Re: cvs "grouping" feature

[Russ Sherk]

On 8/31/05, Derek Price <address@hidden> wrote:
> Russ Sherk wrote:
> 
> >On 8/31/05, Derek Price <address@hidden> wrote:
> >
> >
> >>Russ Sherk wrote:
> >>
> >>
> >>
> >>>>Can the new functionality be extended to allow a 'commit session'?
> >>>>-----
> >>>>cvs ci-startsession (create commitid cid1)
> >>>>cvs ci filea  (cid1)
> >>>>cvs ci /sompe/other/sandbox/fileb (cid1)
> >>>>cvs ci-endsession (close commitid cid1)
> >>>>----
> >>>>
> >>>>
> >>Why not just: `cvs ci filea /some/other/sandbox/fileb'?
> >>
> >>
> >
> >To allow for 'functional' groupings.  It seems though that it would
> >only be a convenience function that keeps track of committed files and
> >applies the same tag to them (instead of the user keeping track and
> >tagging them manually).
> >
> >Just a thought.
> >
> >
> 
> The command I suggested should do the same thing.  Also, it seems to me
> that you could create your 'functional' grouping using `cvs tag -r
> commitid1 functag; cvs tag -r commitid2 functag' easily enough after the
> two commits, at least once the newtags branch is completed and merged.
> 
> I only object because auto-tagging individual commits is a commonly
> requested feature and I don't really want to blur the concept by
> confusing commitids with 'functional' tags, as I thought their use would
> be mutually-exclusive as you presented it.  I suppose if someone wanted
> to code a -t <tag> option or the like for commit that auto-tagged the

Yep.  This did come to mind.  In general it would behave like the -m
option except it'd have to properly handle existing tags (i.e. need to
add -F option to ci...) as well as apply tags instead of log messages.

> committed files with the specified additional tag, that might also
> produce something like the functionality you are requesting.
> 
> Okay, I also object because I have no plans to code this without
> funding.  If you thought it was useful enough for CVS to autogenerate an
> additional 'functional' tag via your suggested transactions or the like,
> and were to code it and contribute the patch yourself, it might be taken
> more seriously.  :)

Hmmm...  so if I paid you, you'd code it up?  I've often wondered how
the funding portion of these 'open source' projects work.  But that is
another story for another time I guess.  I'll take a look at the code
and see what I can come up with.  Assuming it should be for 1.12
branch?

--Russ

> 
> Regards,
> 
> Derek
> 
> --
> Derek R. Price
> CVS Solutions Architect
> Ximbiot <http://ximbiot.com>
> v: +1 717.579.6168
> f: +1 717.234.3125
> <mailto:address@hidden>
> 
> 
>