I guess you could simply analyze the output of a `rake test` task.

Maybe create a rake task that calls the :test task and processes its output?

-- 
Michael Sokol
On lundi 18 avril 2011 at 17:48, milki wrote: 
> Hi,
> 
> I'm building an integration tool between software projects and issue
> trackers. What would be the best way to report test success/failures
> from a ruby project?
> 
> I would need to be able to individually identify which tests
> failed, passed, or are pending (success on expected failures and fail if
> passing on an expected failure).
> I would also like to have minimal modifications if I wanted to add this
> tool to an existing ruby project.
> A report could either be written to a file or execute custom code to
> interact with another software package.
> 
> From what I have found, it would be best to hook into the rake system
> either via a custom runner or formatter. There isn't too much
> documentation on either of these so I'm not sure if this is the right
> way.
> 
> Any suggestions?
> 
> Thanks.
> milki
>