CVS branches RFC

Christopher Faylor cgf@redhat.com
Thu Oct 4 13:30:00 GMT 2001


On Thu, Oct 04, 2001 at 09:43:51PM +1000, Robert Collins wrote:
>----- Original Message -----
>From: "Earnie Boyd" <earnie_boyd@yahoo.com>
>
>If you want to see more doco about the scripts, goto
> http://devel.squid-cache.org , browse around a bit. Also look (via
>viewscvs) at the CVSROOT module - there are extra scripts there to sync
>up the two repositories - across _all_ branches.
>
>>>Which, BTW is why I suggested these scripts - small they are, but
>>>powerful too :}.
>>
>>Yep.  And Chris could have created a branch for his changes so as to
>>not destroy a working snapshot.  ;D.
>
>
>Quite possibly ;].  All the folk who have commit access to the squid
>production repository (there are 2 - one dev, one production) and
>commit sweeping changes do the prepatory work on the dev repository
>under various branches.  Trivial stuff isn't worth making a branch for
>though - and sometimes its hard to tell in advance how significant the
>work is :].  So if I fix a format bug or something, I'll throw it in
>HEAD on the production CVS, otherwise it gets branched.
>
>Mind you, cygwin HEAD and squid HEAD are run somwhat differently.  The
>squid HEAD aims to _never_ get broken, whereas cygwin seems to be
>allowed to be broken - temporarily.

I am, again, lost.  I doubt that anyone purposely checks in known broken
code.  If someone is doing this then they should stop doing it.

Can I suggest that you just move back to discussing how wonderful the
scripts are and leave my name out of the discussion?  You are both
somewhat confused if you think my work should have been done on a
branch.

cgf



More information about the Cygwin-developers mailing list