Merge branch 'master_gh'

.github/CONTRIBUTING.md

@@ -2,7 +2,7 @@
 
 There's lots to do, and we're working hard, so any help is welcome!
 
-- :speech_balloon: Join us on [Slack](https://join.slack.com/t/betterproto/shared_invite/zt-f0n0uolx-iN8gBNrkPxtKHTLpG3o1OQ)!
+- :speech_balloon: Join us on [Discord](https://discord.gg/DEVteTupPb)!
 
 What can you do?
 
@@ -15,9 +15,9 @@ What can you do?
     - File a bug (please check its not a duplicate)
     - Propose an enhancement
 - :white_check_mark: Create a PR:
-    - [Creating a failing test-case](https://github.com/danielgtaylor/python-betterproto/blob/master/betterproto/tests/README.md) to make bug-fixing easier
+    - [Creating a failing test-case](https://github.com/danielgtaylor/python-betterproto/blob/master/tests/README.md) to make bug-fixing easier
     - Fix any of the open issues
         - [Good first issues](https://github.com/danielgtaylor/python-betterproto/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22)
         - [Issues with tests](https://github.com/danielgtaylor/python-betterproto/issues?q=is%3Aissue+is%3Aopen+label%3A%22has+test%22)
     - New bugfix or idea
-        - If you'd like to discuss your idea first, join us on Slack!
+        - If you'd like to discuss your idea first, join us on Discord!

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -8,7 +8,7 @@ body:
       value: >
         Thanks for taking the time to fill out a bug report!
         
-        If you're not sure it's a bug and you just have a question, the [community Slack channel](https://join.slack.com/t/betterproto/shared_invite/zt-f0n0uolx-iN8gBNrkPxtKHTLpG3o1OQ) is a better place for general questions than a GitHub issue.
+        If you're not sure it's a bug and you just have a question, the [community Discord channel](https://discord.gg/DEVteTupPb) is a better place for general questions than a GitHub issue.
 
   - type: input
     attributes:

.github/ISSUE_TEMPLATE/config.yml

@@ -2,5 +2,5 @@ name:
 description:
 contact_links:
   - name: For questions about the library
-    about: Support questions are better answered in our Slack group.
-    url: https://join.slack.com/t/betterproto/shared_invite/zt-f0n0uolx-iN8gBNrkPxtKHTLpG3o1OQ
+    about: Support questions are better answered in our Discord group.
+    url: https://discord.gg/DEVteTupPb

.github/workflows/ci.yml

@@ -16,7 +16,7 @@ jobs:
       fail-fast: false
       matrix:
         os: [Ubuntu, MacOS, Windows]
-        python-version: ['3.8', '3.9', '3.10', '3.11', '3.12']
+        python-version: ['3.9', '3.10', '3.11', '3.12', '3.13']
     steps:
       - uses: actions/checkout@v4
 
@@ -41,7 +41,7 @@ jobs:
         run: poetry config virtualenvs.in-project true
 
       - name: Set up cache
-        uses: actions/cache@v3
+        uses: actions/cache@v4
         id: cache
         with:
           path: .venv

.github/workflows/code-quality.yml

@@ -13,6 +13,6 @@ jobs:
     name: Check code/doc formatting
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v3
-    - uses: actions/setup-python@v4
-    - uses: pre-commit/action@v2.0.3
+    - uses: actions/checkout@v4
+    - uses: actions/setup-python@v5
+    - uses: pre-commit/action@v3.0.1

.github/workflows/codeql-analysis.yml

@@ -25,11 +25,11 @@ jobs:
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v2
+      uses: github/codeql-action/init@v3
       with:
         languages: ${{ matrix.language }}
         # If you wish to specify custom queries, you can do so here or in a config file.
@@ -40,7 +40,7 @@ jobs:
         # queries: security-extended,security-and-quality
 
     - name: Autobuild
-      uses: github/codeql-action/autobuild@v2
+      uses: github/codeql-action/autobuild@v3
 
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v2
+      uses: github/codeql-action/analyze@v3

.github/workflows/release.yml

@@ -15,11 +15,11 @@ jobs:
     name: Distribution
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
-      - name: Set up Python 3.8
-        uses: actions/setup-python@v4
+      - uses: actions/checkout@v4
+      - name: Set up Python 3.9
+        uses: actions/setup-python@v5
         with:
-          python-version: 3.8
+          python-version: 3.9
       - name: Install poetry
         run: python -m pip install poetry
       - name: Build package

.gitignore

@@ -18,3 +18,4 @@ output
 .asv
 venv
 .devcontainer
+.ruff_cache

.pre-commit-config.yaml

@@ -2,16 +2,13 @@ ci:
   autofix_prs: false
 
 repos:
-  - repo: https://github.com/pycqa/isort
-    rev: 5.11.5
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.1
     hooks:
-      - id: isort
-
-  - repo: https://github.com/psf/black
-    rev: 23.1.0
-    hooks:
-      - id: black
-        args: ["--target-version", "py310"]
+      - id: ruff-format
+        args: ["--diff", "src", "tests"]
+      - id: ruff
+        args: ["--select", "I", "src", "tests"]
 
   - repo: https://github.com/PyCQA/doc8
     rev: 0.10.1
@@ -21,7 +18,7 @@ repos:
           - toml
 
   - repo: https://github.com/macisamuele/language-formatters-pre-commit-hooks
-    rev: v2.10.0
+    rev: v2.14.0
     hooks:
       - id: pretty-format-java
         args: [--autofix, --aosp]

README.md

@@ -439,7 +439,7 @@ value3: str | int = 1
 
 ## Development
 
-- _Join us on [Slack](https://join.slack.com/t/betterproto/shared_invite/zt-f0n0uolx-iN8gBNrkPxtKHTLpG3o1OQ)!_
+- _Join us on [Discord](https://discord.gg/DEVteTupPb)!_
 - _See how you can help → [Contributing](.github/CONTRIBUTING.md)_
 
 ### Requirements
@@ -575,7 +575,7 @@ protoc \
 
 ## Community
 
-Join us on [Slack](https://join.slack.com/t/betterproto/shared_invite/zt-f0n0uolx-iN8gBNrkPxtKHTLpG3o1OQ)!
+Join us on [Discord](https://discord.gg/DEVteTupPb)!
 
 ## License
 

pyproject.toml

@@ -1,8 +1,10 @@
-[tool.poetry]
+[project]
 name = "betterproto"
 version = "2.0.0b7"
 description = "A better Protobuf / gRPC generator & library"
-authors = ["Daniel G. Taylor <danielgtaylor@gmail.com>"]
+authors = [
+    {name = "Daniel G. Taylor", email = "danielgtaylor@gmail.com"}
+]
 readme = "README.md"
 repository = "https://github.com/danielgtaylor/python-betterproto"
 keywords = ["protobuf", "gRPC"]
@@ -10,46 +12,54 @@ license = "MIT"
 packages = [
     { include = "betterproto", from = "src" }
 ]
+requires-python = ">=3.9,<4.0"
+dynamic = ["dependencies"]
 
 [tool.poetry.dependencies]
-python = "^3.8"
-black = { version = ">=23.1.0", optional = true }
+# The Ruff version is pinned. To update it, also update it in .pre-commit-config.yaml
+ruff = { version = "~0.9.1", optional = true }
 grpclib = "^0.4.1"
 jinja2 = { version = ">=3.0.3", optional = true }
 python-dateutil = "^2.8"
-isort = { version = "^5.11.5", optional = true }
 typing-extensions = "^4.7.1"
 betterproto-rust-codec = { version = "0.1.1", optional = true }
 
 [tool.poetry.group.dev.dependencies]
-asv = "^0.4.2"
-bpython = "^0.19"
+asv = "^0.6.4"
+bpython = "^0.24"
 jinja2 = ">=3.0.3"
 mypy = "^1.11.2"
-sphinx = "3.1.2"
-sphinx-rtd-theme = "0.5.0"
-pre-commit = "^2.17.0"
+sphinx = "7.4.7"
+sphinx-rtd-theme = "3.0.2"
+pre-commit = "^4.0.1"
 grpcio-tools = "^1.54.2"
 tox = "^4.0.0"
 
 [tool.poetry.group.test.dependencies]
 poethepoet = ">=0.9.0"
-pytest = "^6.2.5"
-pytest-asyncio = "^0.12.0"
-pytest-cov = "^2.9.0"
+pytest = "^7.4.4"
+pytest-asyncio = "^0.23.8"
+pytest-cov = "^6.0.0"
 pytest-mock = "^3.1.1"
 pydantic = ">=2.0,<3"
-protobuf = "^4"
-cachelib = "^0.10.2"
+protobuf = "^5"
+cachelib = "^0.13.0"
 tomlkit = ">=0.7.0"
 
-[tool.poetry.scripts]
+[project.scripts]
 protoc-gen-python_betterproto = "betterproto.plugin:main"
 
-[tool.poetry.extras]
-compiler = ["black", "isort", "jinja2"]
+[project.optional-dependencies]
+compiler = ["ruff", "jinja2"]
 rust-codec = ["betterproto-rust-codec"]
 
+[tool.ruff]
+extend-exclude = ["tests/output_*"]
+target-version = "py38"
+
+[tool.ruff.lint.isort]
+combine-as-imports = true
+lines-after-imports = 2
 
 # Dev workflow tasks
 
@@ -65,13 +75,29 @@ help = "Run tests"
 cmd  = "mypy src --ignore-missing-imports"
 help = "Check types with mypy"
 
-[tool.poe.tasks]
-_black = "black . --exclude tests/output_ --target-version py310"
-_isort = "isort . --extend-skip-glob 'tests/output_*/**/*'"
-
 [tool.poe.tasks.format]
-sequence  = ["_black", "_isort"]
-help = "Apply black and isort formatting to source code"
+sequence = ["_format", "_sort-imports"]
+help = "Format the source code, and sort the imports"
+
+[tool.poe.tasks.check]
+sequence = ["_check-format", "_check-imports"]
+help = "Check that the source code is formatted and the imports sorted"
+
+[tool.poe.tasks._format]
+cmd = "ruff format src tests"
+help = "Format the source code without sorting the imports"
+
+[tool.poe.tasks._sort-imports]
+cmd = "ruff check --select I --fix src tests"
+help = "Sort the imports"
+
+[tool.poe.tasks._check-format]
+cmd = "ruff format --diff src tests"
+help = "Check that the source code is formatted"
+
+[tool.poe.tasks._check-imports]
+cmd = "ruff check --select I src tests"
+help = "Check that the imports are sorted"
 
 [tool.poe.tasks.docs]
 cmd  = "sphinx-build docs docs/build"
@@ -106,23 +132,6 @@ help = "Regenerate the types in betterproto.lib.std.google"
 shell = "poe generate && tox"
 help = "Run tests with multiple pythons"
 
-[tool.poe.tasks.check-style]
-cmd = "black . --check --diff"
-help = "Check if code style is correct"
-
-[tool.isort]
-py_version = 37
-profile = "black"
-force_single_line = false
-combine_as_imports = true
-lines_after_imports = 2
-include_trailing_comma = true
-force_grid_wrap = 2
-src_paths = ["src", "tests"]
-
-[tool.black]
-target-version = ['py37']
-
 [tool.doc8]
 paths = ["docs"]
 max_line_length = 88
@@ -156,5 +165,5 @@ require_poetry = true
 """
 
 [build-system]
-requires = ["poetry-core>=1.0.0,<2"]
+requires = ["poetry-core>=2.0.0,<3"]
 build-backend = "poetry.core.masonry.api"

src/betterproto/__init__.py

@@ -66,8 +66,7 @@ if sys.version_info >= (3, 10):
     from types import UnionType as _types_UnionType
 else:
 
-    class _types_UnionType:
-        ...
+    class _types_UnionType: ...
 
 
 # Proto 3 data types
@@ -2014,10 +2013,10 @@ class _Timestamp(Timestamp):
             return f"{result}Z"
         if (nanos % 1e6) == 0:
             # Serialize 3 fractional digits.
-            return f"{result}.{int(nanos // 1e6) :03d}Z"
+            return f"{result}.{int(nanos // 1e6):03d}Z"
         if (nanos % 1e3) == 0:
             # Serialize 6 fractional digits.
-            return f"{result}.{int(nanos // 1e3) :06d}Z"
+            return f"{result}.{int(nanos // 1e3):06d}Z"
         # Serialize 9 fractional digits.
         return f"{result}.{nanos:09d}"
 

src/betterproto/enum.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-import sys
 from enum import (
     EnumMeta,
     IntEnum,
@@ -90,15 +89,8 @@ class EnumType(EnumMeta if TYPE_CHECKING else type):
         def __iter__(cls) -> Generator[Enum, None, None]:
             yield from cls._member_map_.values()
 
-        if sys.version_info >= (3, 8):  # 3.8 added __reversed__ to dict_values
-
-            def __reversed__(cls) -> Generator[Enum, None, None]:
-                yield from reversed(cls._member_map_.values())
-
-        else:
-
-            def __reversed__(cls) -> Generator[Enum, None, None]:
-                yield from reversed(tuple(cls._member_map_.values()))
+        def __reversed__(cls) -> Generator[Enum, None, None]:
+            yield from reversed(cls._member_map_.values())
 
         def __getitem__(cls, key: str) -> Enum:
             return cls._member_map_[key]
@@ -140,6 +132,9 @@ class Enum(IntEnum if TYPE_CHECKING else int, metaclass=EnumType):
             super().__setattr__(self, "value", value)
             return self
 
+    def __getnewargs_ex__(self) -> Tuple[Tuple[()], Dict[str, Any]]:
+        return (), {"name": self.name, "value": self.value}
+
     def __str__(self) -> str:
         return self.name or "None"
 

src/betterproto/lib/pydantic/google/protobuf/__init__.py

@@ -1204,9 +1204,9 @@ class EnumDescriptorProto(betterproto.Message):
     name: str = betterproto.string_field(1)
     value: List["EnumValueDescriptorProto"] = betterproto.message_field(2)
     options: "EnumOptions" = betterproto.message_field(3)
-    reserved_range: List[
-        "EnumDescriptorProtoEnumReservedRange"
-    ] = betterproto.message_field(4)
+    reserved_range: List["EnumDescriptorProtoEnumReservedRange"] = (
+        betterproto.message_field(4)
+    )
     """
     Range of reserved numeric values. Reserved numeric values may not be used
      by enum values in the same enum declaration. Reserved ranges may not
@@ -1792,9 +1792,9 @@ class FeatureSetDefaults(betterproto.Message):
      for the closest matching edition, followed by proto merges.
     """
 
-    defaults: List[
-        "FeatureSetDefaultsFeatureSetEditionDefault"
-    ] = betterproto.message_field(1)
+    defaults: List["FeatureSetDefaultsFeatureSetEditionDefault"] = (
+        betterproto.message_field(1)
+    )
     minimum_edition: "Edition" = betterproto.enum_field(4)
     """
     The minimum supported edition (inclusive) when this was constructed.

src/betterproto/lib/pydantic/google/protobuf/compiler/__init__.py

@@ -53,9 +53,9 @@ class CodeGeneratorRequest(betterproto.Message):
     parameter: str = betterproto.string_field(2)
     """The generator parameter passed on the command-line."""
 
-    proto_file: List[
-        "betterproto_lib_pydantic_google_protobuf.FileDescriptorProto"
-    ] = betterproto.message_field(15)
+    proto_file: List["betterproto_lib_pydantic_google_protobuf.FileDescriptorProto"] = (
+        betterproto.message_field(15)
+    )
     """
     FileDescriptorProtos for all files in files_to_generate and everything
      they import.  The files will appear in topological order, so each file
@@ -195,9 +195,9 @@ class CodeGeneratorResponseFile(betterproto.Message):
     content: str = betterproto.string_field(15)
     """The file contents."""
 
-    generated_code_info: (
-        "betterproto_lib_pydantic_google_protobuf.GeneratedCodeInfo"
-    ) = betterproto.message_field(16)
+    generated_code_info: "betterproto_lib_pydantic_google_protobuf.GeneratedCodeInfo" = betterproto.message_field(
+        16
+    )
     """
     Information describing the file content being inserted. If an insertion
      point is used, this information will be appropriately offset and inserted

src/betterproto/lib/std/google/protobuf/__init__.py

@@ -1064,9 +1064,9 @@ class EnumDescriptorProto(betterproto.Message):
     name: str = betterproto.string_field(1)
     value: List["EnumValueDescriptorProto"] = betterproto.message_field(2)
     options: "EnumOptions" = betterproto.message_field(3)
-    reserved_range: List[
-        "EnumDescriptorProtoEnumReservedRange"
-    ] = betterproto.message_field(4)
+    reserved_range: List["EnumDescriptorProtoEnumReservedRange"] = (
+        betterproto.message_field(4)
+    )
     """
     Range of reserved numeric values. Reserved numeric values may not be used
      by enum values in the same enum declaration. Reserved ranges may not
@@ -1688,9 +1688,9 @@ class FeatureSetDefaults(betterproto.Message):
      for the closest matching edition, followed by proto merges.
     """
 
-    defaults: List[
-        "FeatureSetDefaultsFeatureSetEditionDefault"
-    ] = betterproto.message_field(1)
+    defaults: List["FeatureSetDefaultsFeatureSetEditionDefault"] = (
+        betterproto.message_field(1)
+    )
     minimum_edition: "Edition" = betterproto.enum_field(4)
     """
     The minimum supported edition (inclusive) when this was constructed.

src/betterproto/lib/std/google/protobuf/compiler/__init__.py

@@ -46,9 +46,9 @@ class CodeGeneratorRequest(betterproto.Message):
     parameter: str = betterproto.string_field(2)
     """The generator parameter passed on the command-line."""
 
-    proto_file: List[
-        "betterproto_lib_google_protobuf.FileDescriptorProto"
-    ] = betterproto.message_field(15)
+    proto_file: List["betterproto_lib_google_protobuf.FileDescriptorProto"] = (
+        betterproto.message_field(15)
+    )
     """
     FileDescriptorProtos for all files in files_to_generate and everything
      they import.  The files will appear in topological order, so each file

src/betterproto/plugin/compiler.py

@@ -1,4 +1,5 @@
 import os.path
+import subprocess
 import sys
 
 from .module_validation import ModuleValidator
@@ -6,8 +7,6 @@ from .module_validation import ModuleValidator
 
 try:
     # betterproto[compiler] specific dependencies
-    import black
-    import isort.api
     import jinja2
 except ImportError as err:
     print(
@@ -32,6 +31,7 @@ def outputfile_compiler(output_file: OutputTemplate) -> str:
         trim_blocks=True,
         lstrip_blocks=True,
         loader=jinja2.FileSystemLoader(templates_folder),
+        undefined=jinja2.StrictUndefined,
     )
     # Load the body first so we have a compleate list of imports needed.
     body_template = env.get_template("template.py.j2")
@@ -39,20 +39,17 @@ def outputfile_compiler(output_file: OutputTemplate) -> str:
 
     code = body_template.render(output_file=output_file)
     code = header_template.render(output_file=output_file) + code
-    code = isort.api.sort_code_string(
-        code=code,
-        show_diff=False,
-        py_version=37,
-        profile="black",
-        combine_as_imports=True,
-        lines_after_imports=2,
-        quiet=True,
-        force_grid_wrap=2,
-        known_third_party=["grpclib", "betterproto"],
+
+    # Sort imports, delete unused ones
+    code = subprocess.check_output(
+        ["ruff", "check", "--select", "I,F401", "--fix", "--silent", "-"],
+        input=code,
+        encoding="utf-8",
     )
-    code = black.format_str(
-        src_contents=code,
-        mode=black.Mode(),
+
+    # Format the code
+    code = subprocess.check_output(
+        ["ruff", "format", "-"], input=code, encoding="utf-8"
     )
 
     # Validate the generated code.

src/betterproto/plugin/models.py

@@ -153,11 +153,33 @@ def get_comment(
 ) -> str:
     pad = " " * indent
     for sci_loc in proto_file.source_code_info.location:
-        if list(sci_loc.path) == path and sci_loc.leading_comments:
-            lines = sci_loc.leading_comments.strip().split("\n")
+        if list(sci_loc.path) == path:
+            all_comments = list(sci_loc.leading_detached_comments)
+            if sci_loc.leading_comments:
+                all_comments.append(sci_loc.leading_comments)
+            if sci_loc.trailing_comments:
+                all_comments.append(sci_loc.trailing_comments)
+
+            lines = []
+
+            for comment in all_comments:
+                lines += comment.split("\n")
+                lines.append("")
+
+            # Remove consecutive empty lines
+            lines = [
+                line for i, line in enumerate(lines) if line or (i == 0 or lines[i - 1])
+            ]
+
+            if lines and not lines[-1]:
+                lines.pop()  # Remove the last empty line
+
+            # It is common for one line comments to start with a space, for example: // comment
+            # We don't add this space to the generated file.
+            lines = [line[1:] if line and line[0] == " " else line for line in lines]
+
             # This is a field, message, enum, service, or method
             if len(lines) == 1 and len(lines[0]) < 79 - indent - 6:
-                lines[0] = lines[0].strip('"')
                 return f'{pad}"""{lines[0]}"""'
             else:
                 joined = f"\n{pad}".join(lines)
@@ -238,7 +260,7 @@ class OutputTemplate:
     parent_request: PluginRequestCompiler
     package_proto_obj: FileDescriptorProto
     input_files: List[str] = field(default_factory=list)
-    imports: Set[str] = field(default_factory=set)
+    imports_end: Set[str] = field(default_factory=set)
     datetime_imports: Set[str] = field(default_factory=set)
     pydantic_imports: Set[str] = field(default_factory=set)
     builtins_import: bool = False
@@ -328,12 +350,6 @@ class MessageCompiler(ProtoContentBase):
     def py_name(self) -> str:
         return pythonize_class_name(self.proto_name)
 
-    @property
-    def annotation(self) -> str:
-        if self.repeated:
-            return self.typing_compiler.list(self.py_name)
-        return self.py_name
-
     @property
     def deprecated_fields(self) -> Iterator[str]:
         for f in self.fields:
@@ -484,13 +500,6 @@ class FieldCompiler(MessageCompiler):
     def optional(self) -> bool:
         return self.proto_obj.proto3_optional
 
-    @property
-    def mutable(self) -> bool:
-        """True if the field is a mutable type, otherwise False."""
-        return self.annotation.startswith(
-            ("typing.List[", "typing.Dict[", "dict[", "list[", "Dict[", "List[")
-        )
-
     @property
     def field_type(self) -> str:
         """String representation of proto field type."""
@@ -532,7 +541,7 @@ class FieldCompiler(MessageCompiler):
             # Type referencing another defined Message or a named enum
             return get_type_reference(
                 package=self.output_file.package,
-                imports=self.output_file.imports,
+                imports=self.output_file.imports_end,
                 source_type=self.proto_obj.type_name,
                 typing_compiler=self.typing_compiler,
                 pydantic=self.output_file.pydantic_dataclasses,
@@ -661,6 +670,7 @@ class EnumDefinitionCompiler(MessageCompiler):
 
 @dataclass
 class ServiceCompiler(ProtoContentBase):
+    source_file: FileDescriptorProto
     parent: OutputTemplate = PLACEHOLDER
     proto_obj: DescriptorProto = PLACEHOLDER
     path: List[int] = PLACEHOLDER
@@ -682,6 +692,7 @@ class ServiceCompiler(ProtoContentBase):
 
 @dataclass
 class ServiceMethodCompiler(ProtoContentBase):
+    source_file: FileDescriptorProto
     parent: ServiceCompiler
     proto_obj: MethodDescriptorProto
     path: List[int] = PLACEHOLDER
@@ -730,7 +741,7 @@ class ServiceMethodCompiler(ProtoContentBase):
         """
         return get_type_reference(
             package=self.output_file.package,
-            imports=self.output_file.imports,
+            imports=self.output_file.imports_end,
             source_type=self.proto_obj.input_type,
             typing_compiler=self.output_file.typing_compiler,
             unwrap=False,
@@ -760,7 +771,7 @@ class ServiceMethodCompiler(ProtoContentBase):
         """
         return get_type_reference(
             package=self.output_file.package,
-            imports=self.output_file.imports,
+            imports=self.output_file.imports_end,
             source_type=self.proto_obj.output_type,
             typing_compiler=self.output_file.typing_compiler,
             unwrap=False,

src/betterproto/plugin/parser.py

@@ -143,7 +143,7 @@ def generate_code(request: CodeGeneratorRequest) -> CodeGeneratorResponse:
     for output_package_name, output_package in request_data.output_packages.items():
         for proto_input_file in output_package.input_files:
             for index, service in enumerate(proto_input_file.service):
-                read_protobuf_service(service, index, output_package)
+                read_protobuf_service(proto_input_file, service, index, output_package)
 
     # Generate output files
     output_paths: Set[pathlib.Path] = set()
@@ -249,12 +249,21 @@ def read_protobuf_type(
 
 
 def read_protobuf_service(
-    service: ServiceDescriptorProto, index: int, output_package: OutputTemplate
+    source_file: FileDescriptorProto,
+    service: ServiceDescriptorProto,
+    index: int,
+    output_package: OutputTemplate,
 ) -> None:
     service_data = ServiceCompiler(
-        parent=output_package, proto_obj=service, path=[6, index]
+        source_file=source_file,
+        parent=output_package,
+        proto_obj=service,
+        path=[6, index],
     )
     for j, method in enumerate(service.method):
         ServiceMethodCompiler(
-            parent=service_data, proto_obj=method, path=[6, index, 2, j]
+            source_file=source_file,
+            parent=service_data,
+            proto_obj=method,
+            path=[6, index, 2, j],
         )

src/betterproto/templates/header.py.j2

@@ -2,13 +2,26 @@
 # sources: {{ ', '.join(output_file.input_filenames) }}
 # plugin: python-betterproto
 # This file has been @generated
+
+__all__ = (
+    {%- for enum in output_file.enums -%}
+        "{{ enum.py_name }}",
+    {%- endfor -%}
+    {%- for message in output_file.messages -%}
+        "{{ message.py_name }}",
+    {%- endfor -%}
+    {%- for service in output_file.services -%}
+        "{{ service.py_name }}Stub",
+        "{{ service.py_name }}Base",
+    {%- endfor -%}
+)
+
 {% for i in output_file.python_module_imports|sort %}
 import {{ i }}
 {% endfor %}
 
 {% if output_file.pydantic_dataclasses %}
 from pydantic.dataclasses import dataclass
-from pydantic.dataclasses import rebuild_dataclass
 {%- else -%}
 from dataclasses import dataclass
 {% endif %}
@@ -35,10 +48,6 @@ from betterproto.grpc.grpclib_server import ServiceBase
 import grpclib
 {% endif %}
 
-{% for i in output_file.imports|sort %}
-{{ i }}
-{% endfor %}
-
 {% if output_file.imports_type_checking_only %}
 from typing import TYPE_CHECKING
 

src/betterproto/templates/template.py.j2

@@ -77,14 +77,14 @@ class {{ service.py_name }}Stub(betterproto.ServiceStub):
             , {{ method.py_input_message_param }}: "{{ method.py_input_message_type }}"
         {%- else -%}
             {# Client streaming: need a request iterator instead #}
-            , {{ method.py_input_message_param }}_iterator: {{ output_file.typing_compiler.union(output_file.typing_compiler.async_iterable(method.py_input_message_type), output_file.typing_compiler.iterable(method.py_input_message_type)) }}
+            , {{ method.py_input_message_param }}_iterator: "{{ output_file.typing_compiler.union(output_file.typing_compiler.async_iterable(method.py_input_message_type), output_file.typing_compiler.iterable(method.py_input_message_type)) }}"
         {%- endif -%}
             ,
             *
             , timeout: {{ output_file.typing_compiler.optional("float") }} = None
             , deadline: {{ output_file.typing_compiler.optional('"Deadline"') }} = None
             , metadata: {{ output_file.typing_compiler.optional('"MetadataLike"') }} = None
-            ) -> {% if method.server_streaming %}{{ output_file.typing_compiler.async_iterator(method.py_output_message_type ) }}{% else %}"{{ method.py_output_message_type }}"{% endif %}:
+            ) -> "{% if method.server_streaming %}{{ output_file.typing_compiler.async_iterator(method.py_output_message_type ) }}{% else %}{{ method.py_output_message_type }}{% endif %}":
         {% if method.comment %}
 {{ method.comment }}
 
@@ -143,6 +143,10 @@ class {{ service.py_name }}Stub(betterproto.ServiceStub):
     {% endfor %}
 {% endfor %}
 
+{% for i in output_file.imports_end %}
+{{ i }}
+{% endfor %}
+
 {% for service in output_file.services %}
 class {{ service.py_name }}Base(ServiceBase):
     {% if service.comment %}
@@ -211,11 +215,3 @@ class {{ service.py_name }}Base(ServiceBase):
         }
 
 {% endfor %}
-
-{% if output_file.pydantic_dataclasses %}
-{% for message in output_file.messages %}
-{% if message.has_message_field %}
-rebuild_dataclass({{ message.py_name }})  # type: ignore
-{% endif %}
-{% endfor %}
-{% endif %}

tests/conftest.py

@@ -4,17 +4,6 @@ import sys
 import pytest
 
 
-def pytest_addoption(parser):
-    parser.addoption(
-        "--repeat", type=int, default=1, help="repeat the operation multiple times"
-    )
-
-
-@pytest.fixture(scope="session")
-def repeat(request):
-    return request.config.getoption("repeat")
-
-
 @pytest.fixture
 def reset_sys_path():
     original = copy.deepcopy(sys.path)

tests/grpc/test_grpclib_client.py

@@ -1,5 +1,4 @@
 import asyncio
-import sys
 import uuid
 
 import grpclib
@@ -27,12 +26,12 @@ async def _test_client(client: ThingServiceClient, name="clean room", **kwargs):
 
 def _assert_request_meta_received(deadline, metadata):
     def server_side_test(stream):
-        assert stream.deadline._timestamp == pytest.approx(
-            deadline._timestamp, 1
-        ), "The provided deadline should be received serverside"
-        assert (
-            stream.metadata["authorization"] == metadata["authorization"]
-        ), "The provided authorization metadata should be received serverside"
+        assert stream.deadline._timestamp == pytest.approx(deadline._timestamp, 1), (
+            "The provided deadline should be received serverside"
+        )
+        assert stream.metadata["authorization"] == metadata["authorization"], (
+            "The provided authorization metadata should be received serverside"
+        )
 
     return server_side_test
 
@@ -91,9 +90,6 @@ async def test_trailer_only_error_stream_unary(
 
 
 @pytest.mark.asyncio
-@pytest.mark.skipif(
-    sys.version_info < (3, 8), reason="async mock spy does works for python3.8+"
-)
 async def test_service_call_mutable_defaults(mocker):
     async with ChannelFor([ThingService()]) as channel:
         client = ThingServiceClient(channel)
@@ -269,9 +265,9 @@ async def test_async_gen_for_stream_stream_request():
             else:
                 # No more things to send make sure channel is closed
                 request_chan.close()
-        assert response_index == len(
-            expected_things
-        ), "Didn't receive all expected responses"
+        assert response_index == len(expected_things), (
+            "Didn't receive all expected responses"
+        )
 
 
 @pytest.mark.asyncio

tests/inputs/casing/test_casing.py

@@ -4,20 +4,20 @@ from tests.output_betterproto.casing import Test
 
 def test_message_attributes():
     message = Test()
-    assert hasattr(
-        message, "snake_case_message"
-    ), "snake_case field name is same in python"
+    assert hasattr(message, "snake_case_message"), (
+        "snake_case field name is same in python"
+    )
     assert hasattr(message, "camel_case"), "CamelCase field is snake_case in python"
     assert hasattr(message, "uppercase"), "UPPERCASE field is lowercase in python"
 
 
 def test_message_casing():
-    assert hasattr(
-        casing, "SnakeCaseMessage"
-    ), "snake_case Message name is converted to CamelCase in python"
+    assert hasattr(casing, "SnakeCaseMessage"), (
+        "snake_case Message name is converted to CamelCase in python"
+    )
 
 
 def test_enum_casing():
-    assert hasattr(
-        casing, "MyEnum"
-    ), "snake_case Enum name is converted to CamelCase in python"
+    assert hasattr(casing, "MyEnum"), (
+        "snake_case Enum name is converted to CamelCase in python"
+    )

tests/inputs/casing_inner_class/test_casing_inner_class.py

@@ -2,13 +2,13 @@ import tests.output_betterproto.casing_inner_class as casing_inner_class
 
 
 def test_message_casing_inner_class_name():
-    assert hasattr(
-        casing_inner_class, "TestInnerClass"
-    ), "Inline defined Message is correctly converted to CamelCase"
+    assert hasattr(casing_inner_class, "TestInnerClass"), (
+        "Inline defined Message is correctly converted to CamelCase"
+    )
 
 
 def test_message_casing_inner_class_attributes():
     message = casing_inner_class.Test()
-    assert hasattr(
-        message.inner, "old_exp"
-    ), "Inline defined Message attribute is snake_case"
+    assert hasattr(message.inner, "old_exp"), (
+        "Inline defined Message attribute is snake_case"
+    )

tests/inputs/casing_message_field_uppercase/casing_message_field_uppercase.py

@@ -3,12 +3,12 @@ from tests.output_betterproto.casing_message_field_uppercase import Test
 
 def test_message_casing():
     message = Test()
-    assert hasattr(
-        message, "uppercase"
-    ), "UPPERCASE attribute is converted to 'uppercase' in python"
-    assert hasattr(
-        message, "uppercase_v2"
-    ), "UPPERCASE_V2 attribute is converted to 'uppercase_v2' in python"
-    assert hasattr(
-        message, "upper_camel_case"
-    ), "UPPER_CAMEL_CASE attribute is converted to upper_camel_case in python"
+    assert hasattr(message, "uppercase"), (
+        "UPPERCASE attribute is converted to 'uppercase' in python"
+    )
+    assert hasattr(message, "uppercase_v2"), (
+        "UPPERCASE_V2 attribute is converted to 'uppercase_v2' in python"
+    )
+    assert hasattr(message, "upper_camel_case"), (
+        "UPPER_CAMEL_CASE attribute is converted to upper_camel_case in python"
+    )

tests/inputs/documentation/documentation.proto

@@ -0,0 +1,44 @@
+syntax = "proto3";
+package documentation;
+
+// Documentation of message 1
+// other line 1
+
+// Documentation of message 2
+// other line 2
+message Test { // Documentation of message 3
+    // Documentation of field 1
+    // other line 1
+
+    // Documentation of field 2
+    // other line 2
+    uint32 x = 1; // Documentation of field 3
+}
+
+// Documentation of enum 1
+// other line 1
+
+// Documentation of enum 2
+// other line 2
+enum Enum { // Documentation of enum 3
+    // Documentation of variant 1
+    // other line 1
+
+    // Documentation of variant 2
+    // other line 2
+    Enum_Variant = 0; // Documentation of variant 3
+}
+
+// Documentation of service 1
+// other line 1
+
+// Documentation of service 2
+// other line 2
+service Service { // Documentation of service 3
+    // Documentation of method 1
+    // other line 1
+
+    // Documentation of method 2
+    // other line 2
+    rpc get(Test) returns (Test); // Documentation of method 3
+}

tests/inputs/enum/test_enum.py

@@ -27,9 +27,9 @@ def test_enum_is_comparable_with_int():
 
 
 def test_enum_to_dict():
-    assert (
-        "choice" not in Test(choice=Choice.ZERO).to_dict()
-    ), "Default enum value is not serialized"
+    assert "choice" not in Test(choice=Choice.ZERO).to_dict(), (
+        "Default enum value is not serialized"
+    )
     assert (
         Test(choice=Choice.ZERO).to_dict(include_default_values=True)["choice"]
         == "ZERO"

tests/inputs/import_circular_dependency/import_circular_dependency.proto

@@ -26,5 +26,5 @@ import "other.proto";
 //           (root: Test & RootPackageMessage) <-------> (other: OtherPackageMessage)
 message Test {
   RootPackageMessage message = 1;
-  other.OtherPackageMessage other = 2;
+  other.OtherPackageMessage other_value = 2;
 }

tests/test_all_definition.py

@@ -0,0 +1,19 @@
+def test_all_definition():
+    """
+    Check that a compiled module defines __all__ with the right value.
+
+    These modules have been chosen since they contain messages, services and enums.
+    """
+    import tests.output_betterproto.enum as enum
+    import tests.output_betterproto.service as service
+
+    assert service.__all__ == (
+        "ThingType",
+        "DoThingRequest",
+        "DoThingResponse",
+        "GetThingRequest",
+        "GetThingResponse",
+        "TestStub",
+        "TestBase",
+    )
+    assert enum.__all__ == ("Choice", "ArithmeticOperator", "Test")

tests/test_documentation.py

@@ -0,0 +1,37 @@
+import ast
+import inspect
+
+
+def check(generated_doc: str, type: str) -> None:
+    assert f"Documentation of {type} 1" in generated_doc
+    assert "other line 1" in generated_doc
+    assert f"Documentation of {type} 2" in generated_doc
+    assert "other line 2" in generated_doc
+    assert f"Documentation of {type} 3" in generated_doc
+
+
+def test_documentation() -> None:
+    from .output_betterproto.documentation import (
+        Enum,
+        ServiceBase,
+        ServiceStub,
+        Test,
+    )
+
+    check(Test.__doc__, "message")
+
+    source = inspect.getsource(Test)
+    tree = ast.parse(source)
+    check(tree.body[0].body[2].value.value, "field")
+
+    check(Enum.__doc__, "enum")
+
+    source = inspect.getsource(Enum)
+    tree = ast.parse(source)
+    check(tree.body[0].body[2].value.value, "variant")
+
+    check(ServiceBase.__doc__, "service")
+    check(ServiceBase.get.__doc__, "method")
+
+    check(ServiceStub.__doc__, "service")
+    check(ServiceStub.get.__doc__, "method")

tests/test_features.py

@@ -621,9 +621,7 @@ iso_candidates = """2009-12-12T12:34
 2010-02-18T16:00:00.23334444
 2010-02-18T16:00:00,2283
 2009-05-19 143922
-2009-05-19 1439""".split(
-    "\n"
-)
+2009-05-19 1439""".split("\n")
 
 
 def test_iso_datetime():

tests/test_get_ref_type.py

@@ -56,9 +56,9 @@ def test_reference_google_wellknown_types_non_wrappers(
     )
 
     assert name == expected_name
-    assert imports.__contains__(
-        expected_import
-    ), f"{expected_import} not found in {imports}"
+    assert imports.__contains__(expected_import), (
+        f"{expected_import} not found in {imports}"
+    )
 
 
 @pytest.mark.parametrize(
@@ -102,9 +102,9 @@ def test_reference_google_wellknown_types_non_wrappers_pydantic(
     )
 
     assert name == expected_name
-    assert imports.__contains__(
-        expected_import
-    ), f"{expected_import} not found in {imports}"
+    assert imports.__contains__(expected_import), (
+        f"{expected_import} not found in {imports}"
+    )
 
 
 @pytest.mark.parametrize(

tests/test_inputs.py

@@ -174,22 +174,21 @@ def test_message_equality(test_data: TestData) -> None:
 
 
 @pytest.mark.parametrize("test_data", test_cases.messages_with_json, indirect=True)
-def test_message_json(repeat, test_data: TestData) -> None:
+def test_message_json(test_data: TestData) -> None:
     plugin_module, _, json_data = test_data
 
-    for _ in range(repeat):
-        for sample in json_data:
-            if sample.belongs_to(test_input_config.non_symmetrical_json):
-                continue
+    for sample in json_data:
+        if sample.belongs_to(test_input_config.non_symmetrical_json):
+            continue
 
-            message: betterproto.Message = plugin_module.Test()
+        message: betterproto.Message = plugin_module.Test()
 
-            message.from_json(sample.json)
-            message_json = message.to_json(0)
+        message.from_json(sample.json)
+        message_json = message.to_json(0)
 
-            assert dict_replace_nans(json.loads(message_json)) == dict_replace_nans(
-                json.loads(sample.json)
-            )
+        assert dict_replace_nans(json.loads(message_json)) == dict_replace_nans(
+            json.loads(sample.json)
+        )
 
 
 @pytest.mark.parametrize("test_data", test_cases.services, indirect=True)
@@ -198,28 +197,27 @@ def test_service_can_be_instantiated(test_data: TestData) -> None:
 
 
 @pytest.mark.parametrize("test_data", test_cases.messages_with_json, indirect=True)
-def test_binary_compatibility(repeat, test_data: TestData) -> None:
+def test_binary_compatibility(test_data: TestData) -> None:
     plugin_module, reference_module, json_data = test_data
 
     for sample in json_data:
         reference_instance = Parse(sample.json, reference_module().Test())
         reference_binary_output = reference_instance.SerializeToString()
 
-        for _ in range(repeat):
-            plugin_instance_from_json: betterproto.Message = (
-                plugin_module.Test().from_json(sample.json)
-            )
-            plugin_instance_from_binary = plugin_module.Test.FromString(
-                reference_binary_output
-            )
+        plugin_instance_from_json: betterproto.Message = plugin_module.Test().from_json(
+            sample.json
+        )
+        plugin_instance_from_binary = plugin_module.Test.FromString(
+            reference_binary_output
+        )
 
-            # Generally this can't be relied on, but here we are aiming to match the
-            # existing Python implementation and aren't doing anything tricky.
-            # https://developers.google.com/protocol-buffers/docs/encoding#implications
-            assert bytes(plugin_instance_from_json) == reference_binary_output
-            assert bytes(plugin_instance_from_binary) == reference_binary_output
+        # Generally this can't be relied on, but here we are aiming to match the
+        # existing Python implementation and aren't doing anything tricky.
+        # https://developers.google.com/protocol-buffers/docs/encoding#implications
+        assert bytes(plugin_instance_from_json) == reference_binary_output
+        assert bytes(plugin_instance_from_binary) == reference_binary_output
 
-            assert plugin_instance_from_json == plugin_instance_from_binary
-            assert dict_replace_nans(
-                plugin_instance_from_json.to_dict()
-            ) == dict_replace_nans(plugin_instance_from_binary.to_dict())
+        assert plugin_instance_from_json == plugin_instance_from_binary
+        assert dict_replace_nans(
+            plugin_instance_from_json.to_dict()
+        ) == dict_replace_nans(plugin_instance_from_binary.to_dict())

tests/test_pickling.py

@@ -57,6 +57,11 @@ class Complex(betterproto.Message):
     )
 
 
+class BetterprotoEnum(betterproto.Enum):
+    UNSPECIFIED = 0
+    ONE = 1
+
+
 def complex_msg():
     return Complex(
         foo_str="yep",
@@ -201,3 +206,11 @@ def test_message_can_be_cached():
             .string_value
             == "world"
         )
+
+
+def test_pickle_enum():
+    enum = BetterprotoEnum.ONE
+    assert unpickled(enum) == enum
+
+    enum = BetterprotoEnum.UNSPECIFIED
+    assert unpickled(enum) == enum

tests/test_streams.py

@@ -62,7 +62,7 @@ def test_load_varint_file():
         stream.read(2)  # Skip until first multi-byte
         assert betterproto.load_varint(stream) == (
             123456789,
-            b"\x95\x9A\xEF\x3A",
+            b"\x95\x9a\xef\x3a",
         )  # Multi-byte varint
 
 
@@ -338,7 +338,7 @@ def run_java_single_varint(value: int, tmp_path) -> int:
 
 def test_single_varint(compile_jar, tmp_path):
     single_byte = (1, b"\x01")
-    multi_byte = (123456789, b"\x95\x9A\xEF\x3A")
+    multi_byte = (123456789, b"\x95\x9a\xef\x3a")
 
     # Write a single-byte varint to a file and have Java read it back
     returned = run_java_single_varint(single_byte[0], tmp_path)
@@ -351,8 +351,8 @@ def test_single_varint(compile_jar, tmp_path):
 
 def test_multiple_varints(compile_jar, tmp_path):
     single_byte = (1, b"\x01")
-    multi_byte = (123456789, b"\x95\x9A\xEF\x3A")
-    over32 = (3000000000, b"\x80\xBC\xC1\x96\x0B")
+    multi_byte = (123456789, b"\x95\x9a\xef\x3a")
+    over32 = (3000000000, b"\x80\xbc\xc1\x96\x0b")
 
     # Write two varints to the same file
     with open(tmp_path / "py_multiple_varints.out", "wb") as stream:

tests/test_version.py

@@ -11,6 +11,6 @@ PROJECT_TOML = Path(__file__).joinpath("..", "..", "pyproject.toml").resolve()
 def test_version():
     with PROJECT_TOML.open() as toml_file:
         project_config = tomlkit.loads(toml_file.read())
-    assert (
-        __version__ == project_config["tool"]["poetry"]["version"]
-    ), "Project version should match in package and package config"
+    assert __version__ == project_config["project"]["version"], (
+        "Project version should match in package and package config"
+    )