openmedialibrary_platform_d.../lib/python3.7/glob.py

172 lines
5.5 KiB
Python
Raw Permalink Normal View History

2016-02-06 09:36:57 +00:00
"""Filename globbing utility."""
import os
import re
import fnmatch
2018-12-31 23:25:26 +00:00
__all__ = ["glob", "iglob", "escape"]
2016-02-06 09:36:57 +00:00
def glob(pathname, *, recursive=False):
"""Return a list of paths matching a pathname pattern.
The pattern may contain simple shell-style wildcards a la
fnmatch. However, unlike fnmatch, filenames starting with a
dot are special cases that are not matched by '*' and '?'
patterns.
If recursive is true, the pattern '**' will match any files and
zero or more directories and subdirectories.
"""
return list(iglob(pathname, recursive=recursive))
def iglob(pathname, *, recursive=False):
"""Return an iterator which yields the paths matching a pathname pattern.
The pattern may contain simple shell-style wildcards a la
fnmatch. However, unlike fnmatch, filenames starting with a
dot are special cases that are not matched by '*' and '?'
patterns.
If recursive is true, the pattern '**' will match any files and
zero or more directories and subdirectories.
"""
2018-12-31 23:25:26 +00:00
it = _iglob(pathname, recursive, False)
if recursive and _isrecursive(pathname):
s = next(it) # skip empty string
assert not s
return it
def _iglob(pathname, recursive, dironly):
2016-02-06 09:36:57 +00:00
dirname, basename = os.path.split(pathname)
if not has_magic(pathname):
2018-12-31 23:25:26 +00:00
assert not dironly
2016-02-06 09:36:57 +00:00
if basename:
if os.path.lexists(pathname):
yield pathname
else:
# Patterns ending with a slash should match only directories
if os.path.isdir(dirname):
yield pathname
return
if not dirname:
if recursive and _isrecursive(basename):
2018-12-31 23:25:26 +00:00
yield from _glob2(dirname, basename, dironly)
2016-02-06 09:36:57 +00:00
else:
2018-12-31 23:25:26 +00:00
yield from _glob1(dirname, basename, dironly)
2016-02-06 09:36:57 +00:00
return
# `os.path.split()` returns the argument itself as a dirname if it is a
# drive or UNC path. Prevent an infinite recursion if a drive or UNC path
# contains magic characters (i.e. r'\\?\C:').
if dirname != pathname and has_magic(dirname):
2018-12-31 23:25:26 +00:00
dirs = _iglob(dirname, recursive, True)
2016-02-06 09:36:57 +00:00
else:
dirs = [dirname]
if has_magic(basename):
if recursive and _isrecursive(basename):
2018-12-31 23:25:26 +00:00
glob_in_dir = _glob2
2016-02-06 09:36:57 +00:00
else:
2018-12-31 23:25:26 +00:00
glob_in_dir = _glob1
2016-02-06 09:36:57 +00:00
else:
2018-12-31 23:25:26 +00:00
glob_in_dir = _glob0
2016-02-06 09:36:57 +00:00
for dirname in dirs:
2018-12-31 23:25:26 +00:00
for name in glob_in_dir(dirname, basename, dironly):
2016-02-06 09:36:57 +00:00
yield os.path.join(dirname, name)
# These 2 helper functions non-recursively glob inside a literal directory.
2018-12-31 23:25:26 +00:00
# They return a list of basenames. _glob1 accepts a pattern while _glob0
2016-02-06 09:36:57 +00:00
# takes a literal basename (so it only has to check for its existence).
2018-12-31 23:25:26 +00:00
def _glob1(dirname, pattern, dironly):
names = list(_iterdir(dirname, dironly))
2016-02-06 09:36:57 +00:00
if not _ishidden(pattern):
2018-12-31 23:25:26 +00:00
names = (x for x in names if not _ishidden(x))
2016-02-06 09:36:57 +00:00
return fnmatch.filter(names, pattern)
2018-12-31 23:25:26 +00:00
def _glob0(dirname, basename, dironly):
2016-02-06 09:36:57 +00:00
if not basename:
# `os.path.split()` returns an empty basename for paths ending with a
# directory separator. 'q*x/' should match only directories.
if os.path.isdir(dirname):
return [basename]
else:
if os.path.lexists(os.path.join(dirname, basename)):
return [basename]
return []
2018-12-31 23:25:26 +00:00
# Following functions are not public but can be used by third-party code.
def glob0(dirname, pattern):
return _glob0(dirname, pattern, False)
def glob1(dirname, pattern):
return _glob1(dirname, pattern, False)
2016-02-06 09:36:57 +00:00
# This helper function recursively yields relative pathnames inside a literal
# directory.
2018-12-31 23:25:26 +00:00
def _glob2(dirname, pattern, dironly):
2016-02-06 09:36:57 +00:00
assert _isrecursive(pattern)
2018-12-31 23:25:26 +00:00
yield pattern[:0]
yield from _rlistdir(dirname, dironly)
2016-02-06 09:36:57 +00:00
2018-12-31 23:25:26 +00:00
# If dironly is false, yields all file names inside a directory.
# If dironly is true, yields only directory names.
def _iterdir(dirname, dironly):
2016-02-06 09:36:57 +00:00
if not dirname:
if isinstance(dirname, bytes):
dirname = bytes(os.curdir, 'ASCII')
else:
dirname = os.curdir
try:
2018-12-31 23:25:26 +00:00
with os.scandir(dirname) as it:
for entry in it:
try:
if not dironly or entry.is_dir():
yield entry.name
except OSError:
pass
except OSError:
2016-02-06 09:36:57 +00:00
return
2018-12-31 23:25:26 +00:00
# Recursively yields relative pathnames inside a literal directory.
def _rlistdir(dirname, dironly):
names = list(_iterdir(dirname, dironly))
2016-02-06 09:36:57 +00:00
for x in names:
if not _ishidden(x):
yield x
path = os.path.join(dirname, x) if dirname else x
2018-12-31 23:25:26 +00:00
for y in _rlistdir(path, dironly):
2016-02-06 09:36:57 +00:00
yield os.path.join(x, y)
magic_check = re.compile('([*?[])')
magic_check_bytes = re.compile(b'([*?[])')
def has_magic(s):
if isinstance(s, bytes):
match = magic_check_bytes.search(s)
else:
match = magic_check.search(s)
return match is not None
def _ishidden(path):
return path[0] in ('.', b'.'[0])
def _isrecursive(pattern):
if isinstance(pattern, bytes):
return pattern == b'**'
else:
return pattern == '**'
def escape(pathname):
"""Escape all special characters.
"""
# Escaping is done by wrapping any of "*?[" between square brackets.
# Metacharacters do not work in the drive part and shouldn't be escaped.
drive, pathname = os.path.splitdrive(pathname)
if isinstance(pathname, bytes):
pathname = magic_check_bytes.sub(br'[\1]', pathname)
else:
pathname = magic_check.sub(r'[\1]', pathname)
return drive + pathname