https://github.com/python/cpython/commit/a60520da3cb0c571a361b9d0b8d998e78f38b836 commit: a60520da3cb0c571a361b9d0b8d998e78f38b836 branch: main author: Victor Stinner <[email protected]> committer: encukou <[email protected]> date: 2026-04-30T18:25:32+02:00 summary:
gh-147991: Speed up tomllib import time (GH-147992) - Use lazy import for regular expressions. - Use frozendict for string escapes Co-authored-by: Taneli Hukkinen <[email protected]> Co-authored-by: Petr Viktorin <[email protected]> files: A Misc/NEWS.d/next/Library/2026-04-02-05-06-34.gh-issue-147991.2ANtR5.rst M Lib/test/test_tomllib/test_misc.py M Lib/tomllib/_parser.py diff --git a/Lib/test/test_tomllib/test_misc.py b/Lib/test/test_tomllib/test_misc.py index 118fde24d88521..abd0842d10b254 100644 --- a/Lib/test/test_tomllib/test_misc.py +++ b/Lib/test/test_tomllib/test_misc.py @@ -9,8 +9,10 @@ from pathlib import Path import sys import tempfile +import textwrap import unittest from test import support +from test.support.script_helper import assert_python_ok from . import tomllib @@ -124,3 +126,20 @@ def test_types_import(self): never imported by tests. """ importlib.import_module(f"{tomllib.__name__}._types") + + def test_lazy_import(self): + # Test the TOML file can be parsed without importing regular + # expressions (tomllib._re) + code = textwrap.dedent(""" + import sys, tomllib, textwrap + document = textwrap.dedent(''' + [metadata] + key = "text" + array = ["array", "of", "text"] + booleans = [true, false] + ''') + tomllib.loads(document) + print("lazy import?", 'tomllib._re' not in sys.modules) + """) + proc = assert_python_ok("-c", code) + self.assertIn(b"lazy import? True", proc.out) diff --git a/Lib/tomllib/_parser.py b/Lib/tomllib/_parser.py index b59d0f7d54bdc3..8aa01301dcea32 100644 --- a/Lib/tomllib/_parser.py +++ b/Lib/tomllib/_parser.py @@ -4,7 +4,11 @@ from __future__ import annotations -from types import MappingProxyType +# Defer loading regular expressions until we actually need them in +# parse_value(). +__lazy_modules__ = ["tomllib._re"] + +import sys from ._re import ( RE_DATETIME, @@ -15,6 +19,9 @@ match_to_number, ) +if sys.version_info < (3, 15): + from types import MappingProxyType as frozendict + TYPE_CHECKING = False if TYPE_CHECKING: from collections.abc import Iterable @@ -42,7 +49,7 @@ KEY_INITIAL_CHARS: Final = BARE_KEY_CHARS | frozenset("\"'") HEXDIGIT_CHARS: Final = frozenset("abcdef" "ABCDEF" "0123456789") -BASIC_STR_ESCAPE_REPLACEMENTS: Final = MappingProxyType( +BASIC_STR_ESCAPE_REPLACEMENTS: Final = frozendict( { "\\b": "\u0008", # backspace "\\t": "\u0009", # tab diff --git a/Misc/NEWS.d/next/Library/2026-04-02-05-06-34.gh-issue-147991.2ANtR5.rst b/Misc/NEWS.d/next/Library/2026-04-02-05-06-34.gh-issue-147991.2ANtR5.rst new file mode 100644 index 00000000000000..581c52926c3565 --- /dev/null +++ b/Misc/NEWS.d/next/Library/2026-04-02-05-06-34.gh-issue-147991.2ANtR5.rst @@ -0,0 +1,2 @@ +Improve :mod:`tomllib` import time (up to 10x faster). Patch by Victor +Stinner. _______________________________________________ Python-checkins mailing list -- [email protected] To unsubscribe send an email to [email protected] https://mail.python.org/mailman3//lists/python-checkins.python.org Member address: [email protected]
