Closed GoogleCodeExporter closed 9 years ago
[deleted comment]
--- tmp/tmpAjDQ-T-meld/fs_impl/fs_gnomevfs.py
+++
home/kai/src/hotwire/hotwire-shell-read-only/hotwire/sysdep/fs_impl/fs_gnomevfs.
py
@@ -47,7 +47,7 @@
self.target_vfsstat = None
self.target_vfsstat_error = None
- def test_directory(self, follow_link=False):
+ def test_directory(self, follow_link=True):
if not self.vfsstat:
return False
if follow_link and self.vfsstat.type == gnomevfs.FILE_TYPE_SYMBOLIC_LINK:
--- tmp/tmpAjDQ-T-meld/fs.py
+++ home/kai/src/hotwire/hotwire-shell-read-only/hotwire/sysdep/fs.py
@@ -187,7 +187,7 @@
self.target_stat = None
self.stat_error = None
- def test_directory(self, follow_link=False):
+ def test_directory(self, follow_link=True):
if not self.stat:
return False
if follow_link and stat.S_ISLNK(self.stat.st_mode):
Original comment by KaiSchro...@gmail.com
on 24 Jan 2008 at 5:09
Ok, I see the bug. I'm a bit nervous about this change though - the
test_directory
is used in a number of places and we'd need to double-check all of the callers.
I'm
pretty sure there is at least a case or two in the code where we don't want to
follow
the link.
I'll take a closer look tonight.
Original comment by cgwalt...@gmail.com
on 24 Jan 2008 at 6:20
Note that not only all users of test_directory() had to be audited, but all
users of
"is_directory" as well, because it changed too with this patch.
I did a pass through the code and didn't see anything that should break (and I
noticed this patch fixed one other bug where we added a space after a symlink
to a
directory).
Still though we should keep an eye out for any regressions. The current test
suite
isn't yet very comprehensive.
Anyways, committed. Thank you for the patch!
Committed r885
M hotwire/sysdep/fs.py
M hotwire/sysdep/fs_impl/fs_gnomevfs.py
r885 = 2faa0e634098b1345061813164e929896c403528 (git-svn)
Original comment by cgwalt...@gmail.com
on 26 Jan 2008 at 3:21
Original issue reported on code.google.com by
KaiSchro...@gmail.com
on 24 Jan 2008 at 5:03