[pull] main from python:main

Doc/library/warnings.rst

@@ -487,7 +487,14 @@ Available Functions
    ignored.
 
    *module*, if supplied, should be the module name.
-   If no module is passed, the filename with ``.py`` stripped is used.
+   If no module is passed, the module regular expression in
+   :ref:`warnings filter <warning-filter>` will be tested against the module
+   names constructed from the path components starting from all parent
+   directories (with ``/__init__.py``, ``.py`` and, on Windows, ``.pyw``
+   stripped) and against the filename with ``.py`` stripped.
+   For example, when the filename is ``'/path/to/package/module.py'``, it will
+   be tested against  ``'path.to.package.module'``, ``'to.package.module'``
+   ``'package.module'``, ``'module'``, and ``'/path/to/package/module'``.
 
    *registry*, if supplied, should be the ``__warningregistry__`` dictionary
    of the module.
@@ -506,6 +513,10 @@ Available Functions
    .. versionchanged:: 3.6
       Add the *source* parameter.
 
+   .. versionchanged:: next
+      If no module is passed, test the filter regular expression against
+      module names created from the path, not only the path itself.
+
 
 .. function:: showwarning(message, category, filename, lineno, file=None, line=None)
 

Doc/whatsnew/3.15.rst

@@ -611,6 +611,18 @@ unittest
   (Contributed by Garry Cairns in :gh:`134567`.)
 
 
+warnings
+--------
+
+* Improve filtering by module in :func:`warnings.warn_explicit` if no *module*
+  argument is passed.
+  It now tests the module regular expression in the warnings filter not only
+  against the filename with ``.py`` stripped, but also against module names
+  constructed starting from different parent directories of the filename
+  (with ``/__init__.py``, ``.py`` and, on Windows, ``.pyw`` stripped).
+  (Contributed by Serhiy Storchaka in :gh:`135801`.)
+
+
 venv
 ----
 

Include/internal/pycore_unicodeobject.h

@@ -17,6 +17,46 @@ extern "C" {
 
 
 extern int _PyUnicode_IsModifiable(PyObject *unicode);
+extern void _PyUnicodeWriter_InitWithBuffer(
+    _PyUnicodeWriter *writer,
+    PyObject *buffer);
+extern PyObject* _PyUnicode_Result(PyObject *unicode);
+extern int _PyUnicode_DecodeUTF8Writer(
+    _PyUnicodeWriter *writer,
+    const char *s,
+    Py_ssize_t size,
+    _Py_error_handler error_handler,
+    const char *errors,
+    Py_ssize_t *consumed);
+extern PyObject* _PyUnicode_ResizeCompact(
+    PyObject *unicode,
+    Py_ssize_t length);
+extern PyObject* _PyUnicode_GetEmpty(void);
+
+
+/* Generic helper macro to convert characters of different types.
+   from_type and to_type have to be valid type names, begin and end
+   are pointers to the source characters which should be of type
+   "from_type *".  to is a pointer of type "to_type *" and points to the
+   buffer where the result characters are written to. */
+#define _PyUnicode_CONVERT_BYTES(from_type, to_type, begin, end, to) \
+    do {                                                \
+        to_type *_to = (to_type *)(to);                 \
+        const from_type *_iter = (const from_type *)(begin);\
+        const from_type *_end = (const from_type *)(end);\
+        Py_ssize_t n = (_end) - (_iter);                \
+        const from_type *_unrolled_end =                \
+            _iter + _Py_SIZE_ROUND_DOWN(n, 4);          \
+        while (_iter < (_unrolled_end)) {               \
+            _to[0] = (to_type) _iter[0];                \
+            _to[1] = (to_type) _iter[1];                \
+            _to[2] = (to_type) _iter[2];                \
+            _to[3] = (to_type) _iter[3];                \
+            _iter += 4; _to += 4;                       \
+        }                                               \
+        while (_iter < (_end))                          \
+            *_to++ = (to_type) *_iter++;                \
+    } while (0)
 
 
 static inline void

Lib/_py_warnings.py

@@ -520,20 +520,50 @@ def warn(message, category=None, stacklevel=1, source=None,
     )
 
 
+def _match_filename(pattern, filename, *, MS_WINDOWS=(sys.platform == 'win32')):
+    if not filename:
+        return pattern.match('<unknown>') is not None
+    if filename[0] == '<' and filename[-1] == '>':
+        return pattern.match(filename) is not None
+
+    is_py = (filename[-3:].lower() == '.py'
+             if MS_WINDOWS else
+             filename.endswith('.py'))
+    if is_py:
+        filename = filename[:-3]
+    if pattern.match(filename):  # for backward compatibility
+        return True
+    if MS_WINDOWS:
+        if not is_py and filename[-4:].lower() == '.pyw':
+            filename = filename[:-4]
+            is_py = True
+        if is_py and filename[-9:].lower() in (r'\__init__', '/__init__'):
+            filename = filename[:-9]
+        filename = filename.replace('\\', '/')
+    else:
+        if is_py and filename.endswith('/__init__'):
+            filename = filename[:-9]
+    filename = filename.replace('/', '.')
+    i = 0
+    while True:
+        if pattern.match(filename, i):
+            return True
+        i = filename.find('.', i) + 1
+        if not i:
+            return False
+
+
 def warn_explicit(message, category, filename, lineno,
                   module=None, registry=None, module_globals=None,
                   source=None):
     lineno = int(lineno)
-    if module is None:
-        module = filename or "<unknown>"
-        if module[-3:].lower() == ".py":
-            module = module[:-3] # XXX What about leading pathname?
     if isinstance(message, Warning):
         text = str(message)
         category = message.__class__
     else:
         text = message
         message = category(message)
+    modules = None
     key = (text, category, lineno)
     with _wm._lock:
         if registry is None:
@@ -549,9 +579,11 @@ def warn_explicit(message, category, filename, lineno,
             action, msg, cat, mod, ln = item
             if ((msg is None or msg.match(text)) and
                 issubclass(category, cat) and
-                (mod is None or mod.match(module)) and
-                (ln == 0 or lineno == ln)):
-                break
+                (ln == 0 or lineno == ln) and
+                (mod is None or (_match_filename(mod, filename)
+                                 if module is None else
+                                 mod.match(module)))):
+                    break
         else:
             action = _wm.defaultaction
         # Early exit actions

Lib/_pyrepl/utils.py

@@ -63,6 +63,12 @@ class ColorSpan(NamedTuple):
 def str_width(c: str) -> int:
     if ord(c) < 128:
         return 1
+    # gh-139246 for zero-width joiner and combining characters
+    if unicodedata.combining(c):
+        return 0
+    category = unicodedata.category(c)
+    if category == "Cf" and c != "\u00ad":
+        return 0
     w = unicodedata.east_asian_width(c)
     if w in ("N", "Na", "H", "A"):
         return 1

Lib/encodings/__init__.py

@@ -30,6 +30,7 @@
 
 import codecs
 import sys
+from _codecs import _normalize_encoding
 from . import aliases
 
 _cache = {}
@@ -55,18 +56,7 @@ def normalize_encoding(encoding):
     if isinstance(encoding, bytes):
         encoding = str(encoding, "ascii")
 
-    chars = []
-    punct = False
-    for c in encoding:
-        if c.isalnum() or c == '.':
-            if punct and chars:
-                chars.append('_')
-            if c.isascii():
-                chars.append(c)
-            punct = False
-        else:
-            punct = True
-    return ''.join(chars)
+    return _normalize_encoding(encoding)
 
 def search_function(encoding):
 

Lib/test/test_ast/test_ast.py

@@ -13,6 +13,7 @@
 import textwrap
 import types
 import unittest
+import warnings
 import weakref
 from io import StringIO
 from pathlib import Path
@@ -1069,6 +1070,19 @@ def test_tstring(self):
         self.assertIsInstance(tree.body[0].value.values[0], ast.Constant)
         self.assertIsInstance(tree.body[0].value.values[1], ast.Interpolation)
 
+    def test_filter_syntax_warnings_by_module(self):
+        filename = support.findfile('test_import/data/syntax_warnings.py')
+        with open(filename, 'rb') as f:
+            source = f.read()
+        with warnings.catch_warnings(record=True) as wlog:
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=r'<unknown>\z')
+            ast.parse(source)
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10])
+        for wm in wlog:
+            self.assertEqual(wm.filename, '<unknown>')
+            self.assertIs(wm.category, SyntaxWarning)
+
 
 class CopyTests(unittest.TestCase):
     """Test copying and pickling AST nodes."""

Lib/test/test_builtin.py

@@ -1088,6 +1088,28 @@ def four_freevars():
             three_freevars.__globals__,
             closure=my_closure)
 
+    def test_exec_filter_syntax_warnings_by_module(self):
+        filename = support.findfile('test_import/data/syntax_warnings.py')
+        with open(filename, 'rb') as f:
+            source = f.read()
+        with warnings.catch_warnings(record=True) as wlog:
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=r'<string>\z')
+            exec(source, {})
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10, 13, 14, 21])
+        for wm in wlog:
+            self.assertEqual(wm.filename, '<string>')
+            self.assertIs(wm.category, SyntaxWarning)
+
+        with warnings.catch_warnings(record=True) as wlog:
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=r'<string>\z')
+            exec(source, {'__name__': 'package.module', '__file__': filename})
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10, 13, 14, 21])
+        for wm in wlog:
+            self.assertEqual(wm.filename, '<string>')
+            self.assertIs(wm.category, SyntaxWarning)
+
 
     def test_filter(self):
         self.assertEqual(list(filter(lambda c: 'a' <= c <= 'z', 'Hello World')), list('elloorld'))

Lib/test/test_cmd_line_script.py

@@ -810,6 +810,19 @@ def test_script_as_dev_fd(self):
                 out, err = p.communicate()
                 self.assertEqual(out, b"12345678912345678912345\n")
 
+    def test_filter_syntax_warnings_by_module(self):
+        filename = support.findfile('test_import/data/syntax_warnings.py')
+        rc, out, err = assert_python_ok(
+            '-Werror',
+            '-Walways:::test.test_import.data.syntax_warnings',
+            filename)
+        self.assertEqual(err.count(b': SyntaxWarning: '), 6)
+
+        rc, out, err = assert_python_ok(
+            '-Werror',
+            '-Walways:::syntax_warnings',
+            filename)
+        self.assertEqual(err.count(b': SyntaxWarning: '), 6)
 
 
 def tearDownModule():

Lib/test/test_compile.py

@@ -1745,6 +1745,20 @@ def test_compile_warning_in_finally(self):
             self.assertEqual(wm.category, SyntaxWarning)
             self.assertIn("\"is\" with 'int' literal", str(wm.message))
 
+    def test_filter_syntax_warnings_by_module(self):
+        filename = support.findfile('test_import/data/syntax_warnings.py')
+        with open(filename, 'rb') as f:
+            source = f.read()
+        module_re = r'test\.test_import\.data\.syntax_warnings\z'
+        with warnings.catch_warnings(record=True) as wlog:
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=module_re)
+            compile(source, filename, 'exec')
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10, 13, 14, 21])
+        for wm in wlog:
+            self.assertEqual(wm.filename, filename)
+            self.assertIs(wm.category, SyntaxWarning)
+
     @support.subTests('src', [
         textwrap.dedent("""
             def f():

Lib/test/test_import/__init__.py

@@ -15,6 +15,7 @@
 import os
 import py_compile
 import random
+import re
 import shutil
 import stat
 import subprocess
@@ -23,6 +24,7 @@
 import threading
 import time
 import types
+import warnings
 import unittest
 from unittest import mock
 import _imp
@@ -51,7 +53,7 @@
     TESTFN, rmtree, temp_umask, TESTFN_UNENCODABLE)
 from test.support import script_helper
 from test.support import threading_helper
-from test.test_importlib.util import uncache
+from test.test_importlib.util import uncache, temporary_pycache_prefix
 from types import ModuleType
 try:
     import _testsinglephase
@@ -412,7 +414,6 @@ def test_from_import_missing_attr_path_is_canonical(self):
         self.assertIsNotNone(cm.exception)
 
     def test_from_import_star_invalid_type(self):
-        import re
         with ready_to_import() as (name, path):
             with open(path, 'w', encoding='utf-8') as f:
                 f.write("__all__ = [b'invalid_type']")
@@ -1250,6 +1251,35 @@ class Spec2:
                 origin = "a\x00b"
             _imp.create_dynamic(Spec2())
 
+    def test_filter_syntax_warnings_by_module(self):
+        module_re = r'test\.test_import\.data\.syntax_warnings\z'
+        unload('test.test_import.data.syntax_warnings')
+        with (os_helper.temp_dir() as tmpdir,
+              temporary_pycache_prefix(tmpdir),
+              warnings.catch_warnings(record=True) as wlog):
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=module_re)
+            import test.test_import.data.syntax_warnings
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10, 13, 14, 21])
+        filename = test.test_import.data.syntax_warnings.__file__
+        for wm in wlog:
+            self.assertEqual(wm.filename, filename)
+            self.assertIs(wm.category, SyntaxWarning)
+
+        module_re = r'syntax_warnings\z'
+        unload('test.test_import.data.syntax_warnings')
+        with (os_helper.temp_dir() as tmpdir,
+              temporary_pycache_prefix(tmpdir),
+              warnings.catch_warnings(record=True) as wlog):
+            warnings.simplefilter('error')
+            warnings.filterwarnings('always', module=module_re)
+            import test.test_import.data.syntax_warnings
+        self.assertEqual(sorted(wm.lineno for wm in wlog), [4, 7, 10, 13, 14, 21])
+        filename = test.test_import.data.syntax_warnings.__file__
+        for wm in wlog:
+            self.assertEqual(wm.filename, filename)
+            self.assertIs(wm.category, SyntaxWarning)
+
 
 @skip_if_dont_write_bytecode
 class FilePermissionTests(unittest.TestCase):

Lib/test/test_import/data/syntax_warnings.py

@@ -0,0 +1,21 @@
+# Syntax warnings emitted in different parts of the Python compiler.
+
+# Parser/lexer/lexer.c
+x = 1or 0  # line 4
+
+# Parser/tokenizer/helpers.c
+'\z'  # line 7
+
+# Parser/string_parser.c
+'\400'  # line 10
+
+# _PyCompile_Warn() in Python/codegen.c
+assert(x, 'message')  # line 13
+x is 1  # line 14
+
+# _PyErr_EmitSyntaxWarning() in Python/ast_preprocess.c
+def f():
+    try:
+        pass
+    finally:
+        return 42  # line 21