Загрузить файлы в «venv/Lib/site-packages/pip/_internal/utils»

This commit is contained in:
2026-07-02 19:11:14 +00:00
parent 9e07fc9c4d
commit f9a4c20c40
5 changed files with 1064 additions and 0 deletions

View File

@@ -0,0 +1,248 @@
from __future__ import annotations
import logging
import os
import shlex
import subprocess
from collections.abc import Iterable, Mapping
from typing import Any, Callable, Literal, Union
from pip._vendor.rich.markup import escape
from pip._internal.cli.spinners import SpinnerInterface, open_spinner
from pip._internal.exceptions import InstallationSubprocessError
from pip._internal.utils.logging import VERBOSE, subprocess_logger
from pip._internal.utils.misc import HiddenText
CommandArgs = list[Union[str, HiddenText]]
def make_command(*args: str | HiddenText | CommandArgs) -> CommandArgs:
"""
Create a CommandArgs object.
"""
command_args: CommandArgs = []
for arg in args:
# Check for list instead of CommandArgs since CommandArgs is
# only known during type-checking.
if isinstance(arg, list):
command_args.extend(arg)
else:
# Otherwise, arg is str or HiddenText.
command_args.append(arg)
return command_args
def format_command_args(args: list[str] | CommandArgs) -> str:
"""
Format command arguments for display.
"""
# For HiddenText arguments, display the redacted form by calling str().
# Also, we don't apply str() to arguments that aren't HiddenText since
# this can trigger a UnicodeDecodeError in Python 2 if the argument
# has type unicode and includes a non-ascii character. (The type
# checker doesn't ensure the annotations are correct in all cases.)
return " ".join(
shlex.quote(str(arg)) if isinstance(arg, HiddenText) else shlex.quote(arg)
for arg in args
)
def reveal_command_args(args: list[str] | CommandArgs) -> list[str]:
"""
Return the arguments in their raw, unredacted form.
"""
return [arg.secret if isinstance(arg, HiddenText) else arg for arg in args]
def call_subprocess(
cmd: list[str] | CommandArgs,
show_stdout: bool = False,
cwd: str | None = None,
on_returncode: Literal["raise", "warn", "ignore"] = "raise",
extra_ok_returncodes: Iterable[int] | None = None,
extra_environ: Mapping[str, Any] | None = None,
unset_environ: Iterable[str] | None = None,
spinner: SpinnerInterface | None = None,
log_failed_cmd: bool | None = True,
stdout_only: bool | None = False,
*,
command_desc: str,
) -> str:
"""
Args:
show_stdout: if true, use INFO to log the subprocess's stderr and
stdout streams. Otherwise, use DEBUG. Defaults to False.
extra_ok_returncodes: an iterable of integer return codes that are
acceptable, in addition to 0. Defaults to None, which means [].
unset_environ: an iterable of environment variable names to unset
prior to calling subprocess.Popen().
log_failed_cmd: if false, failed commands are not logged, only raised.
stdout_only: if true, return only stdout, else return both. When true,
logging of both stdout and stderr occurs when the subprocess has
terminated, else logging occurs as subprocess output is produced.
"""
if extra_ok_returncodes is None:
extra_ok_returncodes = []
if unset_environ is None:
unset_environ = []
# Most places in pip use show_stdout=False. What this means is--
#
# - We connect the child's output (combined stderr and stdout) to a
# single pipe, which we read.
# - We log this output to stderr at DEBUG level as it is received.
# - If DEBUG logging isn't enabled (e.g. if --verbose logging wasn't
# requested), then we show a spinner so the user can still see the
# subprocess is in progress.
# - If the subprocess exits with an error, we log the output to stderr
# at ERROR level if it hasn't already been displayed to the console
# (e.g. if --verbose logging wasn't enabled). This way we don't log
# the output to the console twice.
#
# If show_stdout=True, then the above is still done, but with DEBUG
# replaced by INFO.
if show_stdout:
# Then log the subprocess output at INFO level.
log_subprocess: Callable[..., None] = subprocess_logger.info
used_level = logging.INFO
else:
# Then log the subprocess output using VERBOSE. This also ensures
# it will be logged to the log file (aka user_log), if enabled.
log_subprocess = subprocess_logger.verbose
used_level = VERBOSE
# Whether the subprocess will be visible in the console.
showing_subprocess = subprocess_logger.getEffectiveLevel() <= used_level
# Only use the spinner if we're not showing the subprocess output
# and we have a spinner.
use_spinner = not showing_subprocess and spinner is not None
log_subprocess("Running command %s", command_desc)
env = os.environ.copy()
if extra_environ:
env.update(extra_environ)
for name in unset_environ:
env.pop(name, None)
try:
proc = subprocess.Popen(
# Convert HiddenText objects to the underlying str.
reveal_command_args(cmd),
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT if not stdout_only else subprocess.PIPE,
cwd=cwd,
env=env,
errors="backslashreplace",
)
except Exception as exc:
if log_failed_cmd:
subprocess_logger.critical(
"Error %s while executing command %s",
exc,
command_desc,
)
raise
all_output = []
if not stdout_only:
assert proc.stdout
assert proc.stdin
proc.stdin.close()
# In this mode, stdout and stderr are in the same pipe.
while True:
line: str = proc.stdout.readline()
if not line:
break
line = line.rstrip()
all_output.append(line + "\n")
# Show the line immediately.
log_subprocess(line)
# Update the spinner.
if use_spinner:
assert spinner
spinner.spin()
try:
proc.wait()
finally:
if proc.stdout:
proc.stdout.close()
output = "".join(all_output)
else:
# In this mode, stdout and stderr are in different pipes.
# We must use communicate() which is the only safe way to read both.
out, err = proc.communicate()
# log line by line to preserve pip log indenting
for out_line in out.splitlines():
log_subprocess(out_line)
all_output.append(out)
for err_line in err.splitlines():
log_subprocess(err_line)
all_output.append(err)
output = out
proc_had_error = proc.returncode and proc.returncode not in extra_ok_returncodes
if use_spinner:
assert spinner
if proc_had_error:
spinner.finish("error")
else:
spinner.finish("done")
if proc_had_error:
if on_returncode == "raise":
error = InstallationSubprocessError(
command_description=command_desc,
exit_code=proc.returncode,
output_lines=all_output if not showing_subprocess else None,
)
if log_failed_cmd:
subprocess_logger.error("%s", error, extra={"rich": True})
subprocess_logger.verbose(
"[bold magenta]full command[/]: [blue]%s[/]",
escape(format_command_args(cmd)),
extra={"markup": True},
)
subprocess_logger.verbose(
"[bold magenta]cwd[/]: %s",
escape(cwd or "[inherit]"),
extra={"markup": True},
)
raise error
elif on_returncode == "warn":
subprocess_logger.warning(
'Command "%s" had error code %s in %s',
command_desc,
proc.returncode,
cwd,
)
elif on_returncode == "ignore":
pass
else:
raise ValueError(f"Invalid value: on_returncode={on_returncode!r}")
return output
def runner_with_spinner_message(message: str) -> Callable[..., None]:
"""Provide a subprocess_runner that shows a spinner message.
Intended for use with for BuildBackendHookCaller. Thus, the runner has
an API that matches what's expected by BuildBackendHookCaller.subprocess_runner.
"""
def runner(
cmd: list[str],
cwd: str | None = None,
extra_environ: Mapping[str, Any] | None = None,
) -> None:
with open_spinner(message) as spinner:
call_subprocess(
cmd,
command_desc=message,
cwd=cwd,
extra_environ=extra_environ,
spinner=spinner,
)
return runner

View File

@@ -0,0 +1,294 @@
from __future__ import annotations
import errno
import itertools
import logging
import os.path
import tempfile
import traceback
from collections.abc import Generator
from contextlib import ExitStack, contextmanager
from pathlib import Path
from typing import (
Any,
Callable,
TypeVar,
)
from pip._internal.utils.misc import enum, rmtree
logger = logging.getLogger(__name__)
_T = TypeVar("_T", bound="TempDirectory")
# Kinds of temporary directories. Only needed for ones that are
# globally-managed.
tempdir_kinds = enum(
BUILD_ENV="build-env",
EPHEM_WHEEL_CACHE="ephem-wheel-cache",
REQ_BUILD="req-build",
)
_tempdir_manager: ExitStack | None = None
@contextmanager
def global_tempdir_manager() -> Generator[None, None, None]:
global _tempdir_manager
with ExitStack() as stack:
old_tempdir_manager, _tempdir_manager = _tempdir_manager, stack
try:
yield
finally:
_tempdir_manager = old_tempdir_manager
class TempDirectoryTypeRegistry:
"""Manages temp directory behavior"""
def __init__(self) -> None:
self._should_delete: dict[str, bool] = {}
def set_delete(self, kind: str, value: bool) -> None:
"""Indicate whether a TempDirectory of the given kind should be
auto-deleted.
"""
self._should_delete[kind] = value
def get_delete(self, kind: str) -> bool:
"""Get configured auto-delete flag for a given TempDirectory type,
default True.
"""
return self._should_delete.get(kind, True)
_tempdir_registry: TempDirectoryTypeRegistry | None = None
@contextmanager
def tempdir_registry() -> Generator[TempDirectoryTypeRegistry, None, None]:
"""Provides a scoped global tempdir registry that can be used to dictate
whether directories should be deleted.
"""
global _tempdir_registry
old_tempdir_registry = _tempdir_registry
_tempdir_registry = TempDirectoryTypeRegistry()
try:
yield _tempdir_registry
finally:
_tempdir_registry = old_tempdir_registry
class _Default:
pass
_default = _Default()
class TempDirectory:
"""Helper class that owns and cleans up a temporary directory.
This class can be used as a context manager or as an OO representation of a
temporary directory.
Attributes:
path
Location to the created temporary directory
delete
Whether the directory should be deleted when exiting
(when used as a contextmanager)
Methods:
cleanup()
Deletes the temporary directory
When used as a context manager, if the delete attribute is True, on
exiting the context the temporary directory is deleted.
"""
def __init__(
self,
path: str | None = None,
delete: bool | None | _Default = _default,
kind: str = "temp",
globally_managed: bool = False,
ignore_cleanup_errors: bool = True,
):
super().__init__()
if delete is _default:
if path is not None:
# If we were given an explicit directory, resolve delete option
# now.
delete = False
else:
# Otherwise, we wait until cleanup and see what
# tempdir_registry says.
delete = None
# The only time we specify path is in for editables where it
# is the value of the --src option.
if path is None:
path = self._create(kind)
self._path = path
self._deleted = False
self.delete = delete
self.kind = kind
self.ignore_cleanup_errors = ignore_cleanup_errors
if globally_managed:
assert _tempdir_manager is not None
_tempdir_manager.enter_context(self)
@property
def path(self) -> str:
assert not self._deleted, f"Attempted to access deleted path: {self._path}"
return self._path
def __repr__(self) -> str:
return f"<{self.__class__.__name__} {self.path!r}>"
def __enter__(self: _T) -> _T:
return self
def __exit__(self, exc: Any, value: Any, tb: Any) -> None:
if self.delete is not None:
delete = self.delete
elif _tempdir_registry:
delete = _tempdir_registry.get_delete(self.kind)
else:
delete = True
if delete:
self.cleanup()
def _create(self, kind: str) -> str:
"""Create a temporary directory and store its path in self.path"""
# We realpath here because some systems have their default tmpdir
# symlinked to another directory. This tends to confuse build
# scripts, so we canonicalize the path by traversing potential
# symlinks here.
path = os.path.realpath(tempfile.mkdtemp(prefix=f"pip-{kind}-"))
logger.debug("Created temporary directory: %s", path)
return path
def cleanup(self) -> None:
"""Remove the temporary directory created and reset state"""
self._deleted = True
if not os.path.exists(self._path):
return
errors: list[BaseException] = []
def onerror(
func: Callable[..., Any],
path: Path,
exc_val: BaseException,
) -> None:
"""Log a warning for a `rmtree` error and continue"""
formatted_exc = "\n".join(
traceback.format_exception_only(type(exc_val), exc_val)
)
formatted_exc = formatted_exc.rstrip() # remove trailing new line
if func in (os.unlink, os.remove, os.rmdir):
logger.debug(
"Failed to remove a temporary file '%s' due to %s.\n",
path,
formatted_exc,
)
else:
logger.debug("%s failed with %s.", func.__qualname__, formatted_exc)
errors.append(exc_val)
if self.ignore_cleanup_errors:
try:
# first try with @retry; retrying to handle ephemeral errors
rmtree(self._path, ignore_errors=False)
except OSError:
# last pass ignore/log all errors
rmtree(self._path, onexc=onerror)
if errors:
logger.warning(
"Failed to remove contents in a temporary directory '%s'.\n"
"You can safely remove it manually.",
self._path,
)
else:
rmtree(self._path)
class AdjacentTempDirectory(TempDirectory):
"""Helper class that creates a temporary directory adjacent to a real one.
Attributes:
original
The original directory to create a temp directory for.
path
After calling create() or entering, contains the full
path to the temporary directory.
delete
Whether the directory should be deleted when exiting
(when used as a contextmanager)
"""
# The characters that may be used to name the temp directory
# We always prepend a ~ and then rotate through these until
# a usable name is found.
# pkg_resources raises a different error for .dist-info folder
# with leading '-' and invalid metadata
LEADING_CHARS = "-~.=%0123456789"
def __init__(self, original: str, delete: bool | None = None) -> None:
self.original = original.rstrip("/\\")
super().__init__(delete=delete)
@classmethod
def _generate_names(cls, name: str) -> Generator[str, None, None]:
"""Generates a series of temporary names.
The algorithm replaces the leading characters in the name
with ones that are valid filesystem characters, but are not
valid package names (for both Python and pip definitions of
package).
"""
for i in range(1, len(name)):
for candidate in itertools.combinations_with_replacement(
cls.LEADING_CHARS, i - 1
):
new_name = "~" + "".join(candidate) + name[i:]
if new_name != name:
yield new_name
# If we make it this far, we will have to make a longer name
for i in range(len(cls.LEADING_CHARS)):
for candidate in itertools.combinations_with_replacement(
cls.LEADING_CHARS, i
):
new_name = "~" + "".join(candidate) + name
if new_name != name:
yield new_name
def _create(self, kind: str) -> str:
root, name = os.path.split(self.original)
for candidate in self._generate_names(name):
path = os.path.join(root, candidate)
try:
os.mkdir(path)
except OSError as ex:
# Continue if the name exists already
if ex.errno != errno.EEXIST:
raise
else:
path = os.path.realpath(path)
break
else:
# Final fallback on the default behavior.
path = os.path.realpath(tempfile.mkdtemp(prefix=f"pip-{kind}-"))
logger.debug("Created temporary directory: %s", path)
return path

View File

@@ -0,0 +1,362 @@
"""Utilities related archives."""
from __future__ import annotations
import logging
import os
import shutil
import stat
import sys
import tarfile
import zipfile
from collections.abc import Iterable
from zipfile import ZipInfo
from pip._internal.exceptions import InstallationError
from pip._internal.utils.filetypes import (
BZ2_EXTENSIONS,
TAR_EXTENSIONS,
XZ_EXTENSIONS,
ZIP_EXTENSIONS,
)
from pip._internal.utils.misc import ensure_dir
logger = logging.getLogger(__name__)
SUPPORTED_EXTENSIONS = ZIP_EXTENSIONS + TAR_EXTENSIONS
try:
import bz2 # noqa
SUPPORTED_EXTENSIONS += BZ2_EXTENSIONS
except ImportError:
logger.debug("bz2 module is not available")
try:
# Only for Python 3.3+
import lzma # noqa
SUPPORTED_EXTENSIONS += XZ_EXTENSIONS
except ImportError:
logger.debug("lzma module is not available")
def current_umask() -> int:
"""Get the current umask which involves having to set it temporarily."""
mask = os.umask(0)
os.umask(mask)
return mask
def split_leading_dir(path: str) -> list[str]:
path = path.lstrip("/").lstrip("\\")
if "/" in path and (
("\\" in path and path.find("/") < path.find("\\")) or "\\" not in path
):
return path.split("/", 1)
elif "\\" in path:
return path.split("\\", 1)
else:
return [path, ""]
def has_leading_dir(paths: Iterable[str]) -> bool:
"""Returns true if all the paths have the same leading path name
(i.e., everything is in one subdirectory in an archive)"""
common_prefix = None
for path in paths:
prefix, rest = split_leading_dir(path)
if not prefix:
return False
elif common_prefix is None:
common_prefix = prefix
elif prefix != common_prefix:
return False
return True
def is_within_directory(directory: str, target: str) -> bool:
"""
Return true if the absolute path of target is within the directory
"""
abs_directory = os.path.abspath(directory)
abs_target = os.path.abspath(target)
prefix = os.path.commonpath([abs_directory, abs_target])
return prefix == abs_directory
def _get_default_mode_plus_executable() -> int:
return 0o777 & ~current_umask() | 0o111
def set_extracted_file_to_default_mode_plus_executable(path: str) -> None:
"""
Make file present at path have execute for user/group/world
(chmod +x) is no-op on windows per python docs
"""
os.chmod(path, _get_default_mode_plus_executable())
def zip_item_is_executable(info: ZipInfo) -> bool:
mode = info.external_attr >> 16
# if mode and regular file and any execute permissions for
# user/group/world?
return bool(mode and stat.S_ISREG(mode) and mode & 0o111)
def unzip_file(filename: str, location: str, flatten: bool = True) -> None:
"""
Unzip the file (with path `filename`) to the destination `location`. All
files are written based on system defaults and umask (i.e. permissions are
not preserved), except that regular file members with any execute
permissions (user, group, or world) have "chmod +x" applied after being
written. Note that for windows, any execute changes using os.chmod are
no-ops per the python docs.
"""
ensure_dir(location)
zipfp = open(filename, "rb")
try:
zip = zipfile.ZipFile(zipfp, allowZip64=True)
leading = has_leading_dir(zip.namelist()) and flatten
for info in zip.infolist():
name = info.filename
fn = name
if leading:
fn = split_leading_dir(name)[1]
fn = os.path.join(location, fn)
dir = os.path.dirname(fn)
if not is_within_directory(location, fn):
message = (
"The zip file ({}) has a file ({}) trying to install "
"outside target directory ({})"
)
raise InstallationError(message.format(filename, fn, location))
if fn.endswith(("/", "\\")):
# A directory
ensure_dir(fn)
else:
ensure_dir(dir)
# Don't use read() to avoid allocating an arbitrarily large
# chunk of memory for the file's content
fp = zip.open(name)
try:
with open(fn, "wb") as destfp:
shutil.copyfileobj(fp, destfp)
finally:
fp.close()
if zip_item_is_executable(info):
set_extracted_file_to_default_mode_plus_executable(fn)
finally:
zipfp.close()
def untar_file(filename: str, location: str) -> None:
"""
Untar the file (with path `filename`) to the destination `location`.
All files are written based on system defaults and umask (i.e. permissions
are not preserved), except that regular file members with any execute
permissions (user, group, or world) have "chmod +x" applied on top of the
default. Note that for windows, any execute changes using os.chmod are
no-ops per the python docs.
"""
ensure_dir(location)
if filename.lower().endswith(".gz") or filename.lower().endswith(".tgz"):
mode = "r:gz"
elif filename.lower().endswith(BZ2_EXTENSIONS):
mode = "r:bz2"
elif filename.lower().endswith(XZ_EXTENSIONS):
mode = "r:xz"
elif filename.lower().endswith(".tar"):
mode = "r"
else:
logger.warning(
"Cannot determine compression type for file %s",
filename,
)
mode = "r:*"
tar = tarfile.open(filename, mode, encoding="utf-8") # type: ignore
try:
leading = has_leading_dir([member.name for member in tar.getmembers()])
# PEP 706 added `tarfile.data_filter`, and made some other changes to
# Python's tarfile module (see below). The features were backported to
# security releases.
try:
data_filter = tarfile.data_filter
except AttributeError:
_untar_without_filter(filename, location, tar, leading)
else:
default_mode_plus_executable = _get_default_mode_plus_executable()
if leading:
# Strip the leading directory from all files in the archive,
# including hardlink targets (which are relative to the
# unpack location).
for member in tar.getmembers():
name_lead, name_rest = split_leading_dir(member.name)
member.name = name_rest
if member.islnk():
lnk_lead, lnk_rest = split_leading_dir(member.linkname)
if lnk_lead == name_lead:
member.linkname = lnk_rest
def pip_filter(member: tarfile.TarInfo, path: str) -> tarfile.TarInfo:
orig_mode = member.mode
try:
try:
member = data_filter(member, location)
except tarfile.LinkOutsideDestinationError:
if sys.version_info[:3] in {
(3, 9, 17),
(3, 10, 12),
(3, 11, 4),
}:
# The tarfile filter in specific Python versions
# raises LinkOutsideDestinationError on valid input
# (https://github.com/python/cpython/issues/107845)
# Ignore the error there, but do use the
# more lax `tar_filter`
member = tarfile.tar_filter(member, location)
else:
raise
except tarfile.TarError as exc:
message = "Invalid member in the tar file {}: {}"
# Filter error messages mention the member name.
# No need to add it here.
raise InstallationError(
message.format(
filename,
exc,
)
)
if member.isfile() and orig_mode & 0o111:
member.mode = default_mode_plus_executable
else:
# See PEP 706 note above.
# The PEP changed this from `int` to `Optional[int]`,
# where None means "use the default". Mypy doesn't
# know this yet.
member.mode = None # type: ignore [assignment]
return member
tar.extractall(location, filter=pip_filter)
finally:
tar.close()
def is_symlink_target_in_tar(tar: tarfile.TarFile, tarinfo: tarfile.TarInfo) -> bool:
"""Check if the file pointed to by the symbolic link is in the tar archive"""
linkname = os.path.join(os.path.dirname(tarinfo.name), tarinfo.linkname)
linkname = os.path.normpath(linkname)
linkname = linkname.replace("\\", "/")
try:
tar.getmember(linkname)
return True
except KeyError:
return False
def _untar_without_filter(
filename: str,
location: str,
tar: tarfile.TarFile,
leading: bool,
) -> None:
"""Fallback for Python without tarfile.data_filter"""
# NOTE: This function can be removed once pip requires CPython ≥ 3.12.
# PEP 706 added tarfile.data_filter, made tarfile extraction operations more secure.
# This feature is fully supported from CPython 3.12 onward.
for member in tar.getmembers():
fn = member.name
if leading:
fn = split_leading_dir(fn)[1]
path = os.path.join(location, fn)
if not is_within_directory(location, path):
message = (
"The tar file ({}) has a file ({}) trying to install "
"outside target directory ({})"
)
raise InstallationError(message.format(filename, path, location))
if member.isdir():
ensure_dir(path)
elif member.issym():
if not is_symlink_target_in_tar(tar, member):
message = (
"The tar file ({}) has a file ({}) trying to install "
"outside target directory ({})"
)
raise InstallationError(
message.format(filename, member.name, member.linkname)
)
try:
tar._extract_member(member, path)
except Exception as exc:
# Some corrupt tar files seem to produce this
# (specifically bad symlinks)
logger.warning(
"In the tar file %s the member %s is invalid: %s",
filename,
member.name,
exc,
)
continue
else:
try:
fp = tar.extractfile(member)
except (KeyError, AttributeError) as exc:
# Some corrupt tar files seem to produce this
# (specifically bad symlinks)
logger.warning(
"In the tar file %s the member %s is invalid: %s",
filename,
member.name,
exc,
)
continue
ensure_dir(os.path.dirname(path))
assert fp is not None
with open(path, "wb") as destfp:
shutil.copyfileobj(fp, destfp)
fp.close()
# Update the timestamp (useful for cython compiled files)
tar.utime(member, path)
# member have any execute permissions for user/group/world?
if member.mode & 0o111:
set_extracted_file_to_default_mode_plus_executable(path)
def unpack_file(
filename: str,
location: str,
content_type: str | None = None,
) -> None:
filename = os.path.realpath(filename)
if (
content_type == "application/zip"
or filename.lower().endswith(ZIP_EXTENSIONS)
or zipfile.is_zipfile(filename)
):
unzip_file(filename, location, flatten=not filename.endswith(".whl"))
elif (
content_type == "application/x-gzip"
or tarfile.is_tarfile(filename)
or filename.lower().endswith(TAR_EXTENSIONS + BZ2_EXTENSIONS + XZ_EXTENSIONS)
):
untar_file(filename, location)
else:
# FIXME: handle?
# FIXME: magic signatures?
logger.critical(
"Cannot unpack file %s (downloaded from %s, content-type: %s); "
"cannot detect archive format",
filename,
location,
content_type,
)
raise InstallationError(f"Cannot determine archive format of {location}")

View File

@@ -0,0 +1,55 @@
import os
import string
import urllib.parse
import urllib.request
from .compat import WINDOWS
def path_to_url(path: str) -> str:
"""
Convert a path to a file: URL. The path will be made absolute and have
quoted path parts.
"""
path = os.path.normpath(os.path.abspath(path))
url = urllib.parse.urljoin("file://", urllib.request.pathname2url(path))
return url
def url_to_path(url: str) -> str:
"""
Convert a file: URL to a path.
"""
assert url.startswith(
"file:"
), f"You can only turn file: urls into filenames (not {url!r})"
_, netloc, path, _, _ = urllib.parse.urlsplit(url)
if not netloc or netloc == "localhost":
# According to RFC 8089, same as empty authority.
netloc = ""
elif WINDOWS:
# If we have a UNC path, prepend UNC share notation.
netloc = "\\\\" + netloc
else:
raise ValueError(
f"non-local file URIs are not supported on this platform: {url!r}"
)
path = urllib.request.url2pathname(netloc + path)
# On Windows, urlsplit parses the path as something like "/C:/Users/foo".
# This creates issues for path-related functions like io.open(), so we try
# to detect and strip the leading slash.
if (
WINDOWS
and not netloc # Not UNC.
and len(path) >= 3
and path[0] == "/" # Leading slash to strip.
and path[1] in string.ascii_letters # Drive letter.
and path[2:4] in (":", ":/") # Colon + end of string, or colon + absolute path.
):
path = path[1:]
return path

View File

@@ -0,0 +1,105 @@
from __future__ import annotations
import logging
import os
import re
import site
import sys
logger = logging.getLogger(__name__)
_INCLUDE_SYSTEM_SITE_PACKAGES_REGEX = re.compile(
r"include-system-site-packages\s*=\s*(?P<value>true|false)"
)
def _running_under_venv() -> bool:
"""Checks if sys.base_prefix and sys.prefix match.
This handles PEP 405 compliant virtual environments.
"""
return sys.prefix != getattr(sys, "base_prefix", sys.prefix)
def _running_under_legacy_virtualenv() -> bool:
"""Checks if sys.real_prefix is set.
This handles virtual environments created with pypa's virtualenv.
"""
# pypa/virtualenv case
return hasattr(sys, "real_prefix")
def running_under_virtualenv() -> bool:
"""True if we're running inside a virtual environment, False otherwise."""
return _running_under_venv() or _running_under_legacy_virtualenv()
def _get_pyvenv_cfg_lines() -> list[str] | None:
"""Reads {sys.prefix}/pyvenv.cfg and returns its contents as list of lines
Returns None, if it could not read/access the file.
"""
pyvenv_cfg_file = os.path.join(sys.prefix, "pyvenv.cfg")
try:
# Although PEP 405 does not specify, the built-in venv module always
# writes with UTF-8. (pypa/pip#8717)
with open(pyvenv_cfg_file, encoding="utf-8") as f:
return f.read().splitlines() # avoids trailing newlines
except OSError:
return None
def _no_global_under_venv() -> bool:
"""Check `{sys.prefix}/pyvenv.cfg` for system site-packages inclusion
PEP 405 specifies that when system site-packages are not supposed to be
visible from a virtual environment, `pyvenv.cfg` must contain the following
line:
include-system-site-packages = false
Additionally, log a warning if accessing the file fails.
"""
cfg_lines = _get_pyvenv_cfg_lines()
if cfg_lines is None:
# We're not in a "sane" venv, so assume there is no system
# site-packages access (since that's PEP 405's default state).
logger.warning(
"Could not access 'pyvenv.cfg' despite a virtual environment "
"being active. Assuming global site-packages is not accessible "
"in this environment."
)
return True
for line in cfg_lines:
match = _INCLUDE_SYSTEM_SITE_PACKAGES_REGEX.match(line)
if match is not None and match.group("value") == "false":
return True
return False
def _no_global_under_legacy_virtualenv() -> bool:
"""Check if "no-global-site-packages.txt" exists beside site.py
This mirrors logic in pypa/virtualenv for determining whether system
site-packages are visible in the virtual environment.
"""
site_mod_dir = os.path.dirname(os.path.abspath(site.__file__))
no_global_site_packages_file = os.path.join(
site_mod_dir,
"no-global-site-packages.txt",
)
return os.path.exists(no_global_site_packages_file)
def virtualenv_no_global() -> bool:
"""Returns a boolean, whether running in venv with no system site-packages."""
# PEP 405 compliance needs to be checked first since virtualenv >=20 would
# return True for both checks, but is only able to use the PEP 405 config.
if _running_under_venv():
return _no_global_under_venv()
if _running_under_legacy_virtualenv():
return _no_global_under_legacy_virtualenv()
return False