Well "cvs up -kk -j <branch-tag-1> -j <branch-tag-2>" seems to work okay for me to avoid problems with merging the keywords (I think we've only used $Id$, though).
-- Jamie Wellnitz On Mon, Nov 03, 2003 at 09:05:49AM -0500, Donald Sharp wrote: > Ick. I hate keyword expansions. The minute you ever merge > between branches you have to deal with those keywords. > > donald > > > On Sat, Nov 01, 2003 at 02:18:45PM +0100, Spiro Trikaliotis wrote: > > Hello, > > > > On Fri, Oct 31, 2003 at 04:53:33PM -0500, Robert Clark wrote: > > > > > If you are not too worried a few odd merge conflicts, you could always add the > > > $Revision$ or $Id$ keywords to a comment in your source code and the write > > > the canonical 5 lines of Perl to snag the revision number from the file. Or > > > use ident (which is part of RCS) to do that for you. > > > > Sorry for coming to this discussion this late, but IMHO, using the "RCS > > way" (ident with $Id$ in the source code *and* the binary) seems to me > > to be the best solution for what Katherine needs. > > > > 'ident'ing the resulting binary gives you exactly what you need, the > > list of all revisions of the sources used to compile that binary. > > > > Spiro. > > > > > > _______________________________________________ > > Info-cvs mailing list > > [EMAIL PROTECTED] > > http://mail.gnu.org/mailman/listinfo/info-cvs > > > _______________________________________________ > Info-cvs mailing list > [EMAIL PROTECTED] > http://mail.gnu.org/mailman/listinfo/info-cvs _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
