Are we ok with using rspec instead of cukes? Any objections?
If so, could we just start using it at this point, setup the little bit of infra required (run from buildr task, etc) and deal with existing cuke tests piecemeal over time? The lost time should be quite minimal and I'm finding it much quicker to test with anyhow. I would propose that we invest a little time in code reviews early on to make sure we're all on the same page.
Does this sound ok?
On Wed, Jul 21, 2010 at 9:13 AM, Bryan Kearney bkearney@redhat.com wrote:
On 07/21/2010 07:55 AM, Devan Goodwin wrote:
On Tue, Jul 20, 2010 at 8:48 AM, Bryan Kearneybkearney@redhat.com wrote:
On 07/19/2010 11:56 AM, Devan Goodwin wrote:
As before, my opinion is strongly that we are essentially re-inventing variables and namespaces because cuke has chopped out most everything we've ever known about organizing and re-using test code with classes, inheritance, fixtures, methods, and return values. I still can't
Have you or anyone else posted this to any cuke mailing list? Is this the way it is with Cuke, or how we are using it?
-- bk
Got a little feedback in that thread, I do not think it is the way we are using cukes. Surely we could clean some things up but it looks like the core complaints I have are just the way it is.
I am concerned a bit about schedule, and I like the idea of external functional tests. I would be open to changing this if the pain (read time) is not too huge.
-- bk