combined patches from http://bugs.python.org/issue3187

Lib/posixpath.py

 """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.
 """
 
 import os
+import sys
 import stat
 import genericpath
 from genericpath import *
 
 __all__ = ["normcase","isabs","join","splitdrive","split","splitext",
            "basename","dirname","commonprefix","getsize","getmtime",
            "getatime","getctime","islink","exists","lexists","isdir","isfile",
            "ismount", "expanduser","expandvars","normpath","abspath",
            "samefile","sameopenfile","samestat",
            "curdir","pardir","sep","pathsep","defpath","altsep","extsep",
@@ skipping 28 matching lines @@
 
 
 # Join pathnames.
 # Ignore the previous parts if a part is absolute.
 # Insert a '/' unless the first part is empty or already ends in '/'.
 
 def join(a, *p):
     """Join two or more pathname components, inserting '/' as needed.
     If any component is an absolute path, all previous path components
     will be discarded."""
+    use_bytes = any( isinstance(part, bytes) for part in (a,)+p )

[GvR, 2008/08/22 19:04:51]

Style nit: don't put spaces within the ( parentheses ).

     path = a
+    sep = '/'
+    if use_bytes:

[GvR, 2008/08/22 19:04:51]

Again, I'd recommend changing this so that the arguments must either be all str
or all bytes.

+        charset = sys.getfilesystemencoding()
+        sep = sep.encode(charset)
+        if use_bytes and isinstance(path, str):
+            path = path.encode(charset)
     for b in p:
-        if b.startswith('/'):
+        if use_bytes and isinstance(b, str):
+            b = b.encode(charset)
+        if b.startswith(sep):
             path = b
-        elif path == '' or path.endswith('/'):
+        elif path == '' or path.endswith(sep):
             path +=  b
         else:
-            path += '/' + b
+            path += sep + b
     return path
 
 
 # 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.
 
 def split(p):
     """Split a pathname.  Returns tuple "(head, tail)" where "tail" is
@@ skipping 276 matching lines @@
     start_list = abspath(start).split(sep)
     path_list = abspath(path).split(sep)
 
     # 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:]
     if not rel_list:
         return curdir
     return join(*rel_list)