On Wednesday, 10 May 2006 at 08:31, Bryan O'Sullivan wrote:
> On Wed, 2006-05-10 at 15:55 +0200, Lele Gaifax wrote:
> 
> > Ouch, I've seen the same kind of messages coming out from the HG 
> > internals running the cvs2hg test yesterday.
> 
> Is this a regression?

My guess is that it's always been there in the hglib (direct calls to
the mercurial python modules) code. Only in the most recent version
has that version replaced the command-line-based back end.

> > Any comment from HG guys?
> 
> The "already tracked" message is harmless.  It just means you tried to
> add a file that hg was already tracking.
> 
> The "not tracked" message is the problem; you're probably trying to
> commit a file that never got added.
> 
> I'm not aware of any changes to Mercurial that should affect this.

I haven't attempted to reproduce this yet, but is it possible that
hglib is being asked to do a 'replace' and is executing the steps in
the wrong order? eg it is doing the add of the new file before the
remove of the old?
_______________________________________________
Tailor mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/tailor

Reply via email to