Re: [Quilt-dev] [patch] Quilt support for committing patches to CVS.

[Joe Green]

Dean Roehrich wrote:
> Your patch expects CVS to manage the series file and the individual patch
> files rather than the files being patched.

Yes, that's right.  I manage the tar files and patches in source 
control, and use quilt to build trees when I need to work on them.

> Jason's patch presumably requires that "quilt commit" be manually run for each
> file that will be patched by the next "quilt push".

It looks like he specifies a particular patch to commit (defaulting to 
the top patch), and looks at all the files in the patch to determine CVS 
operations to do.  The patch must be applied, so it has already been pushed.

> My patch expects the SCM to manage the files being patched, and currently does
> not expect the series file or the individual patch files to be managed by the
> SCM.  I have hooks in add/apatch/patchfns to make quilt coordinate with the
> SCM, and it also determines which SCM is being used so it can issue the
> appropriate commands.

Yes, it looks like you do it on push.

> There's not a lot of overlap here; we're almost solving three different
> problems and I think we have at least two very different ideas of what a
> "commit" is all about.

Yes, the difference in commit behaviors might be difficult to resolve to 
everyone's satisfaction.  To me it seems natural for "quilt commit" to 
commit the patches/series, since that where quilt keeps its long-term 
state information, but then I would say that :).

Putting the commit issue aside for the moment, perhaps the right thing 
for the other hooks is to have separate functions like 
"scm_series_modify", "scm_patch_add", "scm_file_add", etc., and allow 
people to extend the ones they want?

--
Joe Green <address@hidden>
MontaVista Software, Inc.