httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <dgau...@hotwired.com>
Subject Re: Voting Summary
Date Wed, 06 Dec 1995 18:48:01 GMT
In message <199512061507.PAA03603@cadair.elsevier.co.uk>, Paul Richards writes:
>Any gnu archive site wil have it [cvs]. You'll need to grab rcs as well since
>cvs sits on top of it.

And gnu diffutils (2.7 or later) and gnu patch (2.1 or later) for best
effect.  Patch is used by CVS when doing 'remote cvs' via rsh/krsh.
Which, um, REWLZ.  Except for the total insecurity of using rsh that is.
But I've used remote cvs on a 7 programmer project (120k lines) where
we're spread all over North America and Finland.

Be sure you install diffutils and patch first, then rcs, then cvs.
If you don't want gnu diffutils/patch/rcs in your default path then
ensure that you tweak rcs and cvs configuration files to hard code paths
to the executables.  It's best to hard code at any rate... so that you
never get the wrong diff/rcs.

We're able to do sweeping changes without stepping on each other too much.
It does take some co-ordination if you are changing core data structures.
But usually that amounts to a message one day saying 'please don't
commit today' then next day saying 'ok to commit, but be warned that the
following has seriously changed: ...'.  I push some of the work off onto
the compiler -- if something has changed in a backwards incompatible way
then I change the structure name/member names so that stuff written the
old way will fail to compile.

Here's a URL for CVS: http://www.loria.fr/~molli/cvs-index.html
Not sure how good it is, the old one I used has been discontinued
because the auther doesn't have time to maintain the pages.

Dean

Mime
View raw message