windows: check target type before actual unlinking to follow POSIX semantics stable

windows: check target type before actual unlinking to follow POSIX semantics

Creation and writing into target file via vfs (a.k.a opener) is done
after "unlink()" target file, if it exists.

For example, it is assumed that the revision X consists of file 'A',
and the revision Y consists of file 'A/B'. Merging revision X into Y
tries to "unlink()" on directory 'A' of 'A/B', before creation of file
'A'.

On POSIX environment, directories should be removed by "rmdir(2)", and
"unlink(2)" on directories fails. "unlink()" of Mercurial (and Python)
uses "unlink(2)" directly, so unlinking in the merge case above would
fail.

In the other hand, on Windows environment, "unlink()" of Mercurial
tries to rename before actual unlinking, to follow POSIX semantics:
already opened file can be unlinked safely.

This causes unexpected success in unlinking in the merge case above,
even though directory 'A' is renamed to another. This confuses users.

This patch checks whether target is directory or not before renaming,
and raises IOError(errno.EPERM) if so, to follow POSIX semantics.

--- a/mercurial/win32.py	Thu May 09 21:09:58 2013 +0900
+++ b/mercurial/win32.py	Tue May 07 05:04:11 2013 +0900
@@ -344,6 +344,12 @@
 def unlink(f):
     '''try to implement POSIX' unlink semantics on Windows'''
 
+    if os.path.isdir(f):
+        # use EPERM because it is POSIX prescribed value, even though
+        # unlink(2) on directories returns EISDIR on Linux
+        raise IOError(errno.EPERM,
+                      "Unlinking directory not permitted: '%s'" % f)
+
     # POSIX allows to unlink and rename open files. Windows has serious
     # problems with doing that:
     # - Calling os.unlink (or os.rename) on a file f fails if f or any