On Wed, Feb 20, 2008 at 12:35:46PM +1100, Richard Beare wrote: > Hi Everyone, > > I'd like to ask advice concerning the last couple of issues before we > can make our local savane instance active. > > Firstly, are there any explanations about how a project's web pages > should be managed using svn? I'm not talking about the ssh component, > but rather the layout of the svn repository and how any changes > commited to svn are passed through the system. Sorry if this is an > obvious question, but so far I haven't found any explanation of how to > go about it.
I think there's a cron job at Gna! that list all projects in the database (SELECT unix_group_name FROM groups WHERE status='A') then for each of them: - check if (say) /var/www/checkouts/$unix_group_name exists - if yes, svn update - if no, svn co svn://svn.gna.org/.../$unix_group_name/webpages $unix_group_name There's something similar at Savannah with CVS, it's on a machine where I don't have access though, I don't have the code for that. > My second question relates to the use of git for source code > management. I've noticed that the > savannah.gnu.org has instructions concerning the use of git, but I > haven't seen any mention of server side configuration for savane. Does > savane support the use of git? If so how? Savane supports Git if you use the clean-up branch located at: https://savannah.nongnu.org/p/savane-cleanup If you already have Savane running, you can install it like other Savane upgrades. -- Sylvain _______________________________________________ Savane-help mailing list [email protected] https://mail.gna.org/listinfo/savane-help
