Zach, the "cle" variable you ask for should have been read as "key". The
word "clé" is the French translation for the word "key". I changed it, but
obviously forgot instances, for you guys to understand the code since you
might not be familiar with French. ;-)
Thank you all for your good advices. I am looking forward to learn how to
work with rspec properly !
Mathieu
2009/2/28 Zach Dennis <zach.den...@gmail.com>
> On Fri, Feb 27, 2009 at 3:35 PM, MathLef <mlefrancoi...@gmail.com> wrote:
> > Hi,
>
> Hi Mathieu,
>
> >
> > I need help to spec a local variable in a controller. I have been
> > wandering how I can do that since all I have done so far is useless.
> > Note that the code is in a controller and that it is not yet finished
> > since I am trying to spec it as I develop it. Here is what the code
> > should look like when it is finished :
> >
> > ...
> > crits = {}
> > params.each {
> > |key, val|
> > crits[key.to_sym] = StringUtils.remove_chars(val).upcase if not
> > (cle == "controller" || cle == "action")
> > }
> >
> > if crits[:name].nil?
> > add_where(where_clause, crits[:name])
> > end
> > ...
> >
> > I am trying to mock the crits local variable. Here is what the rspec
> > looks like so far :
> >
> > @mock_crits = mock(Hash, { :name => 'Steve' })
> > assigns(:crits) = @mock_crits
> >
> > First, I can not seem to have control on the params array. How can I
> > mock this array ?
>
> You can control "params" by passing in a hash representative of query
> parameters from a request.
>
> ie: get :action_name, :foo => "bar", :bar => "baz", :baz => "banana"
>
> > Also, I also can not seem to have control on the
> > crits local variable when I try to spec the if crits[:name].nil?. Is
> > this a bad way to code ? Can anyone explain this ?
>
> I think you have some room to adjust the design a little to make your
> controller responsible for doing less, and making your model
> responsible for doing more. This will open up less complicated ways of
> spec'ing both your controller and the model. It's hard to give a
> concrete example because the term "crits" means nothing to me. What is
> the intention behind building up the crits hash?
>
> Also, what does 'cle == "controller" || cle == "action"' represent?
>
> Not knowing anything else I would probably try to make your controller
> code look like:
>
> def your_action
> crits = SomeOtherObject.parse_crits params.except(:controller, :action)
> FooModel.find_stuff_by_crits crits
> end
>
> My only disclaimer on this is that I don't know anything about your
> code other than what you posted, so it is up to you to find a good
> home for the appropriate methods and logic.
>
> This approach lets you spec the parsing of "crits" separate from the
> controller action while being able to ensure the controller action did
> in fact parse them. Here's a sample spec for the #your_action I posted
> above:
>
> before(:each) do
> SomeOtherObject.stub!(:parse_crits)
> end
>
> it "should parse crits from params" do
> SomeOtherObject.should_receive(:parse_crits).with("foo" => "bar")
> get :your_action, "foo" => "bar"
> end
>
> it "should have find whatever it is you're looking for" do
> crits_stub = stub("crits")
> SomeOtherObject.stub!(:parse_crits).and_return stub("crits")
> FooModel.should_receive(:find_stuff_by_crits).with(crits_stub)
> get :your_action
> end
>
> This makes the controller spec much easier to gr0k and write! Although
> I urge you to find better names for "crits" and to hide some of the
> functionality behind intention-revealing interfaces. It not only makes
> your code easier for you to read and maintain, but it does that for
> anyone who has to look at, either on this mailing list, or other
> developers who will come back and maintain the code.
>
> >
> > Many thanks in advance.
> > Mathieu
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users@rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
>
>
>
> --
> Zach Dennis
> http://www.continuousthinking.com
> http://www.mutuallyhuman.com
> _______________________________________________
> rspec-users mailing list
> rspec-users@rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>
_______________________________________________
rspec-users mailing list
rspec-users@rubyforge.org
http://rubyforge.org/mailman/listinfo/rspec-users
