traverseda
/
cpython
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Issue #25596: Optimized glob() and iglob() functions in the 

glob module; they are now about 3--6 times faster.
This commit is contained in:
Serhiy Storchaka 2016-09-06 22:33:41 +03:00
parent ea525a2d1a
commit 28ab634fa6
4 changed files with 49 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Doc/library/glob.rst
View File

@ -14,7 +14,7 @@ The :mod:`glob` module finds all the pathnames matching a specified pattern
according to the rules used by the Unix shell, although results are returned in
arbitrary order. No tilde expansion is done, but ``*``, ``?``, and character
ranges expressed with ``[]`` will be correctly matched. This is done by using
the :func:`os.listdir` and :func:`fnmatch.fnmatch` functions in concert, and
the :func:`os.scandir` and :func:`fnmatch.fnmatch` functions in concert, and
not by actually invoking a subshell. Note that unlike :func:`fnmatch.fnmatch`,
:mod:`glob` treats filenames beginning with a dot (``.``) as special cases.
(For tilde and shell variable expansion, use :func:`os.path.expanduser` and

4
Doc/whatsnew/3.6.rst
View File

@ -767,6 +767,10 @@ Optimizations
Argument Clinic this overhead is significantly decreased.
(Contributed by Serhiy Storchaka in :issue:`27574`).
* Optimized :func:`~glob.glob` and :func:`~glob.iglob` functions in the
:mod:`glob` module; they are now about 3--6 times faster.
(Contributed by Serhiy Storchaka in :issue:`25596`).
Build and C API Changes
=======================

70
Lib/glob.py
View File

@ -30,15 +30,16 @@ def iglob(pathname, *, recursive=False):
If recursive is true, the pattern '**' will match any files and
zero or more directories and subdirectories.
"""
it = _iglob(pathname, recursive)
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):
def _iglob(pathname, recursive, dironly):
dirname, basename = os.path.split(pathname)
if not has_magic(pathname):
assert not dironly
if basename:
if os.path.lexists(pathname):
yield pathname
@ -49,47 +50,39 @@ def _iglob(pathname, recursive):
return
if not dirname:
if recursive and _isrecursive(basename):
yield from glob2(dirname, basename)
yield from _glob2(dirname, basename, dironly)
else:
yield from glob1(dirname, basename)
yield from _glob1(dirname, basename, dironly)
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):
dirs = _iglob(dirname, recursive)
dirs = _iglob(dirname, recursive, True)
else:
dirs = [dirname]
if has_magic(basename):
if recursive and _isrecursive(basename):
glob_in_dir = glob2
glob_in_dir = _glob2
else:
glob_in_dir = glob1
glob_in_dir = _glob1
else:
glob_in_dir = glob0
glob_in_dir = _glob0
for dirname in dirs:
for name in glob_in_dir(dirname, basename):
for name in glob_in_dir(dirname, basename, dironly):
yield os.path.join(dirname, name)
# These 2 helper functions non-recursively glob inside a literal directory.
# They return a list of basenames. `glob1` accepts a pattern while `glob0`
# They return a list of basenames. _glob1 accepts a pattern while _glob0
# takes a literal basename (so it only has to check for its existence).
def glob1(dirname, pattern):
if not dirname:
if isinstance(pattern, bytes):
dirname = bytes(os.curdir, 'ASCII')
else:
dirname = os.curdir
try:
names = os.listdir(dirname)
except OSError:
return []
def _glob1(dirname, pattern, dironly):
names = list(_iterdir(dirname, dironly))
if not _ishidden(pattern):
names = [x for x in names if not _ishidden(x)]
names = (x for x in names if not _ishidden(x))
return fnmatch.filter(names, pattern)
def glob0(dirname, basename):
def _glob0(dirname, basename, dironly):
if not basename:
# `os.path.split()` returns an empty basename for paths ending with a
# directory separator. 'q*x/' should match only directories.
@ -100,30 +93,49 @@ def glob0(dirname, basename):
return [basename]
return []
# 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)
# This helper function recursively yields relative pathnames inside a literal
# directory.
def glob2(dirname, pattern):
def _glob2(dirname, pattern, dironly):
assert _isrecursive(pattern)
yield pattern[:0]
yield from _rlistdir(dirname)
yield from _rlistdir(dirname, dironly)
# Recursively yields relative pathnames inside a literal directory.
def _rlistdir(dirname):
# If dironly is false, yields all file names inside a directory.
# If dironly is true, yields only directory names.
def _iterdir(dirname, dironly):
if not dirname:
if isinstance(dirname, bytes):
dirname = bytes(os.curdir, 'ASCII')
else:
dirname = os.curdir
try:
names = os.listdir(dirname)
except os.error:
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:
return
# Recursively yields relative pathnames inside a literal directory.
def _rlistdir(dirname, dironly):
names = list(_iterdir(dirname, dironly))
for x in names:
if not _ishidden(x):
yield x
path = os.path.join(dirname, x) if dirname else x
for y in _rlistdir(path):
for y in _rlistdir(path, dironly):
yield os.path.join(x, y)

3
Misc/NEWS
View File

@ -89,6 +89,9 @@ Core and Builtins
Library
-------
- Issue #25596: Optimized glob() and iglob() functions in the
glob module; they are now about 3--6 times faster.
- Issue #27928: Add scrypt (password-based key derivation function) to
hashlib module (requires OpenSSL 1.1.0).