Replace and deprecate modutils.is_standard_module() (#2015)

3 files changed, 406 insertions, 3 deletions

diff --git a/astroid/_backport_stdlib_names.py b/astroid/_backport_stdlib_names.py
new file mode 100644
index 00000000..51d6957d
--- /dev/null
+++ b/astroid/_backport_stdlib_names.py
@@ -0,0 +1,356 @@
+# Licensed under the LGPL: https://www.gnu.org/licenses/old-licenses/lgpl-2.1.en.html
+# For details: https://github.com/PyCQA/astroid/blob/main/LICENSE
+# Copyright (c) https://github.com/PyCQA/astroid/blob/main/CONTRIBUTORS.txt
+
+"""
+Shim to support Python versions < 3.10 that don't have sys.stdlib_module_names
+
+These values were created by cherry-picking the commits from
+https://bugs.python.org/issue42955 into each version, but may be updated
+manually if changes are needed.
+"""
+
+import sys
+
+# TODO: Remove this file when Python 3.9 is no longer supported
+
+PY_3_7 = frozenset(
+    {
+        "__future__",
+        "_abc",
+        "_ast",
+        "_asyncio",
+        "_bisect",
+        "_blake2",
+        "_bootlocale",
+        "_bz2",
+        "_codecs",
+        "_codecs_cn",
+        "_codecs_hk",
+        "_codecs_iso2022",
+        "_codecs_jp",
+        "_codecs_kr",
+        "_codecs_tw",
+        "_collections",
+        "_collections_abc",
+        "_compat_pickle",
+        "_compression",
+        "_contextvars",
+        "_crypt",
+        "_csv",
+        "_ctypes",
+        "_curses",
+        "_curses_panel",
+        "_datetime",
+        "_dbm",
+        "_decimal",
+        "_dummy_thread",
+        "_elementtree",
+        "_functools",
+        "_gdbm",
+        "_hashlib",
+        "_heapq",
+        "_imp",
+        "_io",
+        "_json",
+        "_locale",
+        "_lsprof",
+        "_lzma",
+        "_markupbase",
+        "_md5",
+        "_msi",
+        "_multibytecodec",
+        "_multiprocessing",
+        "_opcode",
+        "_operator",
+        "_osx_support",
+        "_pickle",
+        "_posixsubprocess",
+        "_py_abc",
+        "_pydecimal",
+        "_pyio",
+        "_queue",
+        "_random",
+        "_sha1",
+        "_sha256",
+        "_sha3",
+        "_sha512",
+        "_signal",
+        "_sitebuiltins",
+        "_socket",
+        "_sqlite3",
+        "_sre",
+        "_ssl",
+        "_stat",
+        "_string",
+        "_strptime",
+        "_struct",
+        "_symtable",
+        "_thread",
+        "_threading_local",
+        "_tkinter",
+        "_tracemalloc",
+        "_uuid",
+        "_warnings",
+        "_weakref",
+        "_weakrefset",
+        "_winapi",
+        "abc",
+        "aifc",
+        "antigravity",
+        "argparse",
+        "array",
+        "ast",
+        "asynchat",
+        "asyncio",
+        "asyncore",
+        "atexit",
+        "audioop",
+        "base64",
+        "bdb",
+        "binascii",
+        "binhex",
+        "bisect",
+        "builtins",
+        "bz2",
+        "cProfile",
+        "calendar",
+        "cgi",
+        "cgitb",
+        "chunk",
+        "cmath",
+        "cmd",
+        "code",
+        "codecs",
+        "codeop",
+        "collections",
+        "colorsys",
+        "compileall",
+        "concurrent",
+        "configparser",
+        "contextlib",
+        "contextvars",
+        "copy",
+        "copyreg",
+        "crypt",
+        "csv",
+        "ctypes",
+        "curses",
+        "dataclasses",
+        "datetime",
+        "dbm",
+        "decimal",
+        "difflib",
+        "dis",
+        "distutils",
+        "doctest",
+        "dummy_threading",
+        "email",
+        "encodings",
+        "ensurepip",
+        "enum",
+        "errno",
+        "faulthandler",
+        "fcntl",
+        "filecmp",
+        "fileinput",
+        "fnmatch",
+        "formatter",
+        "fractions",
+        "ftplib",
+        "functools",
+        "gc",
+        "genericpath",
+        "getopt",
+        "getpass",
+        "gettext",
+        "glob",
+        "grp",
+        "gzip",
+        "hashlib",
+        "heapq",
+        "hmac",
+        "html",
+        "http",
+        "idlelib",
+        "imaplib",
+        "imghdr",
+        "imp",
+        "importlib",
+        "inspect",
+        "io",
+        "ipaddress",
+        "itertools",
+        "json",
+        "keyword",
+        "lib2to3",
+        "linecache",
+        "locale",
+        "logging",
+        "lzma",
+        "macpath",
+        "mailbox",
+        "mailcap",
+        "marshal",
+        "math",
+        "mimetypes",
+        "mmap",
+        "modulefinder",
+        "msilib",
+        "msvcrt",
+        "multiprocessing",
+        "netrc",
+        "nis",
+        "nntplib",
+        "nt",
+        "ntpath",
+        "nturl2path",
+        "numbers",
+        "opcode",
+        "operator",
+        "optparse",
+        "os",
+        "ossaudiodev",
+        "parser",
+        "pathlib",
+        "pdb",
+        "pickle",
+        "pickletools",
+        "pipes",
+        "pkgutil",
+        "platform",
+        "plistlib",
+        "poplib",
+        "posix",
+        "posixpath",
+        "pprint",
+        "profile",
+        "pstats",
+        "pty",
+        "pwd",
+        "py_compile",
+        "pyclbr",
+        "pydoc",
+        "pydoc_data",
+        "pyexpat",
+        "queue",
+        "quopri",
+        "random",
+        "re",
+        "readline",
+        "reprlib",
+        "resource",
+        "rlcompleter",
+        "runpy",
+        "sched",
+        "secrets",
+        "select",
+        "selectors",
+        "shelve",
+        "shlex",
+        "shutil",
+        "signal",
+        "site",
+        "smtpd",
+        "smtplib",
+        "sndhdr",
+        "socket",
+        "socketserver",
+        "spwd",
+        "sqlite3",
+        "sre_compile",
+        "sre_constants",
+        "sre_parse",
+        "ssl",
+        "stat",
+        "statistics",
+        "string",
+        "stringprep",
+        "struct",
+        "subprocess",
+        "sunau",
+        "symbol",
+        "symtable",
+        "sys",
+        "sysconfig",
+        "syslog",
+        "tabnanny",
+        "tarfile",
+        "telnetlib",
+        "tempfile",
+        "termios",
+        "textwrap",
+        "this",
+        "threading",
+        "time",
+        "timeit",
+        "tkinter",
+        "token",
+        "tokenize",
+        "trace",
+        "traceback",
+        "tracemalloc",
+        "tty",
+        "turtle",
+        "turtledemo",
+        "types",
+        "typing",
+        "unicodedata",
+        "unittest",
+        "urllib",
+        "uu",
+        "uuid",
+        "venv",
+        "warnings",
+        "wave",
+        "weakref",
+        "webbrowser",
+        "winreg",
+        "winsound",
+        "wsgiref",
+        "xdrlib",
+        "xml",
+        "xmlrpc",
+        "zipapp",
+        "zipfile",
+        "zipimport",
+        "zlib",
+    }
+)
+
+PY_3_8 = frozenset(
+    PY_3_7
+    - {
+        "macpath",
+    }
+    | {
+        "_posixshmem",
+        "_statistics",
+        "_xxsubinterpreters",
+    }
+)
+
+PY_3_9 = frozenset(
+    PY_3_8
+    - {
+        "_dummy_thread",
+        "dummy_threading",
+    }
+    | {
+        "_aix_support",
+        "_bootsubprocess",
+        "_peg_parser",
+        "_zoneinfo",
+        "graphlib",
+        "zoneinfo",
+    }
+)
+
+if sys.version_info[:2] == (3, 7):
+    stdlib_module_names = PY_3_7
+elif sys.version_info[:2] == (3, 8):
+    stdlib_module_names = PY_3_8
+elif sys.version_info[:2] == (3, 9):
+    stdlib_module_names = PY_3_9
+else:
+    raise AssertionError("This module is only intended as a backport for Python <= 3.9")
diff --git a/astroid/manager.py b/astroid/manager.py
index 8a5b05c7..b0cecc30 100644
--- a/astroid/manager.py
+++ b/astroid/manager.py
@@ -30,7 +30,7 @@ from astroid.modutils import (
     get_source_file,
     is_module_name_part_of_extension_package_whitelist,
     is_python_source,
-    is_standard_module,
+    is_stdlib_module,
     load_module_from_name,
     modpath_from_file,
 )
@@ -154,7 +154,7 @@ class AstroidManager:
     def _can_load_extension(self, modname: str) -> bool:
         if self.always_load_extensions:
             return True
-        if is_standard_module(modname):
+        if is_stdlib_module(modname):
             return True
         return is_module_name_part_of_extension_package_whitelist(
             modname, self.extension_package_whitelist
diff --git a/astroid/modutils.py b/astroid/modutils.py
index 1b5057f5..f05b5f89 100644
--- a/astroid/modutils.py
+++ b/astroid/modutils.py
@@ -26,14 +26,20 @@ import os
 import sys
 import sysconfig
 import types
+import warnings
 from collections.abc import Callable, Iterable, Sequence
 from contextlib import redirect_stderr, redirect_stdout
 from functools import lru_cache
 from pathlib import Path
 
-from astroid.const import IS_JYTHON, IS_PYPY
+from astroid.const import IS_JYTHON, IS_PYPY, PY310_PLUS
 from astroid.interpreter._import import spec, util
 
+if PY310_PLUS:
+    from sys import stdlib_module_names
+else:
+    from astroid._backport_stdlib_names import stdlib_module_names
+
 logger = logging.getLogger(__name__)
 
 
@@ -510,6 +516,41 @@ def is_python_source(filename: str | None) -> bool:
     return os.path.splitext(filename)[1][1:] in PY_SOURCE_EXTS
 
 
+def is_stdlib_module(modname: str) -> bool:
+    """Return: True if the modname is in the standard library"""
+    return modname.split(".")[0] in stdlib_module_names
+
+
+def module_in_path(modname: str, path: str | Iterable[str]) -> bool:
+    """Try to determine if a module is imported from one of the specified paths
+
+    :param modname: name of the module
+
+    :param path: paths to consider
+
+    :return:
+      true if the module:
+      - is located on the path listed in one of the directory in `paths`
+    """
+
+    modname = modname.split(".")[0]
+    try:
+        filename = file_from_modpath([modname])
+    except ImportError:
+        # Import failed, we can't check path if we don't know it
+        return False
+
+    if filename is None:
+        # No filename likely means it's compiled in, or potentially a namespace
+        return False
+    filename = _normalize_path(filename)
+
+    if isinstance(path, str):
+        return filename.startswith(_cache_normalize_path(path))
+
+    return any(filename.startswith(_cache_normalize_path(entry)) for entry in path)
+
+
 def is_standard_module(modname: str, std_path: Iterable[str] | None = None) -> bool:
     """Try to guess if a module is a standard python module (by default,
     see `std_path` parameter's description).
@@ -523,6 +564,12 @@ def is_standard_module(modname: str, std_path: Iterable[str] | None = None) -> b
       - is located on the path listed in one of the directory in `std_path`
       - is a built-in module
     """
+    warnings.warn(
+        "is_standard_module() is deprecated. Use, is_stdlib_module() or module_in_path() instead",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
     modname = modname.split(".")[0]
     try:
         filename = file_from_modpath([modname])