Sylvain Beucler <[EMAIL PROTECTED]> writes: > Can you tel me exactly how you use GNU Arch and then integrate your > changes in the CVS repository? Generic documentation on that subject > would also be welcome.
I use "tla-cvs-sync", from my tla-tools package, to (bi-directionally) synchronize various Emacs CVS branches with their arch counterpart. I run it once or twice a day. tla-cvs-sync works by essentially just replaying each arch change and committing it to CVS (each arch revision results in a single CVS commit), and then committing all outstanding CVS changes to the arch branch as a single changeset. For more info about tla-cvs-sync, see: http://wiki.gnuarch.org/tla-cvs-sync To make the log message used for committing to CVS, tla-cvs-sync feeds the arch revision log through "tla-log-to-cvs-log" (also in tla-tools). The output from tla-log-to-cvs-log consists of some "pseudo headers", a blank line, the summary (with the "Summary: " prefix removed), a blank line, and the body of the arch log: Revision: ARCHIVE/REVISION Creator: CREATOR_ID SUMMARY BODY [The "Creator:" pseudo-header is omitted if it seems to be the same as the person running the CVS syncing process.] Thanks, -Miles -- We have met the enemy, and he is us. -- Pogo _______________________________________________ Savannah-help-public mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/savannah-hackers
