cpython/Lib/fnmatch.py

193 lines
6.0 KiB
Python
Raw Normal View History

1995-01-26 22:41:45 -04:00
"""Filename matching with shell patterns.
1992-01-12 19:29:29 -04:00
1995-01-26 22:41:45 -04:00
fnmatch(FILENAME, PATTERN) matches according to the local convention.
fnmatchcase(FILENAME, PATTERN) always takes case in account.
1992-01-12 19:29:29 -04:00
1995-01-26 22:41:45 -04:00
The functions operate by translating the pattern into a regular
expression. They cache the compiled regular expressions for speed.
The function translate(PATTERN) returns a regular expression
corresponding to PATTERN. (It does not compile it.)
"""
import os
import posixpath
import re
2010-08-13 13:26:40 -03:00
import functools
2010-08-13 13:26:40 -03:00
__all__ = ["filter", "fnmatch", "fnmatchcase", "translate"]
1991-01-01 14:11:14 -04:00
def fnmatch(name, pat):
2001-01-14 19:36:06 -04:00
"""Test whether FILENAME matches PATTERN.
Patterns are Unix shell style:
* matches everything
? matches any single character
[seq] matches any character in seq
[!seq] matches any char not in seq
An initial period in FILENAME is not special.
Both FILENAME and PATTERN are first case-normalized
if the operating system requires it.
If you don't want this, use fnmatchcase(FILENAME, PATTERN).
"""
name = os.path.normcase(name)
pat = os.path.normcase(pat)
return fnmatchcase(name, pat)
1995-01-26 22:41:45 -04:00
@functools.lru_cache(maxsize=32768, typed=True)
2011-10-20 13:22:10 -03:00
def _compile_pattern(pat):
if isinstance(pat, bytes):
2010-08-13 13:26:40 -03:00
pat_str = str(pat, 'ISO-8859-1')
res_str = translate(pat_str)
res = bytes(res_str, 'ISO-8859-1')
else:
res = translate(pat)
return re.compile(res).match
def filter(names, pat):
"""Construct a list from those elements of the iterable NAMES that match PAT."""
result = []
pat = os.path.normcase(pat)
2011-10-20 13:22:10 -03:00
match = _compile_pattern(pat)
if os.path is posixpath:
# normcase on posix is NOP. Optimize it away from the loop.
for name in names:
if match(name):
result.append(name)
else:
for name in names:
if match(os.path.normcase(name)):
result.append(name)
return result
1995-01-26 22:41:45 -04:00
def fnmatchcase(name, pat):
2001-01-14 19:36:06 -04:00
"""Test whether FILENAME matches PATTERN, including case.
This is a version of fnmatch() which doesn't case-normalize
its arguments.
"""
2011-10-20 13:22:10 -03:00
match = _compile_pattern(pat)
return match(name) is not None
1991-01-01 14:11:14 -04:00
1992-01-12 19:29:29 -04:00
def translate(pat):
2001-01-14 19:36:06 -04:00
"""Translate a shell PATTERN to a regular expression.
There is no way to quote meta-characters.
"""
STAR = object()
parts = _translate(pat, STAR, '.')
return _join_translated_parts(parts, STAR)
def _translate(pat, STAR, QUESTION_MARK):
res = []
add = res.append
2001-01-14 19:36:06 -04:00
i, n = 0, len(pat)
while i < n:
c = pat[i]
i = i+1
if c == '*':
# compress consecutive `*` into one
if (not res) or res[-1] is not STAR:
add(STAR)
2001-01-14 19:36:06 -04:00
elif c == '?':
add(QUESTION_MARK)
2001-01-14 19:36:06 -04:00
elif c == '[':
j = i
if j < n and pat[j] == '!':
j = j+1
if j < n and pat[j] == ']':
j = j+1
while j < n and pat[j] != ']':
j = j+1
if j >= n:
add('\\[')
2001-01-14 19:36:06 -04:00
else:
stuff = pat[i:j]
if '-' not in stuff:
stuff = stuff.replace('\\', r'\\')
else:
chunks = []
k = i+2 if pat[i] == '!' else i+1
while True:
k = pat.find('-', k, j)
if k < 0:
break
chunks.append(pat[i:k])
i = k+1
k = k+3
chunk = pat[i:j]
if chunk:
chunks.append(chunk)
else:
chunks[-1] += '-'
# Remove empty ranges -- invalid in RE.
for k in range(len(chunks)-1, 0, -1):
if chunks[k-1][-1] > chunks[k][0]:
chunks[k-1] = chunks[k-1][:-1] + chunks[k][1:]
del chunks[k]
# Escape backslashes and hyphens for set difference (--).
# Hyphens that create ranges shouldn't be escaped.
stuff = '-'.join(s.replace('\\', r'\\').replace('-', r'\-')
for s in chunks)
# Escape set operations (&&, ~~ and ||).
stuff = re.sub(r'([&~|])', r'\\\1', stuff)
2001-01-14 19:36:06 -04:00
i = j+1
if not stuff:
# Empty range: never match.
add('(?!)')
elif stuff == '!':
# Negated empty range: match any character.
add('.')
else:
if stuff[0] == '!':
stuff = '^' + stuff[1:]
elif stuff[0] in ('^', '['):
stuff = '\\' + stuff
add(f'[{stuff}]')
2001-01-14 19:36:06 -04:00
else:
add(re.escape(c))
assert i == n
return res
def _join_translated_parts(inp, STAR):
# Deal with STARs.
res = []
add = res.append
i, n = 0, len(inp)
# Fixed pieces at the start?
while i < n and inp[i] is not STAR:
add(inp[i])
i += 1
# Now deal with STAR fixed STAR fixed ...
# For an interior `STAR fixed` pairing, we want to do a minimal
# .*? match followed by `fixed`, with no possibility of backtracking.
# Atomic groups ("(?>...)") allow us to spell that directly.
# Note: people rely on the undocumented ability to join multiple
# translate() results together via "|" to build large regexps matching
# "one of many" shell patterns.
while i < n:
assert inp[i] is STAR
i += 1
if i == n:
add(".*")
break
assert inp[i] is not STAR
fixed = []
while i < n and inp[i] is not STAR:
fixed.append(inp[i])
i += 1
fixed = "".join(fixed)
if i == n:
add(".*")
add(fixed)
else:
add(f"(?>.*?{fixed})")
assert i == n
res = "".join(res)
return fr'(?s:{res})\Z'