tree conflict infos

[Neels Janosch Hofmeyr-2]

Hi guys,

I notice we don't have a single test to verify the "Source left" and "Source
right" paths and peg revs shown on "a tree-conflicted" `svn info'...

~Neels

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2414219

[Stefan Sperling-7]

On Tue, Nov 03, 2009 at 10:02:21PM +0100, Neels J Hofmeyr wrote:
> Hi guys,
>
> I notice we don't have a single test to verify the "Source left" and "Source
> right" paths and peg revs shown on "a tree-conflicted" `svn info'...

While we're on the topic of what we're not testing:
We don't seem to have any tests which verify copyfrom information
as being recorded correctly either...

Stefan

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2414682

[Neels Janosch Hofmeyr-2]

Stefan Sperling wrote:
> On Tue, Nov 03, 2009 at 10:02:21PM +0100, Neels J Hofmeyr wrote:
>> Hi guys,
>>
>> I notice we don't have a single test to verify the "Source left" and "Source
>> right" paths and peg revs shown on "a tree-conflicted" `svn info'...
>
> While we're on the topic of what we're not testing:
> We don't seem to have any tests which verify copyfrom information
> as being recorded correctly either...

sheesh, who needs tests anyway.

Seriously though, it would be great if the svn power users out there would
model their use case cycles in tests. I've got one on the other line trying,
and he's quite disappointed with the current lack of tests.

The other way would be to try and find every loathsome permutation of
everything and write all of those into tests -- not trivial at all. That
would need a serious sponsor.

So my guess is, if corner-case-hitters would document their workflow in
cmdline tests, we'd have a pretty optimal coverage. It's definitely in the
interest of power-user corner-case-hitters to do so.

~Neels

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2414732

[Stefan Sperling-7]

On Thu, Nov 05, 2009 at 03:48:28PM +0100, Neels Janosch Hofmeyr wrote:

> Stefan Sperling wrote:
> > On Tue, Nov 03, 2009 at 10:02:21PM +0100, Neels J Hofmeyr wrote:
> >> Hi guys,
> >>
> >> I notice we don't have a single test to verify the "Source left" and "Source
> >> right" paths and peg revs shown on "a tree-conflicted" `svn info'...
> >
> > While we're on the topic of what we're not testing:
> > We don't seem to have any tests which verify copyfrom information
> > as being recorded correctly either...
>
> sheesh, who needs tests anyway.

Bert has added a copyfrom test, but it does not cover the copyfrom
case I have broken on trunk. I'll expand the test, and fix the breakage,
shortly.

Stefan

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2414745