collection.save(doc, safe=safe)
+                object_id = doc['_id']
+                updates, removals = self._delta()
+                # Need to add shard key to query, or you get an error
+                select_dict = {'_id': object_id}
+                shard_key = self.__class__._meta.get('shard_key', tuple())
+                for k in shard_key:
+                    actual_key = self._db_field_map.get(k, k)
+                    select_dict[actual_key] = doc[actual_key]
+
+                def is_new_object(last_error):
+                    if last_error is not None:
+                        updated = last_error.get("updatedExisting")
+                        if updated is not None:
+                            return not updated
+                    return created
+
+                upsert = self._created
+                update_query = {}
+
+                if updates:
+                    update_query["$set"] = updates
+                if removals:
+                    update_query["$unset"] = removals
+                if updates or removals:
+                    last_error = collection.update(select_dict, update_query,
+                                    upsert=upsert, safe=safe, **write_options)
+                    created = is_new_object(last_error)
+
+            warn_cascade = not cascade and 'cascade' not in self._meta
+            cascade = (self._meta.get('cascade', True)
+                       if cascade is None else cascade)
+            if cascade:
+                kwargs = {
+                    "safe": safe,
+                    "force_insert": force_insert,
+                    "validate": validate,
+                    "write_options": write_options,
+                    "cascade": cascade
+                }
+                if cascade_kwargs:  # Allow granular control over cascades
+                    kwargs.update(cascade_kwargs)
+                kwargs['_refs'] = _refs
+                self.cascade_save(warn_cascade=warn_cascade, **kwargs)
+
         except pymongo.errors.OperationFailure, err:
             message = 'Could not save document (%s)'
-            if u'duplicate key' in unicode(err):
+            if re.match('^E1100[01] duplicate key', unicode(err)):
+                # E11000 - duplicate key error index
+                # E11001 - duplicate key on update
                 message = u'Tried to save duplicate unique keys (%s)'
+                raise NotUniqueError(message % unicode(err))
             raise OperationError(message % unicode(err))
         id_field = self._meta['id_field']
-        self[id_field] = self._fields[id_field].to_python(object_id)
+        if id_field not in self._meta.get('shard_key', []):
+            self[id_field] = self._fields[id_field].to_python(object_id)
+
+        self._clear_changed_fields()
+        self._created = False
+        signals.post_save.send(self.__class__, document=self, created=created)
+        return self
+
+    def cascade_save(self, warn_cascade=None, *args, **kwargs):
+        """Recursively saves any references /
+           generic references on an objects"""
+        import fields
+        _refs = kwargs.get('_refs', []) or []
+
+        for name, cls in self._fields.items():
+            if not isinstance(cls, (fields.ReferenceField,
+                                    fields.GenericReferenceField)):
+                continue
+
+            ref = getattr(self, name)
+            if not ref or isinstance(ref, DBRef):
+                continue
+
+            if not getattr(ref, '_changed_fields', True):
+                continue
+
+            ref_id = "%s,%s" % (ref.__class__.__name__, str(ref._data))
+            if ref and ref_id not in _refs:
+                if warn_cascade:
+                    msg = ("Cascading saves will default to off in 0.8, "
+                          "please  explicitly set `.save(cascade=True)`")
+                    warnings.warn(msg, FutureWarning)
+                _refs.append(ref_id)
+                kwargs["_refs"] = _refs
+                ref.save(**kwargs)
+                ref._changed_fields = []
+
+    @property
+    def _qs(self):
+        """
+        Returns the queryset to use for updating / reloading / deletions
+        """
+        if not hasattr(self, '__objects'):
+            self.__objects = QuerySet(self, self._get_collection())
+        return self.__objects
+
+    @property
+    def _object_key(self):
+        """Dict to identify object in collection
+        """
+        select_dict = {'pk': self.pk}
+        shard_key = self.__class__._meta.get('shard_key', tuple())
+        for k in shard_key:
+            select_dict[k] = getattr(self, k)
+        return select_dict
+
+    def update(self, **kwargs):
+        """Performs an update on the :class:`~mongoengine.Document`
+        A convenience wrapper to :meth:`~mongoengine.QuerySet.update`.
+
+        Raises :class:`OperationError` if called on an object that has not yet
+        been saved.
+        """
+        if not self.pk:
+            raise OperationError('attempt to update a document not yet saved')
+
+        # Need to add shard key to query, or you get an error
+        return self._qs.filter(**self._object_key).update_one(**kwargs)
 
     def delete(self, safe=False):
         """Delete the :class:`~mongoengine.Document` from the database. This
@@ -90,60 +350,301 @@ class Document(BaseDocument):
 
         :param safe: check if the operation succeeded before returning
         """
-        id_field = self._meta['id_field']
+        signals.pre_delete.send(self.__class__, document=self)
-        object_id = self._fields[id_field].to_mongo(self[id_field])
+
         try:
-            self.__class__.objects(**{id_field: object_id}).delete(safe=safe)
+            self._qs.filter(**self._object_key).delete(safe=safe)
         except pymongo.errors.OperationFailure, err:
             message = u'Could not delete document (%s)' % err.message
             raise OperationError(message)
 
-    def reload(self):
+        signals.post_delete.send(self.__class__, document=self)
+
+    def switch_db(self, db_alias):
+        """
+        Temporarily switch the database for a document instance.
+
+        Only really useful for archiving off data and calling `save()`::
+
+            user = User.objects.get(id=user_id)
+            user.switch_db('archive-db')
+            user.save()
+
+        If you need to read from another database see
+        :class:`~mongoengine.context_managers.switch_db`
+
+        :param db_alias: The database alias to use for saving the document
+        """
+        with switch_db(self.__class__, db_alias) as cls:
+            collection = cls._get_collection()
+            db = cls._get_db
+        self._get_collection = lambda: collection
+        self._get_db = lambda: db
+        self._collection = collection
+        self._created = True
+        self.__objects = self._qs
+        self.__objects._collection_obj = collection
+        return self
+
+    def switch_collection(self, collection_name):
+        """
+        Temporarily switch the collection for a document instance.
+
+        Only really useful for archiving off data and calling `save()`::
+
+            user = User.objects.get(id=user_id)
+            user.switch_collection('old-users')
+            user.save()
+
+        If you need to read from another database see
+        :class:`~mongoengine.context_managers.switch_collection`
+
+        :param collection_name: The database alias to use for saving the
+            document
+        """
+        with switch_collection(self.__class__, collection_name) as cls:
+            collection = cls._get_collection()
+        self._get_collection = lambda: collection
+        self._collection = collection
+        self._created = True
+        self.__objects = self._qs
+        self.__objects._collection_obj = collection
+        return self
+
+    def select_related(self, max_depth=1):
+        """Handles dereferencing of :class:`~bson.dbref.DBRef` objects to
+        a maximum depth in order to cut down the number queries to mongodb.
+
+        .. versionadded:: 0.5
+        """
+        import dereference
+        self._data = dereference.DeReference()(self._data, max_depth)
+        return self
+
+    def reload(self, max_depth=1):
         """Reloads all attributes from the database.
 
         .. versionadded:: 0.1.2
+        .. versionchanged:: 0.6  Now chainable
         """
         id_field = self._meta['id_field']
-        obj = self.__class__.objects(**{id_field: self[id_field]}).first()
+        obj = self._qs.filter(
+                **{id_field: self[id_field]}
+              ).limit(1).select_related(max_depth=max_depth)
+        if obj:
+            obj = obj[0]
+        else:
+            msg = "Reloaded document has been deleted"
+            raise OperationError(msg)
         for field in self._fields:
-            setattr(self, field, obj[field])
+            setattr(self, field, self._reload(field, obj[field]))
+        if self._dynamic:
+            for name in self._dynamic_fields.keys():
+                setattr(self, name, self._reload(name, obj._data[name]))
+        self._changed_fields = obj._changed_fields
+        self._created = False
+        return obj
+
+    def _reload(self, key, value):
+        """Used by :meth:`~mongoengine.Document.reload` to ensure the
+        correct instance is linked to self.
+        """
+        if isinstance(value, BaseDict):
+            value = [(k, self._reload(k, v)) for k, v in value.items()]
+            value = BaseDict(value, self, key)
+        elif isinstance(value, BaseList):
+            value = [self._reload(key, v) for v in value]
+            value = BaseList(value, self, key)
+        elif isinstance(value, (EmbeddedDocument, DynamicEmbeddedDocument)):
+            value._changed_fields = []
+        return value
+
+    def to_dbref(self):
+        """Returns an instance of :class:`~bson.dbref.DBRef` useful in
+        `__raw__` queries."""
+        if not self.pk:
+            msg = "Only saved documents can have a valid dbref"
+            raise OperationError(msg)
+        return DBRef(self.__class__._get_collection_name(), self.pk)
+
+    @classmethod
+    def register_delete_rule(cls, document_cls, field_name, rule):
+        """This method registers the delete rules to apply when removing this
+        object.
+        """
+        classes = [get_document(class_name)
+                    for class_name in cls._subclasses
+                    if class_name != cls.__name__] + [cls]
+        documents = [get_document(class_name)
+                     for class_name in document_cls._subclasses
+                     if class_name != document_cls.__name__] + [document_cls]
+
+        for cls in classes:
+            for document_cls in documents:
+                delete_rules = cls._meta.get('delete_rules') or {}
+                delete_rules[(document_cls, field_name)] = rule
+                cls._meta['delete_rules'] = delete_rules
 
     @classmethod
     def drop_collection(cls):
         """Drops the entire collection associated with this
         :class:`~mongoengine.Document` type from the database.
         """
-        db = _get_db()
+        cls._collection = None
-        db.drop_collection(cls._meta['collection'])
+        db = cls._get_db()
+        db.drop_collection(cls._get_collection_name())
+
+    @classmethod
+    def ensure_index(cls, key_or_list, drop_dups=False, background=False,
+        **kwargs):
+        """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_spec = cls._build_index_spec(key_or_list)
+        index_spec = index_spec.copy()
+        fields = index_spec.pop('fields')
+        index_spec['drop_dups'] = drop_dups
+        index_spec['background'] = background
+        index_spec.update(kwargs)
+
+        return cls._get_collection().ensure_index(fields, **index_spec)
+
+    @classmethod
+    def ensure_indexes(cls):
+        """Checks the document meta data and ensures all the indexes exist.
+
+        .. note:: You can disable automatic index creation by setting
+                  `auto_create_index` to False in the documents meta data
+        """
+        background = cls._meta.get('index_background', False)
+        drop_dups = cls._meta.get('index_drop_dups', False)
+        index_opts = cls._meta.get('index_opts') or {}
+        index_cls = cls._meta.get('index_cls', True)
+
+        collection = cls._get_collection()
+
+        # determine if an index which we are creating includes
+        # _cls as its first field; if so, we can avoid creating
+        # an extra index on _cls, as mongodb will use the existing
+        # index to service queries against _cls
+        cls_indexed = False
+
+        def includes_cls(fields):
+            first_field = None
+            if len(fields):
+                if isinstance(fields[0], basestring):
+                    first_field = fields[0]
+                elif isinstance(fields[0], (list, tuple)) and len(fields[0]):
+                    first_field = fields[0][0]
+            return first_field == '_cls'
+
+        # Ensure document-defined indexes are created
+        if cls._meta['index_specs']:
+            index_spec = cls._meta['index_specs']
+            for spec in index_spec:
+                spec = spec.copy()
+                fields = spec.pop('fields')
+                cls_indexed = cls_indexed or includes_cls(fields)
+                opts = index_opts.copy()
+                opts.update(spec)
+                collection.ensure_index(fields, background=background,
+                                        drop_dups=drop_dups, **opts)
+
+        # If _cls is being used (for polymorphism), it needs an index,
+        # only if another index doesn't begin with _cls
+        if (index_cls and not cls_indexed and
+            cls._meta.get('allow_inheritance', ALLOW_INHERITANCE) == True):
+            collection.ensure_index('_cls', background=background,
+                                    **index_opts)
+
+
+class DynamicDocument(Document):
+    """A Dynamic Document class allowing flexible, expandable and uncontrolled
+    schemas.  As a :class:`~mongoengine.Document` subclass, acts in the same
+    way as an ordinary document but has expando style properties.  Any data
+    passed or set against the :class:`~mongoengine.DynamicDocument` that is
+    not a field is automatically converted into a
+    :class:`~mongoengine.DynamicField` and data can be attributed to that
+    field.
+
+    .. note::
+
+        There is one caveat on Dynamic Documents: fields cannot start with `_`
+    """
+
+    # The __metaclass__ attribute is removed by 2to3 when running with Python3
+    # my_metaclass is defined so that metaclass can be queried in Python 2 & 3
+    my_metaclass  = TopLevelDocumentMetaclass
+    __metaclass__ = TopLevelDocumentMetaclass
+
+    _dynamic = True
+
+    def __delattr__(self, *args, **kwargs):
+        """Deletes the attribute by setting to None and allowing _delta to unset
+        it"""
+        field_name = args[0]
+        if field_name in self._dynamic_fields:
+            setattr(self, field_name, None)
+        else:
+            super(DynamicDocument, self).__delattr__(*args, **kwargs)
+
+
+class DynamicEmbeddedDocument(EmbeddedDocument):
+    """A Dynamic Embedded Document class allowing flexible, expandable and
+    uncontrolled schemas. See :class:`~mongoengine.DynamicDocument` for more
+    information about dynamic documents.
+    """
+
+    # The __metaclass__ attribute is removed by 2to3 when running with Python3
+    # my_metaclass is defined so that metaclass can be queried in Python 2 & 3
+    my_metaclass  = DocumentMetaclass
+    __metaclass__ = DocumentMetaclass
+
+    _dynamic = True
+
+    def __delattr__(self, *args, **kwargs):
+        """Deletes the attribute by setting to None and allowing _delta to unset
+        it"""
+        field_name = args[0]
+        if field_name in self._fields:
+            default = self._fields[field_name].default
+            if callable(default):
+                default = default()
+            setattr(self, field_name, default)
+        else:
+            setattr(self, field_name, None)
 
 
 class MapReduceDocument(object):
     """A document returned from a map/reduce query.
-    
+
     :param collection: An instance of :class:`~pymongo.Collection`
-    :param key: Document/result key, often an instance of 
+    :param key: Document/result key, often an instance of
-                :class:`~pymongo.objectid.ObjectId`. If supplied as 
+                :class:`~bson.objectid.ObjectId`. If supplied as
-                an ``ObjectId`` found in the given ``collection``, 
+                an ``ObjectId`` found in the given ``collection``,
                 the object can be accessed via the ``object`` property.
     :param value: The result(s) for this key.
-    
+
     .. versionadded:: 0.3
     """
-    
+
     def __init__(self, document, collection, key, value):
         self._document = document
         self._collection = collection
         self.key = key
         self.value = value
-    
+
     @property
     def object(self):
-        """Lazy-load the object referenced by ``self.key``. ``self.key`` 
+        """Lazy-load the object referenced by ``self.key``. ``self.key``
         should be the ``primary_key``.
         """
         id_field = self._document()._meta['id_field']
         id_field_type = type(id_field)
-        
+
         if not isinstance(self.key, id_field_type):
             try:
                 self.key = id_field_type(self.key)

mongoengine/errors.py

@@ -0,0 +1,124 @@
+from collections import defaultdict
+
+from mongoengine.python_support import txt_type
+
+
+__all__ = ('NotRegistered', 'InvalidDocumentError', 'ValidationError')
+
+
+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 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):
+        self.errors = kwargs.get('errors', {})
+        self.field_name = kwargs.get('field_name')
+        self.message = message
+
+    def __str__(self):
+        return txt_type(self.message)
+
+    def __repr__(self):
+        return '%s(%s,)' % (self.__class__.__name__, self.message)
+
+    def __getattribute__(self, name):
+        message = super(ValidationError, self).__getattribute__(name)
+        if name == 'message':
+            if self.field_name:
+                message = '%s' % message
+            if self.errors:
+                message = '%s(%s)' % (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 not source:
+                return errors_dict
+            if isinstance(source, dict):
+                for field_name, error in source.iteritems():
+                    errors_dict[field_name] = build_dict(error)
+            elif isinstance(source, ValidationError) and source.errors:
+                return build_dict(source.errors)
+            else:
+                return unicode(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])
+            if isinstance(value, dict):
+                value = ' '.join(
+                        [generate_key(v, k) for k, v in value.iteritems()])
+
+            results = "%s.%s" % (prefix, value) if prefix else value
+            return results
+
+        error_dict = defaultdict(list)
+        for k, v in self.to_dict().iteritems():
+            error_dict[generate_key(v)].append(k)
+        return ' '.join(["%s: %s" % (k, v) for k, v in error_dict.iteritems()])

mongoengine/python_support.py

@@ -0,0 +1,61 @@
+"""Helper functions and types to aid with Python 2.5 - 3 support."""
+
+import sys
+
+PY3 = sys.version_info[0] == 3
+PY25 = sys.version_info[:2] == (2, 5)
+UNICODE_KWARGS = int(''.join([str(x) for x in sys.version_info[:3]])) > 264
+
+if PY3:
+    import codecs
+    from io import BytesIO as StringIO
+    # return s converted to binary.  b('test') should be equivalent to b'test'
+    def b(s):
+        return codecs.latin_1_encode(s)[0]
+
+    bin_type = bytes
+    txt_type   = str
+else:
+    try:
+        from cStringIO import StringIO
+    except ImportError:
+        from StringIO import StringIO
+
+    # Conversion to binary only necessary in Python 3
+    def b(s):
+        return s
+
+    bin_type = str
+    txt_type = unicode
+
+str_types = (bin_type, txt_type)
+
+if PY25:
+    def product(*args, **kwds):
+        pools = map(tuple, args) * kwds.get('repeat', 1)
+        result = [[]]
+        for pool in pools:
+            result = [x + [y] for x in result for y in pool]
+        for prod in result:
+            yield tuple(prod)
+    reduce = reduce
+else:
+    from itertools import product
+    from functools import reduce
+
+
+# For use with Python 2.5
+# converts all keys from unicode to str for d and all nested dictionaries
+def to_str_keys_recursive(d):
+    if isinstance(d, list):
+        for val in d:
+            if isinstance(val, (dict, list)):
+                to_str_keys_recursive(val)
+    elif isinstance(d, dict):
+        for key, val in d.items():
+            if isinstance(val, (dict, list)):
+                to_str_keys_recursive(val)
+            if isinstance(key, unicode):
+                d[str(key)] = d.pop(key)
+    else:
+        raise ValueError("non list/dict parameter not allowed")

mongoengine/queryset.py

@@ -1,941 +0,0 @@
-from connection import _get_db
-
-import pymongo
-import re
-import copy
-
-
-__all__ = ['queryset_manager', 'Q', 'InvalidQueryError',
-           'InvalidCollectionError']
-
-# The maximum number of items to display in a QuerySet.__repr__
-REPR_OUTPUT_SIZE = 20
-
-
-class DoesNotExist(Exception):
-    pass
-
-
-class MultipleObjectsReturned(Exception):
-    pass
-
-
-class InvalidQueryError(Exception):
-    pass
-
-
-class OperationError(Exception):
-    pass
-
-
-RE_TYPE = type(re.compile(''))
-
-
-class Q(object):
-
-    OR = '||'
-    AND = '&&'
-    OPERATORS = {
-        'eq': 'this.%(field)s == %(value)s',
-        'ne': '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': '%(value)s.indexOf(this.%(field)s) != -1',
-        'nin': '%(value)s.indexOf(this.%(field)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',
-        'regex_eq': '%(value)s.test(this.%(field)s)',
-        'regex_ne': '!%(value)s.test(this.%(field)s)',
-    }
-
-    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)
-                    # Construct the JS that uses this op
-                    value, operation_js = self._build_op_js(op, key, value,
-                                                            op_value_name)
-                    # Update the js scope with the value for this op
-                    js_scope[op_value_name] = value
-                    js.append(operation_js)
-            else:
-                # Construct the JS for this field
-                value, field_js = self._build_op_js(op, key, value, value_name)
-                js_scope[value_name] = value
-                js.append(field_js)
-        return ' && '.join(js)
-
-    def _build_op_js(self, op, key, value, value_name):
-        """Substitute the values in to the correct chunk of Javascript.
-        """
-        if isinstance(value, RE_TYPE):
-            # Regexes are handled specially
-            if op.strip('$') == 'ne':
-                op_js = Q.OPERATORS['regex_ne']
-            else:
-                op_js = Q.OPERATORS['regex_eq']
-        else:
-            op_js = Q.OPERATORS[op.strip('$')]
-
-        # Comparing two ObjectIds in Javascript doesn't work..
-        if isinstance(value, pymongo.objectid.ObjectId):
-            value = str(value)
-
-        # Perform the substitution
-        operation_js = op_js % {
-            'field': key, 
-            'value': value_name
-        }
-        return value, operation_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_clause = None
-        self._loaded_fields = []
-        self._ordering = []
-        
-        # 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
-        self._limit = None
-        self._skip = 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.db_field 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_obj=None, **query):
-        """Filter the selected documents by calling the
-        :class:`~mongoengine.queryset.QuerySet` with a query.
-
-        :param q_obj: a :class:`~mongoengine.queryset.Q` object to be used in
-            the query; the :class:`~mongoengine.queryset.QuerySet` is filtered
-            multiple times with different :class:`~mongoengine.queryset.Q`
-            objects, only the last one will be used
-        :param query: Django-style query keyword arguments
-        """
-        if q_obj:
-            self._where_clause = q_obj.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 self._cursor_obj is None:
-            cursor_args = {}
-            if self._loaded_fields:
-                cursor_args = {'fields': self._loaded_fields}
-            self._cursor_obj = self._collection.find(self._query, 
-                                                     **cursor_args)
-            # Apply where clauses to cursor
-            if self._where_clause:
-                self._cursor_obj.where(self._where_clause)
-
-            # 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.db_field 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 = ['ne', 'gt', 'gte', 'lt', 'lte', 'in', 'nin', 'mod',
-                     'all', 'size', 'exists']
-        match_operators = ['contains', 'icontains', 'startswith', 
-                           'istartswith', 'endswith', 'iendswith']
-
-        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 + match_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.db_field for field in fields]
-
-                # Convert value to proper value
-                field = fields[-1]
-                singular_ops = [None, 'ne', 'gt', 'gte', 'lt', 'lte']
-                singular_ops += match_operators
-                if op in singular_ops:
-                    value = field.prepare_query_value(op, value)
-                elif op in ('in', 'nin', 'all'):
-                    # 'in', 'nin' and 'all' require a list of values
-                    value = [field.prepare_query_value(op, v) for v in value]
-
-                if field.__class__.__name__ == 'GenericReferenceField':
-                    parts.append('_ref')
-
-            if op and op not in match_operators:
-                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 get(self, *q_objs, **query):
-        """Retrieve the the matching object raising
-        :class:`~mongoengine.queryset.MultipleObjectsReturned` or
-        :class:`~mongoengine.queryset.DoesNotExist` exceptions if multiple or
-        no results are found.
-
-        .. versionadded:: 0.3
-        """
-        self.__call__(*q_objs, **query)
-        count = self.count()
-        if count == 1:
-            return self[0]
-        elif count > 1:
-            message = u'%d items returned, instead of 1' % count
-            raise MultipleObjectsReturned(message)
-        else:
-            raise DoesNotExist('Document not found')
-
-    def get_or_create(self, *q_objs, **query):
-        """Retreive unique object or create, if it doesn't exist. Raises
-        :class:`~mongoengine.queryset.MultipleObjectsReturned` if multiple
-        results are found. A new document will be created if the document
-        doesn't exists; a dictionary of default values for the new document
-        may be provided as a keyword argument called :attr:`defaults`.
-
-        .. versionadded:: 0.3
-        """
-        defaults = query.get('defaults', {})
-        if 'defaults' in query:
-            del query['defaults']
-
-        self.__call__(*q_objs, **query)
-        count = self.count()
-        if count == 0:
-            query.update(defaults)
-            doc = self._document(**query)
-            doc.save()
-            return doc
-        elif count == 1:
-            return self.first()
-        else:
-            message = u'%d items returned, instead of 1' % count
-            raise MultipleObjectsReturned(message)
-
-    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 in_bulk(self, object_ids):
-        """Retrieve a set of documents by their ids.
-        
-        :param object_ids: a list or tuple of ``ObjectId``\ s
-        :rtype: dict of ObjectIds as keys and collection-specific
-                Document subclasses as values.
-
-        .. versionadded:: 0.3
-        """
-        doc_map = {}
-
-        docs = self._collection.find({'_id': {'$in': object_ids}})
-        for doc in docs:
-            doc_map[doc['_id']] = self._document._from_son(doc)
- 
-        return doc_map
-
-    def next(self):
-        """Wrap the result in a :class:`~mongoengine.Document` object.
-        """
-        try:
-            if self._limit == 0:
-                raise StopIteration
-            return self._document._from_son(self._cursor.next())
-        except StopIteration, e:
-            self.rewind()
-            raise e
-
-    def rewind(self):
-        """Rewind the cursor to its unevaluated state.
-
-        .. versionadded:: 0.3
-        """
-        self._cursor.rewind()
-
-    def count(self):
-        """Count the selected elements in the query.
-        """
-        if self._limit == 0:
-            return 0
-        return self._cursor.count(with_limit_and_skip=True)
-
-    def __len__(self):
-        return self.count()
-
-    def map_reduce(self, map_f, reduce_f, finalize_f=None, limit=None,
-                   scope=None, keep_temp=False):
-        """Perform a map/reduce query using the current query spec
-        and ordering. While ``map_reduce`` respects ``QuerySet`` chaining,
-        it must be the last call made, as it does not return a maleable
-        ``QuerySet``.
-
-        See the :meth:`~mongoengine.tests.QuerySetTest.test_map_reduce`
-        and :meth:`~mongoengine.tests.QuerySetTest.test_map_advanced`
-        tests in ``tests.queryset.QuerySetTest`` for usage examples.
-
-        :param map_f: map function, as :class:`~pymongo.code.Code` or string
-        :param reduce_f: reduce function, as
-                         :class:`~pymongo.code.Code` or string
-        :param finalize_f: finalize function, an optional function that
-                           performs any post-reduction processing.
-        :param scope: values to insert into map/reduce global scope. Optional.
-        :param limit: number of objects from current query to provide
-                      to map/reduce method
-        :param keep_temp: keep temporary table (boolean, default ``True``)
-
-        Returns an iterator yielding
-        :class:`~mongoengine.document.MapReduceDocument`.
-
-        .. note:: Map/Reduce requires server version **>= 1.1.1**. The PyMongo
-           :meth:`~pymongo.collection.Collection.map_reduce` helper requires
-           PyMongo version **>= 1.2**.
-
-        .. versionadded:: 0.3
-        """
-        from document import MapReduceDocument
-
-        if not hasattr(self._collection, "map_reduce"):
-            raise NotImplementedError("Requires MongoDB >= 1.1.1")
-
-        map_f_scope = {}
-        if isinstance(map_f, pymongo.code.Code):
-            map_f_scope = map_f.scope
-            map_f = str(map_f)
-        map_f = pymongo.code.Code(self._sub_js_fields(map_f), map_f_scope)
-
-        reduce_f_scope = {}
-        if isinstance(reduce_f, pymongo.code.Code):
-            reduce_f_scope = reduce_f.scope
-            reduce_f = str(reduce_f)
-        reduce_f_code = self._sub_js_fields(reduce_f)
-        reduce_f = pymongo.code.Code(reduce_f_code, reduce_f_scope)
-
-        mr_args = {'query': self._query, 'keeptemp': keep_temp}
-
-        if finalize_f:
-            finalize_f_scope = {}
-            if isinstance(finalize_f, pymongo.code.Code):
-                finalize_f_scope = finalize_f.scope
-                finalize_f = str(finalize_f)
-            finalize_f_code = self._sub_js_fields(finalize_f)
-            finalize_f = pymongo.code.Code(finalize_f_code, finalize_f_scope)
-            mr_args['finalize'] = finalize_f
-
-        if scope:
-            mr_args['scope'] = scope
-
-        if limit:
-            mr_args['limit'] = limit
-
-        results = self._collection.map_reduce(map_f, reduce_f, **mr_args)
-        results = results.find()
-
-        if self._ordering:
-            results = results.sort(self._ordering)
-
-        for doc in results:
-            yield MapReduceDocument(self._document, self._collection,
-                                    doc['_id'], doc['value'])
-
-    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
-        """
-        if n == 0:
-            self._cursor.limit(1)
-        else:
-            self._cursor.limit(n)
-        self._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)
-        self._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]
-                self._skip, self._limit = key.start, key.stop
-            except IndexError, err:
-                # PyMongo raises an error if key.start == key.stop, catch it,
-                # bin it, kill it. 
-                start = key.start or 0
-                if start >= 0 and key.stop >= 0 and key.step is None:
-                    if start == key.stop:
-                        self.limit(0)
-                        self._skip, self._limit = key.start, key.stop - start
-                        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 only(self, *fields):
-        """Load only a subset of this document's fields. ::
-        
-            post = BlogPost.objects(...).only("title")
-        
-        :param fields: fields to include
-
-        .. versionadded:: 0.3
-        """
-        self._loaded_fields = []
-        for field in fields:
-            if '.' in field:
-                raise InvalidQueryError('Subfields cannot be used as '
-                                        'arguments to QuerySet.only')
-            # Translate field name
-            field = QuerySet._lookup_field(self._document, field)[-1].db_field
-            self._loaded_fields.append(field)
-
-        # _cls is needed for polymorphism
-        if self._document._meta.get('allow_inheritance'):
-            self._loaded_fields += ['_cls']
-        return self
-
-    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._ordering = key_list
-        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.db_field 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(op, value)
-                elif op in ('pushAll', 'pullAll'):
-                    value = [field.prepare_query_value(op, 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, upsert=False, **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, 
-                                    upsert=upsert, multi=True)
-        except pymongo.errors.OperationFailure, err:
-            if unicode(err) == u'multi not coded yet':
-                message = u'update() method requires MongoDB 1.1.3+'
-                raise OperationError(message)
-            raise OperationError(u'Update failed (%s)' % unicode(err))
-
-    def update_one(self, safe_update=True, upsert=False, **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, 
-                                        upsert=upsert, 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 _sub_js_fields(self, code):
-        """When fields are specified with [~fieldname] syntax, where 
-        *fieldname* is the Python name of a field, *fieldname* will be 
-        substituted for the MongoDB name of the field (specified using the
-        :attr:`name` keyword argument in a field's constructor).
-        """
-        def field_sub(match):
-            # Extract just the field name, and look up the field objects
-            field_name = match.group(1).split('.')
-            fields = QuerySet._lookup_field(self._document, field_name)
-            # Substitute the correct name for the field into the javascript
-            return '["%s"]' % fields[-1].db_field
-
-        return re.sub('\[\s*~([A-z_][A-z_0-9.]+?)\s*\]', field_sub, code)
-
-    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.
-
-        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, use square-bracket notation, and prefix the MongoEngine field
-        name with a tilde (~).
-
-        :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)
-        """
-        code = self._sub_js_fields(code)
-
-        fields = [QuerySet._translate_field_name(self._document, f)
-                  for f in fields]
-        collection = self._document._meta['collection']
-
-        scope = {
-            'collection': collection,
-            'options': options or {},
-        }
-
-        query = self._query
-        if self._where_clause:
-            query['$where'] = self._where_clause
-
-        scope['query'] = query
-        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):
-        limit = REPR_OUTPUT_SIZE + 1
-        if self._limit is not None and self._limit < limit:
-            limit = self._limit
-        data = list(self[self._skip:limit])
-        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:
-            if self._manager_func.func_code.co_argcount == 1:
-                queryset = self._manager_func(queryset)
-            else:
-                queryset = self._manager_func(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.
-    """
-    if func.func_code.co_argcount == 1:
-        import warnings
-        msg = 'Methods decorated with queryset_manager should take 2 arguments'
-        warnings.warn(msg, DeprecationWarning)
-    return QuerySetManager(func)

mongoengine/queryset/__init__.py

@@ -0,0 +1,11 @@
+from mongoengine.errors import (DoesNotExist, MultipleObjectsReturned,
+                                InvalidQueryError, OperationError,
+                                NotUniqueError)
+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 *
+
+__all__ = (field_list.__all__ + manager.__all__ + queryset.__all__ +
+           transform.__all__ + visitor.__all__)

mongoengine/queryset/field_list.py

@@ -0,0 +1,70 @@
+
+__all__ = ('QueryFieldList',)
+
+
+class QueryFieldList(object):
+    """Object that handles combinations of .only() and .exclude() calls"""
+    ONLY = 1
+    EXCLUDE = 0
+
+    def __init__(self, fields=[], value=ONLY, always_include=[]):
+        self.value = value
+        self.fields = set(fields)
+        self.always_include = set(always_include)
+        self._id = None
+        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()
+            self.fields = self.fields.intersection(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:
+                self.fields = self.fields.union(self.always_include)
+            else:
+                self.fields -= self.always_include
+        return self
+
+    def __nonzero__(self):
+        return bool(self.fields)
+
+    def as_dict(self):
+        field_list = dict((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(object):
+    """
+    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 class being used rather than a document object
+            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.func_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/transform.py

@@ -0,0 +1,252 @@
+from collections import defaultdict
+
+from bson import SON
+
+from mongoengine.common import _import_class
+from mongoengine.errors import InvalidQueryError, LookUpError
+
+__all__ = ('query', 'update')
+
+
+COMPARISON_OPERATORS = ('ne', 'gt', 'gte', 'lt', 'lte', 'in', 'nin', 'mod',
+                          'all', 'size', 'exists', 'not')
+GEO_OPERATORS        = ('within_distance', 'within_spherical_distance',
+                        'within_box', 'within_polygon', 'near', 'near_sphere',
+                        'max_distance')
+STRING_OPERATORS     = ('contains', 'icontains', 'startswith',
+                        'istartswith', 'endswith', 'iendswith',
+                        'exact', 'iexact')
+CUSTOM_OPERATORS     = ('match',)
+MATCH_OPERATORS      = (COMPARISON_OPERATORS + GEO_OPERATORS +
+                        STRING_OPERATORS + CUSTOM_OPERATORS)
+
+UPDATE_OPERATORS     = ('set', 'unset', 'inc', 'dec', 'pop', 'push',
+                        'push_all', 'pull', 'pull_all', 'add_to_set')
+
+
+def query(_doc_cls=None, _field_operation=False, **query):
+    """Transform a query from Django-style format to Mongo format.
+    """
+    mongo_query = {}
+    merge_query = defaultdict(list)
+    for key, value in sorted(query.items()):
+        if key == "__raw__":
+            mongo_query.update(value)
+            continue
+
+        parts = key.split('__')
+        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 parts[-1] in MATCH_OPERATORS:
+            op = parts.pop()
+
+        negate = False
+        if 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, e:
+                raise InvalidQueryError(e)
+            parts = []
+
+            cleaned_fields = []
+            for field in fields:
+                append_field = True
+                if isinstance(field, basestring):
+                    parts.append(field)
+                    append_field = False
+                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:
+                if isinstance(field, basestring):
+                    if (op in STRING_OPERATORS and
+                        isinstance(value, basestring)):
+                        StringField = _import_class('StringField')
+                        value = StringField.prepare_query_value(op, value)
+                    else:
+                        value = field
+                else:
+                    value = field.prepare_query_value(op, value)
+            elif op in ('in', 'nin', 'all', 'near'):
+                # 'in', 'nin' and 'all' require a list of values
+                value = [field.prepare_query_value(op, v) for v in value]
+
+        # if op and op not in COMPARISON_OPERATORS:
+        if op:
+            if op in GEO_OPERATORS:
+                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}}
+                elif op == "max_distance":
+                    value = {'$maxDistance': value}
+                else:
+                    raise NotImplementedError("Geo method '%s' has not "
+                                              "been implemented" % op)
+            elif op in CUSTOM_OPERATORS:
+                if op == 'match':
+                    value = {"$elemMatch": value}
+                else:
+                    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 op is None or key not in mongo_query:
+            mongo_query[key] = value
+        elif key in mongo_query:
+            if key in mongo_query and isinstance(mongo_query[key], dict):
+                mongo_query[key].update(value)
+                # $maxDistance needs to come last - convert to SON
+                if '$maxDistance' in mongo_query[key]:
+                    value_dict = mongo_query[key]
+                    value_son = SON()
+                    for k, v in value_dict.iteritems():
+                        if k == '$maxDistance':
+                            continue
+                        value_son[k] = v
+                    value_son['$maxDistance'] = value_dict['$maxDistance']
+                    mongo_query[key] = value_son
+            else:
+                # Store for manually merging later
+                merge_query[key].append(value)
+
+    # The queryset has been filter in such a way we must manually merge
+    for k, v in merge_query.items():
+        merge_query[k].append(mongo_query[k])
+        del mongo_query[k]
+        if isinstance(v, list):
+            value = [{k:val} for val in v]
+            if '$and' in mongo_query.keys():
+                mongo_query['$and'].append(value)
+            else:
+                mongo_query['$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('__')
+        # 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
+            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
+            elif op == 'add_to_set':
+                op = op.replace('_to_set', 'ToSet')
+
+        match = None
+        if parts[-1] in COMPARISON_OPERATORS:
+            match = 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, e:
+                raise InvalidQueryError(e)
+            parts = []
+
+            cleaned_fields = []
+            for field in fields:
+                append_field = True
+                if isinstance(field, basestring):
+                    # Convert the S operator to $
+                    if field == 'S':
+                        field = '$'
+                    parts.append(field)
+                    append_field = False
+                else:
+                    parts.append(field.db_field)
+                if append_field:
+                    cleaned_fields.append(field)
+
+            # Convert value to proper value
+            field = cleaned_fields[-1]
+
+            if op in (None, 'set', 'push', 'pull'):
+                if field.required or value is not None:
+                    value = field.prepare_query_value(op, value)
+            elif op in ('pushAll', 'pullAll'):
+                value = [field.prepare_query_value(op, v) for v in value]
+            elif op == 'addToSet':
+                if isinstance(value, (list, tuple, set)):
+                    value = [field.prepare_query_value(op, v) for v in value]
+                elif field.required or value is not None:
+                    value = field.prepare_query_value(op, value)
+
+        if match:
+            match = '$' + match
+            value = {match: value}
+
+        key = '.'.join(parts)
+
+        if not op:
+            raise InvalidQueryError("Updates must supply an operation "
+                                    "eg: set__FIELD=value")
+
+        if 'pull' in op and '.' in key:
+            # Dot operators don't work on pull operations
+            # it uses nested dict syntax
+            if op == 'pullAll':
+                raise InvalidQueryError("pullAll operations only support "
+                                        "a single field depth")
+
+            parts.reverse()
+            for key in parts:
+                value = {key: value}
+        elif op == 'addToSet' and isinstance(value, list):
+            value = {key: {"$each": value}}
+        else:
+            value = {key: value}
+        key = '$' + 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

mongoengine/queryset/visitor.py

@@ -0,0 +1,155 @@
+import copy
+
+from mongoengine.errors import InvalidQueryError
+from mongoengine.python_support import product, reduce
+
+from mongoengine.queryset import transform
+
+__all__ = ('Q',)
+
+
+class QNodeVisitor(object):
+    """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 SimplificationVisitor(QNodeVisitor):
+    """Simplifies query trees by combinging 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]
+                return Q(**self._query_conjunction(queries))
+        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:
+                msg = 'Duplicate query conditions: '
+                raise InvalidQueryError(msg + ', '.join(intersection))
+
+            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(object):
+    """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 getattr(other, 'empty', True):
+            return self
+
+        if self.empty:
+            return other
+
+        return QCombination(operation, [self, other])
+
+    @property
+    def empty(self):
+        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
+                self.children.append(node)
+            else:
+                self.children.append(node)
+
+    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):
+        return not bool(self.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 accept(self, visitor):
+        return visitor.visit_query(self)
+
+    @property
+    def empty(self):
+        return not bool(self.query)

mongoengine/signals.py

@@ -0,0 +1,46 @@
+# -*- coding: utf-8 -*-
+
+__all__ = ['pre_init', 'post_init', 'pre_save', 'post_save',
+           'pre_delete', 'post_delete']
+
+signals_available = False
+try:
+    from blinker import Namespace
+    signals_available = True
+except ImportError:
+    class Namespace(object):
+        def signal(self, name, doc=None):
+            return _FakeSignal(name, doc)
+
+    class _FakeSignal(object):
+        """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
+        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')
+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:        0.7.9
+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://readthedocs.org/docs/mongoengine-odm/en/latest/changelog.html

requirements.txt

@@ -0,0 +1 @@
+pymongo

setup.cfg

@@ -0,0 +1,11 @@
+[nosetests]
+verbosity = 3
+detailed-errors = 1
+#with-coverage = 1
+#cover-erase = 1
+#cover-html = 1
+#cover-html-dir = ../htmlcov
+#cover-package = mongoengine
+py3where = build
+where = tests
+#tests =  document/__init__.py

setup.py

@@ -1,27 +1,35 @@
-from setuptools import setup, find_packages
 import os
+import sys
+from setuptools import setup, find_packages
 
-DESCRIPTION = "A Python Document-Object Mapper for working with MongoDB"
+# Hack to silence atexit traceback in newer python versions
+try:
+    import multiprocessing
+except ImportError:
+    pass
 
+DESCRIPTION = 'MongoEngine is a Python Object-Document ' + \
+'Mapper for working with MongoDB.'
 LONG_DESCRIPTION = None
 try:
     LONG_DESCRIPTION = open('README.rst').read()
 except:
     pass
 
-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
 
-# Dirty hack to get version number from monogengine/__init__.py - we can't 
+def get_version(version_tuple):
+    if not isinstance(version_tuple[-1], int):
+        return '.'.join(map(str, version_tuple[:-1])) + version_tuple[-1]
+    return '.'.join(map(str, version_tuple))
+
+# 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_line = list(filter(lambda l: l.startswith('VERSION'), open(init)))[0]
+
 VERSION = get_version(eval(version_line.split('=')[-1]))
-print VERSION
+print(VERSION)
 
 CLASSIFIERS = [
     'Development Status :: 4 - Beta',
@@ -29,16 +37,38 @@ CLASSIFIERS = [
     'License :: OSI Approved :: MIT License',
     'Operating System :: OS Independent',
     'Programming Language :: Python',
+    "Programming Language :: Python :: 2",
+    "Programming Language :: Python :: 2.5",
+    "Programming Language :: Python :: 2.6",
+    "Programming Language :: Python :: 2.7",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.1",
+    "Programming Language :: Python :: 3.2",
+    "Programming Language :: Python :: Implementation :: CPython",
     'Topic :: Database',
     'Topic :: Software Development :: Libraries :: Python Modules',
 ]
 
+extra_opts = {}
+if sys.version_info[0] == 3:
+    extra_opts['use_2to3'] = True
+    extra_opts['tests_require'] = ['nose', 'coverage', 'blinker']
+    extra_opts['packages'] = find_packages(exclude=('tests',))
+    if "test" in sys.argv or "nosetests" in sys.argv:
+        extra_opts['packages'].append("tests")
+        extra_opts['package_data'] = {"tests": ["fields/mongoengine.png"]}
+else:
+    extra_opts['tests_require'] = ['nose', 'coverage', 'blinker', 'django==1.4.2', 'PIL']
+    extra_opts['packages'] = find_packages(exclude=('tests',))
+
 setup(name='mongoengine',
       version=VERSION,
-      packages=find_packages(),
       author='Harry Marr',
       author_email='harry.marr@{nospam}gmail.com',
-      url='http://hmarr.com/mongoengine/',
+      maintainer="Ross Lawley",
+      maintainer_email="ross.lawley@{nospam}gmail.com",
+      url='http://mongoengine.org/',
+      download_url='https://github.com/MongoEngine/mongoengine/tarball/master',
       license='MIT',
       include_package_data=True,
       description=DESCRIPTION,
@@ -46,5 +76,6 @@ setup(name='mongoengine',
       platforms=['any'],
       classifiers=CLASSIFIERS,
       install_requires=['pymongo'],
-      test_suite='tests',
+      test_suite='nose.collector',
+      **extra_opts
 )

tests/__init__.py

@@ -0,0 +1,3 @@
+from all_warnings import AllWarnings
+from document import *
+from queryset import *

tests/all_warnings/__init__.py

@@ -0,0 +1,94 @@
+"""
+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 sys
+sys.path[0:0] = [""]
+import unittest
+import warnings
+
+from mongoengine import *
+
+
+__all__ = ('AllWarnings', )
+
+
+class AllWarnings(unittest.TestCase):
+
+    def setUp(self):
+        conn = 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_dbref_reference_field_future_warning(self):
+
+        class Person(Document):
+            name = StringField()
+            parent = ReferenceField('self')
+
+        Person.drop_collection()
+
+        p1 = Person()
+        p1.parent = None
+        p1.save()
+
+        p2 = Person(name="Wilson Jr")
+        p2.parent = p1
+        p2.save(cascade=False)
+
+        self.assertTrue(len(self.warning_list) > 0)
+        warning = self.warning_list[0]
+        self.assertEqual(FutureWarning, warning["category"])
+        self.assertTrue("ReferenceFields will default to using ObjectId"
+                        in str(warning["message"]))
+
+    def test_document_save_cascade_future_warning(self):
+
+        class Person(Document):
+            name = StringField()
+            parent = ReferenceField('self')
+
+        Person.drop_collection()
+
+        p1 = Person(name="Wilson Snr")
+        p1.parent = None
+        p1.save()
+
+        p2 = Person(name="Wilson Jr")
+        p2.parent = p1
+        p2.parent.name = "Poppa Wilson"
+        p2.save()
+
+        self.assertTrue(len(self.warning_list) > 0)
+        if len(self.warning_list) > 1:
+            print self.warning_list
+        warning = self.warning_list[0]
+        self.assertEqual(FutureWarning, warning["category"])
+        self.assertTrue("Cascading saves will default to off in 0.8"
+                        in str(warning["message"]))
+
+    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]
+        self.assertEqual(SyntaxWarning, warning["category"])
+        self.assertEqual('non_abstract_base',
+                         InheritedDocumentFailTest._get_collection_name())
+
+import sys
+sys.path[0:0] = [""]

tests/document.py

@@ -1,555 +0,0 @@
-import unittest
-from datetime 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)
-        
-        # Test the same for embedded documents
-        class Comment(EmbeddedDocument):
-            content = StringField()
-            meta = {'allow_inheritance': False}
-
-        def create_special_comment():
-            class SpecialComment(Comment):
-                pass
-        self.assertRaises(ValueError, create_special_comment)
-
-        comment = Comment(content='test')
-        self.assertFalse('_cls' in comment.to_mongo())
-        self.assertFalse('_types' in comment.to_mongo())
-
-    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.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.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(db_field='addDate', default=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(db_field='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'].db_field, '_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('collection' in 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.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(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/__init__.py

@@ -0,0 +1,15 @@
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+from class_methods import *
+from delta import *
+from dynamic import *
+from indexes import *
+from inheritance import *
+from instance import *
+from json_serialisation import *
+from validation import *
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/class_methods.py

@@ -0,0 +1,187 @@
+# -*- coding: utf-8 -*-
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+from mongoengine import *
+
+from mongoengine.queryset import NULLIFY
+from mongoengine.connection import get_db
+
+__all__ = ("ClassMethodsTest", )
+
+
+class ClassMethodsTest(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 self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            self.db.drop_collection(collection)
+
+    def test_definition(self):
+        """Ensure that document may be defined using fields.
+        """
+        self.assertEqual(['age', 'id', 'name'],
+                         sorted(self.Person._fields.keys()))
+        self.assertEqual(["IntField", "ObjectIdField", "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()
+        self.assertEqual(self.db, db)
+
+    def test_get_collection_name(self):
+        """Ensure that get_collection_name returns the expected collection
+        name.
+        """
+        collection_name = 'person'
+        self.assertEqual(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()
+        self.assertEqual(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()
+        self.assertTrue(collection_name in self.db.collection_names())
+
+        self.Person.drop_collection()
+        self.assertFalse(collection_name in self.db.collection_names())
+
+    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)
+
+        self.assertEqual(self.Person._meta.get('delete_rules'), None)
+
+        self.Person.register_delete_rule(Job, 'employee', NULLIFY)
+        self.assertEqual(self.Person._meta['delete_rules'],
+                         {(Job, 'employee'): NULLIFY})
+
+    def test_collection_naming(self):
+        """Ensure that a collection with a specified name may be used.
+        """
+
+        class DefaultNamingTest(Document):
+            pass
+        self.assertEqual('default_naming_test',
+                         DefaultNamingTest._get_collection_name())
+
+        class CustomNamingTest(Document):
+            meta = {'collection': 'pimp_my_collection'}
+
+        self.assertEqual('pimp_my_collection',
+                         CustomNamingTest._get_collection_name())
+
+        class DynamicNamingTest(Document):
+            meta = {'collection': lambda c: "DYNAMO"}
+        self.assertEqual('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
+        self.assertEqual('oldnamingconvention',
+                         OldNamingConvention._get_collection_name())
+
+        class InheritedAbstractNamingTest(BaseDocument):
+            meta = {'collection': 'wibble'}
+        self.assertEqual('wibble',
+                         InheritedAbstractNamingTest._get_collection_name())
+
+        # Mixin tests
+        class BaseMixin(object):
+            meta = {
+                'collection': lambda c: c.__name__.lower()
+            }
+
+        class OldMixinNamingConvention(Document, BaseMixin):
+            pass
+        self.assertEqual('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
+
+        self.assertEqual('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()
+        self.assertTrue(collection_name in self.db.collection_names())
+
+        user_obj = self.db[collection_name].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_name in self.db.collection_names())
+
+    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()
+        self.assertEqual(user_obj.name, "Test User")
+
+        Person.drop_collection()
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/delta.py

@@ -0,0 +1,689 @@
+# -*- coding: utf-8 -*-
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+from mongoengine import *
+from mongoengine.connection import get_db
+
+__all__ = ("DeltaTest",)
+
+
+class DeltaTest(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 self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            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()
+        self.assertEqual(doc._get_changed_fields(), [])
+        self.assertEqual(doc._delta(), ({}, {}))
+
+        doc.string_field = 'hello'
+        self.assertEqual(doc._get_changed_fields(), ['string_field'])
+        self.assertEqual(doc._delta(), ({'string_field': 'hello'}, {}))
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        self.assertEqual(doc._get_changed_fields(), ['int_field'])
+        self.assertEqual(doc._delta(), ({'int_field': 1}, {}))
+
+        doc._changed_fields = []
+        dict_value = {'hello': 'world', 'ping': 'pong'}
+        doc.dict_field = dict_value
+        self.assertEqual(doc._get_changed_fields(), ['dict_field'])
+        self.assertEqual(doc._delta(), ({'dict_field': dict_value}, {}))
+
+        doc._changed_fields = []
+        list_value = ['1', 2, {'hello': 'world'}]
+        doc.list_field = list_value
+        self.assertEqual(doc._get_changed_fields(), ['list_field'])
+        self.assertEqual(doc._delta(), ({'list_field': list_value}, {}))
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        self.assertEqual(doc._get_changed_fields(), ['dict_field'])
+        self.assertEqual(doc._delta(), ({}, {'dict_field': 1}))
+
+        doc._changed_fields = []
+        doc.list_field = []
+        self.assertEqual(doc._get_changed_fields(), ['list_field'])
+        self.assertEqual(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):
+            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()
+        self.assertEqual(doc._get_changed_fields(), [])
+        self.assertEqual(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
+
+        self.assertEqual(doc._get_changed_fields(), ['embedded_field'])
+
+        embedded_delta = {
+            'string_field': 'hello',
+            'int_field': 1,
+            'dict_field': {'hello': 'world'},
+            'list_field': ['1', 2, {'hello': 'world'}]
+        }
+        self.assertEqual(doc.embedded_field._delta(), (embedded_delta, {}))
+        self.assertEqual(doc._delta(),
+            ({'embedded_field': embedded_delta}, {}))
+
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.dict_field = {}
+        self.assertEqual(doc._get_changed_fields(),
+                        ['embedded_field.dict_field'])
+        self.assertEqual(doc.embedded_field._delta(), ({}, {'dict_field': 1}))
+        self.assertEqual(doc._delta(), ({}, {'embedded_field.dict_field': 1}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(doc.embedded_field.dict_field, {})
+
+        doc.embedded_field.list_field = []
+        self.assertEqual(doc._get_changed_fields(),
+            ['embedded_field.list_field'])
+        self.assertEqual(doc.embedded_field._delta(), ({}, {'list_field': 1}))
+        self.assertEqual(doc._delta(), ({}, {'embedded_field.list_field': 1}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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]
+        self.assertEqual(doc._get_changed_fields(),
+            ['embedded_field.list_field'])
+
+        self.assertEqual(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'}],
+            }]
+        }, {}))
+
+        self.assertEqual(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)
+
+        self.assertEqual(doc.embedded_field.list_field[0], '1')
+        self.assertEqual(doc.embedded_field.list_field[1], 2)
+        for k in doc.embedded_field.list_field[2]._fields:
+            self.assertEqual(doc.embedded_field.list_field[2][k],
+                             embedded_2[k])
+
+        doc.embedded_field.list_field[2].string_field = 'world'
+        self.assertEqual(doc._get_changed_fields(),
+            ['embedded_field.list_field.2.string_field'])
+        self.assertEqual(doc.embedded_field._delta(),
+            ({'list_field.2.string_field': 'world'}, {}))
+        self.assertEqual(doc._delta(),
+            ({'embedded_field.list_field.2.string_field': 'world'}, {}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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]
+        self.assertEqual(doc._get_changed_fields(),
+            ['embedded_field.list_field'])
+        self.assertEqual(doc.embedded_field._delta(), ({
+            'list_field': ['1', 2, {
+            '_cls': 'Embedded',
+            'string_field': 'hello world',
+            'int_field': 1,
+            'list_field': ['1', 2, {'hello': 'world'}],
+            'dict_field': {'hello': 'world'}}]}, {}))
+        self.assertEqual(doc._delta(), ({
+            'embedded_field.list_field': ['1', 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)
+        self.assertEqual(doc.embedded_field.list_field[2].string_field,
+                        'hello world')
+
+        # Test list native methods
+        doc.embedded_field.list_field[2].list_field.pop(0)
+        self.assertEqual(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)
+        self.assertEqual(doc._delta(),
+            ({'embedded_field.list_field.2.list_field':
+                [2, {'hello': 'world'}, 1]}, {}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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)
+        self.assertEqual(doc.embedded_field.list_field[2].list_field,
+            [1, 2, {'hello': 'world'}])
+
+        del(doc.embedded_field.list_field[2].list_field[2]['hello'])
+        self.assertEqual(doc._delta(),
+            ({'embedded_field.list_field.2.list_field': [1, 2, {}]}, {}))
+        doc.save()
+        doc = doc.reload(10)
+
+        del(doc.embedded_field.list_field[2].list_field)
+        self.assertEqual(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'
+        self.assertEqual(doc._get_changed_fields(),
+            ['dict_field.Embedded.string_field'])
+        self.assertEqual(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 = Person(name="owner")
+        person.save()
+        organization = Organization(name="company")
+        organization.save()
+
+        person.owns.append(organization)
+        organization.owner = person
+
+        person.save()
+        organization.save()
+
+        p = Person.objects[0].select_related()
+        o = Organization.objects.first()
+        self.assertEqual(p.owns[0], o)
+        self.assertEqual(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):
+
+        class Person(DocClass1):
+            name = StringField()
+            owns = ListField(ReferenceField('Organization'))
+            employer = ReferenceField('Organization')
+
+        class Organization(DocClass2):
+            name = StringField()
+            owner = ReferenceField('Person')
+            employees = ListField(ReferenceField('Person'))
+
+        Person.drop_collection()
+        Organization.drop_collection()
+
+        person = Person(name="owner")
+        person.save()
+
+        employee = Person(name="employee")
+        employee.save()
+
+        organization = Organization(name="company")
+        organization.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()
+
+        self.assertEqual(p.owns[0], o)
+        self.assertEqual(o.owner, p)
+        self.assertEqual(e.employer, o)
+
+    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()
+        self.assertEqual(doc._get_changed_fields(), [])
+        self.assertEqual(doc._delta(), ({}, {}))
+
+        doc.string_field = 'hello'
+        self.assertEqual(doc._get_changed_fields(), ['db_string_field'])
+        self.assertEqual(doc._delta(), ({'db_string_field': 'hello'}, {}))
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        self.assertEqual(doc._get_changed_fields(), ['db_int_field'])
+        self.assertEqual(doc._delta(), ({'db_int_field': 1}, {}))
+
+        doc._changed_fields = []
+        dict_value = {'hello': 'world', 'ping': 'pong'}
+        doc.dict_field = dict_value
+        self.assertEqual(doc._get_changed_fields(), ['db_dict_field'])
+        self.assertEqual(doc._delta(), ({'db_dict_field': dict_value}, {}))
+
+        doc._changed_fields = []
+        list_value = ['1', 2, {'hello': 'world'}]
+        doc.list_field = list_value
+        self.assertEqual(doc._get_changed_fields(), ['db_list_field'])
+        self.assertEqual(doc._delta(), ({'db_list_field': list_value}, {}))
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        self.assertEqual(doc._get_changed_fields(), ['db_dict_field'])
+        self.assertEqual(doc._delta(), ({}, {'db_dict_field': 1}))
+
+        doc._changed_fields = []
+        doc.list_field = []
+        self.assertEqual(doc._get_changed_fields(), ['db_list_field'])
+        self.assertEqual(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)
+
+        self.assertEqual(doc.string_field, 'hello')
+        self.assertEqual(doc.int_field, 1)
+        self.assertEqual(doc.dict_field, {'hello': 'world'})
+        self.assertEqual(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()
+        self.assertEqual(doc._get_changed_fields(), [])
+        self.assertEqual(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
+
+        self.assertEqual(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'}]
+        }
+        self.assertEqual(doc.embedded_field._delta(), (embedded_delta, {}))
+        self.assertEqual(doc._delta(),
+            ({'db_embedded_field': embedded_delta}, {}))
+
+        doc.save()
+        doc = doc.reload(10)
+
+        doc.embedded_field.dict_field = {}
+        self.assertEqual(doc._get_changed_fields(),
+            ['db_embedded_field.db_dict_field'])
+        self.assertEqual(doc.embedded_field._delta(),
+            ({}, {'db_dict_field': 1}))
+        self.assertEqual(doc._delta(),
+            ({}, {'db_embedded_field.db_dict_field': 1}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(doc.embedded_field.dict_field, {})
+
+        doc.embedded_field.list_field = []
+        self.assertEqual(doc._get_changed_fields(),
+            ['db_embedded_field.db_list_field'])
+        self.assertEqual(doc.embedded_field._delta(),
+            ({}, {'db_list_field': 1}))
+        self.assertEqual(doc._delta(),
+            ({}, {'db_embedded_field.db_list_field': 1}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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]
+        self.assertEqual(doc._get_changed_fields(),
+            ['db_embedded_field.db_list_field'])
+        self.assertEqual(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'}],
+            }]
+        }, {}))
+
+        self.assertEqual(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)
+
+        self.assertEqual(doc.embedded_field.list_field[0], '1')
+        self.assertEqual(doc.embedded_field.list_field[1], 2)
+        for k in doc.embedded_field.list_field[2]._fields:
+            self.assertEqual(doc.embedded_field.list_field[2][k],
+                             embedded_2[k])
+
+        doc.embedded_field.list_field[2].string_field = 'world'
+        self.assertEqual(doc._get_changed_fields(),
+            ['db_embedded_field.db_list_field.2.db_string_field'])
+        self.assertEqual(doc.embedded_field._delta(),
+            ({'db_list_field.2.db_string_field': 'world'}, {}))
+        self.assertEqual(doc._delta(),
+            ({'db_embedded_field.db_list_field.2.db_string_field': 'world'},
+             {}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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]
+        self.assertEqual(doc._get_changed_fields(),
+            ['db_embedded_field.db_list_field'])
+        self.assertEqual(doc.embedded_field._delta(), ({
+            'db_list_field': ['1', 2, {
+            '_cls': 'Embedded',
+            'db_string_field': 'hello world',
+            'db_int_field': 1,
+            'db_list_field': ['1', 2, {'hello': 'world'}],
+            'db_dict_field': {'hello': 'world'}}]}, {}))
+        self.assertEqual(doc._delta(), ({
+            'db_embedded_field.db_list_field': ['1', 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)
+        self.assertEqual(doc.embedded_field.list_field[2].string_field,
+                        'hello world')
+
+        # Test list native methods
+        doc.embedded_field.list_field[2].list_field.pop(0)
+        self.assertEqual(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)
+        self.assertEqual(doc._delta(),
+            ({'db_embedded_field.db_list_field.2.db_list_field':
+                [2, {'hello': 'world'}, 1]}, {}))
+        doc.save()
+        doc = doc.reload(10)
+        self.assertEqual(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)
+        self.assertEqual(doc.embedded_field.list_field[2].list_field,
+            [1, 2, {'hello': 'world'}])
+
+        del(doc.embedded_field.list_field[2].list_field[2]['hello'])
+        self.assertEqual(doc._delta(),
+            ({'db_embedded_field.db_list_field.2.db_list_field':
+                [1, 2, {}]}, {}))
+        doc.save()
+        doc = doc.reload(10)
+
+        del(doc.embedded_field.list_field[2].list_field)
+        self.assertEqual(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)
+        self.assertEqual(p._delta(), ({'age': 34, 'name': 'James',
+                                       '_cls': 'Person'}, {}))
+
+        p.doc = 123
+        del(p.doc)
+        self.assertEqual(p._delta(), ({'age': 34, 'name': 'James',
+                                       '_cls': 'Person'}, {'doc': 1}))
+
+        p = Person()
+        p.name = "Dean"
+        p.age = 22
+        p.save()
+
+        p.age = 24
+        self.assertEqual(p.age, 24)
+        self.assertEqual(p._get_changed_fields(), ['age'])
+        self.assertEqual(p._delta(), ({'age': 24}, {}))
+
+        p = self.Person.objects(age=22).get()
+        p.age = 24
+        self.assertEqual(p.age, 24)
+        self.assertEqual(p._get_changed_fields(), ['age'])
+        self.assertEqual(p._delta(), ({'age': 24}, {}))
+
+        p.save()
+        self.assertEqual(1, self.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()
+        self.assertEqual(doc._get_changed_fields(), [])
+        self.assertEqual(doc._delta(), ({}, {}))
+
+        doc.string_field = 'hello'
+        self.assertEqual(doc._get_changed_fields(), ['string_field'])
+        self.assertEqual(doc._delta(), ({'string_field': 'hello'}, {}))
+
+        doc._changed_fields = []
+        doc.int_field = 1
+        self.assertEqual(doc._get_changed_fields(), ['int_field'])
+        self.assertEqual(doc._delta(), ({'int_field': 1}, {}))
+
+        doc._changed_fields = []
+        dict_value = {'hello': 'world', 'ping': 'pong'}
+        doc.dict_field = dict_value
+        self.assertEqual(doc._get_changed_fields(), ['dict_field'])
+        self.assertEqual(doc._delta(), ({'dict_field': dict_value}, {}))
+
+        doc._changed_fields = []
+        list_value = ['1', 2, {'hello': 'world'}]
+        doc.list_field = list_value
+        self.assertEqual(doc._get_changed_fields(), ['list_field'])
+        self.assertEqual(doc._delta(), ({'list_field': list_value}, {}))
+
+        # Test unsetting
+        doc._changed_fields = []
+        doc.dict_field = {}
+        self.assertEqual(doc._get_changed_fields(), ['dict_field'])
+        self.assertEqual(doc._delta(), ({}, {'dict_field': 1}))
+
+        doc._changed_fields = []
+        doc.list_field = []
+        self.assertEqual(doc._get_changed_fields(), ['list_field'])
+        self.assertEqual(doc._delta(), ({}, {'list_field': 1}))
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/dynamic.py

@@ -0,0 +1,296 @@
+import unittest
+import sys
+sys.path[0:0] = [""]
+
+from mongoengine import *
+from mongoengine.connection import get_db
+
+__all__ = ("DynamicTest", )
+
+
+class DynamicTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+        self.db = get_db()
+
+        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
+
+        self.assertEqual(p.to_mongo(), {"_cls": "Person", "name": "James",
+                                        "age": 34})
+
+        p.save()
+
+        self.assertEqual(self.Person.objects.first().age, 34)
+
+        # Confirm no changes to self.Person
+        self.assertFalse(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()
+        self.assertEqual(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()
+        self.assertEqual(p.misc, {'hello': 'world'})
+        collection = self.db[self.Person._get_collection_name()]
+        obj = collection.find_one()
+        self.assertEqual(sorted(obj.keys()), ['_cls', '_id', 'misc', 'name'])
+
+        del(p.misc)
+        p.save()
+
+        p = self.Person.objects.get()
+        self.assertFalse(hasattr(p, 'misc'))
+
+        obj = collection.find_one()
+        self.assertEqual(sorted(obj.keys()), ['_cls', '_id', 'name'])
+
+    def test_dynamic_document_queries(self):
+        """Ensure we can query dynamic fields"""
+        p = self.Person()
+        p.name = "Dean"
+        p.age = 22
+        p.save()
+
+        self.assertEqual(1, self.Person.objects(age=22).count())
+        p = self.Person.objects(age=22)
+        p = p.get()
+        self.assertEqual(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()
+
+        self.assertEqual(Person.objects(age__icontains='ten').count(), 2)
+        self.assertEqual(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()
+
+        self.assertEqual(1, self.Person.objects(misc__hello='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')
+        self.assertRaises(ValidationError, embedded_doc_2.validate)
+
+        doc.embedded_field_1 = embedded_doc_1
+        doc.embedded_field_2 = embedded_doc_2
+        self.assertRaises(ValidationError, doc.validate)
+
+    def test_inheritance(self):
+        """Ensure that dynamic document plays nice with inheritance"""
+        class Employee(self.Person):
+            salary = IntField()
+
+        Employee.drop_collection()
+
+        self.assertTrue('name' in Employee._fields)
+        self.assertTrue('salary' in Employee._fields)
+        self.assertEqual(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()
+
+        self.assertEqual(1, self.Person.objects(age=20).count())
+        self.assertEqual(1, Employee.objects(age=20).count())
+
+        joe_bloggs = self.Person.objects.first()
+        self.assertTrue(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
+
+        self.assertEqual(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()
+        self.assertEqual(doc.embedded_field.__class__, Embedded)
+        self.assertEqual(doc.embedded_field.string_field, "hello")
+        self.assertEqual(doc.embedded_field.int_field, 1)
+        self.assertEqual(doc.embedded_field.dict_field, {'hello': 'world'})
+        self.assertEqual(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
+
+        self.assertEqual(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()
+        self.assertEqual(doc.embedded_field.__class__, Embedded)
+        self.assertEqual(doc.embedded_field.string_field, "hello")
+        self.assertEqual(doc.embedded_field.int_field, 1)
+        self.assertEqual(doc.embedded_field.dict_field, {'hello': 'world'})
+        self.assertEqual(doc.embedded_field.list_field[0], '1')
+        self.assertEqual(doc.embedded_field.list_field[1], 2)
+
+        embedded_field = doc.embedded_field.list_field[2]
+
+        self.assertEqual(embedded_field.__class__, Embedded)
+        self.assertEqual(embedded_field.string_field, "hello")
+        self.assertEqual(embedded_field.int_field, 1)
+        self.assertEqual(embedded_field.dict_field, {'hello': 'world'})
+        self.assertEqual(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()
+
+        self.assertEqual(Person.objects.first().address.city, "Lundenne")
+
+        person = Person.objects.first()
+        person.address = Address(city="Londinium")
+        person.save()
+
+        self.assertEqual(Person.objects.first().address.city, "Londinium")
+
+        person = Person.objects.first()
+        person.age = 35
+        person.save()
+        self.assertEqual(Person.objects.first().age, 35)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/indexes.py

@@ -0,0 +1,730 @@
+# -*- coding: utf-8 -*-
+from __future__ import with_statement
+import unittest
+import sys
+sys.path[0:0] = [""]
+
+import os
+import pymongo
+
+from nose.plugins.skip import SkipTest
+from datetime import datetime
+
+from mongoengine import *
+from mongoengine.connection import get_db, get_connection
+
+__all__ = ("IndexesTest", )
+
+
+class IndexesTest(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 self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            self.db.drop_collection(collection)
+
+    def test_indexes_document(self):
+        """Ensure that indexes are used when meta[indexes] is specified for
+        Documents
+        """
+        self._index_test(Document)
+
+    def test_indexes_dynamic_document(self):
+        """Ensure that indexes are used when meta[indexes] is specified for
+        Dynamic Documents
+        """
+        self._index_test(DynamicDocument)
+
+    def _index_test(self, InheritFrom):
+
+        class BlogPost(InheritFrom):
+            date = DateTimeField(db_field='addDate', default=datetime.now)
+            category = StringField()
+            tags = ListField(StringField())
+            meta = {
+                'indexes': [
+                    '-date',
+                    'tags',
+                    ('category', '-date')
+                ]
+            }
+
+        expected_specs = [{'fields': [('addDate', -1)]},
+                          {'fields': [('tags', 1)]},
+                          {'fields': [('category', 1), ('addDate', -1)]}]
+        self.assertEqual(expected_specs, BlogPost._meta['index_specs'])
+
+        BlogPost.ensure_indexes()
+        info = BlogPost.objects._collection.index_information()
+        # _id, '-date', 'tags', ('cat', 'date')
+        self.assertEqual(len(info), 4)
+        info = [value['key'] for key, value in info.iteritems()]
+        for expected in expected_specs:
+            self.assertTrue(expected['fields'] in info)
+
+    def _index_test_inheritance(self, InheritFrom):
+
+        class BlogPost(InheritFrom):
+            date = DateTimeField(db_field='addDate', default=datetime.now)
+            category = StringField()
+            tags = ListField(StringField())
+            meta = {
+                'indexes': [
+                    '-date',
+                    'tags',
+                    ('category', '-date')
+                ],
+                'allow_inheritance': True
+            }
+
+        expected_specs = [{'fields': [('_cls', 1), ('addDate', -1)]},
+                          {'fields': [('_cls', 1), ('tags', 1)]},
+                          {'fields': [('_cls', 1), ('category', 1),
+                                      ('addDate', -1)]}]
+        self.assertEqual(expected_specs, BlogPost._meta['index_specs'])
+
+        BlogPost.ensure_indexes()
+        info = BlogPost.objects._collection.index_information()
+        # _id, '-date', 'tags', ('cat', 'date')
+        # NB: there is no index on _cls by itself, since
+        # the indices on -date and tags will both contain
+        # _cls as first element in the key
+        self.assertEqual(len(info), 4)
+        info = [value['key'] for key, value in info.iteritems()]
+        for expected in expected_specs:
+            self.assertTrue(expected['fields'] in info)
+
+        class ExtendedBlogPost(BlogPost):
+            title = StringField()
+            meta = {'indexes': ['title']}
+
+        expected_specs.append({'fields': [('_cls', 1), ('title', 1)]})
+        self.assertEqual(expected_specs, ExtendedBlogPost._meta['index_specs'])
+
+        BlogPost.drop_collection()
+
+        ExtendedBlogPost.ensure_indexes()
+        info = ExtendedBlogPost.objects._collection.index_information()
+        info = [value['key'] for key, value in info.iteritems()]
+        for expected in expected_specs:
+            self.assertTrue(expected['fields'] in info)
+
+    def test_indexes_document_inheritance(self):
+        """Ensure that indexes are used when meta[indexes] is specified for
+        Documents
+        """
+        self._index_test_inheritance(Document)
+
+    def test_indexes_dynamic_document_inheritance(self):
+        """Ensure that indexes are used when meta[indexes] is specified for
+        Dynamic Documents
+        """
+        self._index_test_inheritance(DynamicDocument)
+
+    def test_inherited_index(self):
+        """Ensure index specs are inhertited correctly"""
+
+        class A(Document):
+            title = StringField()
+            meta = {
+                'indexes': [
+                        {
+                        'fields': ('title',),
+                        },
+                ],
+                'allow_inheritance': True,
+                }
+
+        class B(A):
+            description = StringField()
+
+        self.assertEqual(A._meta['index_specs'], B._meta['index_specs'])
+        self.assertEqual([{'fields': [('_cls', 1), ('title', 1)]}],
+                         A._meta['index_specs'])
+
+    def test_build_index_spec_is_not_destructive(self):
+
+        class MyDoc(Document):
+            keywords = StringField()
+
+            meta = {
+                'indexes': ['keywords'],
+                'allow_inheritance': False
+            }
+
+        self.assertEqual(MyDoc._meta['index_specs'],
+                         [{'fields': [('keywords', 1)]}])
+
+        # Force index creation
+        MyDoc.ensure_indexes()
+
+        self.assertEqual(MyDoc._meta['index_specs'],
+                        [{'fields': [('keywords', 1)]}])
+
+    def test_embedded_document_index_meta(self):
+        """Ensure that embedded document indexes are created explicitly
+        """
+        class Rank(EmbeddedDocument):
+            title = StringField(required=True)
+
+        class Person(Document):
+            name = StringField(required=True)
+            rank = EmbeddedDocumentField(Rank, required=False)
+
+            meta = {
+                'indexes': [
+                    'rank.title',
+                ],
+                'allow_inheritance': False
+            }
+
+        self.assertEqual([{'fields': [('rank.title', 1)]}],
+                        Person._meta['index_specs'])
+
+        Person.drop_collection()
+
+        # Indexes are lazy so use list() to perform query
+        list(Person.objects)
+        info = Person.objects._collection.index_information()
+        info = [value['key'] for key, value in info.iteritems()]
+        self.assertTrue([('rank.title', 1)] in info)
+
+    def test_explicit_geo2d_index(self):
+        """Ensure that geo2d indexes work when created via meta[indexes]
+        """
+        class Place(Document):
+            location = DictField()
+            meta = {
+                'allow_inheritance': True,
+                'indexes': [
+                    '*location.point',
+                ]
+            }
+
+        self.assertEqual([{'fields': [('location.point', '2d')]}],
+                        Place._meta['index_specs'])
+
+        Place.ensure_indexes()
+        info = Place._get_collection().index_information()
+        info = [value['key'] for key, value in info.iteritems()]
+        self.assertTrue([('location.point', '2d')] in info)
+
+    def test_explicit_geo2d_index_embedded(self):
+        """Ensure that geo2d indexes work when created via meta[indexes]
+        """
+        class EmbeddedLocation(EmbeddedDocument):
+            location = DictField()
+
+        class Place(Document):
+            current = DictField(
+                    field=EmbeddedDocumentField('EmbeddedLocation'))
+            meta = {
+                'allow_inheritance': True,
+                'indexes': [
+                    '*current.location.point',
+                ]
+            }
+
+        self.assertEqual([{'fields': [('current.location.point', '2d')]}],
+                        Place._meta['index_specs'])
+
+        Place.ensure_indexes()
+        info = Place._get_collection().index_information()
+        info = [value['key'] for key, value in info.iteritems()]
+        self.assertTrue([('current.location.point', '2d')] in info)
+
+    def test_dictionary_indexes(self):
+        """Ensure that indexes are used when meta[indexes] contains
+        dictionaries instead of lists.
+        """
+        class BlogPost(Document):
+            date = DateTimeField(db_field='addDate', default=datetime.now)
+            category = StringField()
+            tags = ListField(StringField())
+            meta = {
+                'indexes': [
+                    {'fields': ['-date'], 'unique': True, 'sparse': True},
+                ],
+            }
+
+        self.assertEqual([{'fields': [('addDate', -1)], 'unique': True,
+                          'sparse': True}],
+                        BlogPost._meta['index_specs'])
+
+        BlogPost.drop_collection()
+
+        info = BlogPost.objects._collection.index_information()
+        # _id, '-date'
+        self.assertEqual(len(info), 2)
+
+        # Indexes are lazy so use list() to perform query
+        list(BlogPost.objects)
+        info = BlogPost.objects._collection.index_information()
+        info = [(value['key'],
+                 value.get('unique', False),
+                 value.get('sparse', False))
+                for key, value in info.iteritems()]
+        self.assertTrue(([('addDate', -1)], True, True) in info)
+
+        BlogPost.drop_collection()
+
+    def test_abstract_index_inheritance(self):
+
+        class UserBase(Document):
+            user_guid = StringField(required=True)
+            meta = {
+                'abstract': True,
+                'indexes': ['user_guid'],
+                'allow_inheritance': True
+            }
+
+        class Person(UserBase):
+            name = StringField()
+
+            meta = {
+                'indexes': ['name'],
+            }
+        Person.drop_collection()
+
+        Person(name="test", user_guid='123').save()
+
+        self.assertEqual(1, Person.objects.count())
+        info = Person.objects._collection.index_information()
+        self.assertEqual(sorted(info.keys()),
+                         ['_cls_1_name_1', '_cls_1_user_guid_1', '_id_'])
+
+    def test_disable_index_creation(self):
+        """Tests setting auto_create_index to False on the connection will
+        disable any index generation.
+        """
+        class User(Document):
+            meta = {
+                'indexes': ['user_guid'],
+                'auto_create_index': False
+            }
+            user_guid = StringField(required=True)
+
+        User.drop_collection()
+
+        u = User(user_guid='123')
+        u.save()
+
+        self.assertEqual(1, User.objects.count())
+        info = User.objects._collection.index_information()
+        self.assertEqual(info.keys(), ['_id_'])
+        User.drop_collection()
+
+    def test_embedded_document_index(self):
+        """Tests settings an index on an embedded document
+        """
+        class Date(EmbeddedDocument):
+            year = IntField(db_field='yr')
+
+        class BlogPost(Document):
+            title = StringField()
+            date = EmbeddedDocumentField(Date)
+
+            meta = {
+                'indexes': [
+                    '-date.year'
+                ],
+            }
+
+        BlogPost.drop_collection()
+
+        info = BlogPost.objects._collection.index_information()
+        self.assertEqual(sorted(info.keys()), ['_id_', 'date.yr_-1'])
+        BlogPost.drop_collection()
+
+    def test_list_embedded_document_index(self):
+        """Ensure list embedded documents can be indexed
+        """
+        class Tag(EmbeddedDocument):
+            name = StringField(db_field='tag')
+
+        class BlogPost(Document):
+            title = StringField()
+            tags = ListField(EmbeddedDocumentField(Tag))
+
+            meta = {
+                'indexes': [
+                    'tags.name'
+                ]
+            }
+
+        BlogPost.drop_collection()
+
+        info = BlogPost.objects._collection.index_information()
+        # we don't use _cls in with list fields by default
+        self.assertEqual(sorted(info.keys()), ['_id_', 'tags.tag_1'])
+
+        post1 = BlogPost(title="Embedded Indexes tests in place",
+                        tags=[Tag(name="about"), Tag(name="time")]
+                )
+        post1.save()
+        BlogPost.drop_collection()
+
+    def test_recursive_embedded_objects_dont_break_indexes(self):
+
+        class RecursiveObject(EmbeddedDocument):
+            obj = EmbeddedDocumentField('self')
+
+        class RecursiveDocument(Document):
+            recursive_obj = EmbeddedDocumentField(RecursiveObject)
+            meta = {'allow_inheritance': True}
+
+        RecursiveDocument.ensure_indexes()
+        info = RecursiveDocument._get_collection().index_information()
+        self.assertEqual(sorted(info.keys()), ['_cls_1', '_id_'])
+
+    def test_geo_indexes_recursion(self):
+
+        class Location(Document):
+            name = StringField()
+            location = GeoPointField()
+
+        class Parent(Document):
+            name = StringField()
+            location = ReferenceField(Location, dbref=False)
+
+        Location.drop_collection()
+        Parent.drop_collection()
+
+        list(Parent.objects)
+
+        collection = Parent._get_collection()
+        info = collection.index_information()
+
+        self.assertFalse('location_2d' in info)
+
+        self.assertEqual(len(Parent._geo_indices()), 0)
+        self.assertEqual(len(Location._geo_indices()), 1)
+
+    def test_covered_index(self):
+        """Ensure that covered indexes can be used
+        """
+
+        class Test(Document):
+            a = IntField()
+
+            meta = {
+                'indexes': ['a'],
+                'allow_inheritance': False
+                }
+
+        Test.drop_collection()
+
+        obj = Test(a=1)
+        obj.save()
+
+        # Need to be explicit about covered indexes as mongoDB doesn't know if
+        # the documents returned might have more keys in that here.
+        query_plan = Test.objects(id=obj.id).exclude('a').explain()
+        self.assertFalse(query_plan['indexOnly'])
+
+        query_plan = Test.objects(id=obj.id).only('id').explain()
+        self.assertTrue(query_plan['indexOnly'])
+
+        query_plan = Test.objects(a=1).only('a').exclude('id').explain()
+        self.assertTrue(query_plan['indexOnly'])
+
+    def test_index_on_id(self):
+
+        class BlogPost(Document):
+            meta = {
+                'indexes': [
+                    ['categories', 'id']
+                ]
+            }
+
+            title = StringField(required=True)
+            description = StringField(required=True)
+            categories = ListField()
+
+        BlogPost.drop_collection()
+
+        indexes = BlogPost.objects._collection.index_information()
+        self.assertEqual(indexes['categories_1__id_1']['key'],
+                                 [('categories', 1), ('_id', 1)])
+
+    def test_hint(self):
+
+        class BlogPost(Document):
+            tags = ListField(StringField())
+            meta = {
+                'indexes': [
+                    'tags',
+                ],
+            }
+
+        BlogPost.drop_collection()
+
+        for i in xrange(0, 10):
+            tags = [("tag %i" % n) for n in xrange(0, i % 2)]
+            BlogPost(tags=tags).save()
+
+        self.assertEqual(BlogPost.objects.count(), 10)
+        self.assertEqual(BlogPost.objects.hint().count(), 10)
+        self.assertEqual(BlogPost.objects.hint([('tags', 1)]).count(), 10)
+
+        self.assertEqual(BlogPost.objects.hint([('ZZ', 1)]).count(), 10)
+
+        def invalid_index():
+            BlogPost.objects.hint('tags')
+        self.assertRaises(TypeError, invalid_index)
+
+        def invalid_index_2():
+            return BlogPost.objects.hint(('tags', 1))
+        self.assertRaises(TypeError, invalid_index_2)
+
+    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(NotUniqueError, post2.save)
+
+        # Ensure backwards compatibilty for errors
+        self.assertRaises(OperationError, post2.save)
+
+    def test_unique_with(self):
+        """Ensure that unique_with constraints are applied to fields.
+        """
+        class Date(EmbeddedDocument):
+            year = IntField(db_field='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_unique_embedded_document(self):
+        """Ensure that uniqueness constraints are applied to fields on embedded documents.
+        """
+        class SubDocument(EmbeddedDocument):
+            year = IntField(db_field='yr')
+            slug = StringField(unique=True)
+
+        class BlogPost(Document):
+            title = StringField()
+            sub = EmbeddedDocumentField(SubDocument)
+
+        BlogPost.drop_collection()
+
+        post1 = BlogPost(title='test1',
+                         sub=SubDocument(year=2009, slug="test"))
+        post1.save()
+
+        # sub.slug is different so won't raise exception
+        post2 = BlogPost(title='test2',
+                         sub=SubDocument(year=2010, slug='another-slug'))
+        post2.save()
+
+        # Now there will be two docs with the same sub.slug
+        post3 = BlogPost(title='test3',
+                         sub=SubDocument(year=2010, slug='test'))
+        self.assertRaises(NotUniqueError, post3.save)
+
+        BlogPost.drop_collection()
+
+    def test_unique_with_embedded_document_and_embedded_unique(self):
+        """Ensure that uniqueness constraints are applied to fields on
+        embedded documents.  And work with unique_with as well.
+        """
+        class SubDocument(EmbeddedDocument):
+            year = IntField(db_field='yr')
+            slug = StringField(unique=True)
+
+        class BlogPost(Document):
+            title = StringField(unique_with='sub.year')
+            sub = EmbeddedDocumentField(SubDocument)
+
+        BlogPost.drop_collection()
+
+        post1 = BlogPost(title='test1',
+                         sub=SubDocument(year=2009, slug="test"))
+        post1.save()
+
+        # sub.slug is different so won't raise exception
+        post2 = BlogPost(title='test2',
+                         sub=SubDocument(year=2010, slug='another-slug'))
+        post2.save()
+
+        # Now there will be two docs with the same sub.slug
+        post3 = BlogPost(title='test3',
+                         sub=SubDocument(year=2010, slug='test'))
+        self.assertRaises(NotUniqueError, post3.save)
+
+        # Now there will be two docs with the same title and year
+        post3 = BlogPost(title='test1',
+                         sub=SubDocument(year=2009, slug='test-1'))
+        self.assertRaises(NotUniqueError, post3.save)
+
+        BlogPost.drop_collection()
+
+    def test_ttl_indexes(self):
+
+        class Log(Document):
+            created = DateTimeField(default=datetime.now)
+            meta = {
+                'indexes': [
+                    {'fields': ['created'], 'expireAfterSeconds': 3600}
+                ]
+            }
+
+        Log.drop_collection()
+
+        if pymongo.version_tuple[0] < 2 and pymongo.version_tuple[1] < 3:
+            raise SkipTest('pymongo needs to be 2.3 or higher for this test')
+
+        connection = get_connection()
+        version_array = connection.server_info()['versionArray']
+        if version_array[0] < 2 and version_array[1] < 2:
+            raise SkipTest('MongoDB needs to be 2.2 or higher for this test')
+
+        # Indexes are lazy so use list() to perform query
+        list(Log.objects)
+        info = Log.objects._collection.index_information()
+        self.assertEqual(3600,
+                info['created_1']['expireAfterSeconds'])
+
+    def test_unique_and_indexes(self):
+        """Ensure that 'unique' constraints aren't overridden by
+        meta.indexes.
+        """
+        class Customer(Document):
+            cust_id = IntField(unique=True, required=True)
+            meta = {
+                'indexes': ['cust_id'],
+                'allow_inheritance': False,
+            }
+
+        Customer.drop_collection()
+        cust = Customer(cust_id=1)
+        cust.save()
+
+        cust_dupe = Customer(cust_id=1)
+        try:
+            cust_dupe.save()
+            raise AssertionError("We saved a dupe!")
+        except NotUniqueError:
+            pass
+        Customer.drop_collection()
+
+    def test_unique_and_primary(self):
+        """If you set a field as primary, then unexpected behaviour can occur.
+        You won't create a duplicate but you will update an existing document.
+        """
+
+        class User(Document):
+            name = StringField(primary_key=True, unique=True)
+            password = StringField()
+
+        User.drop_collection()
+
+        user = User(name='huangz', password='secret')
+        user.save()
+
+        user = User(name='huangz', password='secret2')
+        user.save()
+
+        self.assertEqual(User.objects.count(), 1)
+        self.assertEqual(User.objects.get().password, 'secret2')
+
+        User.drop_collection()
+
+    def test_index_with_pk(self):
+        """Ensure you can use `pk` as part of a query"""
+
+        class Comment(EmbeddedDocument):
+            comment_id = IntField(required=True)
+
+        try:
+            class BlogPost(Document):
+                comments = EmbeddedDocumentField(Comment)
+                meta = {'indexes': [
+                            {'fields': ['pk', 'comments.comment_id'],
+                             'unique': True}]}
+        except UnboundLocalError:
+            self.fail('Unbound local error at index + pk definition')
+
+        info = BlogPost.objects._collection.index_information()
+        info = [value['key'] for key, value in info.iteritems()]
+        index_item = [('_id', 1), ('comments.comment_id', 1)]
+        self.assertTrue(index_item in info)
+
+    def test_compound_key_embedded(self):
+
+        class CompoundKey(EmbeddedDocument):
+            name = StringField(required=True)
+            term = StringField(required=True)
+
+        class Report(Document):
+            key = EmbeddedDocumentField(CompoundKey, primary_key=True)
+            text = StringField()
+
+        Report.drop_collection()
+
+        my_key = CompoundKey(name="n", term="ok")
+        report = Report(text="OK", key=my_key).save()
+
+        self.assertEqual({'text': 'OK', '_id': {'term': 'ok', 'name': 'n'}},
+                         report.to_mongo())
+        self.assertEqual(report, Report.objects.get(pk=my_key))
+
+    def test_compound_key_dictfield(self):
+
+        class Report(Document):
+            key = DictField(primary_key=True)
+            text = StringField()
+
+        Report.drop_collection()
+
+        my_key = {"name": "n", "term": "ok"}
+        report = Report(text="OK", key=my_key).save()
+
+        self.assertEqual({'text': 'OK', '_id': {'term': 'ok', 'name': 'n'}},
+                         report.to_mongo())
+        self.assertEqual(report, Report.objects.get(pk=my_key))
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/inheritance.py

@@ -0,0 +1,395 @@
+# -*- coding: utf-8 -*-
+import sys
+sys.path[0:0] = [""]
+import unittest
+import warnings
+
+from datetime import datetime
+
+from tests.fixtures import Base
+
+from mongoengine import Document, EmbeddedDocument, connect
+from mongoengine.connection import get_db
+from mongoengine.fields import (BooleanField, GenericReferenceField,
+                                IntField, StringField)
+
+__all__ = ('InheritanceTest', )
+
+
+class InheritanceTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+        self.db = get_db()
+
+    def tearDown(self):
+        for collection in self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            self.db.drop_collection(collection)
+
+    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
+
+        self.assertEqual(Animal._superclasses, ())
+        self.assertEqual(Fish._superclasses, ('Animal',))
+        self.assertEqual(Guppy._superclasses, ('Animal', 'Animal.Fish'))
+        self.assertEqual(Mammal._superclasses, ('Animal',))
+        self.assertEqual(Dog._superclasses, ('Animal', 'Animal.Mammal'))
+        self.assertEqual(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
+
+        self.assertEqual(Animal._superclasses, ('Base', ))
+        self.assertEqual(Fish._superclasses, ('Base', 'Base.Animal',))
+        self.assertEqual(Guppy._superclasses, ('Base', 'Base.Animal',
+                                               'Base.Animal.Fish'))
+        self.assertEqual(Mammal._superclasses, ('Base', 'Base.Animal',))
+        self.assertEqual(Dog._superclasses, ('Base', 'Base.Animal',
+                                             'Base.Animal.Mammal'))
+        self.assertEqual(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
+
+        self.assertEqual(Animal._subclasses, ('Animal',
+                                         'Animal.Fish',
+                                         'Animal.Fish.Guppy',
+                                         'Animal.Mammal',
+                                         'Animal.Mammal.Dog',
+                                         'Animal.Mammal.Human'))
+        self.assertEqual(Fish._subclasses, ('Animal.Fish',
+                                       'Animal.Fish.Guppy',))
+        self.assertEqual(Guppy._subclasses, ('Animal.Fish.Guppy',))
+        self.assertEqual(Mammal._subclasses, ('Animal.Mammal',
+                                         'Animal.Mammal.Dog',
+                                         'Animal.Mammal.Human'))
+        self.assertEqual(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
+
+        self.assertEqual(Animal._subclasses, ('Base.Animal',
+                                              'Base.Animal.Fish',
+                                              'Base.Animal.Fish.Guppy',
+                                              'Base.Animal.Mammal',
+                                              'Base.Animal.Mammal.Dog',
+                                              'Base.Animal.Mammal.Human'))
+        self.assertEqual(Fish._subclasses, ('Base.Animal.Fish',
+                                            'Base.Animal.Fish.Guppy',))
+        self.assertEqual(Guppy._subclasses, ('Base.Animal.Fish.Guppy',))
+        self.assertEqual(Mammal._subclasses, ('Base.Animal.Mammal',
+                                              'Base.Animal.Mammal.Dog',
+                                              'Base.Animal.Mammal.Human'))
+        self.assertEqual(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}
+
+        self.assertEqual(Animal._superclasses, ())
+        self.assertEqual(Animal._subclasses, ('Animal',))
+
+        # Test dynamically adding a class changes the meta data
+        class Fish(Animal):
+            pass
+
+        self.assertEqual(Animal._superclasses, ())
+        self.assertEqual(Animal._subclasses, ('Animal', 'Animal.Fish'))
+
+        self.assertEqual(Fish._superclasses, ('Animal', ))
+        self.assertEqual(Fish._subclasses, ('Animal.Fish',))
+
+        # Test dynamically adding an inherited class changes the meta data
+        class Pike(Fish):
+            pass
+
+        self.assertEqual(Animal._superclasses, ())
+        self.assertEqual(Animal._subclasses, ('Animal', 'Animal.Fish',
+                                         'Animal.Fish.Pike'))
+
+        self.assertEqual(Fish._superclasses, ('Animal', ))
+        self.assertEqual(Fish._subclasses, ('Animal.Fish', 'Animal.Fish.Pike'))
+
+        self.assertEqual(Pike._superclasses, ('Animal', 'Animal.Fish'))
+        self.assertEqual(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()
+
+        self.assertEqual(['age', 'id', 'name', 'salary'],
+                         sorted(Employee._fields.keys()))
+        self.assertEqual(Employee._get_collection_name(),
+                         Person._get_collection_name())
+
+
+    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]
+        self.assertEqual(classes, [Animal, Fish, Mammal, Dog, Human])
+
+        classes = [obj.__class__ for obj in Mammal.objects]
+        self.assertEqual(classes, [Mammal, Dog, Human])
+
+        classes = [obj.__class__ for obj in Human.objects]
+        self.assertEqual(classes, [Human])
+
+
+    def test_allow_inheritance(self):
+        """Ensure that inheritance may be disabled on simple classes and that
+        _cls and _subclasses will not be used.
+        """
+
+        class Animal(Document):
+            name = StringField()
+
+        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._get_collection_name()]
+        obj = collection.find_one()
+        self.assertFalse('_cls' 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}
+
+        def create_mammal_class():
+            class Mammal(Animal):
+                meta = {'allow_inheritance': False}
+        self.assertRaises(ValueError, create_mammal_class)
+
+    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()
+
+        def create_mammal_class():
+            class Mammal(Animal):
+                pass
+        self.assertRaises(ValueError, create_mammal_class)
+
+        # Check that _cls isn't present in simple documents
+        doc = Animal(name='dog')
+        self.assertFalse('_cls' in doc.to_mongo())
+
+    def test_allow_inheritance_embedded_document(self):
+        """Ensure embedded documents respect inheritance
+        """
+
+        class Comment(EmbeddedDocument):
+            content = StringField()
+
+        def create_special_comment():
+            class SpecialComment(Comment):
+                pass
+
+        self.assertRaises(ValueError, create_special_comment)
+
+        doc = Comment(content='test')
+        self.assertFalse('_cls' in doc.to_mongo())
+
+        class Comment(EmbeddedDocument):
+            content = StringField()
+            meta = {'allow_inheritance': True}
+
+        doc = Comment(content='test')
+        self.assertTrue('_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:
+            self.assertTrue(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_drop_dups': 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.iteritems():
+            for cls in [Animal, Fish, Guppy]:
+                self.assertEqual(cls._meta[k], v)
+
+        self.assertFalse('collection' in Animal._meta)
+        self.assertFalse('collection' in Mammal._meta)
+
+        self.assertEqual(Animal._get_collection_name(), None)
+        self.assertEqual(Mammal._get_collection_name(), None)
+
+        self.assertEqual(Fish._get_collection_name(), 'fish')
+        self.assertEqual(Guppy._get_collection_name(), 'fish')
+        self.assertEqual(Human._get_collection_name(), 'human')
+
+        def create_bad_abstract():
+            class EvilHuman(Human):
+                evil = BooleanField(default=True)
+                meta = {'abstract': True}
+        self.assertRaises(ValueError, create_bad_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()
+
+        self.assertEqual(Drinker.objects[0].drink.name, red_bull.name)
+        self.assertEqual(Drinker.objects[1].drink.name, beer.name)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/json_serialisation.py

@@ -0,0 +1,81 @@
+import sys
+sys.path[0:0] = [""]
+
+import unittest
+import uuid
+
+from nose.plugins.skip import SkipTest
+from datetime import datetime
+from bson import ObjectId
+
+import pymongo
+
+from mongoengine import *
+
+__all__ = ("TestJson",)
+
+
+class TestJson(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+    def test_json_simple(self):
+
+        class Embedded(EmbeddedDocument):
+            string = StringField()
+
+        class Doc(Document):
+            string = StringField()
+            embedded_field = EmbeddedDocumentField(Embedded)
+
+        doc = Doc(string="Hi", embedded_field=Embedded(string="Hi"))
+
+        self.assertEqual(doc, Doc.from_json(doc.to_json()))
+
+    def test_json_complex(self):
+
+        if pymongo.version_tuple[0] <= 2 and pymongo.version_tuple[1] <= 3:
+            raise SkipTest("Need pymongo 2.4 as has a fix for DBRefs")
+
+        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())
+
+        doc = Doc()
+        self.assertEqual(doc, Doc.from_json(doc.to_json()))
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/document/validation.py

@@ -0,0 +1,146 @@
+# -*- coding: utf-8 -*-
+import sys
+sys.path[0:0] = [""]
+
+import unittest
+from datetime import datetime
+
+from mongoengine import *
+
+__all__ = ("ValidatorErrorTest",)
+
+
+class ValidatorErrorTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+    def test_to_dict(self):
+        """Ensure a ValidationError handles error to_dict correctly.
+        """
+        error = ValidationError('root')
+        self.assertEqual(error.to_dict(), {})
+
+        # 1st level error schema
+        error.errors = {'1st': ValidationError('bad 1st'), }
+        self.assertTrue('1st' in error.to_dict())
+        self.assertEqual(error.to_dict()['1st'], 'bad 1st')
+
+        # 2nd level error schema
+        error.errors = {'1st': ValidationError('bad 1st', errors={
+            '2nd': ValidationError('bad 2nd'),
+        })}
+        self.assertTrue('1st' in error.to_dict())
+        self.assertTrue(isinstance(error.to_dict()['1st'], dict))
+        self.assertTrue('2nd' in error.to_dict()['1st'])
+        self.assertEqual(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'),
+                }),
+            }),
+        })}
+        self.assertTrue('1st' in error.to_dict())
+        self.assertTrue('2nd' in error.to_dict()['1st'])
+        self.assertTrue('3rd' in error.to_dict()['1st']['2nd'])
+        self.assertTrue('4th' in error.to_dict()['1st']['2nd']['3rd'])
+        self.assertEqual(error.to_dict()['1st']['2nd']['3rd']['4th'],
+                         'Inception')
+
+        self.assertEqual(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, e:
+            self.assertTrue("User:None" in e.message)
+            self.assertEqual(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, e:
+            self.assertTrue("User:RossC0" in e.message)
+            self.assertEqual(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)
+        self.assertRaises(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()
+        self.assertRaises(ValidationError, comment.validate)
+
+        comment.content = 'test'
+        comment.validate()
+
+        comment.date = 4
+        self.assertRaises(ValidationError, comment.validate)
+
+        comment.date = datetime.now()
+        comment.validate()
+        self.assertEqual(comment._instance, 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, e:
+            self.assertTrue("SubDoc:None" in e.message)
+            self.assertEqual(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()
+        self.assertEqual(2, len(keys))
+        self.assertTrue('e' in keys)
+        self.assertTrue('id' in keys)
+
+        doc.e.val = "OK"
+        try:
+            doc.save()
+        except ValidationError, e:
+            self.assertTrue("Doc:test" in e.message)
+            self.assertEqual(e.to_dict(), {
+                "e": {'val': 'OK could not be converted to int'}})
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/fields.py

@@ -1,555 +0,0 @@
-import unittest
-import datetime
-from decimal import Decimal
-
-import pymongo
-
-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_url_validation(self):
-        """Ensure that URLFields validate urls properly.
-        """
-        class Link(Document):
-            url = URLField()
-
-        link = Link()
-        link.url = 'google'
-        self.assertRaises(ValidationError, link.validate)
-
-        link.url = 'http://www.google.com:8080'
-        link.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 = '2.0'
-        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_decimal_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 = Person()
-        person.height = Decimal('1.89')
-        person.validate()
-
-        person.height = '2.0'
-        person.validate()
-        person.height = 0.01
-        self.assertRaises(ValidationError, person.validate)
-        person.height = Decimal('0.01')
-        self.assertRaises(ValidationError, person.validate)
-        person.height = Decimal('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_dict_validation(self):
-        """Ensure that dict types work as expected.
-        """
-        class BlogPost(Document):
-            info = DictField()
-
-        post = BlogPost()
-        post.info = 'my post'
-        self.assertRaises(ValidationError, post.validate)
-
-        post.info = ['test', 'test']
-        self.assertRaises(ValidationError, post.validate)
-
-        post.info = {'$title': 'test'}
-        self.assertRaises(ValidationError, post.validate)
-
-        post.info = {'the.title': 'test'}
-        self.assertRaises(ValidationError, post.validate)
-
-        post.info = {'title': 'test'}
-        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(required=True)
-            number = IntField()
-
-        class Person(Document):
-            name = StringField()
-            preferences = EmbeddedDocumentField(PersonPreferences)
-
-        person = Person(name='Test User')
-        person.preferences = 'My Preferences'
-        self.assertRaises(ValidationError, person.validate)
-
-        # Check that only the right embedded doc works
-        person.preferences = Comment(content='Nice blog post...')
-        self.assertRaises(ValidationError, person.validate)
-
-        # Check that the embedded doc is valid
-        person.preferences = PersonPreferences()
-        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()
-    
-    def test_list_item_dereference(self):
-        """Ensure that DBRef items in ListFields are dereferenced.
-        """
-        class User(Document):
-            name = StringField()
-
-        class Group(Document):
-            members = ListField(ReferenceField(User))
-
-        User.drop_collection()
-        Group.drop_collection()
-
-        user1 = User(name='user1')
-        user1.save()
-        user2 = User(name='user2')
-        user2.save()
-
-        group = Group(members=[user1, user2])
-        group.save()
-
-        group_obj = Group.objects.first()
-
-        self.assertEqual(group_obj.members[0].name, user1.name)
-        self.assertEqual(group_obj.members[1].name, user2.name)
-
-        User.drop_collection()
-        Group.drop_collection()
-
-    def test_recursive_reference(self):
-        """Ensure that ReferenceFields can reference their own documents.
-        """
-        class Employee(Document):
-            name = StringField()
-            boss = ReferenceField('self')
-
-        bill = Employee(name='Bill Lumbergh')
-        bill.save()
-        peter = Employee(name='Peter Gibbons', boss=bill)
-        peter.save()
-
-        peter = Employee.objects.with_id(peter.id)
-        self.assertEqual(peter.boss, bill)
-
-    def test_undefined_reference(self):
-        """Ensure that ReferenceFields may reference undefined Documents.
-        """
-        class Product(Document):
-            name = StringField()
-            company = ReferenceField('Company')
-
-        class Company(Document):
-            name = StringField()
-
-        ten_gen = Company(name='10gen')
-        ten_gen.save()
-        mongodb = Product(name='MongoDB', company=ten_gen)
-        mongodb.save()
-
-        obj = Product.objects(company=ten_gen).first()
-        self.assertEqual(obj, mongodb)
-        self.assertEqual(obj.company, ten_gen)
-
-    def test_reference_query_conversion(self):
-        """Ensure that ReferenceFields 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 = ReferenceField(Member)
-
-        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()
-        self.assertEqual(post.id, post1.id)
-
-        post = BlogPost.objects(author=m2).first()
-        self.assertEqual(post.id, post2.id)
-
-        Member.drop_collection()
-        BlogPost.drop_collection()
-        
-    def test_generic_reference(self):
-        """Ensure that a GenericReferenceField properly dereferences items.
-        """
-        class Link(Document):
-            title = StringField()
-            meta = {'allow_inheritance': False}
-            
-        class Post(Document):
-            title = StringField()
-            
-        class Bookmark(Document):
-            bookmark_object = GenericReferenceField()
-            
-        Link.drop_collection()
-        Post.drop_collection()
-        Bookmark.drop_collection()
-    
-        link_1 = Link(title="Pitchfork")
-        link_1.save()
-    
-        post_1 = Post(title="Behind the Scenes of the Pavement Reunion")
-        post_1.save()
-        
-        bm = Bookmark(bookmark_object=post_1)
-        bm.save()
-        
-        bm = Bookmark.objects(bookmark_object=post_1).first()
-        
-        self.assertEqual(bm.bookmark_object, post_1)
-        self.assertTrue(isinstance(bm.bookmark_object, Post))
-        
-        bm.bookmark_object = link_1
-        bm.save()
-        
-        bm = Bookmark.objects(bookmark_object=link_1).first()
-        
-        self.assertEqual(bm.bookmark_object, link_1)
-        self.assertTrue(isinstance(bm.bookmark_object, Link))
-    
-        Link.drop_collection()
-        Post.drop_collection()
-        Bookmark.drop_collection()
-
-    def test_generic_reference_list(self):
-        """Ensure that a ListField properly dereferences generic references.
-        """
-        class Link(Document):
-            title = StringField()
-    
-        class Post(Document):
-            title = StringField()
-    
-        class User(Document):
-            bookmarks = ListField(GenericReferenceField())
-    
-        Link.drop_collection()
-        Post.drop_collection()
-        User.drop_collection()
-    
-        link_1 = Link(title="Pitchfork")
-        link_1.save()
-    
-        post_1 = Post(title="Behind the Scenes of the Pavement Reunion")
-        post_1.save()
-    
-        user = User(bookmarks=[post_1, link_1])
-        user.save()
-        
-        user = User.objects(bookmarks__all=[post_1, link_1]).first()
-        
-        self.assertEqual(user.bookmarks[0], post_1)
-        self.assertEqual(user.bookmarks[1], link_1)
-        
-        Link.drop_collection()
-        Post.drop_collection()
-        User.drop_collection()
-
-    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'
-        MIME_TYPE = 'application/octet-stream'
-
-        Attachment.drop_collection()
-
-        attachment = Attachment(content_type=MIME_TYPE, blob=BLOB)
-        attachment.save()
-
-        attachment_1 = Attachment.objects().first()
-        self.assertEqual(MIME_TYPE, attachment_1.content_type)
-        self.assertEqual(BLOB, attachment_1.blob)
-
-        Attachment.drop_collection()
-
-    def test_binary_validation(self):
-        """Ensure that invalid values cannot be assigned to binary fields.
-        """
-        class Attachment(Document):
-            blob = BinaryField()
-
-        class AttachmentRequired(Document):
-            blob = BinaryField(required=True)
-
-        class AttachmentSizeLimit(Document):
-            blob = BinaryField(max_bytes=4)
-
-        Attachment.drop_collection()
-        AttachmentRequired.drop_collection()
-        AttachmentSizeLimit.drop_collection()
-
-        attachment = Attachment()
-        attachment.validate()
-        attachment.blob = 2
-        self.assertRaises(ValidationError, attachment.validate)
-
-        attachment_required = AttachmentRequired()
-        self.assertRaises(ValidationError, attachment_required.validate)
-        attachment_required.blob = '\xe6\x00\xc4\xff\x07'
-        attachment_required.validate()
-
-        attachment_size_limit = AttachmentSizeLimit(blob='\xe6\x00\xc4\xff\x07')
-        self.assertRaises(ValidationError, attachment_size_limit.validate)
-        attachment_size_limit.blob = '\xe6\x00\xc4\xff'
-        attachment_size_limit.validate()
-
-        Attachment.drop_collection()
-        AttachmentRequired.drop_collection()
-        AttachmentSizeLimit.drop_collection()
-
-if __name__ == '__main__':
-    unittest.main()

tests/fields/__init__.py

@@ -0,0 +1,2 @@
+from fields import *
+from file_tests import *

tests/fields/file_tests.py

@@ -0,0 +1,383 @@
+# -*- coding: utf-8 -*-
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+
+import copy
+import os
+import unittest
+import tempfile
+
+import gridfs
+
+from nose.plugins.skip import SkipTest
+from mongoengine import *
+from mongoengine.connection import get_db
+from mongoengine.python_support import PY3, b, StringIO
+
+TEST_IMAGE_PATH = os.path.join(os.path.dirname(__file__), 'mongoengine.png')
+
+
+class FileTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+        self.db = get_db()
+
+    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 = b('Hello, World!')
+        content_type = 'text/plain'
+
+        putfile = PutFile()
+        putfile.the_file.put(text, content_type=content_type)
+        putfile.save()
+
+        result = PutFile.objects.first()
+        self.assertTrue(putfile == result)
+        self.assertEqual(result.the_file.read(), text)
+        self.assertEqual(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 = StringIO()
+        putstring.write(text)
+        putstring.seek(0)
+        putfile.the_file.put(putstring, content_type=content_type)
+        putfile.save()
+
+        result = PutFile.objects.first()
+        self.assertTrue(putfile == result)
+        self.assertEqual(result.the_file.read(), text)
+        self.assertEqual(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 = b('Hello, World!')
+        more_text = b('Foo Bar')
+        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()
+        self.assertTrue(streamfile == result)
+        self.assertEqual(result.the_file.read(), text + more_text)
+        self.assertEqual(result.the_file.content_type, content_type)
+        result.the_file.seek(0)
+        self.assertEqual(result.the_file.tell(), 0)
+        self.assertEqual(result.the_file.read(len(text)), text)
+        self.assertEqual(result.the_file.tell(), len(text))
+        self.assertEqual(result.the_file.read(len(more_text)), more_text)
+        self.assertEqual(result.the_file.tell(), len(text + more_text))
+        result.the_file.delete()
+
+        # Ensure deleted file returns None
+        self.assertTrue(result.the_file.read() == None)
+
+    def test_file_fields_set(self):
+
+        class SetFile(Document):
+            the_file = FileField()
+
+        text = b('Hello, World!')
+        more_text = b('Foo Bar')
+
+        SetFile.drop_collection()
+
+        setfile = SetFile()
+        setfile.the_file = text
+        setfile.save()
+
+        result = SetFile.objects.first()
+        self.assertTrue(setfile == result)
+        self.assertEqual(result.the_file.read(), text)
+
+        # Try replacing file with new one
+        result.the_file.replace(more_text)
+        result.save()
+
+        result = SetFile.objects.first()
+        self.assertTrue(setfile == result)
+        self.assertEqual(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(b("Hello World!"))
+            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()
+            self.assertNotEqual(doc_b.the_file.grid_id, None)
+
+            # Test it matches
+            doc_c = GridDocument.objects.with_id(doc_b.id)
+            self.assertEqual(doc_b.the_file.grid_id, doc_c.the_file.grid_id)
+
+            # Test with default
+            doc_d = GridDocument(the_file=b(''))
+            doc_d.save()
+
+            doc_e = GridDocument.objects.with_id(doc_d.id)
+            self.assertEqual(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)
+            self.assertEqual(doc_e.the_file.grid_id, doc_f.the_file.grid_id)
+
+        db = GridDocument._get_db()
+        grid_fs = gridfs.GridFS(db)
+        self.assertEqual(['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(b('Hello, World!'))
+        test_file.save()
+
+        # Second instance
+        test_file_dupe = TestFile()
+        data = test_file_dupe.the_file.read()  # Should be None
+
+        self.assertTrue(test_file.name != test_file_dupe.name)
+        self.assertTrue(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 = open(TEST_IMAGE_PATH, 'rb')  # Retrieve a photo from disk
+        marmot.photo.put(marmot_photo, content_type='image/jpeg', foo='bar')
+        marmot.photo.close()
+        marmot.save()
+
+        marmot = Animal.objects.get()
+        self.assertEqual(marmot.photo.content_type, 'image/jpeg')
+        self.assertEqual(marmot.photo.foo, 'bar')
+
+    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()
+        self.assertFalse(bool(test_file.the_file))
+        test_file.the_file.put(b('Hello, World!'), content_type='text/plain')
+        test_file.save()
+        self.assertTrue(bool(test_file.the_file))
+
+        test_file = TestFile.objects.first()
+        self.assertEqual(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()
+        self.assertFalse(test_file.the_file in [{"test": 1}])
+
+    def test_image_field(self):
+        if PY3:
+            raise SkipTest('PIL does not have Python 3 support')
+
+        class TestImage(Document):
+            image = ImageField()
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(open(TEST_IMAGE_PATH, 'rb'))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        self.assertEqual(t.image.format, 'PNG')
+
+        w, h = t.image.size
+        self.assertEqual(w, 371)
+        self.assertEqual(h, 76)
+
+        t.image.delete()
+
+    def test_image_field_resize(self):
+        if PY3:
+            raise SkipTest('PIL does not have Python 3 support')
+
+        class TestImage(Document):
+            image = ImageField(size=(185, 37))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(open(TEST_IMAGE_PATH, 'rb'))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        self.assertEqual(t.image.format, 'PNG')
+        w, h = t.image.size
+
+        self.assertEqual(w, 185)
+        self.assertEqual(h, 37)
+
+        t.image.delete()
+
+    def test_image_field_resize_force(self):
+        if PY3:
+            raise SkipTest('PIL does not have Python 3 support')
+
+        class TestImage(Document):
+            image = ImageField(size=(185, 37, True))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(open(TEST_IMAGE_PATH, 'rb'))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        self.assertEqual(t.image.format, 'PNG')
+        w, h = t.image.size
+
+        self.assertEqual(w, 185)
+        self.assertEqual(h, 37)
+
+        t.image.delete()
+
+    def test_image_field_thumbnail(self):
+        if PY3:
+            raise SkipTest('PIL does not have Python 3 support')
+
+        class TestImage(Document):
+            image = ImageField(thumbnail_size=(92, 18))
+
+        TestImage.drop_collection()
+
+        t = TestImage()
+        t.image.put(open(TEST_IMAGE_PATH, 'rb'))
+        t.save()
+
+        t = TestImage.objects.first()
+
+        self.assertEqual(t.image.thumbnail.format, 'PNG')
+        self.assertEqual(t.image.thumbnail.width, 92)
+        self.assertEqual(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(b('Hello, World!'),
+                          name="hello.txt")
+        test_file.save()
+
+        data = get_db("test_files").macumba.files.find_one()
+        self.assertEqual(data.get('name'), 'hello.txt')
+
+        test_file = TestFile.objects.first()
+        self.assertEqual(test_file.the_file.read(),
+                          b('Hello, World!'))
+
+    def test_copyable(self):
+        class PutFile(Document):
+            the_file = FileField()
+
+        PutFile.drop_collection()
+
+        text = b('Hello, World!')
+        content_type = 'text/plain'
+
+        putfile = PutFile()
+        putfile.the_file.put(text, content_type=content_type)
+        putfile.save()
+
+        class TestFile(Document):
+            name = StringField()
+
+        self.assertEqual(putfile, copy.copy(putfile))
+        self.assertEqual(putfile, copy.deepcopy(putfile))
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/fixtures.py

@@ -0,0 +1,23 @@
+from datetime import datetime
+
+from mongoengine import *
+
+
+class PickleEmbedded(EmbeddedDocument):
+    date = DateTimeField(default=datetime.now)
+
+
+class PickleTest(Document):
+    number = IntField()
+    string = StringField(choices=(('One', '1'), ('Two', '2')))
+    embedded = EmbeddedDocumentField(PickleEmbedded)
+    lists = ListField(StringField())
+    photo = FileField()
+
+
+class Mixin(object):
+    name = StringField()
+
+
+class Base(Document):
+    meta = {'allow_inheritance': True}

tests/migration/__init__.py

@@ -0,0 +1,4 @@
+from turn_off_inheritance import *
+
+if __name__ == '__main__':
+    unittest.main()

tests/migration/test_convert_to_new_inheritance_model.py

@@ -0,0 +1,51 @@
+# -*- coding: utf-8 -*-
+import unittest
+
+from mongoengine import Document, connect
+from mongoengine.connection import get_db
+from mongoengine.fields import StringField
+
+__all__ = ('ConvertToNewInheritanceModel', )
+
+
+class ConvertToNewInheritanceModel(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+        self.db = get_db()
+
+    def tearDown(self):
+        for collection in self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            self.db.drop_collection(collection)
+
+    def test_how_to_convert_to_the_new_inheritance_model(self):
+        """Demonstrates migrating from 0.7 to 0.8
+        """
+
+        # 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.iteritems()
+                           if '_types' in dict(value['key'])]
+        for index in indexes_to_drop:
+            collection.drop_index(index)
+
+        # 5. Recreate indexes
+        Animal.ensure_indexes()

tests/migration/turn_off_inheritance.py

@@ -0,0 +1,62 @@
+# -*- coding: utf-8 -*-
+import unittest
+
+from mongoengine import Document, connect
+from mongoengine.connection import get_db
+from mongoengine.fields import StringField
+
+__all__ = ('TurnOffInheritanceTest', )
+
+
+class TurnOffInheritanceTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+        self.db = get_db()
+
+    def tearDown(self):
+        for collection in self.db.collection_names():
+            if 'system.' in collection:
+                continue
+            self.db.drop_collection(collection)
+
+    def test_how_to_turn_off_inheritance(self):
+        """Demonstrates migrating from allow_inheritance = True to False.
+        """
+
+        # 1. Old declaration of the class
+
+        class Animal(Document):
+            name = StringField()
+            meta = {
+                'allow_inheritance': True,
+                'indexes': ['name']
+            }
+
+        # 2. Turn off inheritance
+        class Animal(Document):
+            name = StringField()
+            meta = {
+                'allow_inheritance': False,
+                'indexes': ['name']
+            }
+
+        # 3. Remove _types and _cls
+        collection = Animal._get_collection()
+        collection.update({}, {"$unset": {"_types": 1, "_cls": 1}}, multi=True)
+
+        # 3. Confirm extra data is removed
+        count = collection.find({"$or": [{'_types': {"$exists": True}},
+                                         {'_cls': {"$exists": True}}]}).count()
+        assert count == 0
+
+        # 4. Remove indexes
+        info = collection.index_information()
+        indexes_to_drop = [key for key, value in info.iteritems()
+                           if '_types' in dict(value['key'])
+                              or '_cls' in dict(value['key'])]
+        for index in indexes_to_drop:
+            collection.drop_index(index)
+
+        # 5. Recreate indexes
+        Animal.ensure_indexes()

tests/queryset/__init__.py

@@ -0,0 +1,5 @@
+
+from transform import *
+from field_list import *
+from queryset import *
+from visitor import *

tests/queryset/field_list.py

@@ -0,0 +1,399 @@
+import sys
+sys.path[0:0] = [""]
+
+import unittest
+
+from mongoengine import *
+from mongoengine.queryset import QueryFieldList
+
+__all__ = ("QueryFieldListTest", "OnlyExcludeAllTest")
+
+
+class QueryFieldListTest(unittest.TestCase):
+
+    def test_empty(self):
+        q = QueryFieldList()
+        self.assertFalse(q)
+
+        q = QueryFieldList(always_include=['_cls'])
+        self.assertFalse(q)
+
+    def test_include_include(self):
+        q = QueryFieldList()
+        q += QueryFieldList(fields=['a', 'b'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'a': True, 'b': True})
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'b': True})
+
+    def test_include_exclude(self):
+        q = QueryFieldList()
+        q += QueryFieldList(fields=['a', 'b'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'a': True, 'b': True})
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.EXCLUDE)
+        self.assertEqual(q.as_dict(), {'a': True})
+
+    def test_exclude_exclude(self):
+        q = QueryFieldList()
+        q += QueryFieldList(fields=['a', 'b'], value=QueryFieldList.EXCLUDE)
+        self.assertEqual(q.as_dict(), {'a': False, 'b': False})
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.EXCLUDE)
+        self.assertEqual(q.as_dict(), {'a': False, 'b': False, 'c': False})
+
+    def test_exclude_include(self):
+        q = QueryFieldList()
+        q += QueryFieldList(fields=['a', 'b'], value=QueryFieldList.EXCLUDE)
+        self.assertEqual(q.as_dict(), {'a': False, 'b': False})
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'c': True})
+
+    def test_always_include(self):
+        q = QueryFieldList(always_include=['x', 'y'])
+        q += QueryFieldList(fields=['a', 'b', 'x'], value=QueryFieldList.EXCLUDE)
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'x': True, 'y': True, 'c': True})
+
+    def test_reset(self):
+        q = QueryFieldList(always_include=['x', 'y'])
+        q += QueryFieldList(fields=['a', 'b', 'x'], value=QueryFieldList.EXCLUDE)
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'x': True, 'y': True, 'c': True})
+        q.reset()
+        self.assertFalse(q)
+        q += QueryFieldList(fields=['b', 'c'], value=QueryFieldList.ONLY)
+        self.assertEqual(q.as_dict(), {'x': True, 'y': True, 'b': True, 'c': True})
+
+    def test_using_a_slice(self):
+        q = QueryFieldList()
+        q += QueryFieldList(fields=['a'], value={"$slice": 5})
+        self.assertEqual(q.as_dict(), {'a': {"$slice": 5}})
+
+
+class OnlyExcludeAllTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+            meta = {'allow_inheritance': True}
+
+        Person.drop_collection()
+        self.Person = Person
+
+    def test_mixing_only_exclude(self):
+
+        class MyDoc(Document):
+            a = StringField()
+            b = StringField()
+            c = StringField()
+            d = StringField()
+            e = StringField()
+            f = StringField()
+
+        include = ['a', 'b', 'c', 'd', 'e']
+        exclude = ['d', 'e']
+        only = ['b', 'c']
+
+        qs = MyDoc.objects.fields(**dict(((i, 1) for i in include)))
+        self.assertEqual(qs._loaded_fields.as_dict(),
+            {'a': 1, 'b': 1, 'c': 1, 'd': 1, 'e': 1})
+        qs = qs.only(*only)
+        self.assertEqual(qs._loaded_fields.as_dict(), {'b': 1, 'c': 1})
+        qs = qs.exclude(*exclude)
+        self.assertEqual(qs._loaded_fields.as_dict(), {'b': 1, 'c': 1})
+
+        qs = MyDoc.objects.fields(**dict(((i, 1) for i in include)))
+        qs = qs.exclude(*exclude)
+        self.assertEqual(qs._loaded_fields.as_dict(), {'a': 1, 'b': 1, 'c': 1})
+        qs = qs.only(*only)
+        self.assertEqual(qs._loaded_fields.as_dict(), {'b': 1, 'c': 1})
+
+        qs = MyDoc.objects.exclude(*exclude)
+        qs = qs.fields(**dict(((i, 1) for i in include)))
+        self.assertEqual(qs._loaded_fields.as_dict(), {'a': 1, 'b': 1, 'c': 1})
+        qs = qs.only(*only)
+        self.assertEqual(qs._loaded_fields.as_dict(), {'b': 1, 'c': 1})
+
+    def test_slicing(self):
+
+        class MyDoc(Document):
+            a = ListField()
+            b = ListField()
+            c = ListField()
+            d = ListField()
+            e = ListField()
+            f = ListField()
+
+        include = ['a', 'b', 'c', 'd', 'e']
+        exclude = ['d', 'e']
+        only = ['b', 'c']
+
+        qs = MyDoc.objects.fields(**dict(((i, 1) for i in include)))
+        qs = qs.exclude(*exclude)
+        qs = qs.only(*only)
+        qs = qs.fields(slice__b=5)
+        self.assertEqual(qs._loaded_fields.as_dict(),
+            {'b': {'$slice': 5}, 'c': 1})
+
+        qs = qs.fields(slice__c=[5, 1])
+        self.assertEqual(qs._loaded_fields.as_dict(),
+            {'b': {'$slice': 5}, 'c': {'$slice': [5, 1]}})
+
+        qs = qs.exclude('c')
+        self.assertEqual(qs._loaded_fields.as_dict(),
+            {'b': {'$slice': 5}})
+
+    def test_only(self):
+        """Ensure that QuerySet.only only returns the requested fields.
+        """
+        person = self.Person(name='test', age=25)
+        person.save()
+
+        obj = self.Person.objects.only('name').get()
+        self.assertEqual(obj.name, person.name)
+        self.assertEqual(obj.age, None)
+
+        obj = self.Person.objects.only('age').get()
+        self.assertEqual(obj.name, None)
+        self.assertEqual(obj.age, person.age)
+
+        obj = self.Person.objects.only('name', 'age').get()
+        self.assertEqual(obj.name, person.name)
+        self.assertEqual(obj.age, person.age)
+
+        # Check polymorphism still works
+        class Employee(self.Person):
+            salary = IntField(db_field='wage')
+
+        employee = Employee(name='test employee', age=40, salary=30000)
+        employee.save()
+
+        obj = self.Person.objects(id=employee.id).only('age').get()
+        self.assertTrue(isinstance(obj, Employee))
+
+        # Check field names are looked up properly
+        obj = Employee.objects(id=employee.id).only('salary').get()
+        self.assertEqual(obj.salary, employee.salary)
+        self.assertEqual(obj.name, None)
+
+    def test_only_with_subfields(self):
+        class User(EmbeddedDocument):
+            name = StringField()
+            email = StringField()
+
+        class Comment(EmbeddedDocument):
+            title = StringField()
+            text = StringField()
+
+        class BlogPost(Document):
+            content = StringField()
+            author = EmbeddedDocumentField(User)
+            comments = ListField(EmbeddedDocumentField(Comment))
+
+        BlogPost.drop_collection()
+
+        post = BlogPost(content='Had a good coffee today...')
+        post.author = User(name='Test User')
+        post.comments = [Comment(title='I aggree', text='Great post!'), Comment(title='Coffee', text='I hate coffee')]
+        post.save()
+
+        obj = BlogPost.objects.only('author.name',).get()
+        self.assertEqual(obj.content, None)
+        self.assertEqual(obj.author.email, None)
+        self.assertEqual(obj.author.name, 'Test User')
+        self.assertEqual(obj.comments, [])
+
+        obj = BlogPost.objects.only('content', 'comments.title',).get()
+        self.assertEqual(obj.content, 'Had a good coffee today...')
+        self.assertEqual(obj.author, None)
+        self.assertEqual(obj.comments[0].title, 'I aggree')
+        self.assertEqual(obj.comments[1].title, 'Coffee')
+        self.assertEqual(obj.comments[0].text, None)
+        self.assertEqual(obj.comments[1].text, None)
+
+        obj = BlogPost.objects.only('comments',).get()
+        self.assertEqual(obj.content, None)
+        self.assertEqual(obj.author, None)
+        self.assertEqual(obj.comments[0].title, 'I aggree')
+        self.assertEqual(obj.comments[1].title, 'Coffee')
+        self.assertEqual(obj.comments[0].text, 'Great post!')
+        self.assertEqual(obj.comments[1].text, 'I hate coffee')
+
+        BlogPost.drop_collection()
+
+    def test_exclude(self):
+        class User(EmbeddedDocument):
+            name = StringField()
+            email = StringField()
+
+        class Comment(EmbeddedDocument):
+            title = StringField()
+            text = StringField()
+
+        class BlogPost(Document):
+            content = StringField()
+            author = EmbeddedDocumentField(User)
+            comments = ListField(EmbeddedDocumentField(Comment))
+
+        BlogPost.drop_collection()
+
+        post = BlogPost(content='Had a good coffee today...')
+        post.author = User(name='Test User')
+        post.comments = [Comment(title='I aggree', text='Great post!'), Comment(title='Coffee', text='I hate coffee')]
+        post.save()
+
+        obj = BlogPost.objects.exclude('author', 'comments.text').get()
+        self.assertEqual(obj.author, None)
+        self.assertEqual(obj.content, 'Had a good coffee today...')
+        self.assertEqual(obj.comments[0].title, 'I aggree')
+        self.assertEqual(obj.comments[0].text, None)
+
+        BlogPost.drop_collection()
+
+    def test_exclude_only_combining(self):
+        class Attachment(EmbeddedDocument):
+            name = StringField()
+            content = StringField()
+
+        class Email(Document):
+            sender = StringField()
+            to = StringField()
+            subject = StringField()
+            body = StringField()
+            content_type = StringField()
+            attachments = ListField(EmbeddedDocumentField(Attachment))
+
+        Email.drop_collection()
+        email = Email(sender='me', to='you', subject='From Russia with Love', body='Hello!', content_type='text/plain')
+        email.attachments = [
+            Attachment(name='file1.doc', content='ABC'),
+            Attachment(name='file2.doc', content='XYZ'),
+        ]
+        email.save()
+
+        obj = Email.objects.exclude('content_type').exclude('body').get()
+        self.assertEqual(obj.sender, 'me')
+        self.assertEqual(obj.to, 'you')
+        self.assertEqual(obj.subject, 'From Russia with Love')
+        self.assertEqual(obj.body, None)
+        self.assertEqual(obj.content_type, None)
+
+        obj = Email.objects.only('sender', 'to').exclude('body', 'sender').get()
+        self.assertEqual(obj.sender, None)
+        self.assertEqual(obj.to, 'you')
+        self.assertEqual(obj.subject, None)
+        self.assertEqual(obj.body, None)
+        self.assertEqual(obj.content_type, None)
+
+        obj = Email.objects.exclude('attachments.content').exclude('body').only('to', 'attachments.name').get()
+        self.assertEqual(obj.attachments[0].name, 'file1.doc')
+        self.assertEqual(obj.attachments[0].content, None)
+        self.assertEqual(obj.sender, None)
+        self.assertEqual(obj.to, 'you')
+        self.assertEqual(obj.subject, None)
+        self.assertEqual(obj.body, None)
+        self.assertEqual(obj.content_type, None)
+
+        Email.drop_collection()
+
+    def test_all_fields(self):
+
+        class Email(Document):
+            sender = StringField()
+            to = StringField()
+            subject = StringField()
+            body = StringField()
+            content_type = StringField()
+
+        Email.drop_collection()
+
+        email = Email(sender='me', to='you', subject='From Russia with Love', body='Hello!', content_type='text/plain')
+        email.save()
+
+        obj = Email.objects.exclude('content_type', 'body').only('to', 'body').all_fields().get()
+        self.assertEqual(obj.sender, 'me')
+        self.assertEqual(obj.to, 'you')
+        self.assertEqual(obj.subject, 'From Russia with Love')
+        self.assertEqual(obj.body, 'Hello!')
+        self.assertEqual(obj.content_type, 'text/plain')
+
+        Email.drop_collection()
+
+    def test_slicing_fields(self):
+        """Ensure that query slicing an array works.
+        """
+        class Numbers(Document):
+            n = ListField(IntField())
+
+        Numbers.drop_collection()
+
+        numbers = Numbers(n=[0,1,2,3,4,5,-5,-4,-3,-2,-1])
+        numbers.save()
+
+        # first three
+        numbers = Numbers.objects.fields(slice__n=3).get()
+        self.assertEqual(numbers.n, [0, 1, 2])
+
+        # last three
+        numbers = Numbers.objects.fields(slice__n=-3).get()
+        self.assertEqual(numbers.n, [-3, -2, -1])
+
+        # skip 2, limit 3
+        numbers = Numbers.objects.fields(slice__n=[2, 3]).get()
+        self.assertEqual(numbers.n, [2, 3, 4])
+
+        # skip to fifth from last, limit 4
+        numbers = Numbers.objects.fields(slice__n=[-5, 4]).get()
+        self.assertEqual(numbers.n, [-5, -4, -3, -2])
+
+        # skip to fifth from last, limit 10
+        numbers = Numbers.objects.fields(slice__n=[-5, 10]).get()
+        self.assertEqual(numbers.n, [-5, -4, -3, -2, -1])
+
+        # skip to fifth from last, limit 10 dict method
+        numbers = Numbers.objects.fields(n={"$slice": [-5, 10]}).get()
+        self.assertEqual(numbers.n, [-5, -4, -3, -2, -1])
+
+    def test_slicing_nested_fields(self):
+        """Ensure that query slicing an embedded array works.
+        """
+
+        class EmbeddedNumber(EmbeddedDocument):
+            n = ListField(IntField())
+
+        class Numbers(Document):
+            embedded = EmbeddedDocumentField(EmbeddedNumber)
+
+        Numbers.drop_collection()
+
+        numbers = Numbers()
+        numbers.embedded = EmbeddedNumber(n=[0,1,2,3,4,5,-5,-4,-3,-2,-1])
+        numbers.save()
+
+        # first three
+        numbers = Numbers.objects.fields(slice__embedded__n=3).get()
+        self.assertEqual(numbers.embedded.n, [0, 1, 2])
+
+        # last three
+        numbers = Numbers.objects.fields(slice__embedded__n=-3).get()
+        self.assertEqual(numbers.embedded.n, [-3, -2, -1])
+
+        # skip 2, limit 3
+        numbers = Numbers.objects.fields(slice__embedded__n=[2, 3]).get()
+        self.assertEqual(numbers.embedded.n, [2, 3, 4])
+
+        # skip to fifth from last, limit 4
+        numbers = Numbers.objects.fields(slice__embedded__n=[-5, 4]).get()
+        self.assertEqual(numbers.embedded.n, [-5, -4, -3, -2])
+
+        # skip to fifth from last, limit 10
+        numbers = Numbers.objects.fields(slice__embedded__n=[-5, 10]).get()
+        self.assertEqual(numbers.embedded.n, [-5, -4, -3, -2, -1])
+
+        # skip to fifth from last, limit 10 dict method
+        numbers = Numbers.objects.fields(embedded__n={"$slice": [-5, 10]}).get()
+        self.assertEqual(numbers.embedded.n, [-5, -4, -3, -2, -1])
+
+if __name__ == '__main__':
+    unittest.main()

tests/queryset/transform.py

@@ -0,0 +1,148 @@
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+
+import unittest
+
+from mongoengine import *
+from mongoengine.queryset import Q
+from mongoengine.queryset import transform
+
+__all__ = ("TransformTest",)
+
+
+class TransformTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+    def test_transform_query(self):
+        """Ensure that the _transform_query function operates correctly.
+        """
+        self.assertEqual(transform.query(name='test', age=30),
+                         {'name': 'test', 'age': 30})
+        self.assertEqual(transform.query(age__lt=30),
+                         {'age': {'$lt': 30}})
+        self.assertEqual(transform.query(age__gt=20, age__lt=50),
+                         {'age': {'$gt': 20, '$lt': 50}})
+        self.assertEqual(transform.query(age=20, age__gt=50),
+                         {'$and': [{'age': {'$gt': 50}}, {'age': 20}]})
+        self.assertEqual(transform.query(friend__age__gte=30),
+                         {'friend.age': {'$gte': 30}})
+        self.assertEqual(transform.query(name__exists=True),
+                         {'name': {'$exists': True}})
+
+    def test_query_field_name(self):
+        """Ensure that the correct field name is used when querying.
+        """
+        class Comment(EmbeddedDocument):
+            content = StringField(db_field='commentContent')
+
+        class BlogPost(Document):
+            title = StringField(db_field='postTitle')
+            comments = ListField(EmbeddedDocumentField(Comment),
+                                 db_field='postComments')
+
+        BlogPost.drop_collection()
+
+        data = {'title': 'Post 1', 'comments': [Comment(content='test')]}
+        post = BlogPost(**data)
+        post.save()
+
+        self.assertTrue('postTitle' in
+                        BlogPost.objects(title=data['title'])._query)
+        self.assertFalse('title' in
+                         BlogPost.objects(title=data['title'])._query)
+        self.assertEqual(len(BlogPost.objects(title=data['title'])), 1)
+
+        self.assertTrue('_id' in BlogPost.objects(pk=post.id)._query)
+        self.assertEqual(len(BlogPost.objects(pk=post.id)), 1)
+
+        self.assertTrue('postComments.commentContent' in
+                        BlogPost.objects(comments__content='test')._query)
+        self.assertEqual(len(BlogPost.objects(comments__content='test')), 1)
+
+        BlogPost.drop_collection()
+
+    def test_query_pk_field_name(self):
+        """Ensure that the correct "primary key" field name is used when
+        querying
+        """
+        class BlogPost(Document):
+            title = StringField(primary_key=True, db_field='postTitle')
+
+        BlogPost.drop_collection()
+
+        data = {'title': 'Post 1'}
+        post = BlogPost(**data)
+        post.save()
+
+        self.assertTrue('_id' in BlogPost.objects(pk=data['title'])._query)
+        self.assertTrue('_id' in BlogPost.objects(title=data['title'])._query)
+        self.assertEqual(len(BlogPost.objects(pk=data['title'])), 1)
+
+        BlogPost.drop_collection()
+
+    def test_chaining(self):
+        class A(Document):
+            pass
+
+        class B(Document):
+            a = ReferenceField(A)
+
+        A.drop_collection()
+        B.drop_collection()
+
+        a1 = A().save()
+        a2 = A().save()
+
+        B(a=a1).save()
+
+        # Works
+        q1 = B.objects.filter(a__in=[a1, a2], a=a1)._query
+
+        # Doesn't work
+        q2 = B.objects.filter(a__in=[a1, a2])
+        q2 = q2.filter(a=a1)._query
+
+        self.assertEqual(q1, q2)
+
+    def test_raw_query_and_Q_objects(self):
+        """
+        Test raw plays nicely
+        """
+        class Foo(Document):
+            name = StringField()
+            a = StringField()
+            b = StringField()
+            c = StringField()
+
+            meta = {
+                'allow_inheritance': False
+            }
+
+        query = Foo.objects(__raw__={'$nor': [{'name': 'bar'}]})._query
+        self.assertEqual(query, {'$nor': [{'name': 'bar'}]})
+
+        q1 = {'$or': [{'a': 1}, {'b': 1}]}
+        query = Foo.objects(Q(__raw__=q1) & Q(c=1))._query
+        self.assertEqual(query, {'$or': [{'a': 1}, {'b': 1}], 'c': 1})
+
+    def test_raw_and_merging(self):
+        class Doc(Document):
+            meta = {'allow_inheritance': False}
+
+        raw_query = Doc.objects(__raw__={'deleted': False,
+                                'scraped': 'yes',
+                                '$nor': [{'views.extracted': 'no'},
+                                         {'attachments.views.extracted':'no'}]
+                                })._query
+
+        expected = {'deleted': False, 'scraped': 'yes',
+                    '$nor': [{'views.extracted': 'no'},
+                             {'attachments.views.extracted': 'no'}]}
+        self.assertEqual(expected, raw_query)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/queryset/visitor.py

@@ -0,0 +1,335 @@
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+
+import unittest
+
+from bson import ObjectId
+from datetime import datetime
+
+from mongoengine import *
+from mongoengine.queryset import Q
+from mongoengine.errors import InvalidQueryError
+
+__all__ = ("QTest",)
+
+
+class QTest(unittest.TestCase):
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+            meta = {'allow_inheritance': True}
+
+        Person.drop_collection()
+        self.Person = Person
+
+    def test_empty_q(self):
+        """Ensure that empty Q objects won't hurt.
+        """
+        q1 = Q()
+        q2 = Q(age__gte=18)
+        q3 = Q()
+        q4 = Q(name='test')
+        q5 = Q()
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+
+        query = {'$or': [{'age': {'$gte': 18}}, {'name': 'test'}]}
+        self.assertEqual((q1 | q2 | q3 | q4 | q5).to_query(Person), query)
+
+        query = {'age': {'$gte': 18}, 'name': 'test'}
+        self.assertEqual((q1 & q2 & q3 & q4 & q5).to_query(Person), query)
+
+    def test_q_with_dbref(self):
+        """Ensure Q objects handle DBRefs correctly"""
+        connect(db='mongoenginetest')
+
+        class User(Document):
+            pass
+
+        class Post(Document):
+            created_user = ReferenceField(User)
+
+        user = User.objects.create()
+        Post.objects.create(created_user=user)
+
+        self.assertEqual(Post.objects.filter(created_user=user).count(), 1)
+        self.assertEqual(Post.objects.filter(Q(created_user=user)).count(), 1)
+
+    def test_and_combination(self):
+        """Ensure that Q-objects correctly AND together.
+        """
+        class TestDoc(Document):
+            x = IntField()
+            y = StringField()
+
+        # Check than an error is raised when conflicting queries are anded
+        def invalid_combination():
+            query = Q(x__lt=7) & Q(x__lt=3)
+            query.to_query(TestDoc)
+        self.assertRaises(InvalidQueryError, invalid_combination)
+
+        # Check normal cases work without an error
+        query = Q(x__lt=7) & Q(x__gt=3)
+
+        q1 = Q(x__lt=7)
+        q2 = Q(x__gt=3)
+        query = (q1 & q2).to_query(TestDoc)
+        self.assertEqual(query, {'x': {'$lt': 7, '$gt': 3}})
+
+        # More complex nested example
+        query = Q(x__lt=100) & Q(y__ne='NotMyString')
+        query &= Q(y__in=['a', 'b', 'c']) & Q(x__gt=-100)
+        mongo_query = {
+            'x': {'$lt': 100, '$gt': -100},
+            'y': {'$ne': 'NotMyString', '$in': ['a', 'b', 'c']},
+        }
+        self.assertEqual(query.to_query(TestDoc), mongo_query)
+
+    def test_or_combination(self):
+        """Ensure that Q-objects correctly OR together.
+        """
+        class TestDoc(Document):
+            x = IntField()
+
+        q1 = Q(x__lt=3)
+        q2 = Q(x__gt=7)
+        query = (q1 | q2).to_query(TestDoc)
+        self.assertEqual(query, {
+            '$or': [
+                {'x': {'$lt': 3}},
+                {'x': {'$gt': 7}},
+            ]
+        })
+
+    def test_and_or_combination(self):
+        """Ensure that Q-objects handle ANDing ORed components.
+        """
+        class TestDoc(Document):
+            x = IntField()
+            y = BooleanField()
+
+        TestDoc.drop_collection()
+
+        query = (Q(x__gt=0) | Q(x__exists=False))
+        query &= Q(x__lt=100)
+        self.assertEqual(query.to_query(TestDoc), {'$and': [
+            {'$or': [{'x': {'$gt': 0}},
+                     {'x': {'$exists': False}}]},
+            {'x': {'$lt': 100}}]
+        })
+
+        q1 = (Q(x__gt=0) | Q(x__exists=False))
+        q2 = (Q(x__lt=100) | Q(y=True))
+        query = (q1 & q2).to_query(TestDoc)
+
+        TestDoc(x=101).save()
+        TestDoc(x=10).save()
+        TestDoc(y=True).save()
+
+        self.assertEqual(query,
+        {'$and': [
+            {'$or': [{'x': {'$gt': 0}}, {'x': {'$exists': False}}]},
+            {'$or': [{'x': {'$lt': 100}}, {'y': True}]}
+        ]})
+
+        self.assertEqual(2, TestDoc.objects(q1 & q2).count())
+
+    def test_or_and_or_combination(self):
+        """Ensure that Q-objects handle ORing ANDed ORed components. :)
+        """
+        class TestDoc(Document):
+            x = IntField()
+            y = BooleanField()
+
+        TestDoc.drop_collection()
+        TestDoc(x=-1, y=True).save()
+        TestDoc(x=101, y=True).save()
+        TestDoc(x=99, y=False).save()
+        TestDoc(x=101, y=False).save()
+
+        q1 = (Q(x__gt=0) & (Q(y=True) | Q(y__exists=False)))
+        q2 = (Q(x__lt=100) & (Q(y=False) | Q(y__exists=False)))
+        query = (q1 | q2).to_query(TestDoc)
+
+        self.assertEqual(query,
+            {'$or': [
+                {'$and': [{'x': {'$gt': 0}},
+                          {'$or': [{'y': True}, {'y': {'$exists': False}}]}]},
+                {'$and': [{'x': {'$lt': 100}},
+                          {'$or': [{'y': False}, {'y': {'$exists': False}}]}]}
+            ]}
+        )
+
+        self.assertEqual(2, TestDoc.objects(q1 | q2).count())
+
+    def test_multiple_occurence_in_field(self):
+        class Test(Document):
+            name = StringField(max_length=40)
+            title = StringField(max_length=40)
+
+        q1 = Q(name__contains='te') | Q(title__contains='te')
+        q2 = Q(name__contains='12') | Q(title__contains='12')
+
+        q3 = q1 & q2
+
+        query = q3.to_query(Test)
+        self.assertEqual(query["$and"][0], q1.to_query(Test))
+        self.assertEqual(query["$and"][1], q2.to_query(Test))
+
+    def test_q_clone(self):
+
+        class TestDoc(Document):
+            x = IntField()
+
+        TestDoc.drop_collection()
+        for i in xrange(1, 101):
+            t = TestDoc(x=i)
+            t.save()
+
+        # Check normal cases work without an error
+        test = TestDoc.objects(Q(x__lt=7) & Q(x__gt=3))
+
+        self.assertEqual(test.count(), 3)
+
+        test2 = test.clone()
+        self.assertEqual(test2.count(), 3)
+        self.assertFalse(test2 == test)
+
+        test3 = test2.filter(x=6)
+        self.assertEqual(test3.count(), 1)
+        self.assertEqual(test.count(), 3)
+
+    def test_q(self):
+        """Ensure that Q objects may be used to query for documents.
+        """
+        class BlogPost(Document):
+            title = StringField()
+            publish_date = DateTimeField()
+            published = BooleanField()
+
+        BlogPost.drop_collection()
+
+        post1 = BlogPost(title='Test 1', publish_date=datetime(2010, 1, 8), published=False)
+        post1.save()
+
+        post2 = BlogPost(title='Test 2', publish_date=datetime(2010, 1, 15), published=True)
+        post2.save()
+
+        post3 = BlogPost(title='Test 3', published=True)
+        post3.save()
+
+        post4 = BlogPost(title='Test 4', publish_date=datetime(2010, 1, 8))
+        post4.save()
+
+        post5 = BlogPost(title='Test 1', publish_date=datetime(2010, 1, 15))
+        post5.save()
+
+        post6 = BlogPost(title='Test 1', published=False)
+        post6.save()
+
+        # Check ObjectId lookup works
+        obj = BlogPost.objects(id=post1.id).first()
+        self.assertEqual(obj, post1)
+
+        # Check Q object combination with one does not exist
+        q = BlogPost.objects(Q(title='Test 5') | Q(published=True))
+        posts = [post.id for post in q]
+
+        published_posts = (post2, post3)
+        self.assertTrue(all(obj.id in posts for obj in published_posts))
+
+        q = BlogPost.objects(Q(title='Test 1') | Q(published=True))
+        posts = [post.id for post in q]
+        published_posts = (post1, post2, post3, post5, post6)
+        self.assertTrue(all(obj.id in posts for obj in published_posts))
+
+        # Check Q object combination
+        date = datetime(2010, 1, 10)
+        q = BlogPost.objects(Q(publish_date__lte=date) | Q(published=True))
+        posts = [post.id for post in q]
+
+        published_posts = (post1, post2, post3, post4)
+        self.assertTrue(all(obj.id in posts for obj in published_posts))
+
+        self.assertFalse(any(obj.id in posts for obj in [post5, post6]))
+
+        BlogPost.drop_collection()
+
+        # Check the 'in' operator
+        self.Person(name='user1', age=20).save()
+        self.Person(name='user2', age=20).save()
+        self.Person(name='user3', age=30).save()
+        self.Person(name='user4', age=40).save()
+
+        self.assertEqual(len(self.Person.objects(Q(age__in=[20]))), 2)
+        self.assertEqual(len(self.Person.objects(Q(age__in=[20, 30]))), 3)
+
+        # Test invalid query objs
+        def wrong_query_objs():
+            self.Person.objects('user1')
+        def wrong_query_objs_filter():
+            self.Person.objects('user1')
+        self.assertRaises(InvalidQueryError, wrong_query_objs)
+        self.assertRaises(InvalidQueryError, wrong_query_objs_filter)
+
+    def test_q_regex(self):
+        """Ensure that Q objects can be queried using regexes.
+        """
+        person = self.Person(name='Guido van Rossum')
+        person.save()
+
+        import re
+        obj = self.Person.objects(Q(name=re.compile('^Gui'))).first()
+        self.assertEqual(obj, person)
+        obj = self.Person.objects(Q(name=re.compile('^gui'))).first()
+        self.assertEqual(obj, None)
+
+        obj = self.Person.objects(Q(name=re.compile('^gui', re.I))).first()
+        self.assertEqual(obj, person)
+
+        obj = self.Person.objects(Q(name__not=re.compile('^bob'))).first()
+        self.assertEqual(obj, person)
+
+        obj = self.Person.objects(Q(name__not=re.compile('^Gui'))).first()
+        self.assertEqual(obj, None)
+
+    def test_q_lists(self):
+        """Ensure that Q objects query ListFields correctly.
+        """
+        class BlogPost(Document):
+            tags = ListField(StringField())
+
+        BlogPost.drop_collection()
+
+        BlogPost(tags=['python', 'mongo']).save()
+        BlogPost(tags=['python']).save()
+
+        self.assertEqual(len(BlogPost.objects(Q(tags='mongo'))), 1)
+        self.assertEqual(len(BlogPost.objects(Q(tags='python'))), 2)
+
+        BlogPost.drop_collection()
+
+    def test_q_merge_queries_edge_case(self):
+
+        class User(Document):
+            email = EmailField(required=False)
+            name = StringField()
+
+        User.drop_collection()
+        pk = ObjectId()
+        User(email='example@example.com', pk=pk).save()
+
+        self.assertEqual(1, User.objects.filter(
+                                Q(email='example@example.com') |
+                                Q(name='John Doe')
+                                ).limit(2).filter(pk=pk).count())
+
+if __name__ == '__main__':
+    unittest.main()

tests/test_connection.py

@@ -0,0 +1,101 @@
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+import unittest
+import datetime
+
+import pymongo
+from bson.tz_util import utc
+
+from mongoengine import *
+import mongoengine.connection
+from mongoengine.connection import get_db, get_connection, ConnectionError
+from mongoengine.context_managers import switch_db
+
+
+class ConnectionTest(unittest.TestCase):
+
+    def tearDown(self):
+        mongoengine.connection._connection_settings = {}
+        mongoengine.connection._connections = {}
+        mongoengine.connection._dbs = {}
+
+    def test_connect(self):
+        """Ensure that the connect() method works properly.
+        """
+        connect('mongoenginetest')
+
+        conn = get_connection()
+        self.assertTrue(isinstance(conn, pymongo.connection.Connection))
+
+        db = get_db()
+        self.assertTrue(isinstance(db, pymongo.database.Database))
+        self.assertEqual(db.name, 'mongoenginetest')
+
+        connect('mongoenginetest2', alias='testdb')
+        conn = get_connection('testdb')
+        self.assertTrue(isinstance(conn, pymongo.connection.Connection))
+
+    def test_connect_uri(self):
+        """Ensure that the connect() method works properly with uri's
+        """
+        c = connect(db='mongoenginetest', alias='admin')
+        c.admin.system.users.remove({})
+        c.mongoenginetest.system.users.remove({})
+
+        c.admin.add_user("admin", "password")
+        c.admin.authenticate("admin", "password")
+        c.mongoenginetest.add_user("username", "password")
+
+        self.assertRaises(ConnectionError, connect, "testdb_uri_bad", host='mongodb://test:password@localhost')
+
+        connect("testdb_uri", host='mongodb://username:password@localhost/mongoenginetest')
+
+        conn = get_connection()
+        self.assertTrue(isinstance(conn, pymongo.connection.Connection))
+
+        db = get_db()
+        self.assertTrue(isinstance(db, pymongo.database.Database))
+        self.assertEqual(db.name, 'mongoenginetest')
+
+    def test_register_connection(self):
+        """Ensure that connections with different aliases may be registered.
+        """
+        register_connection('testdb', 'mongoenginetest2')
+
+        self.assertRaises(ConnectionError, get_connection)
+        conn = get_connection('testdb')
+        self.assertTrue(isinstance(conn, pymongo.connection.Connection))
+
+        db = get_db('testdb')
+        self.assertTrue(isinstance(db, pymongo.database.Database))
+        self.assertEqual(db.name, 'mongoenginetest2')
+
+    def test_connection_kwargs(self):
+        """Ensure that connection kwargs get passed to pymongo.
+        """
+        connect('mongoenginetest', alias='t1', tz_aware=True)
+        conn = get_connection('t1')
+
+        self.assertTrue(conn.tz_aware)
+
+        connect('mongoenginetest2', alias='t2')
+        conn = get_connection('t2')
+        self.assertFalse(conn.tz_aware)
+
+    def test_datetime(self):
+        connect('mongoenginetest', tz_aware=True)
+        d = datetime.datetime(2010, 5, 5, tzinfo=utc)
+
+        class DateDoc(Document):
+            the_date = DateTimeField(required=True)
+
+        DateDoc.drop_collection()
+        DateDoc(the_date=d).save()
+
+        date_doc = DateDoc.objects.first()
+        self.assertEqual(d, date_doc.the_date)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/test_context_managers.py

@@ -0,0 +1,156 @@
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+from mongoengine import *
+from mongoengine.connection import get_db
+from mongoengine.context_managers import (switch_db, switch_collection,
+                                          no_dereference, query_counter)
+
+
+class ContextManagersTest(unittest.TestCase):
+
+    def test_switch_db_context_manager(self):
+        connect('mongoenginetest')
+        register_connection('testdb-1', 'mongoenginetest2')
+
+        class Group(Document):
+            name = StringField()
+
+        Group.drop_collection()
+
+        Group(name="hello - default").save()
+        self.assertEqual(1, Group.objects.count())
+
+        with switch_db(Group, 'testdb-1') as Group:
+
+            self.assertEqual(0, Group.objects.count())
+
+            Group(name="hello").save()
+
+            self.assertEqual(1, Group.objects.count())
+
+            Group.drop_collection()
+            self.assertEqual(0, Group.objects.count())
+
+        self.assertEqual(1, Group.objects.count())
+
+    def test_switch_collection_context_manager(self):
+        connect('mongoenginetest')
+        register_connection('testdb-1', 'mongoenginetest2')
+
+        class Group(Document):
+            name = StringField()
+
+        Group.drop_collection()
+        with switch_collection(Group, 'group1') as Group:
+            Group.drop_collection()
+
+        Group(name="hello - group").save()
+        self.assertEqual(1, Group.objects.count())
+
+        with switch_collection(Group, 'group1') as Group:
+
+            self.assertEqual(0, Group.objects.count())
+
+            Group(name="hello - group1").save()
+
+            self.assertEqual(1, Group.objects.count())
+
+            Group.drop_collection()
+            self.assertEqual(0, Group.objects.count())
+
+        self.assertEqual(1, Group.objects.count())
+
+    def test_no_dereference_context_manager_object_id(self):
+        """Ensure that DBRef items in ListFields aren't dereferenced.
+        """
+        connect('mongoenginetest')
+
+        class User(Document):
+            name = StringField()
+
+        class Group(Document):
+            ref = ReferenceField(User, dbref=False)
+            generic = GenericReferenceField()
+            members = ListField(ReferenceField(User, dbref=False))
+
+        User.drop_collection()
+        Group.drop_collection()
+
+        for i in xrange(1, 51):
+            User(name='user %s' % i).save()
+
+        user = User.objects.first()
+        Group(ref=user, members=User.objects, generic=user).save()
+
+        with no_dereference(Group) as NoDeRefGroup:
+            self.assertTrue(Group._fields['members']._auto_dereference)
+            self.assertFalse(NoDeRefGroup._fields['members']._auto_dereference)
+
+        with no_dereference(Group) as Group:
+            group = Group.objects.first()
+            self.assertTrue(all([not isinstance(m, User)
+                                for m in group.members]))
+            self.assertFalse(isinstance(group.ref, User))
+            self.assertFalse(isinstance(group.generic, User))
+
+        self.assertTrue(all([isinstance(m, User)
+                             for m in group.members]))
+        self.assertTrue(isinstance(group.ref, User))
+        self.assertTrue(isinstance(group.generic, User))
+
+    def test_no_dereference_context_manager_dbref(self):
+        """Ensure that DBRef items in ListFields aren't dereferenced.
+        """
+        connect('mongoenginetest')
+
+        class User(Document):
+            name = StringField()
+
+        class Group(Document):
+            ref = ReferenceField(User, dbref=True)
+            generic = GenericReferenceField()
+            members = ListField(ReferenceField(User, dbref=True))
+
+        User.drop_collection()
+        Group.drop_collection()
+
+        for i in xrange(1, 51):
+            User(name='user %s' % i).save()
+
+        user = User.objects.first()
+        Group(ref=user, members=User.objects, generic=user).save()
+
+        with no_dereference(Group) as NoDeRefGroup:
+            self.assertTrue(Group._fields['members']._auto_dereference)
+            self.assertFalse(NoDeRefGroup._fields['members']._auto_dereference)
+
+        with no_dereference(Group) as Group:
+            group = Group.objects.first()
+            self.assertTrue(all([not isinstance(m, User)
+                                for m in group.members]))
+            self.assertFalse(isinstance(group.ref, User))
+            self.assertFalse(isinstance(group.generic, User))
+
+        self.assertTrue(all([isinstance(m, User)
+                             for m in group.members]))
+        self.assertTrue(isinstance(group.ref, User))
+        self.assertTrue(isinstance(group.generic, User))
+
+    def test_query_counter(self):
+        connect('mongoenginetest')
+        db = get_db()
+        db.test.find({})
+
+        with query_counter() as q:
+            self.assertEqual(0, q)
+
+            for i in xrange(1, 51):
+                db.test.find({}).count()
+
+            self.assertEqual(50, q)
+
+if __name__ == '__main__':
+    unittest.main()

tests/test_django.py

@@ -0,0 +1,188 @@
+from __future__ import with_statement
+import sys
+sys.path[0:0] = [""]
+import unittest
+from nose.plugins.skip import SkipTest
+from mongoengine.python_support import PY3
+from mongoengine import *
+
+try:
+    from mongoengine.django.shortcuts import get_document_or_404
+
+    from django.http import Http404
+    from django.template import Context, Template
+    from django.conf import settings
+    from django.core.paginator import Paginator
+
+    settings.configure(USE_TZ=True)
+
+    from django.contrib.sessions.tests import SessionTestsMixin
+    from mongoengine.django.sessions import SessionStore, MongoSession
+except Exception, err:
+    if PY3:
+        SessionTestsMixin = type  # dummy value so no error
+        SessionStore = None  # dummy value so no error
+    else:
+        raise err
+
+
+from datetime import tzinfo, timedelta
+ZERO = timedelta(0)
+
+class FixedOffset(tzinfo):
+    """Fixed offset in minutes east from UTC."""
+
+    def __init__(self, offset, name):
+        self.__offset = timedelta(minutes = offset)
+        self.__name = name
+
+    def utcoffset(self, dt):
+        return self.__offset
+
+    def tzname(self, dt):
+        return self.__name
+
+    def dst(self, dt):
+        return ZERO
+
+
+def activate_timezone(tz):
+    """Activate Django timezone support if it is available.
+    """
+    try:
+        from django.utils import timezone
+        timezone.deactivate()
+        timezone.activate(tz)
+    except ImportError:
+        pass
+
+
+class QuerySetTest(unittest.TestCase):
+
+    def setUp(self):
+        if PY3:
+            raise SkipTest('django does not have Python 3 support')
+        connect(db='mongoenginetest')
+
+        class Person(Document):
+            name = StringField()
+            age = IntField()
+        self.Person = Person
+
+    def test_order_by_in_django_template(self):
+        """Ensure that QuerySets are properly ordered in Django template.
+        """
+        self.Person.drop_collection()
+
+        self.Person(name="A", age=20).save()
+        self.Person(name="D", age=10).save()
+        self.Person(name="B", age=40).save()
+        self.Person(name="C", age=30).save()
+
+        t = Template("{% for o in ol %}{{ o.name }}-{{ o.age }}:{% endfor %}")
+
+        d = {"ol": self.Person.objects.order_by('-name')}
+        self.assertEqual(t.render(Context(d)), u'D-10:C-30:B-40:A-20:')
+        d = {"ol": self.Person.objects.order_by('+name')}
+        self.assertEqual(t.render(Context(d)), u'A-20:B-40:C-30:D-10:')
+        d = {"ol": self.Person.objects.order_by('-age')}
+        self.assertEqual(t.render(Context(d)), u'B-40:C-30:A-20:D-10:')
+        d = {"ol": self.Person.objects.order_by('+age')}
+        self.assertEqual(t.render(Context(d)), u'D-10:A-20:C-30:B-40:')
+
+        self.Person.drop_collection()
+
+    def test_q_object_filter_in_template(self):
+
+        self.Person.drop_collection()
+
+        self.Person(name="A", age=20).save()
+        self.Person(name="D", age=10).save()
+        self.Person(name="B", age=40).save()
+        self.Person(name="C", age=30).save()
+
+        t = Template("{% for o in ol %}{{ o.name }}-{{ o.age }}:{% endfor %}")
+
+        d = {"ol": self.Person.objects.filter(Q(age=10) | Q(name="C"))}
+        self.assertEqual(t.render(Context(d)), 'D-10:C-30:')
+
+        # Check double rendering doesn't throw an error
+        self.assertEqual(t.render(Context(d)), 'D-10:C-30:')
+
+    def test_get_document_or_404(self):
+        p = self.Person(name="G404")
+        p.save()
+
+        self.assertRaises(Http404, get_document_or_404, self.Person, pk='1234')
+        self.assertEqual(p, get_document_or_404(self.Person, pk=p.pk))
+
+    def test_pagination(self):
+        """Ensure that Pagination works as expected
+        """
+        class Page(Document):
+            name = StringField()
+
+        Page.drop_collection()
+
+        for i in xrange(1, 11):
+            Page(name=str(i)).save()
+
+        paginator = Paginator(Page.objects.all(), 2)
+
+        t = Template("{% for i in page.object_list  %}{{ i.name }}:{% endfor %}")
+        for p in paginator.page_range:
+            d = {"page": paginator.page(p)}
+            end = p * 2
+            start = end - 1
+            self.assertEqual(t.render(Context(d)), u'%d:%d:' % (start, end))
+
+    def test_nested_queryset_template_iterator(self):
+        # Try iterating the same queryset twice, nested, in a Django template.
+        names = ['A', 'B', 'C', 'D']
+
+        class User(Document):
+            name = StringField()
+
+            def __unicode__(self):
+                return self.name
+
+        User.drop_collection()
+
+        for name in names:
+            User(name=name).save()
+
+        users = User.objects.all().order_by('name')
+        template = Template("{% for user in users %}{{ user.name }}{% ifequal forloop.counter 2 %} {% for inner_user in users %}{{ inner_user.name }}{% endfor %} {% endifequal %}{% endfor %}")
+        rendered = template.render(Context({'users': users}))
+        self.assertEqual(rendered, 'AB ABCD CD')
+
+class MongoDBSessionTest(SessionTestsMixin, unittest.TestCase):
+    backend = SessionStore
+
+    def setUp(self):
+        if PY3:
+            raise SkipTest('django does not have Python 3 support')
+        connect(db='mongoenginetest')
+        MongoSession.drop_collection()
+        super(MongoDBSessionTest, self).setUp()
+
+    def test_first_save(self):
+        session = SessionStore()
+        session['test'] = True
+        session.save()
+        self.assertTrue('test' in session)
+
+    def test_session_expiration_tz(self):
+        activate_timezone(FixedOffset(60, 'UTC+1'))
+        # create and save new session
+        session = SessionStore()
+        session.set_expiry(600) # expire in 600 seconds
+        session['test_expire'] = True
+        session.save()
+        # reload session with key
+        key = session.session_key
+        session = SessionStore(key)
+        self.assertTrue('test_expire' in session, 'Session has expired before it is expected')
+
+if __name__ == '__main__':
+    unittest.main()

tests/test_replicaset_connection.py

@@ -0,0 +1,35 @@
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+import pymongo
+from pymongo import ReadPreference, ReplicaSetConnection
+
+import mongoengine
+from mongoengine import *
+from mongoengine.connection import get_db, get_connection, ConnectionError
+
+
+class ConnectionTest(unittest.TestCase):
+
+    def tearDown(self):
+        mongoengine.connection._connection_settings = {}
+        mongoengine.connection._connections = {}
+        mongoengine.connection._dbs = {}
+
+    def test_replicaset_uri_passes_read_preference(self):
+        """Requires a replica set called "rs" on port 27017
+        """
+
+        try:
+            conn = connect(db='mongoenginetest', host="mongodb://localhost/mongoenginetest?replicaSet=rs", read_preference=ReadPreference.SECONDARY_ONLY)
+        except ConnectionError, e:
+            return
+
+        if not isinstance(conn, ReplicaSetConnection):
+            return
+
+        self.assertEqual(conn.read_preference, ReadPreference.SECONDARY_ONLY)
+
+if __name__ == '__main__':
+    unittest.main()

tests/test_signals.py

@@ -0,0 +1,264 @@
+# -*- coding: utf-8 -*-
+import sys
+sys.path[0:0] = [""]
+import unittest
+
+from mongoengine import *
+from mongoengine import signals
+
+signal_output = []
+
+
+class SignalTests(unittest.TestCase):
+    """
+    Testing signals before/after saving and deleting.
+    """
+
+    def get_signal_output(self, fn, *args, **kwargs):
+        # Flush any existing signal output
+        global signal_output
+        signal_output = []
+        fn(*args, **kwargs)
+        return signal_output
+
+    def setUp(self):
+        connect(db='mongoenginetest')
+
+        class Author(Document):
+            name = StringField()
+
+            def __unicode__(self):
+                return self.name
+
+            @classmethod
+            def pre_init(cls, sender, document, *args, **kwargs):
+                signal_output.append('pre_init signal, %s' % cls.__name__)
+                signal_output.append(str(kwargs['values']))
+
+            @classmethod
+            def post_init(cls, sender, document, **kwargs):
+                signal_output.append('post_init signal, %s' % document)
+
+            @classmethod
+            def pre_save(cls, sender, document, **kwargs):
+                signal_output.append('pre_save signal, %s' % document)
+
+            @classmethod
+            def post_save(cls, sender, document, **kwargs):
+                signal_output.append('post_save signal, %s' % document)
+                if 'created' in kwargs:
+                    if kwargs['created']:
+                        signal_output.append('Is created')
+                    else:
+                        signal_output.append('Is updated')
+
+            @classmethod
+            def pre_delete(cls, sender, document, **kwargs):
+                signal_output.append('pre_delete signal, %s' % document)
+
+            @classmethod
+            def post_delete(cls, sender, document, **kwargs):
+                signal_output.append('post_delete signal, %s' % document)
+
+            @classmethod
+            def pre_bulk_insert(cls, sender, documents, **kwargs):
+                signal_output.append('pre_bulk_insert signal, %s' % documents)
+
+            @classmethod
+            def post_bulk_insert(cls, sender, documents, **kwargs):
+                signal_output.append('post_bulk_insert signal, %s' % documents)
+                if kwargs.get('loaded', False):
+                    signal_output.append('Is loaded')
+                else:
+                    signal_output.append('Not loaded')
+        self.Author = Author
+
+        class Another(Document):
+            name = StringField()
+
+            def __unicode__(self):
+                return self.name
+
+            @classmethod
+            def pre_init(cls, sender, document, **kwargs):
+                signal_output.append('pre_init Another signal, %s' % cls.__name__)
+                signal_output.append(str(kwargs['values']))
+
+            @classmethod
+            def post_init(cls, sender, document, **kwargs):
+                signal_output.append('post_init Another signal, %s' % document)
+
+            @classmethod
+            def pre_save(cls, sender, document, **kwargs):
+                signal_output.append('pre_save Another signal, %s' % document)
+
+            @classmethod
+            def post_save(cls, sender, document, **kwargs):
+                signal_output.append('post_save Another signal, %s' % document)
+                if 'created' in kwargs:
+                    if kwargs['created']:
+                        signal_output.append('Is created')
+                    else:
+                        signal_output.append('Is updated')
+
+            @classmethod
+            def pre_delete(cls, sender, document, **kwargs):
+                signal_output.append('pre_delete Another signal, %s' % document)
+
+            @classmethod
+            def post_delete(cls, sender, document, **kwargs):
+                signal_output.append('post_delete Another signal, %s' % document)
+
+        self.Another = Another
+
+        class ExplicitId(Document):
+            id = IntField(primary_key=True)
+
+            @classmethod
+            def post_save(cls, sender, document, **kwargs):
+                if 'created' in kwargs:
+                    if kwargs['created']:
+                        signal_output.append('Is created')
+                    else:
+                        signal_output.append('Is updated')
+
+        self.ExplicitId = ExplicitId
+        self.ExplicitId.objects.delete()
+        # Save up the number of connected signals so that we can check at the
+        # end that all the signals we register get properly unregistered
+        self.pre_signals = (
+            len(signals.pre_init.receivers),
+            len(signals.post_init.receivers),
+            len(signals.pre_save.receivers),
+            len(signals.post_save.receivers),
+            len(signals.pre_delete.receivers),
+            len(signals.post_delete.receivers),
+            len(signals.pre_bulk_insert.receivers),
+            len(signals.post_bulk_insert.receivers),
+        )
+
+        signals.pre_init.connect(Author.pre_init, sender=Author)
+        signals.post_init.connect(Author.post_init, sender=Author)
+        signals.pre_save.connect(Author.pre_save, sender=Author)
+        signals.post_save.connect(Author.post_save, sender=Author)
+        signals.pre_delete.connect(Author.pre_delete, sender=Author)
+        signals.post_delete.connect(Author.post_delete, sender=Author)
+        signals.pre_bulk_insert.connect(Author.pre_bulk_insert, sender=Author)
+        signals.post_bulk_insert.connect(Author.post_bulk_insert, sender=Author)
+
+        signals.pre_init.connect(Another.pre_init, sender=Another)
+        signals.post_init.connect(Another.post_init, sender=Another)
+        signals.pre_save.connect(Another.pre_save, sender=Another)
+        signals.post_save.connect(Another.post_save, sender=Another)
+        signals.pre_delete.connect(Another.pre_delete, sender=Another)
+        signals.post_delete.connect(Another.post_delete, sender=Another)
+
+        signals.post_save.connect(ExplicitId.post_save, sender=ExplicitId)
+
+    def tearDown(self):
+        signals.pre_init.disconnect(self.Author.pre_init)
+        signals.post_init.disconnect(self.Author.post_init)
+        signals.post_delete.disconnect(self.Author.post_delete)
+        signals.pre_delete.disconnect(self.Author.pre_delete)
+        signals.post_save.disconnect(self.Author.post_save)
+        signals.pre_save.disconnect(self.Author.pre_save)
+        signals.pre_bulk_insert.disconnect(self.Author.pre_bulk_insert)
+        signals.post_bulk_insert.disconnect(self.Author.post_bulk_insert)
+
+        signals.pre_init.disconnect(self.Another.pre_init)
+        signals.post_init.disconnect(self.Another.post_init)
+        signals.post_delete.disconnect(self.Another.post_delete)
+        signals.pre_delete.disconnect(self.Another.pre_delete)
+        signals.post_save.disconnect(self.Another.post_save)
+        signals.pre_save.disconnect(self.Another.pre_save)
+
+        signals.post_save.disconnect(self.ExplicitId.post_save)
+
+        # Check that all our signals got disconnected properly.
+        post_signals = (
+            len(signals.pre_init.receivers),
+            len(signals.post_init.receivers),
+            len(signals.pre_save.receivers),
+            len(signals.post_save.receivers),
+            len(signals.pre_delete.receivers),
+            len(signals.post_delete.receivers),
+            len(signals.pre_bulk_insert.receivers),
+            len(signals.post_bulk_insert.receivers),
+        )
+
+        self.ExplicitId.objects.delete()
+
+        self.assertEqual(self.pre_signals, post_signals)
+
+    def test_model_signals(self):
+        """ Model saves should throw some signals. """
+
+        def create_author():
+            self.Author(name='Bill Shakespeare')
+
+        def bulk_create_author_with_load():
+            a1 = self.Author(name='Bill Shakespeare')
+            self.Author.objects.insert([a1], load_bulk=True)
+
+        def bulk_create_author_without_load():
+            a1 = self.Author(name='Bill Shakespeare')
+            self.Author.objects.insert([a1], load_bulk=False)
+
+        self.assertEqual(self.get_signal_output(create_author), [
+            "pre_init signal, Author",
+            "{'name': 'Bill Shakespeare'}",
+            "post_init signal, Bill Shakespeare",
+        ])
+
+        a1 = self.Author(name='Bill Shakespeare')
+        self.assertEqual(self.get_signal_output(a1.save), [
+            "pre_save signal, Bill Shakespeare",
+            "post_save signal, Bill Shakespeare",
+            "Is created"
+        ])
+
+        a1.reload()
+        a1.name = 'William Shakespeare'
+        self.assertEqual(self.get_signal_output(a1.save), [
+            "pre_save signal, William Shakespeare",
+            "post_save signal, William Shakespeare",
+            "Is updated"
+        ])
+
+        self.assertEqual(self.get_signal_output(a1.delete), [
+            'pre_delete signal, William Shakespeare',
+            'post_delete signal, William Shakespeare',
+        ])
+
+        signal_output = self.get_signal_output(bulk_create_author_with_load)
+
+        # The output of this signal is not entirely deterministic. The reloaded
+        # object will have an object ID. Hence, we only check part of the output
+        self.assertEqual(signal_output[3],
+            "pre_bulk_insert signal, [<Author: Bill Shakespeare>]")
+        self.assertEqual(signal_output[-2:],
+            ["post_bulk_insert signal, [<Author: Bill Shakespeare>]",
+             "Is loaded",])
+
+        self.assertEqual(self.get_signal_output(bulk_create_author_without_load), [
+            "pre_init signal, Author",
+            "{'name': 'Bill Shakespeare'}",
+            "post_init signal, Bill Shakespeare",
+            "pre_bulk_insert signal, [<Author: Bill Shakespeare>]",
+            "post_bulk_insert signal, [<Author: Bill Shakespeare>]",
+            "Not loaded",
+        ])
+
+        self.Author.objects.delete()
+
+    def test_signals_with_explicit_doc_ids(self):
+        """ Model saves must have a created flag the first time."""
+        ei = self.ExplicitId(id=123)
+        # post save must received the created flag, even if there's already
+        # an object id present
+        self.assertEqual(self.get_signal_output(ei.save), ['Is created'])
+        # second time, it must be an update
+        self.assertEqual(self.get_signal_output(ei.save), ['Is updated'])
+
+if __name__ == '__main__':
+    unittest.main()