On Tue, Mar 9, 2010 at 9:24 AM, Paul Burba <ptburba_at_gmail.com> wrote:
> On Mon, Mar 8, 2010 at 9:55 AM, C. Michael Pilato <cmpilato_at_collab.net> wrote:
>
>> Paul Burba wrote:
>>> Mike,
>>>
>>> Do you recall what version of Subversion you used to do the
>>> REVISIONIST HISTORY work
>>> (http://svn.apache.org/repos/asf/subversion/README)?
>>>
>>> I'm assuming it was probably 1.6.6, which was the current release at
>>> the time, but just wanted to be sure you were not using a trunk build.
>>
>> I don't run trunk Subversion's these days. Trunk can't handle my
>> sparsely-checked-out working copy size.
>>
>> At any rate, it should have been 1.6.6, because I did all that work on
>> svn.collab.net using the Subversion installed there.
>
> That's what I was afraid of. Like I said, I used 1.6.6 in an attempt
> to recreate the problem. Since I couldn't, I'm out of ideas as to
> what went wrong here. I'll work on fixing the mergeinfo in our repos,
> that is about all I can do re the incorrect offsets.
Nothing new to report re what went wrong with the dump-load, but in
http://svn.apache.org/viewvc?view=revision&revision=929295 I fixed the
mergeinfo on the branches that didn't agree with the expected 840074
offset:
'/branches/1.6.x-r40452'
'/branches/1.6.x-r40452/CHANGES'
'/branches/fs-successor-ids'
'/branches/python-3-compatibility'
Note, earlier in this thread I identified '/branches/issue-3242-dev'
as also having the wrong offset, but Mike already fixed this in
http://svn.apache.org/viewvc?view=revision&revision=888932
> I'll also track down the second bug which didn't correctly adjust the
> self-referential offsets.
This problem wasn't really about self-referential mergeinfo per se,
rather it was load bug, see the reopened issue #3020:
http://subversion.tigris.org/issues/show_bug.cgi?id=3020#desc9
These bugs were fixed in
http://svn.apache.org/viewvc?view=revision&revision=927243 and
backported to 1.6.10.
Paul
Received on 2010-03-30 23:24:36 CEST