Charles-François Natali <neolo...@free.fr> added the comment:
Here's a possible walkfd() implementation.
Example:
"""
$ cat /home/cf/testwalkfd.py
import os
import sys
topfd = os.open(sys.argv[1], os.O_RDONLY)
for rootfd, dirs, files in os.walkfd(topfd):
print(rootfd, dirs, files)
$ ./python ~/testwalkfd.py /etc/apt/
3 ['sources.list.d', 'preferences.d', 'trusted.gpg.d', 'apt.conf.d']
['trustdb.gpg', 'trusted.gpg~', 'sources.list', 'trusted.gpg']
4 [] []
4 [] []
4 [] []
4 [] ['70debconf', '01autoremove', '00trustcdrom']
[44194 refs]
"""
AFAICT, a safe rmtree could be implemented simply with
walkfd(topdown=False), but Antoine's remarks make me thing I missed
something.
> Be aware that you have to manage dirfd's lifetime, which can make things
> interesting.
Basically, this means that doing:
for rootfd, dirs, files in walkfd(topfd):
print(fstat(rootfd), dirs, files))
is valid whereas
print([(fstat(rootfd), dirs, files) for (rootfd, dirs, files) in
walkfd(topfd)]) isn't.
> Also be aware that symlinks mean sometimes you won't have a dirfd: if
> you have a symlink that points to another directory, you can't open that
> directory using openat from the symlink's directory. So if you follow
> symlinks (or have an option to do so) you must also take that case into
> account.
I'm not sure I understand this. Why "you can't open that directory
using openat from the symlink's directory". Could you elaborate?
----------
keywords: +patch
Added file: http://bugs.python.org/file24176/walkfd.diff
_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue13734>
_______________________________________diff --git a/Lib/os.py b/Lib/os.py
--- a/Lib/os.py
+++ b/Lib/os.py
@@ -24,6 +24,7 @@
#'
import sys, errno
+import stat as st
_names = sys.builtin_module_names
@@ -151,7 +152,6 @@
try:
mkdir(name, mode)
except OSError as e:
- import stat as st
if not (e.errno == errno.EEXIST and exist_ok and path.isdir(name) and
st.S_IMODE(lstat(name).st_mode) == _get_masked_mode(mode)):
raise
@@ -298,6 +298,78 @@
__all__.append("walk")
+def _are_same_file(stat1, stat2):
+ """Helper function that checks whether two stat results refer to the same
+ file.
+ """
+ return (stat1.st_mode == stat2.st_mode and stat1.st_ino == stat2.st_ino and
+ stat1.st_dev == stat2.st_dev)
+
+def walkfd(topfd, topdown=True, onerror=None, followlinks=False):
+ """Directory tree generator.
+
+ This behaves exactly like walk(), except that it accepts a file descriptor
+ as top directory, and yields a 3-tuple
+
+ dirfd, dirnames, filenames
+
+ dirfd is a file descriptor referring to the directory. dirnames is a list
+ of the names of the subdirectories in dirpath (excluding '.' and '..').
+ filenames is a list of the names of the non-directory files in dirpath.
+
+ The advantage of walkfd() over walk() is that it's safe against symlink
+ races (when followlinks is False).
+ """
+ # fdlistdir() closes the passed FD, hence the dup()
+ fd = dup(topfd)
+ try:
+ names = fdlistdir(fd)
+ except error as err:
+ if onerror is not None:
+ onerror(err)
+ return
+
+ # whether to follow symlinks
+ flag = 0 if followlinks else AT_SYMLINK_NOFOLLOW
+
+ dirs, dirmodes, nondirs = [], [], []
+ for name in names:
+ try:
+ orig_st = fstatat(topfd, name, flag)
+ except error as err:
+ if onerror is not None:
+ onerror(err)
+ return
+ if st.S_ISDIR(orig_st.st_mode):
+ # Store the result of stat to check that the file hasn't been
+ # modified when we call walkfd() recursively (symlink race).
+ dirs.append(name)
+ dirmodes.append(orig_st)
+ else:
+ nondirs.append(name)
+
+ if topdown:
+ yield topfd, dirs, nondirs
+ for name, orig_st in zip(dirs, dirmodes):
+ try:
+ dirfd = openat(topfd, name, O_RDONLY)
+ except error as err:
+ if onerror is not None:
+ onerror(err)
+ return
+ try:
+ # To guard against symlinks race, compare with the original stat
+ # result.
+ if followlinks or _are_same_file(orig_st, fstat(dirfd)):
+ for x in walkfd(dirfd, topdown, onerror, followlinks):
+ yield x
+ finally:
+ close(dirfd)
+ if not topdown:
+ yield topfd, dirs, nondirs
+
+__all__.append("walkfd")
+
# Make sure os.environ exists, at least
try:
environ
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
