Create binary wheels with mypyc (#242)

Co-authored-by: Ofek Lev <[email protected]>

.github/workflows/tests.yaml

@@ -7,6 +7,11 @@ on:
   pull_request:
     branches: [ master ]
 
+env:
+  CIBW_TEST_COMMAND: python -m unittest discover --start-directory {project}
+  CIBW_SKIP: pp*
+  CIBW_ENVIRONMENT_PASS_LINUX: TOMLI_USE_MYPYC
+
 jobs:
 
   linters:
@@ -74,12 +79,102 @@ jobs:
       with:
         token: ${{ secrets.CODECOV_TOKEN }}
 
+  binary-wheels-standard:
+    name: Binary wheels for ${{ startsWith(matrix.os, 'macos-') && 'macOS' || startsWith(matrix.os, 'windows-') && 'Windows' || 'Linux' }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-latest, windows-latest, macos-latest]
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - uses: actions/setup-python@v5
+      with:
+        python-version: '3.x'
+
+    - name: Switch build backend to setuptools
+      run: |
+        pip install -r scripts/requirements.txt
+        python scripts/use_setuptools.py
+
+    - name: Build wheels
+      uses: pypa/[email protected]
+      env:
+        CIBW_ARCHS_MACOS: x86_64 arm64
+        TOMLI_USE_MYPYC: '1'
+
+    - uses: actions/upload-artifact@v4
+      with:
+        name: artifact-standard-${{ matrix.os }}
+        path: wheelhouse/*.whl
+        if-no-files-found: error
+
+  pure-python-wheel-and-sdist:
+    name: Build a pure Python wheel and source distribution
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Install build dependencies
+      run: pip install build
+
+    - name: Build
+      run: python -m build
+
+    - uses: actions/upload-artifact@v4
+      with:
+        name: artifact-pure-python
+        path: dist/*
+        if-no-files-found: error
+
+  binary-wheels-arm:
+    name: Build Linux wheels for ARM
+    runs-on: ubuntu-latest
+    # Very slow (~ 1 hour), no need to run on PRs
+    if: >
+      github.event_name == 'push'
+      &&
+      (github.ref == 'refs/heads/master' || startsWith(github.event.ref, 'refs/tags'))
+    steps:
+    - uses: actions/checkout@v4
+
+    - uses: actions/setup-python@v5
+      with:
+        python-version: '3.x'
+
+    - name: Switch build backend to setuptools
+      run: |
+        pip install -r scripts/requirements.txt
+        python scripts/use_setuptools.py
+
+    - name: Set up QEMU
+      uses: docker/setup-qemu-action@v3
+      with:
+        platforms: arm64
+
+    - name: Build wheels
+      uses: pypa/[email protected]
+      env:
+        CIBW_ARCHS_LINUX: aarch64
+        TOMLI_USE_MYPYC: '1'
+
+    - uses: actions/upload-artifact@v4
+      with:
+        name: artifact-arm-linux
+        path: wheelhouse/*.whl
+        if-no-files-found: error
+
   allgood:
     runs-on: ubuntu-latest
     needs:
     - tests
     - coverage
     - linters
+    - binary-wheels-standard
+    - pure-python-wheel-and-sdist
+    - binary-wheels-arm
     steps:
     - run: echo "Great success!"
 
@@ -89,19 +184,20 @@ jobs:
     if: github.event_name == 'push' && startsWith(github.event.ref, 'refs/tags')
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/download-artifact@v4
+      with:
+        path: dist
+        pattern: artifact-*
+        merge-multiple: true
     - uses: actions/setup-python@v5
       with:
         python-version: '3.x'
-    - name: Install build and publish tools
+    - name: Install twine
       run: |
-        pip install build twine
-    - name: Build and check
+        pip install twine
+    - name: Check and publish
       run: |
-        rm -rf dist/ && python -m build
         twine check --strict dist/*
-    - name: Publish
-      run: |
         twine upload dist/*
       env:
         TWINE_USERNAME: __token__

CHANGELOG.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## 2.2.0
+
+- Added
+  - mypyc generated binary wheels for common platforms
+
 ## 2.1.0
 
 - Deprecated

MANIFEST.in

@@ -0,0 +1 @@
+prune tests/

pyproject.toml

@@ -50,6 +50,7 @@ env_list = ["3.8", "3.9", "3.10", "3.11", "3.12", "3.13"]
 
 [tool.tox.env_run_base]
 description = "run tests against a built package under {base_python}"
+pass_env = ["TOMLI_USE_MYPYC"]
 commands = [
     ["python", "-m", "unittest", { replace = "posargs", extend = true }],
 ]

scripts/requirements.txt

@@ -0,0 +1 @@
+tomli-w

scripts/use_setuptools.py

@@ -0,0 +1,12 @@
+from pathlib import Path
+import tomllib
+
+import tomli_w  # type: ignore[import-not-found]
+
+pyproject_path = Path(__file__).parent.parent / "pyproject.toml"
+data = tomllib.loads(pyproject_path.read_bytes().decode())
+data["build-system"] = {
+    "requires": ["setuptools>=69", "mypy[mypyc]>=1.13"],
+    "build-backend": "setuptools.build_meta",
+}
+pyproject_path.write_bytes(tomli_w.dumps(data).encode())

setup.py

@@ -0,0 +1,15 @@
+import os
+
+from setuptools import setup  # type: ignore[import-untyped]
+
+if os.environ.get("TOMLI_USE_MYPYC") == "1":
+    import glob
+
+    from mypyc.build import mypycify  # type: ignore[import-untyped]
+
+    files = glob.glob("src/**/*.py", recursive=True)
+    ext_modules = mypycify(files)
+else:
+    ext_modules = []
+
+setup(ext_modules=ext_modules)

src/tomli/__init__.py

@@ -6,6 +6,3 @@
 __version__ = "2.1.0"  # DO NOT EDIT THIS LINE MANUALLY. LET bump2version UTILITY DO IT
 
 from ._parser import TOMLDecodeError, load, loads
-
-# Pretend this exception was created here.
-TOMLDecodeError.__module__ = __name__

src/tomli/_parser.py

@@ -6,8 +6,9 @@
 
 from collections.abc import Iterable
 import string
+import sys
 from types import MappingProxyType
-from typing import IO, Any, NamedTuple
+from typing import IO, Any, Final, NamedTuple
 import warnings
 
 from ._re import (
@@ -20,6 +21,17 @@
 )
 from ._types import Key, ParseFloat, Pos
 
+# Inline tables/arrays are implemented using recursion. Pathologically
+# nested documents cause pure Python to raise RecursionError (which is OK),
+# but mypyc binary wheels will crash unrecoverably (not OK). According to
+# mypyc docs this will be fixed in the future:
+# https://mypyc.readthedocs.io/en/latest/differences_from_python.html#stack-overflows
+# Before mypyc's fix is in, recursion needs to be limited by this library.
+# Choosing `sys.getrecursionlimit()` as maximum inline table/array nesting
+# level, as it allows more nesting than pure Python, but still seems a far
+# lower number than where mypyc binaries crash.
+MAX_INLINE_NESTING: Final = sys.getrecursionlimit()
+
 ASCII_CTRL = frozenset(chr(i) for i in range(32)) | frozenset(chr(127))
 
 # Neither of these sets include quotation mark or backslash. They are
@@ -69,9 +81,9 @@ class TOMLDecodeError(ValueError):
 
     def __init__(
         self,
-        msg: str = DEPRECATED_DEFAULT,  # type: ignore[assignment]
-        doc: str = DEPRECATED_DEFAULT,  # type: ignore[assignment]
-        pos: Pos = DEPRECATED_DEFAULT,  # type: ignore[assignment]
+        msg: str | type[DEPRECATED_DEFAULT] = DEPRECATED_DEFAULT,
+        doc: str | type[DEPRECATED_DEFAULT] = DEPRECATED_DEFAULT,
+        pos: Pos | type[DEPRECATED_DEFAULT] = DEPRECATED_DEFAULT,
         *args: Any,
     ):
         if (
@@ -86,11 +98,11 @@ def __init__(
                 DeprecationWarning,
                 stacklevel=2,
             )
-            if pos is not DEPRECATED_DEFAULT:  # type: ignore[comparison-overlap]
+            if pos is not DEPRECATED_DEFAULT:
                 args = pos, *args
-            if doc is not DEPRECATED_DEFAULT:  # type: ignore[comparison-overlap]
+            if doc is not DEPRECATED_DEFAULT:
                 args = doc, *args
-            if msg is not DEPRECATED_DEFAULT:  # type: ignore[comparison-overlap]
+            if msg is not DEPRECATED_DEFAULT:
                 args = msg, *args
             ValueError.__init__(self, *args)
             return
@@ -202,10 +214,10 @@ class Flags:
     """Flags that map to parsed keys/namespaces."""
 
     # Marks an immutable namespace (inline array or inline table).
-    FROZEN = 0
+    FROZEN: Final = 0
     # Marks a nest that has been explicitly created and can no longer
     # be opened using the "[table]" syntax.
-    EXPLICIT_NEST = 1
+    EXPLICIT_NEST: Final = 1
 
     def __init__(self) -> None:
         self._flags: dict[str, dict] = {}
@@ -251,8 +263,8 @@ def is_(self, key: Key, flag: int) -> bool:
             cont = inner_cont["nested"]
         key_stem = key[-1]
         if key_stem in cont:
-            cont = cont[key_stem]
-            return flag in cont["flags"] or flag in cont["recursive_flags"]
+            inner_cont = cont[key_stem]
+            return flag in inner_cont["flags"] or flag in inner_cont["recursive_flags"]
         return False
 
 
@@ -393,7 +405,7 @@ def create_list_rule(src: str, pos: Pos, out: Output) -> tuple[Pos, Key]:
 def key_value_rule(
     src: str, pos: Pos, out: Output, header: Key, parse_float: ParseFloat
 ) -> Pos:
-    pos, key, value = parse_key_value_pair(src, pos, parse_float)
+    pos, key, value = parse_key_value_pair(src, pos, parse_float, nest_lvl=0)
     key_parent, key_stem = key[:-1], key[-1]
     abs_key_parent = header + key_parent
 
@@ -425,7 +437,7 @@ def key_value_rule(
 
 
 def parse_key_value_pair(
-    src: str, pos: Pos, parse_float: ParseFloat
+    src: str, pos: Pos, parse_float: ParseFloat, nest_lvl: int
 ) -> tuple[Pos, Key, Any]:
     pos, key = parse_key(src, pos)
     try:
@@ -436,7 +448,7 @@ def parse_key_value_pair(
         raise TOMLDecodeError("Expected '=' after a key in a key/value pair", src, pos)
     pos += 1
     pos = skip_chars(src, pos, TOML_WS)
-    pos, value = parse_value(src, pos, parse_float)
+    pos, value = parse_value(src, pos, parse_float, nest_lvl)
     return pos, key, value
 
 
@@ -479,15 +491,17 @@ def parse_one_line_basic_str(src: str, pos: Pos) -> tuple[Pos, str]:
     return parse_basic_str(src, pos, multiline=False)
 
 
-def parse_array(src: str, pos: Pos, parse_float: ParseFloat) -> tuple[Pos, list]:
+def parse_array(
+    src: str, pos: Pos, parse_float: ParseFloat, nest_lvl: int
+) -> tuple[Pos, list]:
     pos += 1
     array: list = []
 
     pos = skip_comments_and_array_ws(src, pos)
     if src.startswith("]", pos):
         return pos + 1, array
     while True:
-        pos, val = parse_value(src, pos, parse_float)
+        pos, val = parse_value(src, pos, parse_float, nest_lvl)
         array.append(val)
         pos = skip_comments_and_array_ws(src, pos)
 
@@ -503,7 +517,9 @@ def parse_array(src: str, pos: Pos, parse_float: ParseFloat) -> tuple[Pos, list]
             return pos + 1, array
 
 
-def parse_inline_table(src: str, pos: Pos, parse_float: ParseFloat) -> tuple[Pos, dict]:
+def parse_inline_table(
+    src: str, pos: Pos, parse_float: ParseFloat, nest_lvl: int
+) -> tuple[Pos, dict]:
     pos += 1
     nested_dict = NestedDict()
     flags = Flags()
@@ -512,7 +528,7 @@ def parse_inline_table(src: str, pos: Pos, parse_float: ParseFloat) -> tuple[Pos
     if src.startswith("}", pos):
         return pos + 1, nested_dict.dict
     while True:
-        pos, key, value = parse_key_value_pair(src, pos, parse_float)
+        pos, key, value = parse_key_value_pair(src, pos, parse_float, nest_lvl)
         key_parent, key_stem = key[:-1], key[-1]
         if flags.is_(key, Flags.FROZEN):
             raise TOMLDecodeError(f"Cannot mutate immutable namespace {key}", src, pos)
@@ -654,8 +670,16 @@ def parse_basic_str(src: str, pos: Pos, *, multiline: bool) -> tuple[Pos, str]:
 
 
 def parse_value(  # noqa: C901
-    src: str, pos: Pos, parse_float: ParseFloat
+    src: str, pos: Pos, parse_float: ParseFloat, nest_lvl: int
 ) -> tuple[Pos, Any]:
+    if nest_lvl > MAX_INLINE_NESTING:
+        # Pure Python should have raised RecursionError already.
+        # This ensures mypyc binaries eventually do the same.
+        raise RecursionError(  # pragma: no cover
+            "TOML inline arrays/tables are nested more than the allowed"
+            f" {MAX_INLINE_NESTING} levels"
+        )
+
     try:
         char: str | None = src[pos]
     except IndexError:
@@ -685,11 +709,11 @@ def parse_value(  # noqa: C901
 
     # Arrays
     if char == "[":
-        return parse_array(src, pos, parse_float)
+        return parse_array(src, pos, parse_float, nest_lvl + 1)
 
     # Inline tables
     if char == "{":
-        return parse_inline_table(src, pos, parse_float)
+        return parse_inline_table(src, pos, parse_float, nest_lvl + 1)
 
     # Dates and times
     datetime_match = RE_DATETIME.match(src, pos)