From 75bcd24918b5a619741f9f9eb6690d5180f7faf6 Mon Sep 17 00:00:00 2001 From: Shahar Naveh <50263213+ShaharNaveh@users.noreply.github.com> Date: Fri, 30 Jan 2026 12:24:33 +0200 Subject: [PATCH] Update `py_compile` from 3.14.2 (#6897) --- Lib/py_compile.py | 2 +- Lib/test/test_py_compile.py | 49 +++-- Lib/test/tokenizedata/__init__.py | 0 Lib/test/tokenizedata/bad_coding.py | 1 + Lib/test/tokenizedata/bad_coding2.py | 2 + Lib/test/tokenizedata/badsyntax_3131.py | 2 + Lib/test/tokenizedata/badsyntax_pep3120.py | 1 + Lib/test/tokenizedata/coding20731.py | 4 + ...-latin1-coding-cookie-and-utf8-bom-sig.txt | 13 ++ ...no-coding-cookie-and-utf8-bom-sig-only.txt | 11 + ...utf8-coding-cookie-and-no-utf8-bom-sig.txt | 13 ++ ...ts-utf8-coding-cookie-and-utf8-bom-sig.txt | 12 ++ Lib/test/tokenizedata/tokenize_tests.txt | 189 ++++++++++++++++++ 13 files changed, 286 insertions(+), 13 deletions(-) create mode 100644 Lib/test/tokenizedata/__init__.py create mode 100644 Lib/test/tokenizedata/bad_coding.py create mode 100644 Lib/test/tokenizedata/bad_coding2.py create mode 100644 Lib/test/tokenizedata/badsyntax_3131.py create mode 100644 Lib/test/tokenizedata/badsyntax_pep3120.py create mode 100644 Lib/test/tokenizedata/coding20731.py create mode 100644 Lib/test/tokenizedata/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt create mode 100644 Lib/test/tokenizedata/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt create mode 100644 Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt create mode 100644 Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt create mode 100644 Lib/test/tokenizedata/tokenize_tests.txt diff --git a/Lib/py_compile.py b/Lib/py_compile.py index 388614e51..43d8ec90f 100644 --- a/Lib/py_compile.py +++ b/Lib/py_compile.py @@ -177,7 +177,7 @@ def main(): import argparse description = 'A simple command-line interface for py_compile module.' - parser = argparse.ArgumentParser(description=description) + parser = argparse.ArgumentParser(description=description, color=True) parser.add_argument( '-q', '--quiet', action='store_true', diff --git a/Lib/test/test_py_compile.py b/Lib/test/test_py_compile.py index 54302eba4..3ea5d6d38 100644 --- a/Lib/test/test_py_compile.py +++ b/Lib/test/test_py_compile.py @@ -109,18 +109,22 @@ class PyCompileTestsBase: self.assertTrue(os.path.exists(self.pyc_path)) self.assertFalse(os.path.exists(self.cache_path)) - import platform - @unittest.expectedFailureIf(sys.platform == "darwin" and int(platform.release().split(".")[0]) < 20, "TODO: RUSTPYTHON") + @unittest.expectedFailureIf( + sys.platform == "darwin" and int( + __import__("platform").release().split(".")[0] + ) < 20, + "TODO: RUSTPYTHON" + ) def test_relative_path(self): py_compile.compile(os.path.relpath(self.source_path), os.path.relpath(self.pyc_path)) self.assertTrue(os.path.exists(self.pyc_path)) self.assertFalse(os.path.exists(self.cache_path)) - @unittest.skipIf(hasattr(os, 'geteuid') and os.geteuid() == 0, - 'non-root user required') + @os_helper.skip_if_dac_override @unittest.skipIf(os.name == 'nt', 'cannot control directory permissions on Windows') + @os_helper.skip_unless_working_chmod def test_exceptions_propagate(self): # Make sure that exceptions raised thanks to issues with writing # bytecode. @@ -133,10 +137,11 @@ class PyCompileTestsBase: finally: os.chmod(self.directory, mode.st_mode) - # TODO: RUSTPYTHON - @unittest.expectedFailure + @unittest.expectedFailure # TODO: RUSTPYTHON def test_bad_coding(self): - bad_coding = os.path.join(os.path.dirname(__file__), 'bad_coding2.py') + bad_coding = os.path.join(os.path.dirname(__file__), + 'tokenizedata', + 'bad_coding2.py') with support.captured_stderr(): self.assertIsNone(py_compile.compile(bad_coding, doraise=False)) self.assertFalse(os.path.exists( @@ -198,6 +203,18 @@ class PyCompileTestsBase: fp.read(), 'test', {}) self.assertEqual(flags, 0b1) + @unittest.expectedFailure # TODO: RUSTPYTHON + def test_quiet(self): + bad_coding = os.path.join(os.path.dirname(__file__), + 'tokenizedata', + 'bad_coding2.py') + with support.captured_stderr() as stderr: + self.assertIsNone(py_compile.compile(bad_coding, doraise=False, quiet=2)) + self.assertIsNone(py_compile.compile(bad_coding, doraise=True, quiet=2)) + self.assertEqual(stderr.getvalue(), '') + with self.assertRaises(py_compile.PyCompileError): + py_compile.compile(bad_coding, doraise=True, quiet=1) + class PyCompileTestsWithSourceEpoch(PyCompileTestsBase, unittest.TestCase, @@ -218,27 +235,31 @@ class PyCompileCLITestCase(unittest.TestCase): def setUp(self): self.directory = tempfile.mkdtemp() self.source_path = os.path.join(self.directory, '_test.py') - self.cache_path = importlib.util.cache_from_source(self.source_path) + self.cache_path = importlib.util.cache_from_source(self.source_path, + optimization='' if __debug__ else 1) with open(self.source_path, 'w') as file: file.write('x = 123\n') def tearDown(self): os_helper.rmtree(self.directory) + @support.requires_subprocess() def pycompilecmd(self, *args, **kwargs): # assert_python_* helpers don't return proc object. We'll just use # subprocess.run() instead of spawn_python() and its friends to test # stdin support of the CLI. + opts = '-m' if __debug__ else '-Om' if args and args[0] == '-' and 'input' in kwargs: - return subprocess.run([sys.executable, '-m', 'py_compile', '-'], + return subprocess.run([sys.executable, opts, 'py_compile', '-'], input=kwargs['input'].encode(), capture_output=True) - return script_helper.assert_python_ok('-m', 'py_compile', *args, **kwargs) + return script_helper.assert_python_ok(opts, 'py_compile', *args, **kwargs) def pycompilecmd_failure(self, *args): return script_helper.assert_python_failure('-m', 'py_compile', *args) def test_stdin(self): + self.assertFalse(os.path.exists(self.cache_path)) result = self.pycompilecmd('-', input=self.source_path) self.assertEqual(result.returncode, 0) self.assertEqual(result.stdout, b'') @@ -253,14 +274,18 @@ class PyCompileCLITestCase(unittest.TestCase): self.assertTrue(os.path.exists(self.cache_path)) def test_bad_syntax(self): - bad_syntax = os.path.join(os.path.dirname(__file__), 'badsyntax_3131.py') + bad_syntax = os.path.join(os.path.dirname(__file__), + 'tokenizedata', + 'badsyntax_3131.py') rc, stdout, stderr = self.pycompilecmd_failure(bad_syntax) self.assertEqual(rc, 1) self.assertEqual(stdout, b'') self.assertIn(b'SyntaxError', stderr) def test_bad_syntax_with_quiet(self): - bad_syntax = os.path.join(os.path.dirname(__file__), 'badsyntax_3131.py') + bad_syntax = os.path.join(os.path.dirname(__file__), + 'tokenizedata', + 'badsyntax_3131.py') rc, stdout, stderr = self.pycompilecmd_failure('-q', bad_syntax) self.assertEqual(rc, 1) self.assertEqual(stdout, b'') diff --git a/Lib/test/tokenizedata/__init__.py b/Lib/test/tokenizedata/__init__.py new file mode 100644 index 000000000..e69de29bb diff --git a/Lib/test/tokenizedata/bad_coding.py b/Lib/test/tokenizedata/bad_coding.py new file mode 100644 index 000000000..971b0a8f3 --- /dev/null +++ b/Lib/test/tokenizedata/bad_coding.py @@ -0,0 +1 @@ +# -*- coding: uft-8 -*- diff --git a/Lib/test/tokenizedata/bad_coding2.py b/Lib/test/tokenizedata/bad_coding2.py new file mode 100644 index 000000000..bb2bb7e1e --- /dev/null +++ b/Lib/test/tokenizedata/bad_coding2.py @@ -0,0 +1,2 @@ +#coding: utf8 +print('我') diff --git a/Lib/test/tokenizedata/badsyntax_3131.py b/Lib/test/tokenizedata/badsyntax_3131.py new file mode 100644 index 000000000..901d3744c --- /dev/null +++ b/Lib/test/tokenizedata/badsyntax_3131.py @@ -0,0 +1,2 @@ +# -*- coding: utf-8 -*- +€ = 2 diff --git a/Lib/test/tokenizedata/badsyntax_pep3120.py b/Lib/test/tokenizedata/badsyntax_pep3120.py new file mode 100644 index 000000000..d14b4c96e --- /dev/null +++ b/Lib/test/tokenizedata/badsyntax_pep3120.py @@ -0,0 +1 @@ +print("bse") diff --git a/Lib/test/tokenizedata/coding20731.py b/Lib/test/tokenizedata/coding20731.py new file mode 100644 index 000000000..b0e227ad1 --- /dev/null +++ b/Lib/test/tokenizedata/coding20731.py @@ -0,0 +1,4 @@ +#coding:latin1 + + + diff --git a/Lib/test/tokenizedata/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt b/Lib/test/tokenizedata/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt new file mode 100644 index 000000000..1b5335b64 --- /dev/null +++ b/Lib/test/tokenizedata/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt @@ -0,0 +1,13 @@ +# -*- coding: latin1 -*- +# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf' +# at the start of it. Make sure this is preserved if any changes +# are made! Also note that the coding cookie above conflicts with +# the presence of a utf-8 BOM signature -- this is intended. + +# Arbitrary encoded utf-8 text (stolen from test_doctest2.py). +x = 'ЉЊЈЁЂ' +def y(): + """ + And again in a comment. ЉЊЈЁЂ + """ + pass diff --git a/Lib/test/tokenizedata/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt b/Lib/test/tokenizedata/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt new file mode 100644 index 000000000..23fd2168a --- /dev/null +++ b/Lib/test/tokenizedata/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt @@ -0,0 +1,11 @@ +# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf' +# at the start of it. Make sure this is preserved if any changes +# are made! + +# Arbitrary encoded utf-8 text (stolen from test_doctest2.py). +x = 'ЉЊЈЁЂ' +def y(): + """ + And again in a comment. ЉЊЈЁЂ + """ + pass diff --git a/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt b/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt new file mode 100644 index 000000000..04561e484 --- /dev/null +++ b/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt @@ -0,0 +1,13 @@ +# -*- coding: utf-8 -*- +# IMPORTANT: unlike the other test_tokenize-*.txt files, this file +# does NOT have the utf-8 BOM signature '\xef\xbb\xbf' at the start +# of it. Make sure this is not added inadvertently by your editor +# if any changes are made to this file! + +# Arbitrary encoded utf-8 text (stolen from test_doctest2.py). +x = 'ЉЊЈЁЂ' +def y(): + """ + And again in a comment. ЉЊЈЁЂ + """ + pass diff --git a/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt b/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt new file mode 100644 index 000000000..4b20ff6ad --- /dev/null +++ b/Lib/test/tokenizedata/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt @@ -0,0 +1,12 @@ +# -*- coding: utf-8 -*- +# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf' +# at the start of it. Make sure this is preserved if any changes +# are made! + +# Arbitrary encoded utf-8 text (stolen from test_doctest2.py). +x = 'ЉЊЈЁЂ' +def y(): + """ + And again in a comment. ЉЊЈЁЂ + """ + pass diff --git a/Lib/test/tokenizedata/tokenize_tests.txt b/Lib/test/tokenizedata/tokenize_tests.txt new file mode 100644 index 000000000..c4f5a58a9 --- /dev/null +++ b/Lib/test/tokenizedata/tokenize_tests.txt @@ -0,0 +1,189 @@ +# Tests for the 'tokenize' module. +# Large bits stolen from test_grammar.py. + +# Comments +"#" +#' +#" +#\ + # + # abc +'''# +#''' + +x = 1 # + +# Balancing continuation + +a = (3, 4, + 5, 6) +y = [3, 4, + 5] +z = {'a':5, + 'b':6} +x = (len(repr(y)) + 5*x - a[ + 3 ] + - x + len({ + } + ) + ) + +# Backslash means line continuation: +x = 1 \ ++ 1 + +# Backslash does not means continuation in comments :\ +x = 0 + +# Ordinary integers +0xff != 255 +0o377 != 255 +2147483647 != 0o17777777777 +-2147483647-1 != 0o20000000000 +0o37777777777 != -1 +0xffffffff != -1; 0o37777777777 != -1; -0o1234567 == 0O001234567; 0b10101 == 0B00010101 + +# Long integers +x = 0 +x = 0 +x = 0xffffffffffffffff +x = 0xffffffffffffffff +x = 0o77777777777777777 +x = 0B11101010111111111 +x = 123456789012345678901234567890 +x = 123456789012345678901234567890 + +# Floating-point numbers +x = 3.14 +x = 314. +x = 0.314 +# XXX x = 000.314 +x = .314 +x = 3e14 +x = 3E14 +x = 3e-14 +x = 3e+14 +x = 3.e14 +x = .3e14 +x = 3.1e4 + +# String literals +x = ''; y = ""; +x = '\''; y = "'"; +x = '"'; y = "\""; +x = "doesn't \"shrink\" does it" +y = 'doesn\'t "shrink" does it' +x = "does \"shrink\" doesn't it" +y = 'does "shrink" doesn\'t it' +x = """ +The "quick" +brown fox +jumps over +the 'lazy' dog. +""" +y = '\nThe "quick"\nbrown fox\njumps over\nthe \'lazy\' dog.\n' +y = ''' +The "quick" +brown fox +jumps over +the 'lazy' dog. +'''; +y = "\n\ +The \"quick\"\n\ +brown fox\n\ +jumps over\n\ +the 'lazy' dog.\n\ +"; +y = '\n\ +The \"quick\"\n\ +brown fox\n\ +jumps over\n\ +the \'lazy\' dog.\n\ +'; +x = r'\\' + R'\\' +x = r'\'' + '' +y = r''' +foo bar \\ +baz''' + R''' +foo''' +y = r"""foo +bar \\ baz +""" + R'''spam +''' +x = b'abc' + B'ABC' +y = b"abc" + B"ABC" +x = br'abc' + Br'ABC' + bR'ABC' + BR'ABC' +y = br"abc" + Br"ABC" + bR"ABC" + BR"ABC" +x = rb'abc' + rB'ABC' + Rb'ABC' + RB'ABC' +y = rb"abc" + rB"ABC" + Rb"ABC" + RB"ABC" +x = br'\\' + BR'\\' +x = rb'\\' + RB'\\' +x = br'\'' + '' +x = rb'\'' + '' +y = br''' +foo bar \\ +baz''' + BR''' +foo''' +y = Br"""foo +bar \\ baz +""" + bR'''spam +''' +y = rB"""foo +bar \\ baz +""" + Rb'''spam +''' + +# Indentation +if 1: + x = 2 +if 1: + x = 2 +if 1: + while 0: + if 0: + x = 2 + x = 2 +if 0: + if 2: + while 0: + if 1: + x = 2 + +# Operators + +def d22(a, b, c=1, d=2): pass +def d01v(a=1, *restt, **restd): pass + +(x, y) != ({'a':1}, {'b':2}) + +# comparison +if 1 < 1 > 1 == 1 >= 1 <= 1 != 1 != 1 in 1 not in 1 is 1 is not 1: pass + +# binary +x = 1 & 1 +x = 1 ^ 1 +x = 1 | 1 + +# shift +x = 1 << 1 >> 1 + +# additive +x = 1 - 1 + 1 - 1 + 1 + +# multiplicative +x = 1 / 1 * 1 % 1 + +# unary +x = ~1 ^ 1 & 1 | 1 & 1 ^ -1 +x = -1*1/1 + 1*1 - ---1*1 + +# selector +import sys, time +x = sys.modules['time'].time() + +@staticmethod +def foo(): pass + +@staticmethod +def foo(x:1)->1: pass +