I'm finding another case where I'll get unintended breaks, and it's not so easy to avoid as adding a PENDING to the end of a step matcher.
Scenario: none of these are pending Given a completed feature # green When I submit a bad create post # green Then I should see an error message # green Scenario: There is no matcher for the When Given a feature in work # green When I submit a bad update # yellow, unmatched Then I should see an error message # red Basically, an unmatched step doesn't stop the rest of the steps from running, and the Then matcher fails. I end up having to drop some kind of nasty blocker inside the .feature file in order to avoid that fail, and remember to take it out when I actually implement the second When step ... It's a tough thing though, because I can think of other cases where I'm glad that a single unmatched step doesn't halt the whole chain. Thinking about it though, I'd say it comes down to which type of step it is. If a When is unimplemented, the whole chain should stop. If a Then is unimplemented, just skip it ... On Wed, Sep 17, 2008 at 3:37 AM, Matt Wynne <[EMAIL PROTECTED]> wrote: > We found that we can raise a Cucumber::Pending error in the step matcher to > achieve the same effect, though in practice we're tending to just write > PENDING in the feature and break the link to the step matcher. > > Actually something that would really help us is the ability to comment out > scenarios, as you quite often have code you want to check in but you don't > want to break the build due to a half-finished feature. I'm finding I spend > a lot longer between check-ins since I started using cucumber in earnest, > and that can't be a good thing. > > Maybe just altering that behaviour where the scenario carries on running > when a Given step fails would help. Aslak you never explained why you > designed it like that - how does it help with your workflow? > > > On 17 Sep 2008, at 02:20, David Chelimsky wrote: > > On Tue, Sep 16, 2008 at 8:16 PM, Scott Taylor >> <[EMAIL PROTECTED]> wrote: >> >>> >>> Shouldn't pending 'a message' work in cucumber? Is this a bug? >>> Currently >>> it's raising Rspec's PendingExpectationError. >>> >> >> This came up last week: >> >> http://groups.google.com/group/rspec/browse_thread/thread/ >> 84743f0ef89a82f/ >> >> Didn't really get resolved. >> >> Go ahead and put a ticket in lighthouse if you want (under cucumber >> project: http://rspec.lighthouseapp.com/projects/16211). >> >> Cheers, >> David >> >> SCott >>> >> _______________________________________________ >> rspec-users mailing list >> rspec-users@rubyforge.org >> http://rubyforge.org/mailman/listinfo/rspec-users >> > > cheers, > Matt > ---- > http://blog.mattwynne.net > http://songkick.com > > In case you wondered: The opinions expressed in this email are my own and > do not necessarily reflect the views of any former, current or future > employers of mine. > > > > > _______________________________________________ > rspec-users mailing list > rspec-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/rspec-users > -- // anything worth taking seriously is worth making fun of // http://blog.devcaffeine.com/
_______________________________________________ rspec-users mailing list rspec-users@rubyforge.org http://rubyforge.org/mailman/listinfo/rspec-users
