[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Problems with svn blame -g

From: Mark Phippard <markphip_at_gmail.com>
Date: 2007-08-21 16:58:28 CEST

On 8/21/07, Hyrum K. Wright <hyrum_wright@mail.utexas.edu> wrote:
> Mark Phippard wrote:
> > On 8/20/07, Hyrum K. Wright <hyrum_wright@mail.utexas.edu> wrote:
> >> Mark Phippard wrote:
> >>> Using my sample merge tracking repository that is used in the svn log -g tests:
> >>>
> >>> $ svn blame -g file:///Users/mphippard/repositories/merge-tracking/trunk/index.html
> >>>
> >>> Works OK.
> >>>
> >>> $ svn blame -g file:///Users/mphippard/repositories/merge-tracking/trunk/products/index.html
> >>> svn: Can't open file '/tmp/tmp.2.tmp': No such file or directory
> >> Fixed in r26207.
> >
> > It now runs to completion, but I see user "merger" from r12 in the
> > output which implies there is a bug of some sort. I think this was a
> > case where branchA was merged to branchB in r12 and then to trunk
> > later. So maybe it did not follow the merge back to its source?
> Oh, good point. Instead of just two possible states a particular
> revision can be in (merged and non-merged), we really have three
> possible states. The revision could be a merged revision with content
> changes, a merged revision which is also a merging revision, and a
> non-merged/non-merging revision.
> If the revision is a merging revision, we don't want it to show up as a
> revision which changed content, even if it is also a merging revision.
> I'll need to think about this for a minute...

One thing that is interesting about this file is that the blame -g
does show one line as r11. It makes me wonder if I made some small
change in r12 as part of resolving the merge?

> >>> $ svn blame -g file:///Users/mphippard/repositories/merge-tracking/trunk/products/roadmap.html
> >>> Bus error
> >> Fixed in r26211.
> >
> > This still crashes for me on OSX. I have not tried it on Windows.
> Hmm, I don't have any problems on Linux. Do you know where it crashes?

It also crashes on Windows, this is what JavaHL reports. I thought
the command line had a new crash reporter? I wonder how I could use
that to get you a stack trace?

# An unexpected error has been detected by Java Runtime Environment:
# EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x08163ce0, pid=5680, tid=5000
# Java VM: Java HotSpot(TM) Client VM (1.6.0_01-b06 mixed mode, sharing)
# Problematic frame:
# C [libsvn_client-1.dll+0x3ce0]
# An error report file with more information is saved as hs_err_pid5680.log

Mark Phippard
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Aug 21 16:56:03 2007

This is an archived mail posted to the Subversion Dev mailing list.