tree-conflicts tests pass to leniently

[Neels J Hofmeyr <ne...@elego.de>]

We are still expecting stuff in the update, switch and merge tree-conflicts
tests to pass that is obviously not correct, and we know it. We want a
tree-conflict to appear on every D1 node in those tests, and some still
don't show it because we can't compare an arbitrary repos URL@REV against
the local working copy with mods (aka diff --summarize). We're just playing
along and PASSing the nonsense current behaviour.

This was convenient in the developing phase, but now that we're about to
branch 1.6, it would be good to XFAIL these tests to clearly mark spec
violations. Good?

~Neels

Re: tree-conflicts tests pass to leniently

[Julian Foad <ju...@btopenworld.com>]

Stefan Sperling wrote:
> On Wed, Nov 26, 2008 at 06:31:50AM +0100, Neels J Hofmeyr wrote:
> > We are still expecting stuff in the update, switch and merge tree-conflicts
> > tests to pass that is obviously not correct, and we know it. We want a
> > tree-conflict to appear on every D1 node in those tests, and some still
> > don't show it because we can't compare an arbitrary repos URL@REV against
> > the local working copy with mods (aka diff --summarize). We're just playing
> > along and PASSing the nonsense current behaviour.
> > 
> > This was convenient in the developing phase, but now that we're about to
> > branch 1.6, it would be good to XFAIL these tests to clearly mark spec
> > violations. Good?
> 
> I don't see why not.

+1 on Neels' changes that do this (r34454). Thanks.

- Julian



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: tree-conflicts tests pass to leniently

[Stefan Sperling <st...@elego.de>]

On Wed, Nov 26, 2008 at 06:31:50AM +0100, Neels J Hofmeyr wrote:
> We are still expecting stuff in the update, switch and merge tree-conflicts
> tests to pass that is obviously not correct, and we know it. We want a
> tree-conflict to appear on every D1 node in those tests, and some still
> don't show it because we can't compare an arbitrary repos URL@REV against
> the local working copy with mods (aka diff --summarize). We're just playing
> along and PASSing the nonsense current behaviour.
> 
> This was convenient in the developing phase, but now that we're about to
> branch 1.6, it would be good to XFAIL these tests to clearly mark spec
> violations. Good?

I don't see why not.

Stefan


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org