> -----Original Message-----
> From: Malcolm Rowe [mailto:malcolm-svn-dev@farside.org.uk]
> Sent: Tuesday, May 01, 2007 9:20 AM
> To: dev@subversion.tigris.org
> Subject: Re: svn commit: r24860 - branches/1.4.x
>
> On Tue, May 01, 2007 at 06:15:45AM -0700, malcolm@tigris.org wrote:
> > * STATUS: Temporarily veto r24483, until someone can fix the test.
> >
>
> Just to clarify: I think the change is probable correct (I'm
> +0: it sounds right, but I haven't fully understood it), but
> we obviously can't merge this change when it causes the test to fail.
>
> (I think the problem is that an additional change needs to be
> made to the 1.4 version of the test, rather than there being
> a problem with the code change itself).
You're correct Malcolm, the test itself needed a changed.
r24483 stops mer'G'e notifications on files unchanged by a merge, but
which had local modifications prior to the merge. In
merge_test-1:textual_merges_galore() the merge,
svn merge -r2:3 %URL/A/D/G %WC%.other/A/D/G
no longer reports %WC%.other/A/D/G/pi as mer'G'ed because the merge has
no effect on it. In trunk, r24483 never had to make this change because
of a previous tweak to textual_merges_galore - see r23813 on
/svn/branches/merge-tracking/subversion/tests/cmdline/merge_tests.py
@24027.
I made the fix in r on
http://svn.collab.net/repos/svn/branches/1.4.x-r24483 and the test now
passes.
Paul B.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue May 1 17:29:27 2007