James Carlson <[EMAIL PROTECTED]> writes: (for the benefit of the list at large...)
> Giorgos Keramidas writes: >> The script calls gdiff3, but if `/usr/bin/diff3 -m' works well enough, >> then it should be easy to adapt it. > > There's no "-m" option for /usr/bin/diff3, at least on Solaris. > >> If the automerge creates a file which contains conflict markers, then >> this script will abort. In the rare case when conflict markers *are* OK > > Thanks ... but the problem isn't about automerge in general, it's > about imerge's "merge -a" falling to bits. I have no clue how to > trick hg's imerge extension into using your script to do > non-interactive merges. There doesn't seem to be a configuration > option to say "use this other script for non-interactive" -- it just > disappears somewhere into Mercurials Python "filemerge" logic. :-/ > > Setting the "merge =" attribute doesn't do the trick. That just > affects interactive merges, which already work just fine with gpyfm. > > If I can't get imerge to work at all, then I'll try using this, but I > really want imerge. This is because when imerge is asked to do an automatic merge, it ignores HGMERGE (and hgmerge) entirely, and forcibly attempts to run merge(1), which is part of RCS and not delivered by Solaris. -- Rich _______________________________________________ tools-discuss mailing list tools-discuss@opensolaris.org
