From: Michael Olson Date: Wed, 28 Apr 2010 15:51:12 GMT Subject: Re: [spf:guess,iffy,mismatch] Re: [PATCH/RFC 2/2] git-svn: Don't allow missing commit parent to stop git-svn Message-ID: In-Reply-To: <4BCF8E07.9080507@vilain.net> On Wed, Apr 21, 2010 at 4:45 PM, Sam Vilain wrote: > Michael Olson wrote: >> Interestingly, I see people reporting the same problem in the thread >> "git svn clone of subversion's own code failing".  So it's not just >> related to patch 1/2 of this series.  At least that thread has a >> public-facing repo so the problem should be easier to reproduce for >> a real fix. > > Oh, right ... I remember that issue now. I'm currently testing the > below change to see if it breaks the test suite; any issues and I'll > post an update. > > Subject: [PATCH] git-svn: deal with svn cherry-picks with non-linear history > > The expression 'BASE^..TIP' is only defined when BASE has parents.  It > actually does what it intends only when it has one parent. > > Allow for more complicated revision list arguments than a range by > splitting any space-containing arguments on the way into _rev_list, > and check the parents of the bottom commit so that we may use the > correct revision list specifier. > --- >  git-svn.perl |   22 +++++++++++++++++++--- >  1 files changed, 19 insertions(+), 3 deletions(-) > > [snip] This change works great on my private repos, so it should supersede my version which only did an eval. -- Michael Olson | http://mwolson.org/