Bug report

Tobias Dammers tdammers at gmail.com
Sat Nov 20 13:59:15 CST 2010


No, that doesn't print anything at all.

Commenting out the offending line makes it crash two lines further,
which means quite certainly the "crev" variable is the culprit. As the
exception dump says, it's "None", but as far as I can tell, it should
be numeric.

If anyone feels like trying to reproduce the problem (or tell me what
I did wrong), the offending repo is at
http://bitbucket.org/tdammers/cantroid; I've attached the exact filter
files I'm using.


On Sat, Nov 20, 2010 at 6:26 PM, Dennis Brakhane
<brakhane at googlemail.com> wrote:
> On Sat, Nov 20, 2010 at 5:52 PM, Tobias Dammers <tdammers at gmail.com> wrote:
>> 1.7.1 gives the exact same error, just different line numbers:
>
> I don't know enough of Mercurial's internals to give useful advice, I'm afraid.
>
> You could try the latest version in hg-crew, maybe that fixes the bug (but I
> don't think so).
>
> You could add the following line before the 827 line in
> lib/python/mercurial/localrepo.py:
>
>  print "%s: copy: %s" % (fname, cfname)
>
> And then see what line is printed right before the crash, maybe you
> the name that will be output will help you remember what might be
> different.
>
> Keep in mind that in python indentation matters, be sure to indent it
> exactly like the following self.ui.debug line (also don't use tabs, only spaces
> to indent).
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cantroid_filemap
Type: application/octet-stream
Size: 273 bytes
Desc: not available
URL: <http://selenic.com/pipermail/mercurial/attachments/20101120/2da8dc4c/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cantroid_usermap
Type: application/octet-stream
Size: 159 bytes
Desc: not available
URL: <http://selenic.com/pipermail/mercurial/attachments/20101120/2da8dc4c/attachment-0001.obj>


More information about the Mercurial mailing list