Use dataclass for typing Extension

Author: abravalheri

distutils/core.py

@@ -25,6 +25,7 @@
     DistutilsSetupError,
 )
 from .extension import Extension
+from .extension import _safe as extension_keywords  # noqa  # backwards compatibility
 
 __all__ = ['Distribution', 'Command', 'Extension', 'setup']
 
@@ -74,25 +75,6 @@ def gen_usage(script_name):
     'obsoletes',
 )
 
-# Legal keyword arguments for the Extension constructor
-extension_keywords = (
-    'name',
-    'sources',
-    'include_dirs',
-    'define_macros',
-    'undef_macros',
-    'library_dirs',
-    'libraries',
-    'runtime_library_dirs',
-    'extra_objects',
-    'extra_compile_args',
-    'extra_link_args',
-    'swig_opts',
-    'export_symbols',
-    'depends',
-    'language',
-)
-
 
 def setup(**attrs):  # noqa: C901
     """The gateway to the Distutils: do everything your setup script needs

distutils/extension.py

@@ -8,6 +8,8 @@
 import os
 import warnings
 from collections.abc import Iterable
+from dataclasses import dataclass, field, fields
+from typing import TYPE_CHECKING
 
 # This class is really only used by the "build_ext" command, so it might
 # make sense to put it in distutils.command.build_ext.  However, that
@@ -20,137 +22,159 @@
 # order to do anything.
 
 
-class Extension:
+@dataclass
+class _Extension:
     """Just a collection of attributes that describes an extension
     module and everything needed to build it (hopefully in a portable
     way, but there are hooks that let you be as unportable as you need).
+    """
+
+    # The use of a parent class as a "trick":
+    # - We need to modify __init__ so to achieve backwards compatibility
+    # - But we don't want to throw away the dataclass-generated __init__
+    # - We also want to fool the typechecker to consider the same type
+    #   signature as the dataclass-generated __init__
+
+    name: str
+    """
+    the full name of the extension, including any packages -- ie.
+    *not* a filename or pathname, but Python dotted name
+    """
+
+    sources: Iterable[str | os.PathLike[str]]
+    """
+    iterable of source filenames (except strings, which could be misinterpreted
+    as a single filename), relative to the distribution root (where the setup
+    script lives), in Unix form (slash-separated) for portability. Can be any
+    non-string iterable (list, tuple, set, etc.) containing strings or
+    PathLike objects. Source files may be C, C++, SWIG (.i), platform-specific
+    resource files, or whatever else is recognized by the "build_ext" command
+    as source for a Python extension.
+    """
+
+    include_dirs: list[str] = field(default_factory=list)
+    """
+    list of directories to search for C/C++ header files (in Unix
+    form for portability)
+    """
+
+    define_macros: list[tuple[str, str | None]] = field(default_factory=list)
+    """
+    list of macros to define; each macro is defined using a 2-tuple,
+    where 'value' is either the string to define it to or None to
+    define it without a particular value (equivalent of "#define
+    FOO" in source or -DFOO on Unix C compiler command line)
+    """
+
+    undef_macros: list[str] = field(default_factory=list)
+    """list of macros to undefine explicitly"""
+
+    library_dirs: list[str] = field(default_factory=list)
+    """list of directories to search for C/C++ libraries at link time"""
+
+    libraries: list[str] = field(default_factory=list)
+    """list of library names (not filenames or paths) to link against"""
+
+    runtime_library_dirs: list[str] = field(default_factory=list)
+    """
+    list of directories to search for C/C++ libraries at run time
+    (for shared extensions, this is when the extension is loaded)
+    """
+
+    extra_objects: list[str] = field(default_factory=list)
+    """
+    list of extra files to link with (eg. object files not implied
+    by 'sources', static library that must be explicitly specified,
+    binary resource files, etc.)
+    """
+
+    extra_compile_args: list[str] = field(default_factory=list)
+    """
+    any extra platform- and compiler-specific information to use
+    when compiling the source files in 'sources'.  For platforms and
+    compilers where "command line" makes sense, this is typically a
+    list of command-line arguments, but for other platforms it could
+    be anything.
+    """
+
+    extra_link_args: list[str] = field(default_factory=list)
+    """
+    any extra platform- and compiler-specific information to use
+    when linking object files together to create the extension (or
+    to create a new static Python interpreter).  Similar
+    interpretation as for 'extra_compile_args'.
+    """
+
+    export_symbols: list[str] = field(default_factory=list)
+    """
+    list of symbols to be exported from a shared extension.  Not
+    used on all platforms, and not generally necessary for Python
+    extensions, which typically export exactly one symbol: "init" +
+    extension_name.
+    """
 
-    Instance attributes:
-      name : string
-        the full name of the extension, including any packages -- ie.
-        *not* a filename or pathname, but Python dotted name
-      sources : Iterable[string | os.PathLike]
-        iterable of source filenames (except strings, which could be misinterpreted
-        as a single filename), relative to the distribution root (where the setup
-        script lives), in Unix form (slash-separated) for portability. Can be any
-        non-string iterable (list, tuple, set, etc.) containing strings or
-        PathLike objects. Source files may be C, C++, SWIG (.i), platform-specific
-        resource files, or whatever else is recognized by the "build_ext" command
-        as source for a Python extension.
-      include_dirs : [string]
-        list of directories to search for C/C++ header files (in Unix
-        form for portability)
-      define_macros : [(name : string, value : string|None)]
-        list of macros to define; each macro is defined using a 2-tuple,
-        where 'value' is either the string to define it to or None to
-        define it without a particular value (equivalent of "#define
-        FOO" in source or -DFOO on Unix C compiler command line)
-      undef_macros : [string]
-        list of macros to undefine explicitly
-      library_dirs : [string]
-        list of directories to search for C/C++ libraries at link time
-      libraries : [string]
-        list of library names (not filenames or paths) to link against
-      runtime_library_dirs : [string]
-        list of directories to search for C/C++ libraries at run time
-        (for shared extensions, this is when the extension is loaded)
-      extra_objects : [string]
-        list of extra files to link with (eg. object files not implied
-        by 'sources', static library that must be explicitly specified,
-        binary resource files, etc.)
-      extra_compile_args : [string]
-        any extra platform- and compiler-specific information to use
-        when compiling the source files in 'sources'.  For platforms and
-        compilers where "command line" makes sense, this is typically a
-        list of command-line arguments, but for other platforms it could
-        be anything.
-      extra_link_args : [string]
-        any extra platform- and compiler-specific information to use
-        when linking object files together to create the extension (or
-        to create a new static Python interpreter).  Similar
-        interpretation as for 'extra_compile_args'.
-      export_symbols : [string]
-        list of symbols to be exported from a shared extension.  Not
-        used on all platforms, and not generally necessary for Python
-        extensions, which typically export exactly one symbol: "init" +
-        extension_name.
-      swig_opts : [string]
-        any extra options to pass to SWIG if a source file has the .i
-        extension.
-      depends : [string]
-        list of files that the extension depends on
-      language : string
-        extension language (i.e. "c", "c++", "objc"). Will be detected
-        from the source extensions if not provided.
-      optional : boolean
-        specifies that a build failure in the extension should not abort the
-        build process, but simply not install the failing extension.
+    swig_opts: list[str] = field(default_factory=list)
     """
+    any extra options to pass to SWIG if a source file has the .i
+    extension.
+    """
+
+    depends: list[str] = field(default_factory=list)
+    """list of files that the extension depends on"""
+
+    language: str | None = None
+    """
+    extension language (i.e. "c", "c++", "objc"). Will be detected
+    from the source extensions if not provided.
+    """
+
+    optional: bool = False
+    """
+    specifies that a build failure in the extension should not abort the
+    build process, but simply not install the failing extension.
+    """
+
+
+# Legal keyword arguments for the Extension constructor
+_safe = tuple(f.name for f in fields(_Extension))
+
+
+if TYPE_CHECKING:
+
+    @dataclass
+    class Extension(_Extension):
+        pass
+
+else:
+
+    @dataclass(init=False)
+    class Extension(_Extension):
+        def __init__(self, name, sources, *args, **kwargs):
+            if not isinstance(name, str):
+                raise TypeError("'name' must be a string")
+
+            # handle the string case first; since strings are iterable, disallow them
+            if isinstance(sources, str):
+                raise TypeError(
+                    "'sources' must be an iterable of strings or PathLike objects, not a string"
+                )
+
+            # now we check if it's iterable and contains valid types
+            try:
+                sources = list(map(os.fspath, sources))
+            except TypeError:
+                raise TypeError(
+                    "'sources' must be an iterable of strings or PathLike objects"
+                )
+
+            extra = {repr(k): kwargs.pop(k) for k in tuple(kwargs) if k not in _safe}
+            if extra:
+                warnings.warn(f"Unknown Extension options: {','.join(extra)}")
 
-    # When adding arguments to this constructor, be sure to update
-    # setup_keywords in core.py.
-    def __init__(
-        self,
-        name: str,
-        sources: Iterable[str | os.PathLike[str]],
-        include_dirs: list[str] | None = None,
-        define_macros: list[tuple[str, str | None]] | None = None,
-        undef_macros: list[str] | None = None,
-        library_dirs: list[str] | None = None,
-        libraries: list[str] | None = None,
-        runtime_library_dirs: list[str] | None = None,
-        extra_objects: list[str] | None = None,
-        extra_compile_args: list[str] | None = None,
-        extra_link_args: list[str] | None = None,
-        export_symbols: list[str] | None = None,
-        swig_opts: list[str] | None = None,
-        depends: list[str] | None = None,
-        language: str | None = None,
-        optional: bool | None = None,
-        **kw,  # To catch unknown keywords
-    ):
-        if not isinstance(name, str):
-            raise TypeError("'name' must be a string")
-
-        # handle the string case first; since strings are iterable, disallow them
-        if isinstance(sources, str):
-            raise TypeError(
-                "'sources' must be an iterable of strings or PathLike objects, not a string"
-            )
-
-        # now we check if it's iterable and contains valid types
-        try:
-            self.sources = list(map(os.fspath, sources))
-        except TypeError:
-            raise TypeError(
-                "'sources' must be an iterable of strings or PathLike objects"
-            )
-
-        self.name = name
-        self.include_dirs = include_dirs or []
-        self.define_macros = define_macros or []
-        self.undef_macros = undef_macros or []
-        self.library_dirs = library_dirs or []
-        self.libraries = libraries or []
-        self.runtime_library_dirs = runtime_library_dirs or []
-        self.extra_objects = extra_objects or []
-        self.extra_compile_args = extra_compile_args or []
-        self.extra_link_args = extra_link_args or []
-        self.export_symbols = export_symbols or []
-        self.swig_opts = swig_opts or []
-        self.depends = depends or []
-        self.language = language
-        self.optional = optional
-
-        # If there are unknown keyword options, warn about them
-        if len(kw) > 0:
-            options = [repr(option) for option in kw]
-            options = ', '.join(sorted(options))
-            msg = f"Unknown Extension options: {options}"
-            warnings.warn(msg)
-
-    def __repr__(self):
-        return f'<{self.__class__.__module__}.{self.__class__.__qualname__}({self.name!r}) at {id(self):#x}>'
+            # Ensure default values (e.g. []) are used instead of None:
+            positional = {k: v for k, v in zip(_safe[2:], args) if v is not None}
+            keywords = {k: v for k, v in kwargs.items() if v is not None}
+            super().__init__(name, sources, **positional, **keywords)
 
 
 def read_setup_file(filename):  # noqa: C901

distutils/tests/test_extension.py

@@ -106,7 +106,7 @@ def test_extension_init(self):
             assert getattr(ext, attr) == []
 
         assert ext.language is None
-        assert ext.optional is None
+        assert ext.optional is False
 
         # if there are unknown keyword options, warn about them
         with check_warnings() as w: