On 23/09/2008, at 1:37 AM, Brett Porter wrote:
> Ok, I started doing this on a branch and added some notes about my
> progress. Anyone have any thoughts on that approach?
It's now possible to run the ITs on trunk against 0.14-incubating-
SNAPSHOT, 0.15-incubating and 0.16-incubating-SNAPSHOT independently.
I'm working on the toolchains again next to be able to easily run the
ITs against a non-default configuration, then I expect to merge it
back to trunk.
- Brett
>
>
> I checked in with Milos about the toolchains. He said the copying to
> M2_HOME/lib is not needed if you have the extension - and it seems
> that the first part of the patch on NMAVEN-147 might be aimed to
> address that issue. Milos also agreed that the change to the
> configuration proposed there seemed to make sense so I'll review
> that patch again.
>
> However, the ITs do run without them (as long as everything is in
> the path, 0.14 style), so I've gone towards getting that combo
> running first, then add the toolchains as an extra layer next for now.
>
> Cheers,
> Brett
>
> On 02/09/2008, at 3:52 PM, Brett Porter wrote:
>
>>
>> On 18/08/2008, at 8:15 AM, Carlos Sanchez wrote:
>>
>>> I've been working in the integration tests lately and fixed some
>>> of the problems
>>>
>>> - found a workaround for trunk to work, copying dotnet toolchain to
>>> M2_HOME/lib NMAVEN-186
>>> with this workaround most of the integration tests pass
>>>
>>> - fixed test compilation in trunk NMAVEN-192
>>>
>>> - make integration tests in trunk run against 0.14 NMAVEN-193 and
>>> NMAVEN-186
>>> just needs search and replace and a small patch in 0.14
>>>
>>> - hit problem with NUnit tests execution, no tests are executed
>>> NMAVEN-8 (reopened)
>>>
>>> If NUnit problem NMAVEN-8 is fixed (waiting for comment from Shane)
>>> then I can continue testing. Right now all integration tests but the
>>> ones involving NUnit seem to work against trunk and 0.14.
>>
>> Unless anyone else is planning to, I can start to take a look at
>> these over the next few days.
>>
>>>
>>>
>>> Also is the convention to require NUnit explicitly listed in the pom
>>> to run the tests? seems so as the tests need to be annotated with
>>> NUnit annotations
>>
>> That would make sense to me - same as junit, etc in Java?
>>
>> - Brett
>>
>> --
>> Brett Porter
>>
brett@...
>>
http://blogs.exist.com/bporter/>>
>
> --
> Brett Porter
>
brett@...
>
http://blogs.exist.com/bporter/>
--
Brett Porter
brett@...
http://blogs.exist.com/bporter/