"""Common operations on Posix pathnames.
Instead of importing this module directly, import os and refer to
this module as os.path. The "os.path" name is an alias for this
module on Posix systems; on other systems (e.g. Mac, Windows),
os.path provides the same operations in a manner specific to that
platform, and is an alias to another module (e.g. macpath, ntpath).
Some of this can actually be useful on non-Posix systems too, e.g.
for manipulation of the pathname component of URLs.
from genericpath import *
from genericpath import _unicode
__all__ = ["normcase","isabs","join","splitdrive","split","splitext",
"basename","dirname","commonprefix","getsize","getmtime",
"getatime","getctime","islink","exists","lexists","isdir","isfile",
"ismount","walk","expanduser","expandvars","normpath","abspath",
"samefile","sameopenfile","samestat",
"curdir","pardir","sep","pathsep","defpath","altsep","extsep",
"devnull","realpath","supports_unicode_filenames","relpath"]
# strings representing various path-related bits and pieces
defpath = ':/bin:/usr/bin'
# Normalize the case of a pathname. Trivial in Posix, string.lower on Mac.
# On MS-DOS this may also turn slashes into backslashes; however, other
# normalizations (such as optimizing '../' away) are not allowed
# (another function should be defined to do that).
"""Normalize case of pathname. Has no effect under Posix"""
# Return whether a path is absolute.
# Trivial in Posix, harder on the Mac or MS-DOS.
"""Test whether a path is absolute"""
# Ignore the previous parts if a part is absolute.
# Insert a '/' unless the first part is empty or already ends in '/'.
"""Join two or more pathname components, inserting '/' as needed.
If any component is an absolute path, all previous path components
will be discarded. An empty last part will result in a path that
ends with a separator."""
elif path == '' or path.endswith('/'):
# Split a path in head (everything up to the last '/') and tail (the
# rest). If the path ends in '/', tail will be empty. If there is no
# '/' in the path, head will be empty.
# Trailing '/'es are stripped from head unless it is the root.
"""Split a pathname. Returns tuple "(head, tail)" where "tail" is
everything after the final slash. Either part may be empty."""
head, tail = p[:i], p[i:]
if head and head != '/'*len(head):
# Split a path in root and extension.
# The extension is everything starting at the last dot in the last
# pathname component; the root is everything before that.
# It is always true that root + ext == p.
return genericpath._splitext(p, sep, altsep, extsep)
splitext.__doc__ = genericpath._splitext.__doc__
# Split a pathname into a drive specification and the rest of the
# path. Useful on DOS/Windows/NT; on Unix, the drive is always empty.
"""Split a pathname into drive and path. On Posix, drive is always
# Return the tail (basename) part of a path, same as split(path)[1].
"""Returns the final component of a pathname"""
# Return the head (dirname) part of a path, same as split(path)[0].
"""Returns the directory component of a pathname"""
if head and head != '/'*len(head):
# Is a path a symbolic link?
# This will always return false on systems where os.lstat doesn't exist.
"""Test whether a path is a symbolic link"""
except (os.error, AttributeError):
return stat.S_ISLNK(st.st_mode)
# Being true for dangling symbolic links is also useful.
"""Test whether a path exists. Returns True for broken symbolic links"""
# Are two filenames really pointing to the same file?
"""Test whether two pathnames reference the same actual file"""
# Are two open files really referencing the same file?
# (Not necessarily the same file descriptor!)
def sameopenfile(fp1, fp2):
"""Test whether two open file objects reference the same file"""
# Are two stat buffers (obtained from stat, fstat or lstat)
# describing the same file?
"""Test whether two stat buffers reference the same file"""
return s1.st_ino == s2.st_ino and \
# Is a path a mount point?
# (Does this work for all UNIXes? Is it even guaranteed to work by Posix?)
"""Test whether a path is a mount point"""
# A symlink can never be a mount point
s2 = os.lstat(join(path, '..'))
return False # It doesn't exist -- so not a mount point :-)
return True # path/.. on a different device as path
return True # path/.. is the same i-node as path
# For each directory under top (including top itself, but excluding
# '.' and '..'), func(arg, dirname, filenames) is called, where
# dirname is the name of the directory and filenames is the list
# of files (and subdirectories etc.) in the directory.
# The func may modify the filenames list, to implement a filter,
# or to impose a different order of visiting.
def walk(top, func, arg):
"""Directory tree walk with callback function.
For each directory in the directory tree rooted at top (including top
itself, but excluding '.' and '..'), call func(arg, dirname, fnames).
dirname is the name of the directory, and fnames a list of the names of
the files and subdirectories in dirname (excluding '.' and '..'). func
may modify the fnames list in-place (e.g. via del or slice assignment),
and walk will only recurse into the subdirectories whose names remain in
fnames; this can be used to implement a filter, or to impose a specific
order of visiting. No semantics are defined for, or required of, arg,
beyond that arg is always passed to func. It can be used, e.g., to pass
a filename pattern, or a mutable object designed to accumulate
statistics. Passing None for arg is common."""
warnings.warnpy3k("In 3.x, os.path.walk is removed in favor of os.walk.",
if stat.S_ISDIR(st.st_mode):
# Expand paths beginning with '~' or '~user'.
# '~' means $HOME; '~user' means that user's home directory.
# If the path doesn't begin with '~', or if the user or $HOME is unknown,
# the path is returned unchanged (leaving error reporting to whatever
# function is called with the expanded path as argument).
# See also module 'glob' for expansion of *, ? and [...] in pathnames.
# (A function should also be defined to do full *sh-style environment
"""Expand ~ and ~user constructions. If user or $HOME is unknown,
if not path.startswith('~'):
if 'HOME' not in os.environ:
userhome = pwd.getpwuid(os.getuid()).pw_dir
userhome = os.environ['HOME']
pwent = pwd.getpwnam(path[1:i])
userhome = userhome.rstrip('/')
return (userhome + path[i:]) or '/'
# Expand paths containing shell variable substitutions.
# This expands the forms $variable and ${variable} only.
# Non-existent variables are left unchanged.
"""Expand shell variables of form $var and ${var}. Unknown variables
global _varprog, _uvarprog
if isinstance(path, _unicode):
_uvarprog = re.compile(ur'\$(\w+|\{[^}]*\})', re.UNICODE)
encoding = sys.getfilesystemencoding()
_varprog = re.compile(r'\$(\w+|\{[^}]*\})')
m = varprog.search(path, i)
if name.startswith('{') and name.endswith('}'):
name = name.encode(encoding)
value = value.decode(encoding)
# Normalize a path, e.g. A//B, A/./B and A/foo/../B all become A/B.
# It should be understood that this may change the meaning of the path
# if it contains symbolic links!
"""Normalize path, eliminating double slashes, etc."""
# Preserve unicode (if path is unicode)
slash, dot = (u'/', u'.') if isinstance(path, _unicode) else ('/', '.')
initial_slashes = path.startswith('/')
# POSIX allows one or two initial slashes, but treats three or more
path.startswith('//') and not path.startswith('///')):
if (comp != '..' or (not initial_slashes and not new_comps) or
(new_comps and new_comps[-1] == '..')):
path = slash*initial_slashes + path
"""Return an absolute path."""
if isinstance(path, _unicode):
# Return a canonical path (i.e. the absolute location of a file on the
"""Return the canonical path of the specified filename, eliminating any
symbolic links encountered in the path."""
path, ok = _joinrealpath('', filename, {})
# Join two paths, normalizing and eliminating any symbolic links
# encountered in the second path.
def _joinrealpath(path, rest, seen):
name, _, rest = rest.partition(sep)
if not name or name == curdir:
path = join(path, pardir, pardir)
newpath = join(path, name)
# Resolve the symbolic link
# The symlink is not resolved, so we must have a symlink loop.
# Return already resolved part + rest of the path unchanged.
return join(newpath, rest), False
seen[newpath] = None # not resolved symlink
path, ok = _joinrealpath(path, os.readlink(newpath), seen)
return join(path, rest), False
seen[newpath] = path # resolved symlink
supports_unicode_filenames = (sys.platform == 'darwin')
def relpath(path, start=curdir):
"""Return a relative version of a path"""
raise ValueError("no path specified")
start_list = [x for x in abspath(start).split(sep) if x]
path_list = [x for x in abspath(path).split(sep) if x]
# Work out how much of the filepath is shared by start and path.
i = len(commonprefix([start_list, path_list]))
rel_list = [pardir] * (len(start_list)-i) + path_list[i:]