rspec users - Sep 2008 - cucumber - mark a step as pending

I love the way I can throw a call to pending() in the top of an  
unfinished RSpec example and stop it from failing the build.

Is there a similar way to do such a thing with good ole'' cucumber?

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.



-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/rspec-users/attachments/20080909/a0776ecb/attachment-0001.html>

Have you tried?

On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at mattwynne.net>
wrote:
> I love the way I can throw a call to pending() in the top of an unfinished
> RSpec example and stop it from failing the build.
> Is there a similar way to do such a thing with good ole'' cucumber?
>
> 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 at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

A plain text step with no matching step definition is pending.

Sent from my iPhone

On 9. sep.. 2008, at 19.05, "David Chelimsky" <dchelimsky at
gmail.com>
wrote:
> Have you tried?
>
> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at mattwynne.net>  
> wrote:
>> I love the way I can throw a call to pending() in the top of an  
>> unfinished
>> RSpec example and stop it from failing the build.
>> Is there a similar way to do such a thing with good ole''
cucumber?
>>
>> 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 at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

Aslak - I think Matt is looking for this:

Given /blah/ do
  pending
end

Matt - I''m pretty sure that will work - give it a shot, OK?

On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
<aslak.hellesoy at gmail.com> wrote:
> A plain text step with no matching step definition is pending.
>
> Sent from my iPhone
>
> On 9. sep.. 2008, at 19.05, "David Chelimsky" <dchelimsky at
gmail.com> wrote:
>
>> Have you tried?
>>
>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net> wrote:
>>>
>>> I love the way I can throw a call to pending() in the top of an
>>> unfinished
>>> RSpec example and stop it from failing the build.
>>> Is there a similar way to do such a thing with good ole''
cucumber?
>>>
>>> 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 at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky <dchelimsky at gmail.com>
wrote:
> Aslak - I think Matt is looking for this:
>
> Given /blah/ do
>  pending
> end
>
I''m not crazy about this "feature". We already have a pending
mechanism for steps (not defining them). What does this bring to the
table over just not defining a step?
> Matt - I''m pretty sure that will work - give it a shot, OK?
>
That won''t work. It will raise a "undefined local variable or
method `pending''"

Aslak
> On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
> <aslak.hellesoy at gmail.com> wrote:
>> A plain text step with no matching step definition is pending.
>>
>> Sent from my iPhone
>>
>> On 9. sep.. 2008, at 19.05, "David Chelimsky" <dchelimsky
at gmail.com> wrote:
>>
>>> Have you tried?
>>>
>>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net> wrote:
>>>>
>>>> I love the way I can throw a call to pending() in the top of an
>>>> unfinished
>>>> RSpec example and stop it from failing the build.
>>>> Is there a similar way to do such a thing with good
ole'' cucumber?
>>>>
>>>> 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 at rubyforge.org
>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>
>>> _______________________________________________
>>> rspec-users mailing list
>>> rspec-users at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

Yeah the trouble is it fails with the following error:

Spec::Example::ExamplePendingError

Which cucumber doesn''t understand as anything other than a complete  
fail.

Our workaround we''ve arrived at (thanks Aslak for pointing out the  
obvious that we''d missed!) is to just write PENDING at the end of the  
regex for the step matcher so it doesn''t match on the step in the  
feature file.

@Aslak - Interestingly, when we do this and make the Given step  
pending, cuc still tries to run the When / Then steps for that  
scenario. Sounds like a bug to me - should we raise a ticket on  
lighthouse or did I miss something?


On 9 Sep 2008, at 18:16, David Chelimsky wrote:
> Aslak - I think Matt is looking for this:
>
> Given /blah/ do
>   pending
> end
>
> Matt - I''m pretty sure that will work - give it a shot, OK?
>
> On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
> <aslak.hellesoy at gmail.com> wrote:
>> A plain text step with no matching step definition is pending.
>>
>> Sent from my iPhone
>>
>> On 9. sep.. 2008, at 19.05, "David Chelimsky"  
>> <dchelimsky at gmail.com> wrote:
>>
>>> Have you tried?
>>>
>>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net>
>>> wrote:
>>>>
>>>> I love the way I can throw a call to pending() in the top of an
>>>> unfinished
>>>> RSpec example and stop it from failing the build.
>>>> Is there a similar way to do such a thing with good
ole'' cucumber?
>>>>
>>>> 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 at rubyforge.org
>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>
>>> _______________________________________________
>>> rspec-users mailing list
>>> rspec-users at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at 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.

On Tue, Sep 9, 2008 at 7:52 PM, Matt Wynne <matt at mattwynne.net>
wrote:
> Yeah the trouble is it fails with the following error:
>
> Spec::Example::ExamplePendingError
>
> Which cucumber doesn''t understand as anything other than a
complete fail.
>
> Our workaround we''ve arrived at (thanks Aslak for pointing out the
obvious
> that we''d missed!) is to just write PENDING at the end of the
regex for the
> step matcher so it doesn''t match on the step in the feature file.
>
> @Aslak - Interestingly, when we do this and make the Given step pending,
cuc
> still tries to run the When / Then steps for that scenario. Sounds like a
> bug to me - should we raise a ticket on lighthouse or did I miss something?
>
It''s by design, but I''m open for suggestions. David created a
similar
ticket yesterday:
http://rspec.lighthouseapp.com/projects/16211/tickets/8-all-steps-after-a-failure-are-listed-as-skipped#ticket-8-1

Currently, only a failure (red) will cause subsequent steps to be skipped.

Aslak
>
> On 9 Sep 2008, at 18:16, David Chelimsky wrote:
>
>> Aslak - I think Matt is looking for this:
>>
>> Given /blah/ do
>>  pending
>> end
>>
>> Matt - I''m pretty sure that will work - give it a shot, OK?
>>
>> On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
>> <aslak.hellesoy at gmail.com> wrote:
>>>
>>> A plain text step with no matching step definition is pending.
>>>
>>> Sent from my iPhone
>>>
>>> On 9. sep.. 2008, at 19.05, "David Chelimsky"
<dchelimsky at gmail.com>
>>> wrote:
>>>
>>>> Have you tried?
>>>>
>>>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net> wrote:
>>>>>
>>>>> I love the way I can throw a call to pending() in the top
of an
>>>>> unfinished
>>>>> RSpec example and stop it from failing the build.
>>>>> Is there a similar way to do such a thing with good
ole'' cucumber?
>>>>>
>>>>> 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 at rubyforge.org
>>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>>
>>>> _______________________________________________
>>>> rspec-users mailing list
>>>> rspec-users at rubyforge.org
>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>>> _______________________________________________
>>> rspec-users mailing list
>>> rspec-users at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at 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 at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

> It''s by design, but I''m open for suggestions. David
created a similar
> ticket yesterday:
> http://rspec.lighthouseapp.com/projects/16211/tickets/8-all-steps- 
> after-a-failure-are-listed-as-skipped#ticket-8-1
>
> Currently, only a failure (red) will cause subsequent steps to be  
> skipped.
In case I''ve missed something, would you mind elaborating on the  
intent of that design? Under what circumstances would you want it to  
carry on running the scenario if some of the context (Given...)  
wasn''t set up?

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.

On Tue, Sep 9, 2008 at 12:47 PM, aslak hellesoy
<aslak.hellesoy at gmail.com> wrote:
> On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky <dchelimsky at
gmail.com> wrote:
>> Aslak - I think Matt is looking for this:
>>
>> Given /blah/ do
>>  pending
>> end
>>
>
> I''m not crazy about this "feature". We already have a
pending
> mechanism for steps (not defining them). What does this bring to the
> table over just not defining a step?
>
>> Matt - I''m pretty sure that will work - give it a shot, OK?
>>
>
> That won''t work. It will raise a "undefined local variable or
method `pending''"
I get this:

    Given I am logged in as a registered member
      TODO (Spec::Example::ExamplePendingError)

for this:

Given /I am logged in as a registered member/ do
  pending
end
>
> Aslak
>
>> On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
>> <aslak.hellesoy at gmail.com> wrote:
>>> A plain text step with no matching step definition is pending.
>>>
>>> Sent from my iPhone
>>>
>>> On 9. sep.. 2008, at 19.05, "David Chelimsky"
<dchelimsky at gmail.com> wrote:
>>>
>>>> Have you tried?
>>>>
>>>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net> wrote:
>>>>>
>>>>> I love the way I can throw a call to pending() in the top
of an
>>>>> unfinished
>>>>> RSpec example and stop it from failing the build.
>>>>> Is there a similar way to do such a thing with good
ole'' cucumber?
>>>>>
>>>>> 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 at rubyforge.org
>>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>>
>>>> _______________________________________________
>>>> rspec-users mailing list
>>>> rspec-users at rubyforge.org
>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>>> _______________________________________________
>>> rspec-users mailing list
>>> rspec-users at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On Tue, Sep 9, 2008 at 8:46 PM, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On Tue, Sep 9, 2008 at 12:47 PM, aslak hellesoy
> <aslak.hellesoy at gmail.com> wrote:
>> On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky <dchelimsky at
gmail.com> wrote:
>>> Aslak - I think Matt is looking for this:
>>>
>>> Given /blah/ do
>>>  pending
>>> end
>>>
>>
>> I''m not crazy about this "feature". We already have
a pending
>> mechanism for steps (not defining them). What does this bring to the
>> table over just not defining a step?
>>
>>> Matt - I''m pretty sure that will work - give it a shot,
OK?
>>>
>>
>> That won''t work. It will raise a "undefined local
variable or method `pending''"
>
> I get this:
>
>    Given I am logged in as a registered member
>      TODO (Spec::Example::ExamplePendingError)
>
> for this:
>
> Given /I am logged in as a registered member/ do
>  pending
> end
>
That''s really odd - I don''t get that if I add it to this step:
http://github.com/aslakhellesoy/cucumber/tree/master/examples/tickets/features/steps/tickets_steps.rb

And run "rake features" from examples/tickets

I haven''t added any knowledge about
pending/Spec::Example::ExamplePendingError to Cucumber so I''m
puzzled...

Aslak
>>
>> Aslak
>>
>>> On Tue, Sep 9, 2008 at 12:09 PM, Aslak Helles?y
>>> <aslak.hellesoy at gmail.com> wrote:
>>>> A plain text step with no matching step definition is pending.
>>>>
>>>> Sent from my iPhone
>>>>
>>>> On 9. sep.. 2008, at 19.05, "David Chelimsky"
<dchelimsky at gmail.com> wrote:
>>>>
>>>>> Have you tried?
>>>>>
>>>>> On Tue, Sep 9, 2008 at 12:01 PM, Matt Wynne <matt at
mattwynne.net> wrote:
>>>>>>
>>>>>> I love the way I can throw a call to pending() in the
top of an
>>>>>> unfinished
>>>>>> RSpec example and stop it from failing the build.
>>>>>> Is there a similar way to do such a thing with good
ole'' cucumber?
>>>>>>
>>>>>> 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 at rubyforge.org
>>>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>>>
>>>>> _______________________________________________
>>>>> rspec-users mailing list
>>>>> rspec-users at rubyforge.org
>>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>
>>>> _______________________________________________
>>>> rspec-users mailing list
>>>> rspec-users at rubyforge.org
>>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>>
>>> _______________________________________________
>>> rspec-users mailing list
>>> rspec-users at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/rspec-users
>>>
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 9 Sep 2008, at 21:34, aslak hellesoy wrote:
> On Tue, Sep 9, 2008 at 8:46 PM, David Chelimsky  
> <dchelimsky at gmail.com> wrote:
>> On Tue, Sep 9, 2008 at 12:47 PM, aslak hellesoy
>> <aslak.hellesoy at gmail.com> wrote:
>>> On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky  
>>> <dchelimsky at gmail.com> wrote:
>>>> Aslak - I think Matt is looking for this:
>>>>
>>>> Given /blah/ do
>>>>  pending
>>>> end
>>>>
>>>
>>> I''m not crazy about this "feature". We already
have a pending
>>> mechanism for steps (not defining them). What does this bring to
the
>>> table over just not defining a step?
>>>
>>>> Matt - I''m pretty sure that will work - give it a
shot, OK?
>>>>
>>>
>>> That won''t work. It will raise a "undefined local
variable or
>>> method `pending''"
>>
>> I get this:
>>
>>    Given I am logged in as a registered member
>>      TODO (Spec::Example::ExamplePendingError)
>>
>> for this:
>>
>> Given /I am logged in as a registered member/ do
>>  pending
>> end
>>
>
> That''s really odd - I don''t get that if I add it to this
step:
> http://github.com/aslakhellesoy/cucumber/tree/master/examples/ 
> tickets/features/steps/tickets_steps.rb
>
> And run "rake features" from examples/tickets
>
> I haven''t added any knowledge about
> pending/Spec::Example::ExamplePendingError to Cucumber so I''m
> puzzled...
>
> Aslak
Doesn''t work for us, either. I think David must have some funky rspec  
requirey stuff going on.

We see this:

   Scenario: Can see the profile page for a user who has a great amount
of details specified
     Given there is a user with all conceivable attributes and  
attachments
       implementation of attachments for users
(Spec::Example::ExamplePendingError)
       /home/nfelger/dev/skweb/vendor/plugins/rspec/lib/spec/example/ 
pending.rb:13:in
`pending''
       ./features/steps/profile_steps.rb:14:in `Given "there is a user
with all conceivable attributes and attachments"''
       features/visitor/profile.feature:13:in `Given there is a user
with all conceivable attributes and attachments''


Which is what Aslak expected, I think.
>
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.

"aslak hellesoy" <aslak.hellesoy at gmail.com> writes:
> On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky <dchelimsky at
gmail.com> wrote:
>> Aslak - I think Matt is looking for this:
>>
>> Given /blah/ do
>>  pending
>> end
>>
>
> I''m not crazy about this "feature". We already have a
pending
> mechanism for steps (not defining them). What does this bring to the
> table over just not defining a step?
The main reason is that you can have meaningful messages.  Those are
just good in general.  And I''ll admit that when you''re sitting
down to
write a feature narrative, it may not seem to have value writing a
particular message vs seeing the default pending one...but for
maintenance it can be very helpful.

In fact, here''s an example that happened to me just today (example
framework, not cucumber, but I rarely use pending ever so when I do, I
find it''s because it''s exceptionally handy):

  describe "book" do
    before(:each) do
      @book = Book.new :isbn => "12345678"
    end
  
    it "should know its amazon url" do
      @book.amazon_url.should == "http://amazon.com/ub/12345678"
      pending "need to figure out affiliate url format"
    end
  
    it "should know its borders url" do
      @book.borders_url.should ==
"http://search.borders.com?isbn=12345678"
      pending "need to figure out affiliate url format"
    end
  end

Our initial story was to link to books using an affiliate code.  But for
all providers other than amazon, it was tough finding the right format.
We''re better off waiting a couple days until our accounts are
available,
then we can look at the internal docs for 2 minutes and figure it out.
We got immediate business value today by just linking directly to these
books from within the site.  And now we''ve got these pending statements
to remind us in a couple days when we come back to this story.

Anyway, I have a feeling I could have implemented this feature in the
time it took to write this message.  But I hope it''s clear that
it''s
something that would be valuable to some people, and harmless to the
rest.

Pat

p.s. I believe a more correct action would be to split the story into
two - the first which is to provide a UI for linking to books, and the
second which is for adding the affiliate URLs.  Finish the first story,
and put the second one at the top of the queue, finishing it when the
required information becomes available.  But that''s another discussion.

On Sep 9, 11:16?pm, Pat Maddox <perg... at gmail.com>
wrote:
> "aslak hellesoy" <aslak.helle... at gmail.com> writes:
> > On Tue, Sep 9, 2008 at 7:16 PM, David Chelimsky <dchelim... at
gmail.com> wrote:
> >> Aslak - I think Matt is looking for this:
>
> >> Given /blah/ do
> >> ?pending
> >> end
>
> > I''m not crazy about this "feature". We already have
apending
> > mechanism for steps (not defining them). What does this bring to the
> > table over just not defining a step?
>
> The main reason is that you can have meaningful messages. ?Those are
> just good in general. ?And I''ll admit that when you''re
sitting down to
> write afeaturenarrative, it may not seem to have value writing a
> particular message vs seeing the defaultpendingone...but for
> maintenance it can be very helpful.
>
> In fact, here''s an example that happened to me just today (example
> framework, not cucumber, but I rarely usependingever so when I do, I
> find it''s because it''s exceptionally handy):
>
> ? describe "book" do
> ? ? before(:each) do
> ? ? ? @book = Book.new :isbn => "12345678"
> ? ? end
>
> ? ? it "should know its amazon url" do
> ? ? ? @book.amazon_url.should == "http://amazon.com/ub/12345678"
> ? ? ?pending"need to figure out affiliate url format"
> ? ? end
>
> ? ? it "should know its borders url" do
> ? ? ? @book.borders_url.should ==
"http://search.borders.com?isbn=12345678"
> ? ? ?pending"need to figure out affiliate url format"
> ? ? end
> ? end
>
> Our initial story was to link to books using an affiliate code. ?But for
> all providers other than amazon, it was tough finding the right format.
> We''re better off waiting a couple days until our accounts are
available,
> then we can look at the internal docs for 2 minutes and figure it out.
> We got immediate business value today by just linking directly to these
> books from within the site. ?And now we''ve got
thesependingstatements
> to remind us in a couple days when we come back to this story.
Is the idea that the steps still run but you get a TODO reminder?
That''s a bit different from how we use pending in code examples.

I think that could be useful:

Then /the isbn should be (.*)/ do |isbn|
  @book.borders_url.should == "http://search.borders.com?isbn=#{isbn}"
  todo "need to figure out affiliate url format"
end

....
TODO:
need to figure out affiliate url format
./features/steps/books.rb:3:in `Given /the isbn should be (.*)/

As for the "pending" feature, let''s say you have a new
scenario with 5
new steps. As a matter of process, I''d like to see the output w/ the
snippets once, copy them all in to a ruby file, add pending to each of
them (maybe even have that already in the output) and now I''m just
working w/ code.

I think it makes the process easier.

More thoughts on that?
> Anyway, I have a feeling I could have implemented thisfeaturein the
> time it took to write this message. ?But I hope it''s clear that
it''s
> something that would be valuable to some people, and harmless to the
> rest.
>
> Pat
>
> p.s. I believe a more correct action would be to split the story into
> two - the first which is to provide a UI for linking to books, and the
> second which is for adding the affiliate URLs. ?Finish the first story,
> and put the second one at the top of the queue, finishing it when the
> required information becomes available. ?But that''s another
discussion.
> _______________________________________________
> rspec-users mailing list
> rspec-us... at
rubyforge.orghttp://rubyforge.org/mailman/listinfo/rspec-users