On Wed, 8 Jan 2014 08:27:42 -0500 "Darryl L. Pierce" <mcpierce@xxxxxxxxx> wrote: > I recently took over package maintenance for Audacity. So the first > thing I want to do is clone their development tree, create a branch > from the latest release and star tracking patches against it as > needed. The Audacity team uses Subversion, while I prefer Git, so I > attempted to use Git to clone their tree. > > The problem is that it's been HOURS and Git hasn't finished cloning > the tree. > > The steps I took are: > > 1. mkdir Audacity; cd Audacity > 2. git svn init --prefix=origin/ -b branches -t tags -T trunk > http://audacity.googlecode.com/svn/audacity-src > 3. git svn fetch --log-window-size 10000 > > It's the third step which is taking forever to complete. I left it to > run overnight (10 hours) and it never finished. I paused it, came to > work, resumed it and it's still going. > > Should this process really take this long? Or is there a faster, more > efficient way to do this? I know that part of what'll make it take so > long is the amount of history that has to be extracted from > Subversion. But is that many hours normal or did I do something wrong? in my experience it can take hours to clone large SVN repos (similar to 13k revisions in audacity), I usually use git svn clone --stdlayout <svn_url> but it should be equivalent to your procedure somewhere under .git you should see a file that changes during the fetch phase (IIRC it lists the grabbed revisions), so you will know it does something, also if svn:// url works, it should be faster than http:// Dan -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
