[Svnmerge] [PATCH] Bidirectional patch take 2
blair at orcaware.com
Mon Feb 27 11:09:56 PST 2006
Raman Gupta wrote:
> Blair Zajac wrote:
>>>>> # merged_revs, we do not need to check those.
>>>>> base = 1
>>>>> r = opts["merged_revs"].normalized()
>>>>> if r and r == 1:
>>>>>- base = r
>>>>>+ base = r + 1
>>If Raman or Giovanni want to come up with a test suite for this fix,
>>then we can get this fix in. I didn't commit this one line change.
> I think the existing test cases that already cover this. You can verify
> the code path is exercised by changing it to base = r + 2 -- about
> 3 or 4 tests will fail. At + 1, no tests will fail that aren't failing
> for other reasons (see my note below).
I think the test should be if the same result is generated with the 'base =
r' or the 'base = r + 1', not if it'll break with +2.
>>Committed in rev 18614.
> Note that with the latest versions in the collabnet repo, the
> testBidirectionalMerges test always fails for me, even with the
> non-modified version, at this chunk:
> # There will be directory property conflict on 'test-branch'
> # due to the attempted merge from trunk of the addition of the
> # svnmerge-integrated property, which already exists in the
> # branch since 'svnmerge.py init' was run in it. So just
> # resolve it, as it currently has the correct value.
> self.launch("svn resolved .",
> match=r"Resolved conflicted state of '\.'")
> I don't recall putting this chunk in my patch, so this must be something
> Blair added.
I tried your original patch against Giovanni's repository and needed this fix in
there to get it to work.
More information about the Svnmerge