Forum: RSpec default to just call the method?

Any interest in the following idea?


a.should include?("1:4") # if there's no matcher called include? then
just call include?

or something along those lines?
-r

On Jan 12, 2010, at 10:49 pm, rogerdpack wrote:

> a.should include?("1:4") # if there's no matcher called include? then
> just call include?

Am I right thinking that this would mean writing a method_missing that 
creates a matcher for every unhandled message on the example object 
(whatever scope the #it block runs in)?  If so, that could produce 
surprising behaviour if methods were silently turned into matchers and 
ignored.

Maybe if you restricted it to ? methods it would not be so bad.  But 
then, that's the point of the be_* handler.  The only problem is that 
#include? is a verb not an adjective/noun.  Can't say I've come across 
many of them, so it might be better in these cases to just write a 
method #a_container_for? ?

Ashley

--
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran

On Tue, Jan 12, 2010 at 5:49 PM, rogerdpack <rogerpack2005@gmail.com> 
wrote:
> Any interest in the following idea?
>
>
> a.should include?("1:4") # if there's no matcher called include? then
> just call include?
>
> or something along those lines?

-1

You can already say

a.should include("1:4")

which is clearer IMHO.


--
Rick DeNatale

Blog: http://talklikeaduck.denhaven2.com/
Twitter: http://twitter.com/RickDeNatale
WWR: http://www.workingwithrails.com/person/9021-rick-denatale
LinkedIn: http://www.linkedin.com/in/rickdenatale

On 14 Jan 2010, at 17:02, Rick DeNatale wrote:

> -1
> 
> You can already say
> 
> a.should include("1:4")
> 
> which is clearer IMHO.


I assume Roger was referring to the general case though (which I still 
don't like) - and just happened to pick an example with an existing 
matcher.

--
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran

On Jan 15, 2010, at 6:19 AM, Ashley Moran wrote:

> 
> 
> I assume Roger was referring to the general case though (which I still don't like) - and just happened to pick an example with an existing matcher.

If that's the case then I'd say it would be worth making a little macro 
to try it out.  Something like

define_simple_predicate_matcher :rise_from_the_ashes?

It would need to define a simple matcher, allowing you to do

Phoenix.new.should rise_from_the_ashes?

which would naturally expand to Phoenix.new.rise_from_the_ashes?.should 
be_true

Roger, if you're just talking about include?, follow Rick's suggestion. 
If you're talking about arbitrary predicates, then you can accomplish it 
in your own codebase with some simple metaprogramming, and see if it 
catches on with other folks.

Pat

On Jan 18, 2010, at 9:31 am, Pat Maddox wrote:

> define_simple_predicate_matcher :rise_from_the_ashes?

As an extension, how about:

define_simple_predicate_matcher :risen_from_the_ashes => 
:rise_from_the_ashes?

Also, in general, I think specs look better without ? symbols on 
methods, my preference though

Ashley

--
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran

On Mon, Jan 18, 2010 at 9:03 AM, Ashley Moran
<ashley.moran@patchspace.co.uk> wrote:
>
> On Jan 18, 2010, at 9:31 am, Pat Maddox wrote:
>
>> define_simple_predicate_matcher :rise_from_the_ashes?
>
> As an extension, how about:
>
> define_simple_predicate_matcher :risen_from_the_ashes => :rises_from_the_ashes?

I'd rather not add a new DSL for the few cases in which we want to
essentially delegate a predicate. We can already do this with the
matcher DSL:

Spec::Matchers.define :risen_from_the_ashes do
  match {|actual| actual.risen_from_the_ashes?}
end

This is better aligned w/ other matchers, more clear as to what it
does, and isn't much more code.

FWIW,
David

On Jan 18, 2010, at 3:12 pm, David Chelimsky wrote:

> I'd rather not add a new DSL for the few cases in which we want to
> essentially delegate a predicate. We can already do this with the
> matcher DSL:

I think Pat was just suggesting Roger try this in his own code.  It's 
not something I especially want in RSpec, that was just a suggestion how 
Roger could play with the macro idea.

Ashley

--
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran

On Jan 18, 2010, at 7:12 AM, David Chelimsky wrote:

> 
> I'd rather not add a new DSL for the few cases in which we want to
> essentially delegate a predicate. We can already do this with the
> matcher DSL:
> 
> Spec::Matchers.define :risen_from_the_ashes do
>  match {|actual| actual.risen_from_the_ashes?}
> end
> 
> This is better aligned w/ other matchers, more clear as to what it
> does, and isn't much more code.

This was a suggestion for OP to add to his own codebase, not RSpec, if 
he wanted to make the matcher definition even more concise.

Pat

On Mon, Jan 18, 2010 at 2:33 PM, Pat Maddox <mailinglists@patmaddox.com> 
wrote:
>>>
>> This is better aligned w/ other matchers, more clear as to what it
>> does, and isn't much more code.
>
> This was a suggestion for OP to add to his own codebase, not RSpec, if he wanted to make the matcher definition even more concise.

Understood. Thx.