[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: commit output
From: |
Mark D. Baushke |
Subject: |
Re: commit output |
Date: |
Mon, 10 May 2004 14:21:48 -0700 |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Derek Robert Price <derek@ximbiot.com> writes:
> Mark D. Baushke wrote:
>
> > >I think I would actually like to fix feature to never output the "RCS
> > >file...", "Checking in...", and "done" lines. They are redundant and
> > >unnecessary. Any comments?
> >
> > You might want to drop a note on folks that are doing programatic
> > interfaces to cvs (eg, eclipse.org has a eclipse-dev mailing list). I
> > would suspect that this kind of change may cause some of them grief
> > along the way...
>
>
> Well, the information I'm removing _is_ redundant and unnecessary.
> Helping the Eclipse folks keep up-to-date is a nice thought, but I
> would hope the Eclipse folks would be able to read the NEWS file and
> ignore the incorrect lines.
I agree. I just thought it might warrant a wider audience as to the
upcoming change.
> I simply don't think we want to relinquish that sort of control over
> our output. It would be one thing if we were trying to take away
> information somebody needed, but any reordering, removal of redundant
> info, moving/addition of quotes, etc., I would expect anyone parsing
> CVS's output to be able to handle on their own.
True.
> As a long term goal, it might be nice to export most CVS functionality
> via a convenient library API so that these issues don't crop up so
> often. Another possibility would be to enable an XML output mode for
> external parsers.
Isn't this the purpose of the libcvs project?
> Regardless, I don't want to have to try and get approval for tidying
> output from every project that happens to be parsing our output. It's
> impractical and would hamper development severely. I _don't_ object
> to trying to keep output on "stable" from changing much, but feature
> is another matter.
True enough. I have no objections with putting your suggested changes
into the feature version of cvs. I thought I read your original question
as pertaining to altering the 'stable' version of cvs which I think
requires a bit of a bigger audience for announcing that change...
-- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)
iD8DBQFAn/Js3x41pRYZE/gRAsvwAKCJ1LIr2JWA+QHz+mah/5sMEa+PIwCgvSx8
n8OwO3DIU/lzFNA2LoSWcUA=
=kFLQ
-----END PGP SIGNATURE-----
- commit output, Derek Robert Price, 2004/05/10
- Re: commit output, Mark D. Baushke, 2004/05/10
- Re: commit output, Derek Robert Price, 2004/05/10
- Re: commit output,
Mark D. Baushke <=
- Re: commit output, Derek Robert Price, 2004/05/10
- Re: commit output, Mark D. Baushke, 2004/05/10
- Re: commit output, Derek Robert Price, 2004/05/10
- Re: commit output, Mark D. Baushke, 2004/05/10
- Re: commit output, Derek Robert Price, 2004/05/11
- Re: commit output, Mark D. Baushke, 2004/05/11
- Re: commit output, Alexander Taler, 2004/05/10
- Re: commit output, Derek Robert Price, 2004/05/11
Re: commit output, Larry Jones, 2004/05/10
Re: commit output, Derek Robert Price, 2004/05/10