From pergesu at gmail.com  Mon Jan  1 03:47:53 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Mon, 1 Jan 2007 01:47:53 -0700
Subject: [rspec-users] Validations based on associations
Message-ID: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>

My model is very simple, it's mostly just a join table that represents
which tournaments a user has registered for.

class Registration < ActiveRecord::Base
  belongs_to :user
  belongs_to :tournament

  validates_presence_of :user_id, :tournament_id
  validates_uniqueness_of :user_id, :scope => :tournament_id
end

the validates_uniqueness checks to make sure there's no record that
has the same user_id and tournament_id.

Is there a relatively easy way to specify this behavior?  The only way
that I know is to do something like

context "A Registration with a User and Tournament" do
  setup do
    @user = User.new
    @user.save false
    @tournament = Tournament.new
    @tournament.save false
    @reg = Registration.new(:user => @user, :tournament => @tournament)
  end

  specify "should be valid" do
    @reg.should_be_valid
  end

  specify "should not be valid if a registration with the same details
already exists" do
    @reg.save
    Registration.new(:user => @user, :tournament =>
@tournament).should_not_be_valid
  end
end

Well actually that's obviously pretty easy :)  Is it bad that I'm tied
to the implementations of User and Tournament at all?  ActiveRecord
dictates that, so I guess it's not a problem to have that in my spec
at all.

Anyway this is just the spec I came up with and I'm wondering if
anyone else has a better idea.

Pat

From pergesu at gmail.com  Mon Jan  1 03:49:03 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Mon, 1 Jan 2007 01:49:03 -0700
Subject: [rspec-users] Validations based on associations
In-Reply-To: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
References: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
Message-ID: <810a540e0701010049j755b0304m4d3b8a495f8ac5f9@mail.gmail.com>

oh, I forgot to say that the @user.save(false) means that the user
object gets saved but skips validation.  That's a quick way of not
having to create a completely valid user object that may become out of
date as you change the User model.

and happy new year everyone!

Pat

From pergesu at gmail.com  Mon Jan  1 04:05:02 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Mon, 1 Jan 2007 02:05:02 -0700
Subject: [rspec-users] Another "how do I spec this?"
Message-ID: <810a540e0701010105sb5610e2t1c3c7d94da67fe54@mail.gmail.com>

I wanted to add a convenience method on my User class to see if he was
already signed up for a tournament.  Here's my spec

context "A User signed up for one tournament" do
  setup do
    @t1 = Tournament.new
    @t1.save false
    @t2 = Tournament.new
    @t2.save false
    @user = User.new
    @user.save false
    @user.registrations << Registration.new(:tournament => @t1)
  end

  specify "should be signed up for that tournament" do
    @user.should_be_signed_up_for @t1
  end

  specify "should not be signed up for another tournament" do
    @user.should_not_be_signed_up_for @t2
  end
end

class User < ActiveRecord::Base
  has_many :registrations

  def signed_up_for?(tournament)
    !registrations.find_by_tournament_id(tournament).nil?
  end
end

Are there any potential improvements, or is that the best way to spec it?

Pat

From dchelimsky at gmail.com  Mon Jan  1 08:33:57 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 1 Jan 2007 08:33:57 -0500
Subject: [rspec-users] Validations based on associations
In-Reply-To: <810a540e0701010049j755b0304m4d3b8a495f8ac5f9@mail.gmail.com>
References: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
	<810a540e0701010049j755b0304m4d3b8a495f8ac5f9@mail.gmail.com>
Message-ID: <57c63afe0701010533m4d9bc67ei7b244ab58a1d7de3@mail.gmail.com>

On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> oh, I forgot to say that the @user.save(false) means that the user
> object gets saved but skips validation.  That's a quick way of not
> having to create a completely valid user object that may become out of
> date as you change the User model.

That's the kind of stuff you should be blogging! Very useful tip.

Thanks Pat.

>
> and happy new year everyone!

You too!

David

>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Mon Jan  1 09:07:18 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 1 Jan 2007 09:07:18 -0500
Subject: [rspec-users] Validations based on associations
In-Reply-To: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
References: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
Message-ID: <57c63afe0701010607m3bcab306k761a8a3156b1e0ec@mail.gmail.com>

On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> My model is very simple, it's mostly just a join table that represents
> which tournaments a user has registered for.
>
> class Registration < ActiveRecord::Base
>   belongs_to :user
>   belongs_to :tournament
>
>   validates_presence_of :user_id, :tournament_id
>   validates_uniqueness_of :user_id, :scope => :tournament_id
> end
>
> the validates_uniqueness checks to make sure there's no record that
> has the same user_id and tournament_id.
>
> Is there a relatively easy way to specify this behavior?  The only way
> that I know is to do something like
>
> context "A Registration with a User and Tournament" do
>   setup do
>     @user = User.new
>     @user.save false
>     @tournament = Tournament.new
>     @tournament.save false
>     @reg = Registration.new(:user => @user, :tournament => @tournament)
>   end
>
>   specify "should be valid" do
>     @reg.should_be_valid
>   end
>
>   specify "should not be valid if a registration with the same details
> already exists" do
>     @reg.save
>     Registration.new(:user => @user, :tournament =>
> @tournament).should_not_be_valid
>   end
> end

Geez, ActiveRecord models are SUCH a violation of SRP! I realize that
this is the rails way, and that there are great productivity benefits
we glean from AR, but having a model that validates both at both the
instance level (validates_presence_of) and the class level
(validates_uniqueness_of) is a pain in the ass. It's confusing because
both sorts of validations look the same, but they are really
fundamentally different.

Ranting aside, there are a couple of different ways you can view this.
To be the most clear, I'd probably spec the class level validations
separately from instance level validations. The uniqueness spec might
look like this:

context "The Registration model class" do
  setup do
    @user1 = User.new
    @user2 = User.new
    @tournament = Tournament.new

    [@user1, @user2, @tournament].each {|model| model.save(false)}
  end

  specify "should permit registrations with different users" do
    Registration.create(:user => @user1, :tournament => @tournament)
    Registration.new(:user => @user2, :tournament =>
@tournament).should_be_valid
  end

  specify "should deny multiple registrations with the same user" do
    Registration.create(:user => @user1, :tournament => @tournament)
    Registration.new(:user => @user1, :tournament =>
@tournament).should_not_be_valid
  end
end

WDYT?

>
> Well actually that's obviously pretty easy :)  Is it bad that I'm tied
> to the implementations of User and Tournament at all?  ActiveRecord
> dictates that, so I guess it's not a problem to have that in my spec
> at all.
>
> Anyway this is just the spec I came up with and I'm wondering if
> anyone else has a better idea.
>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Mon Jan  1 09:16:05 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 1 Jan 2007 09:16:05 -0500
Subject: [rspec-users] Another "how do I spec this?"
In-Reply-To: <810a540e0701010105sb5610e2t1c3c7d94da67fe54@mail.gmail.com>
References: <810a540e0701010105sb5610e2t1c3c7d94da67fe54@mail.gmail.com>
Message-ID: <57c63afe0701010616x15c2e61rca9a2459d0ddd548@mail.gmail.com>

On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> I wanted to add a convenience method on my User class to see if he was
> already signed up for a tournament.  Here's my spec
>
> context "A User signed up for one tournament" do
>   setup do
>     @t1 = Tournament.new
>     @t1.save false
>     @t2 = Tournament.new
>     @t2.save false
>     @user = User.new
>     @user.save false
>     @user.registrations << Registration.new(:tournament => @t1)
>   end
>
>   specify "should be signed up for that tournament" do
>     @user.should_be_signed_up_for @t1
>   end
>
>   specify "should not be signed up for another tournament" do
>     @user.should_not_be_signed_up_for @t2
>   end
> end
>
> class User < ActiveRecord::Base
>   has_many :registrations
>
>   def signed_up_for?(tournament)
>     !registrations.find_by_tournament_id(tournament).nil?
>   end
> end
>
> Are there any potential improvements, or is that the best way to spec it?

Couple of thoughts. Creating new unvalidated models seems to be
something useful, so...

module ModelSpecUtils
  def unvalidated_model(klass)
    model = klass.new
    model.save(false)
    model
  end
end

Also, this line:

@user.registrations << Registration.new(:tournament => @t1)

is breaking encapsulation a bit. Given that and the utility module
above, I might end up w/ something like this:

context "A User signed up for one tournament" do
  include ModelSpecUtils

  setup do
    @tournament1 = unvalidated_model(Tournament)
    @tournament2 = unvalidated_model(Tournament)
    @user = unvalidated_model(User)
    @user.register_for(@tournament1)
  end

  specify "should be signed up for that tournament" do
    @user.should_be_signed_up_for @tournament1
  end

  specify "should not be signed up for another tournament" do
    @user.should_not_be_signed_up_for @tournament2
  end
end

David

>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From lastobelus at mac.com  Mon Jan  1 21:46:47 2007
From: lastobelus at mac.com (Michael Johnston)
Date: Mon, 1 Jan 2007 18:46:47 -0800
Subject: [rspec-users] rspec_autotest no longer re-runs tests with rspec
	0.7.5
In-Reply-To: <4232AA34-F6EC-42BE-A6D9-044F05D29149@mac.com>
References: <4FF77DF8-CDB3-466D-AC16-925BB4842491@mac.com>
	<f0b861390612211113k3d19d2w2adaa75504bddade@mail.gmail.com>
	<4232AA34-F6EC-42BE-A6D9-044F05D29149@mac.com>
Message-ID: <2E2973DA-BE37-4849-B277-1BF6F479F906@mac.com>

Ok I just took the time to figure this out. It should have been  
obvious to me: turning on redgreen breaks the regex in  
"handle_results" which should create an array of failed tests because  
the color codes that are added to the result string.

I discovered it was also mildly broken without color, because the  
second term in the regex needs to be made non greedy, or it matches  
everything after the first error. But because of the way this is  
used, I don't think this would make much (any) difference to the  
functioning of autotest.

Anyway, so here is the diff with the regex for handle_results  
function of rspec_autotest that works with redgreen turned on (and  
with spec command changed to use the rails_spec_server):

Index: lib/rspec_autotest.rb
===================================================================
--- lib/rspec_autotest.rb       (revision 45)
+++ lib/rspec_autotest.rb       (working copy)
@@ -26,7 +26,7 @@
    attr_accessor :spec_command
    def initialize # :nodoc:
-    @spec_command = "spec --diff unified"
+    @spec_command = "script/rails_spec -C --diff unified"
      super
      @exceptions = %r%^\./(?:coverage|db|doc|log|public|script|vendor)%
    end
@@ -76,7 +76,7 @@
    end
    def handle_results(results)
-    failed = results.scan(/^\d+\)\n(?:.*?Error in )?'([^']*?)'(?:  
FAILED)?\n(.*)\n\n/m)
+    failed = results.scan(/^\d+\)\n(?:\e\[\d*m)?(?:.*?Error  
in )?'([^']*?)'(?: FAILED)?(?:\e\[\d*m)?\n(.*?)\n\n/m)
      @files_to_test = consolidate_failures failed
      unless @files_to_test.empty? then
        hook :red




On 30-Dec-06, at 4:46 AM, Michael Johnston wrote:

> That is what I meant. In my setup, it is NOT rerunning the whole  
> suite after I fix a failing spec.
>
> After a cursory look  at autotest.rb in ZenTest, I haven't figured  
> out why it is broken. But I think I will try to figure that out  
> today, because autotest seems kind of pointless without that  
> behaviour.
>
> It probably has nothing to do with rspec or rspec_autotest; I  
> realized that my version of ZenTest is new too, 3.4.3 instead of  
> 3.4.1.
>
> Cheers,
> Michael
> On 21-Dec-06, at 11:13 AM, Nick Sieger wrote:
>
>> On 12/21/06, Michael Johnston <lastobelus at mac.com> wrote:
>> Has anyone else encountered this problem?
>>
>> What I mean is, rspec_autotest runs all specs, then waits. When you
>> change a spec, it reruns that spec. However, if that spec now passes,
>> it should run the whole suite again, so you can see the next one to
>> work on. It no longer does this. This renders it somewhat useless.
>>
>> Does anyone have autotest working with rspec 0.7.5, or does anyone
>> have any idea of what may be causing this issue or how to go about
>> fixing it?
>>
>> Are you sure this is caused by the 0.7.5 upgrade?  I've seen  
>> similar behavior but I'm still on 0.7.3.  Here's the response  
>> comment I posted to yours:
>>
>> I think this is expected behavior for autotest ? I've already  
>> noticed this and assumed it was standard behavior. The only time  
>> autotest reruns the whole suite is after I've just fixed a failing  
>> test or spec. Is it possible that's what you're observing?
>>
>> Feel free to take any further discussion off-list, thanks.
>>
>> /Nick
>> _______________________________________________
>> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070101/f0575b74/attachment.html 

From yrashk at gmail.com  Mon Jan  1 22:19:23 2007
From: yrashk at gmail.com (Yurii Rashkovskii)
Date: Tue, 2 Jan 2007 05:19:23 +0200
Subject: [rspec-users] acts_as_authenticated plugin rspecs
Message-ID: <FF60F651-C7E9-4C9A-B83D-8D9D21383284@gmail.com>

Hello,

Just thought it could be interesting for some of you. I've written  
rspecs for acts_as_authenticated plugin.

http://rashkovskii.com/articles/2007/01/01/rspec-on-rails- 
acts_as_authenticated

I'm not sure they are 100% ok, but they seem to be :)

Thank you,
Yurii.

From nicksieger at gmail.com  Tue Jan  2 11:03:52 2007
From: nicksieger at gmail.com (Nick Sieger)
Date: Tue, 2 Jan 2007 10:03:52 -0600
Subject: [rspec-users] rspec_autotest no longer re-runs tests with rspec
	0.7.5
In-Reply-To: <2E2973DA-BE37-4849-B277-1BF6F479F906@mac.com>
References: <4FF77DF8-CDB3-466D-AC16-925BB4842491@mac.com>
	<f0b861390612211113k3d19d2w2adaa75504bddade@mail.gmail.com>
	<4232AA34-F6EC-42BE-A6D9-044F05D29149@mac.com>
	<2E2973DA-BE37-4849-B277-1BF6F479F906@mac.com>
Message-ID: <f0b861390701020803s50bb8ccv11949dfb04353814@mail.gmail.com>

On 1/1/07, Michael Johnston <lastobelus at mac.com> wrote:
>
> Ok I just took the time to figure this out. It should have been obvious to
> me: turning on redgreen breaks the regex in "handle_results" which should
> create an array of failed tests because the color codes that are added to
> the result string.
> I discovered it was also mildly broken without color, because the second
> term in the regex needs to be made non greedy, or it matches everything
> after the first error. But because of the way this is used, I don't think
> this would make much (any) difference to the functioning of autotest.
>
> Anyway, so here is the diff with the regex for handle_results function of
> rspec_autotest that works with redgreen turned on (and with spec command
> changed to use the rails_spec_server):
>
> Index: lib/rspec_autotest.rb
> ===================================================================
> --- lib/rspec_autotest.rb       (revision 45)
> +++ lib/rspec_autotest.rb       (working copy)
> @@ -26,7 +26,7 @@
>    attr_accessor :spec_command
>    def initialize # :nodoc:
> -    @spec_command = "spec --diff unified"
> +    @spec_command = "script/rails_spec -C --diff unified"
>      super
>      @exceptions = %r%^\./(?:coverage|db|doc|log|public|script|vendor)%
>    end
> @@ -76,7 +76,7 @@
>    end
>    def handle_results(results)
> -    failed = results.scan(/^\d+\)\n(?:.*?Error in )?'([^']*?)'(?:
> FAILED)?\n(.*)\n\n/m)
> +    failed = results.scan(/^\d+\)\n(?:\e\[\d*m)?(?:.*?Error in
> )?'([^']*?)'(?: FAILED)?(?:\e\[\d*m)?\n(.*?)\n\n/m)
>      @files_to_test = consolidate_failures failed
>      unless @files_to_test.empty? then
>        hook :red
>

Thanks Michael, I committed a slightly modified version of the regex.  I
didn't commit the change to @spec_command because you can easily customize
this for yourself by putting a hook in your ~/.autotest:

  Autotest.add_hook :initialize do |at|
    if at.respond_to? :spec_command
      at.spec_command = "script/rails_spec -C --diff unified"
    end
  end

Cheers,
/Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070102/1e08b634/attachment.html 

From cwdinfo at gmail.com  Wed Jan  3 11:57:44 2007
From: cwdinfo at gmail.com (s.ross)
Date: Wed, 3 Jan 2007 08:57:44 -0800
Subject: [rspec-users] 0.7.5 Doesn't Let My Tests Run
Message-ID: <9EB5B084-7D7E-4765-B36E-FEC9E52ACCD5@gmail.com>

The tests are all running on 0.7.4. In installed 0.7.5 as follows:

- sudo gem install rspec
- script/plugin install svn://rubyforge.org/var/svn/rspec/tags/ 
REL_0_7_5_/vendor/rspec_on_rails/vendor/plugins/rspec
- script/generate rspec

ZenTest (3.4.1) is also installed.

Here is what happens when I run the same tests that ran on 0.7.4. (By  
the way, Rails is build 5065). Any help is appreciated.

$ rake spec
(in /Users/minime/rails/voodoo_work)
/opt/local/bin/ruby -I"/opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/ 
lib" "/opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec" "spec/ 
models/cart_spec.rb"
/Users/minime/rails/voodoo_work/config/../vendor/rails/activerecord/ 
lib/../../activesupport/lib/active_support/dependencies.rb:349:in  
`const_missing': uninitialized constant  
Spec::RailsPlugin::ResponseBody::TagExpectations (NameError)
         from /Users/minime/rails/voodoo_work/vendor/plugins/vendor/ 
plugins/rspec/lib/spec/rails_plugin/response_body.rb:4
         from /opt/local/lib/ruby/site_ruby/1.8/rubygems/ 
custom_require.rb:27:in `gem_original_require'
         from /opt/local/lib/ruby/site_ruby/1.8/rubygems/ 
custom_require.rb:27:in `require'
         from /Users/minime/rails/voodoo_work/config/../vendor/rails/ 
activerecord/lib/../../activesupport/lib/active_support/ 
dependencies.rb:364:in `require'
         from /Users/minime/rails/voodoo_work/config/../vendor/ 
plugins/vendor/plugins/rspec/lib/rspec_on_rails.rb:16
         from /opt/local/lib/ruby/site_ruby/1.8/rubygems/ 
custom_require.rb:27:in `gem_original_require'
         from /opt/local/lib/ruby/site_ruby/1.8/rubygems/ 
custom_require.rb:27:in `require'
         from /Users/minime/rails/voodoo_work/config/../vendor/rails/ 
activerecord/lib/../../activesupport/lib/active_support/ 
dependencies.rb:364:in `require'
         from ./spec/models/../spec_helper.rb:3
         from ./spec/models/cart_spec.rb:1:in `require'
         from ./spec/models/cart_spec.rb:1
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:21:in `load'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:21:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:15:in `each'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:15:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4

From listaccount at e-tobi.net  Thu Jan  4 18:03:36 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Fri, 05 Jan 2007 00:03:36 +0100
Subject: [rspec-users] Common setup code and naming specifications
Message-ID: <459D87C8.1010300@e-tobi.net>

Hello!

I have a lot of contexts for testing Rails controllers, that must do
something like 'session[:logged_in] = true' in their setup. How can this
be refactored? In unit tests I would simply create a
LoggedInControllerTest base class, that all my functional tests would
derive from.

And another small question:

In my controller specifications I often have to decide whether to write:

specify "should provide the first ten items in @items and three pages in
@pages when passing no :page parameter to the :list action"

or

specify "should provide the first ten items and three pages when not
selecting a specific page"

So the decision is, whether to explicitly name parameters, return values
and actions or to use a more abstract phrase. How do you handle this?
Sorry, might be a stupid question, but I'm, still trying to get the
right "feeling" for BDD-style testing.

bye,

Tobias


From dchelimsky at gmail.com  Thu Jan  4 18:12:18 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 4 Jan 2007 18:12:18 -0500
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <459D87C8.1010300@e-tobi.net>
References: <459D87C8.1010300@e-tobi.net>
Message-ID: <57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>

On 1/4/07, Tobias Grimm <listaccount at e-tobi.net> wrote:
> Hello!
>
> I have a lot of contexts for testing Rails controllers, that must do
> something like 'session[:logged_in] = true' in their setup. How can this
> be refactored? In unit tests I would simply create a
> LoggedInControllerTest base class, that all my functional tests would
> derive from.

module MyHelpers
  def set_logged_in
    session[:logged_in] = true
  end
end

context "..." do
  include MyHelpers
  setup do
    set_logged_in
  end
end

>
> And another small question:
>
> In my controller specifications I often have to decide whether to write:
>
> specify "should provide the first ten items in @items and three pages in
> @pages when passing no :page parameter to the :list action"
>
> or
>
> specify "should provide the first ten items and three pages when not
> selecting a specific page"

Definitely the latter. These names should be how the customer would
talk about requirements, not how developers would - UNLESS you're
writing a framework for developers and developers ARE your customers.

David

> So the decision is, whether to explicitly name parameters, return values
> and actions or to use a more abstract phrase. How do you handle this?
> Sorry, might be a stupid question, but I'm, still trying to get the
> right "feeling" for BDD-style testing.
>
> bye,
>
> Tobias
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From bob.cotton at rallydev.com  Thu Jan  4 18:34:14 2007
From: bob.cotton at rallydev.com (Bob Cotton)
Date: Thu, 04 Jan 2007 16:34:14 -0700
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
	(David Chelimsky's message of "Thu, 4 Jan 2007 18:12:18 -0500")
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
Message-ID: <uvc3b6qbc3t.fsf@england.f4tech.com>


"David Chelimsky" <dchelimsky at gmail.com> writes:

> On 1/4/07, Tobias Grimm <listaccount at e-tobi.net> wrote:
>> Hello!
>>
>> I have a lot of contexts for testing Rails controllers, that must do
>> something like 'session[:logged_in] = true' in their setup. How can this
>> be refactored? In unit tests I would simply create a
>> LoggedInControllerTest base class, that all my functional tests would
>> derive from.
>
> module MyHelpers
>   def set_logged_in
>     session[:logged_in] = true
>   end
> end
>
> context "..." do
>   include MyHelpers
>   setup do
>     set_logged_in
>   end
> end
>


What about:

module LoggedIn
  def setup 
    session[:logged_in] = true
  end
end

context "..." do
  include LoggedIn
end

A little less typing, and just as clear.

 -Bob



>>
>> And another small question:
>>
>> In my controller specifications I often have to decide whether to write:
>>
>> specify "should provide the first ten items in @items and three pages in
>> @pages when passing no :page parameter to the :list action"
>>
>> or
>>
>> specify "should provide the first ten items and three pages when not
>> selecting a specific page"
>
> Definitely the latter. These names should be how the customer would
> talk about requirements, not how developers would - UNLESS you're
> writing a framework for developers and developers ARE your customers.
>
> David
>
>> So the decision is, whether to explicitly name parameters, return values
>> and actions or to use a more abstract phrase. How do you handle this?
>> Sorry, might be a stupid question, but I'm, still trying to get the
>> right "feeling" for BDD-style testing.
>>
>> bye,
>>
>> Tobias
>>
>> _______________________________________________
>> 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

-- 
Bob Cotton
Test Architect -- Rally Software -- rallydev.com
http://www.testarchitecture.com/blog

From listaccount at e-tobi.net  Thu Jan  4 19:15:38 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Fri, 05 Jan 2007 01:15:38 +0100
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
Message-ID: <459D98AA.3050003@e-tobi.net>

Thanks for your fast response!

David Chelimsky wrote:
> context "..." do
>   include MyHelpers
>   setup do
>     set_logged_in
>   end
> end
>   

Ok, but this still requires me to call set_logged_in in 99% of my
controller contexts. I can live with it, but it tastes like evil
duplication.

>> specify "should provide the first ten items in @items and three pages in
>> @pages when passing no :page parameter to the :list action"
>>
>> specify "should provide the first ten items and three pages when not
>> selecting a specific page"
>>     
>
> Definitely the latter. These names should be how the customer would
> talk about requirements, not how developers would - UNLESS you're
> writing a framework for developers and developers ARE your customers.
>   

Fine - sounds reasonable! But sometimes it's hard to see, whether you
are working on a framework or not. If I'm at a controller, it's not only
the customer who uses it in some way, it's also the view. And the view
needs to call :list and pass :page and expects to get @items and @pages.
The second of the two specifications above tells nothing about how the
view should interact with the controller.

Tobias


From bryan at osesm.com  Thu Jan  4 19:16:18 2007
From: bryan at osesm.com (Bryan Liles)
Date: Thu, 4 Jan 2007 19:16:18 -0500
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <uvc3b6qbc3t.fsf@england.f4tech.com>
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
	<uvc3b6qbc3t.fsf@england.f4tech.com>
Message-ID: <BD1126A9-F45A-4D13-B131-31D80E9AE754@osesm.com>

On Jan 4, 2007, at 6:34 PM, Bob Cotton wrote:

>
> What about:
>
> module LoggedIn
>   def setup
>     session[:logged_in] = true
>   end
> end
>
> context "..." do
>   include LoggedIn
> end
>

With this method, don't you lose access to setup?  Seems like a lot  
to lose just to gain a a login mixin.

From bryan at osesm.com  Thu Jan  4 21:05:39 2007
From: bryan at osesm.com (Bryan Liles)
Date: Thu, 4 Jan 2007 21:05:39 -0500
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
Message-ID: <CAD71583-F29B-4AE1-9F36-87D489A8D270@osesm.com>

I seem to be missing something because I am at a loss on how to get  
rcov, RSpec, and Rails working together.  Would someone mind dropping  
some hints?

From dchelimsky at gmail.com  Thu Jan  4 22:05:02 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 4 Jan 2007 22:05:02 -0500
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <459D98AA.3050003@e-tobi.net>
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
	<459D98AA.3050003@e-tobi.net>
Message-ID: <57c63afe0701041905sd54b02ew48492c89b3205d11@mail.gmail.com>

On 1/4/07, Tobias Grimm <listaccount at e-tobi.net> wrote:
> Thanks for your fast response!
>
> David Chelimsky wrote:
> > context "..." do
> >   include MyHelpers
> >   setup do
> >     set_logged_in
> >   end
> > end
> >
>
> Ok, but this still requires me to call set_logged_in in 99% of my
> controller contexts. I can live with it, but it tastes like evil
> duplication.

Ah, the DRY battle cry.

There is a difference between duplication of code and duplicated calls
to a no-arg method. While typing the call to set_logged_in may require
a few extra strokes, the meaning of it will never change in different
contexts. So the risks associated w/ code duplication are mitigated.
In fact, the real risk is that you might change its one and only
meaning, in which case you'd have to look at each case and decide if
the new meaning makes sense.

That problem wouldn't go away if you could subclass contexts. In fact,
it would be more sinister because the behaviour is implicit. At least
if you look at set_logged_in in each context you have some sense of
what it means.

There is also a difference between duplication in code and duplication
in tests. Jay Fields put it very well when he put it this way: test
are inherently procedural. The risk of duplication in code is that
when you have to change it in one place you'll miss the other place.
That risk doesn't really apply to this situation.

Alternatively, you could figure out a way to subclass contexts and not
be able to look at a given context and understand why the user keeps
on ending up logged in.

My only slightly solicited 2 cents.

>
> >> specify "should provide the first ten items in @items and three pages in
> >> @pages when passing no :page parameter to the :list action"
> >>
> >> specify "should provide the first ten items and three pages when not
> >> selecting a specific page"
> >>
> >
> > Definitely the latter. These names should be how the customer would
> > talk about requirements, not how developers would - UNLESS you're
> > writing a framework for developers and developers ARE your customers.
> >
>
> Fine - sounds reasonable! But sometimes it's hard to see, whether you
> are working on a framework or not. If I'm at a controller, it's not only
> the customer who uses it in some way, it's also the view. And the view
> needs to call :list and pass :page and expects to get @items and @pages.
> The second of the two specifications above tells nothing about how the
> view should interact with the controller.

Excellent point. There's no simple answer for this. Your best bet is
to keep throwing examples at us and we can discuss them more
explicitly.

Cheers,
David

> Tobias
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From bob.cotton at rallydev.com  Thu Jan  4 22:42:48 2007
From: bob.cotton at rallydev.com (Bob Cotton)
Date: Thu, 04 Jan 2007 20:42:48 -0700
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <BD1126A9-F45A-4D13-B131-31D80E9AE754@osesm.com> (Bryan Liles's
	message of "Thu, 4 Jan 2007 19:16:18 -0500")
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
	<uvc3b6qbc3t.fsf@england.f4tech.com>
	<BD1126A9-F45A-4D13-B131-31D80E9AE754@osesm.com>
Message-ID: <uvcodpe9m13.fsf@england.f4tech.com>

Bryan Liles <bryan at osesm.com> writes:

> On Jan 4, 2007, at 6:34 PM, Bob Cotton wrote:
>
>>
>> What about:
>>
>> module LoggedIn
>>   def setup
>>     session[:logged_in] = true
>>   end
>> end
>>
>> context "..." do
>>   include LoggedIn
>> end
>>
>
> With this method, don't you lose access to setup?  Seems like a lot  
> to lose just to gain a a login mixin.

Sorry, LoggedIn should have been a class:

class One
  def setup
    puts "in setup one"
  end
end

context "context with inherit" do
  inherit One

  setup do
    puts "in context setup"
  end

  specify "some specify" do
    puts "in specify"
  end
end

$ spec foo_spec..rb

in setup one
in context setup
in specify
.


I do agree with David, that some duplication in test code is ok. It
would be easy to get the LoggedIn class buried up the inheritance
hierarchy where it can't be seen.

But if it's the only superclass of the context, and it's visible, and
it named correctly, this will save you some typing.

 -Bob

-- 
Bob Cotton
Test Architect -- Rally Software -- rallydev.com
http://www.testarchitecture.com/blog

From dchelimsky at gmail.com  Thu Jan  4 22:50:01 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 4 Jan 2007 22:50:01 -0500
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <uvcodpe9m13.fsf@england.f4tech.com>
References: <459D87C8.1010300@e-tobi.net>
	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>
	<uvc3b6qbc3t.fsf@england.f4tech.com>
	<BD1126A9-F45A-4D13-B131-31D80E9AE754@osesm.com>
	<uvcodpe9m13.fsf@england.f4tech.com>
Message-ID: <57c63afe0701041950r693df98n4edbadda42837567@mail.gmail.com>

On 1/4/07, Bob Cotton <bob.cotton at rallydev.com> wrote:
> Bryan Liles <bryan at osesm.com> writes:
>
> > On Jan 4, 2007, at 6:34 PM, Bob Cotton wrote:
> >
> >>
> >> What about:
> >>
> >> module LoggedIn
> >>   def setup
> >>     session[:logged_in] = true
> >>   end
> >> end
> >>
> >> context "..." do
> >>   include LoggedIn
> >> end
> >>
> >
> > With this method, don't you lose access to setup?  Seems like a lot
> > to lose just to gain a a login mixin.
>
> Sorry, LoggedIn should have been a class:
>
> class One
>   def setup
>     puts "in setup one"
>   end
> end
>
> context "context with inherit" do
>   inherit One
>
>   setup do
>     puts "in context setup"
>   end
>
>   specify "some specify" do
>     puts "in specify"
>   end
> end
>
> $ spec foo_spec..rb
>
> in setup one
> in context setup
> in specify
> .
>
>
> I do agree with David, that some duplication in test code is ok. It
> would be easy to get the LoggedIn class buried up the inheritance
> hierarchy where it can't be seen.
>
> But if it's the only superclass of the context, and it's visible, and
> it named correctly, this will save you some typing.

Great idea - but unfortunately we already use "inherit" in all of the
rails contexts. That's ultimately how we're able to tap into
Test::Unit::TestCase to take advantage of fixture loading, etc.

>
>  -Bob
>
> --
> Bob Cotton
> Test Architect -- Rally Software -- rallydev.com
> http://www.testarchitecture.com/blog
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From pergesu at gmail.com  Thu Jan  4 23:22:22 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Thu, 4 Jan 2007 21:22:22 -0700
Subject: [rspec-users] Validations based on associations
In-Reply-To: <57c63afe0701010607m3bcab306k761a8a3156b1e0ec@mail.gmail.com>
References: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
	<57c63afe0701010607m3bcab306k761a8a3156b1e0ec@mail.gmail.com>
Message-ID: <810a540e0701042022r4cc726e7p842e595e76c934a@mail.gmail.com>

On 1/1/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> > My model is very simple, it's mostly just a join table that represents
> > which tournaments a user has registered for.
> >
> > class Registration < ActiveRecord::Base
> >   belongs_to :user
> >   belongs_to :tournament
> >
> >   validates_presence_of :user_id, :tournament_id
> >   validates_uniqueness_of :user_id, :scope => :tournament_id
> > end
> >
> > the validates_uniqueness checks to make sure there's no record that
> > has the same user_id and tournament_id.
> >
> > Is there a relatively easy way to specify this behavior?  The only way
> > that I know is to do something like
> >
> > context "A Registration with a User and Tournament" do
> >   setup do
> >     @user = User.new
> >     @user.save false
> >     @tournament = Tournament.new
> >     @tournament.save false
> >     @reg = Registration.new(:user => @user, :tournament => @tournament)
> >   end
> >
> >   specify "should be valid" do
> >     @reg.should_be_valid
> >   end
> >
> >   specify "should not be valid if a registration with the same details
> > already exists" do
> >     @reg.save
> >     Registration.new(:user => @user, :tournament =>
> > @tournament).should_not_be_valid
> >   end
> > end
>
> Geez, ActiveRecord models are SUCH a violation of SRP! I realize that
> this is the rails way, and that there are great productivity benefits
> we glean from AR, but having a model that validates both at both the
> instance level (validates_presence_of) and the class level
> (validates_uniqueness_of) is a pain in the ass. It's confusing because
> both sorts of validations look the same, but they are really
> fundamentally different.
>
> Ranting aside, there are a couple of different ways you can view this.
> To be the most clear, I'd probably spec the class level validations
> separately from instance level validations. The uniqueness spec might
> look like this:
>
> context "The Registration model class" do
>   setup do
>     @user1 = User.new
>     @user2 = User.new
>     @tournament = Tournament.new
>
>     [@user1, @user2, @tournament].each {|model| model.save(false)}
>   end
>
>   specify "should permit registrations with different users" do
>     Registration.create(:user => @user1, :tournament => @tournament)
>     Registration.new(:user => @user2, :tournament =>
> @tournament).should_be_valid
>   end
>
>   specify "should deny multiple registrations with the same user" do
>     Registration.create(:user => @user1, :tournament => @tournament)
>     Registration.new(:user => @user1, :tournament =>
> @tournament).should_not_be_valid
>   end
> end
>
> WDYT?

That makes a lot of sense.  I'm finding that, as productive as I am
with Rails, RSpec is helping me find a lot of deficiencies in the
Rails philosophy, but far more importantly in my own thinking.

I've learned that I have to pay a LOT more attention than I have been
when coding...and I consider myself a mindful programmer to begin
with.  I know it sounds obvious that you have to pay attention, but it
can be tough in practice.

Anyway I'm loving RSpec, if only because I'm asking more design
questions than I was before.  Whenever a spec is awkward to write or
implement, it's time to investigate further.

Thanks a lot for the input.

Pat

From pergesu at gmail.com  Thu Jan  4 23:47:30 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Thu, 4 Jan 2007 21:47:30 -0700
Subject: [rspec-users] Another "how do I spec this?"
In-Reply-To: <57c63afe0701010616x15c2e61rca9a2459d0ddd548@mail.gmail.com>
References: <810a540e0701010105sb5610e2t1c3c7d94da67fe54@mail.gmail.com>
	<57c63afe0701010616x15c2e61rca9a2459d0ddd548@mail.gmail.com>
Message-ID: <810a540e0701042047w1fea7004u339877997f6abe99@mail.gmail.com>

On 1/1/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> > I wanted to add a convenience method on my User class to see if he was
> > already signed up for a tournament.  Here's my spec
> >
> > context "A User signed up for one tournament" do
> >   setup do
> >     @t1 = Tournament.new
> >     @t1.save false
> >     @t2 = Tournament.new
> >     @t2.save false
> >     @user = User.new
> >     @user.save false
> >     @user.registrations << Registration.new(:tournament => @t1)
> >   end
> >
> >   specify "should be signed up for that tournament" do
> >     @user.should_be_signed_up_for @t1
> >   end
> >
> >   specify "should not be signed up for another tournament" do
> >     @user.should_not_be_signed_up_for @t2
> >   end
> > end
> >
> > class User < ActiveRecord::Base
> >   has_many :registrations
> >
> >   def signed_up_for?(tournament)
> >     !registrations.find_by_tournament_id(tournament).nil?
> >   end
> > end
> >
> > Are there any potential improvements, or is that the best way to spec it?
>
> Couple of thoughts. Creating new unvalidated models seems to be
> something useful, so...
>
> module ModelSpecUtils
>   def unvalidated_model(klass)
>     model = klass.new
>     model.save(false)
>     model
>   end
> end
>
> Also, this line:
>
> @user.registrations << Registration.new(:tournament => @t1)
>
> is breaking encapsulation a bit. Given that and the utility module
> above, I might end up w/ something like this:
>
> context "A User signed up for one tournament" do
>   include ModelSpecUtils
>
>   setup do
>     @tournament1 = unvalidated_model(Tournament)
>     @tournament2 = unvalidated_model(Tournament)
>     @user = unvalidated_model(User)
>     @user.register_for(@tournament1)
>   end
>
>   specify "should be signed up for that tournament" do
>     @user.should_be_signed_up_for @tournament1
>   end
>
>   specify "should not be signed up for another tournament" do
>     @user.should_not_be_signed_up_for @tournament2
>   end
> end
>
> David

My original spec is bad, because it doesn't show how the Registration
model is being used.  Right now in my controller I'm just doing
@registration = Registration.new :tournament => @tournament, :user =>
current_user

and then saving it.  It's probably not good then that I'm doing
user.registrations<< in the spec.  I know that breaks encapsulation,
but I'm not doing it in my app.  Either way though the specs should
match the use.

This brings up a more interesting question (to me at least.  It could
be pointless to some).  I've got User and Tournament models, and have
the Registration relationship.  What's the best way to create the
Registration?

Creating User#register_for has a couple upsides
1. Clients only need to know about two classes, user and tournament.
2. It reads really well.  this_user.register_for(some_tourney) is very clear
a. It's "opinionated."  Having an explicit method at least promotes a
single creation mechanism (it gets a letter because I'm not sure if
it's that important)

#2 is the biggest benefit, in my opinion, and it should be obvious why
it's good.

#1 though is interesting to me, because while you want to couple
clients to as few classes as possible, here you lose/diminish the
value of a richer model.  When you create a relationship model like
this, presumably there are some interesting attributes other than the
simple relationship.

In my case, there's nothing else interesting (yet).  I use the join
model because it's more flexible and no more conceptually difficult
than using HABTM.  I prefer the explicit relationship.  But maybe I
need to signify that the relationship doesn't have interesting
attributes by abstracting it away behind the #register_for method.

So here's my hypothesis: when there are no interesting attributes to
the relationship, create a simple method that hides the join class.
When there are interesting attributes, create the relationship object
explicitly.

Of course you can never stick to simple rules like that, but hopefully
it's enough for you to give me another OOP lesson :)

Pat

From dchelimsky at gmail.com  Fri Jan  5 00:06:12 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Fri, 5 Jan 2007 00:06:12 -0500
Subject: [rspec-users] Another "how do I spec this?"
In-Reply-To: <810a540e0701042047w1fea7004u339877997f6abe99@mail.gmail.com>
References: <810a540e0701010105sb5610e2t1c3c7d94da67fe54@mail.gmail.com>
	<57c63afe0701010616x15c2e61rca9a2459d0ddd548@mail.gmail.com>
	<810a540e0701042047w1fea7004u339877997f6abe99@mail.gmail.com>
Message-ID: <57c63afe0701042106r18646bd4uadff72a9de583eee@mail.gmail.com>

On 1/4/07, Pat Maddox <pergesu at gmail.com> wrote:
> On 1/1/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> > > I wanted to add a convenience method on my User class to see if he was
> > > already signed up for a tournament.  Here's my spec
> > >
> > > context "A User signed up for one tournament" do
> > >   setup do
> > >     @t1 = Tournament.new
> > >     @t1.save false
> > >     @t2 = Tournament.new
> > >     @t2.save false
> > >     @user = User.new
> > >     @user.save false
> > >     @user.registrations << Registration.new(:tournament => @t1)
> > >   end
> > >
> > >   specify "should be signed up for that tournament" do
> > >     @user.should_be_signed_up_for @t1
> > >   end
> > >
> > >   specify "should not be signed up for another tournament" do
> > >     @user.should_not_be_signed_up_for @t2
> > >   end
> > > end
> > >
> > > class User < ActiveRecord::Base
> > >   has_many :registrations
> > >
> > >   def signed_up_for?(tournament)
> > >     !registrations.find_by_tournament_id(tournament).nil?
> > >   end
> > > end
> > >
> > > Are there any potential improvements, or is that the best way to spec it?
> >
> > Couple of thoughts. Creating new unvalidated models seems to be
> > something useful, so...
> >
> > module ModelSpecUtils
> >   def unvalidated_model(klass)
> >     model = klass.new
> >     model.save(false)
> >     model
> >   end
> > end
> >
> > Also, this line:
> >
> > @user.registrations << Registration.new(:tournament => @t1)
> >
> > is breaking encapsulation a bit. Given that and the utility module
> > above, I might end up w/ something like this:
> >
> > context "A User signed up for one tournament" do
> >   include ModelSpecUtils
> >
> >   setup do
> >     @tournament1 = unvalidated_model(Tournament)
> >     @tournament2 = unvalidated_model(Tournament)
> >     @user = unvalidated_model(User)
> >     @user.register_for(@tournament1)
> >   end
> >
> >   specify "should be signed up for that tournament" do
> >     @user.should_be_signed_up_for @tournament1
> >   end
> >
> >   specify "should not be signed up for another tournament" do
> >     @user.should_not_be_signed_up_for @tournament2
> >   end
> > end
> >
> > David
>
> My original spec is bad, because it doesn't show how the Registration
> model is being used.  Right now in my controller I'm just doing
> @registration = Registration.new :tournament => @tournament, :user =>
> current_user
>
> and then saving it.  It's probably not good then that I'm doing
> user.registrations<< in the spec.  I know that breaks encapsulation,
> but I'm not doing it in my app.  Either way though the specs should
> match the use.
>
> This brings up a more interesting question (to me at least.  It could
> be pointless to some).  I've got User and Tournament models, and have
> the Registration relationship.  What's the best way to create the
> Registration?
>
> Creating User#register_for has a couple upsides
> 1. Clients only need to know about two classes, user and tournament.
> 2. It reads really well.  this_user.register_for(some_tourney) is very clear
> a. It's "opinionated."  Having an explicit method at least promotes a
> single creation mechanism (it gets a letter because I'm not sure if
> it's that important)
>
> #2 is the biggest benefit, in my opinion, and it should be obvious why
> it's good.
>
> #1 though is interesting to me, because while you want to couple
> clients to as few classes as possible, here you lose/diminish the
> value of a richer model.  When you create a relationship model like
> this, presumably there are some interesting attributes other than the
> simple relationship.
>
> In my case, there's nothing else interesting (yet).  I use the join
> model because it's more flexible and no more conceptually difficult
> than using HABTM.  I prefer the explicit relationship.  But maybe I
> need to signify that the relationship doesn't have interesting
> attributes by abstracting it away behind the #register_for method.
>
> So here's my hypothesis: when there are no interesting attributes to
> the relationship, create a simple method that hides the join class.
> When there are interesting attributes, create the relationship object
> explicitly.

That probably makes sense a lot of the time, though I can imagine a
case where there are interesting attributes in the join but it *still*
makes things most clear to do it through the user. Here's one:

@guest = Guest.new(:smoker => false)
@hotel = Hotel.new
@hotel.should_receive(:reserve_room).with(:start_date => Time.new,
:nights => 3, :smoking_preference => "non-smoking')
@guest.reserve_room_at(@hotel)

In this case, the guest is responsible to let the hotel know that
she's interested in a non-smoking room rather than the controller
being responsible for asking the guest her preference and passing that
on to the hotel.

>
> Of course you can never stick to simple rules like that, but hopefully
> it's enough for you to give me another OOP lesson :)

You're too kind.

>
> Pat

David

> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Fri Jan  5 00:10:02 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Fri, 5 Jan 2007 00:10:02 -0500
Subject: [rspec-users] Validations based on associations
In-Reply-To: <810a540e0701042022r4cc726e7p842e595e76c934a@mail.gmail.com>
References: <810a540e0701010047o49fa5f04ucac203b68aecb08e@mail.gmail.com>
	<57c63afe0701010607m3bcab306k761a8a3156b1e0ec@mail.gmail.com>
	<810a540e0701042022r4cc726e7p842e595e76c934a@mail.gmail.com>
Message-ID: <57c63afe0701042110l5e4ffd88o80730cb0dbb7b483@mail.gmail.com>

On 1/4/07, Pat Maddox <pergesu at gmail.com> wrote:
> On 1/1/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 1/1/07, Pat Maddox <pergesu at gmail.com> wrote:
> > > My model is very simple, it's mostly just a join table that represents
> > > which tournaments a user has registered for.
> > >
> > > class Registration < ActiveRecord::Base
> > >   belongs_to :user
> > >   belongs_to :tournament
> > >
> > >   validates_presence_of :user_id, :tournament_id
> > >   validates_uniqueness_of :user_id, :scope => :tournament_id
> > > end
> > >
> > > the validates_uniqueness checks to make sure there's no record that
> > > has the same user_id and tournament_id.
> > >
> > > Is there a relatively easy way to specify this behavior?  The only way
> > > that I know is to do something like
> > >
> > > context "A Registration with a User and Tournament" do
> > >   setup do
> > >     @user = User.new
> > >     @user.save false
> > >     @tournament = Tournament.new
> > >     @tournament.save false
> > >     @reg = Registration.new(:user => @user, :tournament => @tournament)
> > >   end
> > >
> > >   specify "should be valid" do
> > >     @reg.should_be_valid
> > >   end
> > >
> > >   specify "should not be valid if a registration with the same details
> > > already exists" do
> > >     @reg.save
> > >     Registration.new(:user => @user, :tournament =>
> > > @tournament).should_not_be_valid
> > >   end
> > > end
> >
> > Geez, ActiveRecord models are SUCH a violation of SRP! I realize that
> > this is the rails way, and that there are great productivity benefits
> > we glean from AR, but having a model that validates both at both the
> > instance level (validates_presence_of) and the class level
> > (validates_uniqueness_of) is a pain in the ass. It's confusing because
> > both sorts of validations look the same, but they are really
> > fundamentally different.
> >
> > Ranting aside, there are a couple of different ways you can view this.
> > To be the most clear, I'd probably spec the class level validations
> > separately from instance level validations. The uniqueness spec might
> > look like this:
> >
> > context "The Registration model class" do
> >   setup do
> >     @user1 = User.new
> >     @user2 = User.new
> >     @tournament = Tournament.new
> >
> >     [@user1, @user2, @tournament].each {|model| model.save(false)}
> >   end
> >
> >   specify "should permit registrations with different users" do
> >     Registration.create(:user => @user1, :tournament => @tournament)
> >     Registration.new(:user => @user2, :tournament =>
> > @tournament).should_be_valid
> >   end
> >
> >   specify "should deny multiple registrations with the same user" do
> >     Registration.create(:user => @user1, :tournament => @tournament)
> >     Registration.new(:user => @user1, :tournament =>
> > @tournament).should_not_be_valid
> >   end
> > end
> >
> > WDYT?
>
> That makes a lot of sense.  I'm finding that, as productive as I am
> with Rails, RSpec is helping me find a lot of deficiencies in the
> Rails philosophy, but far more importantly in my own thinking.
>
> I've learned that I have to pay a LOT more attention than I have been
> when coding...and I consider myself a mindful programmer to begin
> with.  I know it sounds obvious that you have to pay attention, but it
> can be tough in practice.

It's REALLY tough in practice. But that's what separates the geniuses
from the rest of us. In programming and in any other craft. I was just
discussing this with a jazz singer in my family - how there are some
people who just live in the zone. The rest of us can only aspire to
that by continuing to practice and pay attention when we do.
Hopefully, eventually, you don't have to pay attention and good stuff
just happens as a matter of course.

> Anyway I'm loving RSpec, if only because I'm asking more design
> questions than I was before.  Whenever a spec is awkward to write or
> implement, it's time to investigate further.

Can I quote you on that?

>
> Thanks a lot for the input.
>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From listaccount at e-tobi.net  Fri Jan  5 04:25:29 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Fri, 05 Jan 2007 10:25:29 +0100
Subject: [rspec-users] Common setup code and naming specifications
In-Reply-To: <57c63afe0701041905sd54b02ew48492c89b3205d11@mail.gmail.com>
References: <459D87C8.1010300@e-tobi.net>	<57c63afe0701041512t6740ff1ax898dffd75753d5f@mail.gmail.com>	<459D98AA.3050003@e-tobi.net>
	<57c63afe0701041905sd54b02ew48492c89b3205d11@mail.gmail.com>
Message-ID: <459E1989.5030306@e-tobi.net>

David Chelimsky wrote:
> Ah, the DRY battle cry.
>   

Yeeehaaa! :)

> There is a difference between duplication of code and duplicated calls
> to a no-arg method. While typing the call to set_logged_in may require
> a few extra strokes,

You're right. At least in this particular case there would be no gain in
having something like a logged_in_context-"base class". It wouldn't save
very much keystrokes and it doesn't communicate better than putting
set_logged_in in the setup. If there's common setup code, it should be
refactored into a helper module and included in the context.

> are inherently procedural. The risk of duplication in code is that
> when you have to change it in one place you'll miss the other place.
> That risk doesn't really apply to this situation.
>   

The only thing I'm constantly missing is putting the logged in code in
the setup, when creating a new context. But a subclassing approach
probably wouldn't make this any better. Maybe I just shouldn't code at
one o'clock in the night :-)

>> Excellent point. There's no simple answer for this. Your best bet is
>> to keep throwing examples at us and we can discuss them more
>> explicitly.
>>     

>From another point of view, one might also see it this way:

Not putting implementation details into the specification title gives
you more freedom to implement it along the way. In this sense I would
describe the anatomy of a specification as:

specify "<title>" do
    <body>
        <usage>
        <verification>
    </body>
end


The <title> should be a hint on what functionality / behaviour is
specified. <body> describes the details and typically consists of a
<usage> part and a <verification> part. The customer does only look at
the title. The developer, who e.g. implements the view, looks mainly at
the <usage> part. This can also look like:

setup do
    <usage>
end

specify "<title>" do
    <verification>
end


Maybe this can be seen as a pair of different languages in a
translation. <title> is a phrase in the human customer language, <body>
is the translation in the developer language. Mmmm... thats still not
the right analogy. It's more like a dictionary where <title> is a term
in one language and <usage>/<verification> is the definition in another
language.

Ok, before I get too philosophical about this, I'll rather go ahead and
write some code to see where this gets me :-)

Regarding the examples - what I'm currently working on is open source
anyway. I'll put it into a public SVN repository these days, so if
anyone is interested he can take a look at it.

bye,

Tobias


From jonathan.tron.mailings at gmail.com  Fri Jan  5 04:47:58 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Fri, 5 Jan 2007 10:47:58 +0100
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
In-Reply-To: <CAD71583-F29B-4AE1-9F36-87D489A8D270@osesm.com>
References: <CAD71583-F29B-4AE1-9F36-87D489A8D270@osesm.com>
Message-ID: <E96C0CFF-9E38-44DA-80B0-C99CC065AA6C@gmail.com>

Le 5 janv. 07 ? 03:05, Bryan Liles a ?crit :

> I seem to be missing something because I am at a loss on how to get
> rcov, RSpec, and Rails working together.  Would someone mind dropping
> some hints?

I don't know what you want to achieve, but if your goal is to  
generate some reports here's what I added to lib/tasks/ 
rspec_additions.rake :

namespace :spec do
   desc "Spec with rspec/rcov reports"
   Spec::Rake::SpecTask.new('reports') do |t|
     t.spec_files = FileList['spec/**/*.rb']
     t.spec_opts = ["--format", "html", "--diff"]
     t.out = 'doc/rspec.html'
     t.rcov = true
     t.rcov_dir = 'doc/coverage'

# Optional :
# Prevent rcov to generate report on specs, environment.rb and boot.rb
#    t.rcov_opts = ['--exclude', 'spec,config\/environment.rb,config\/ 
boot.rb']

     t.fail_on_error = false
     t.failure_message = "The specs failed. Check for the problem at  
http://projects.tron.name/flash007/rspec.html"
   end
end

Run it with rake : spec:reports
It will generate rspec.html report in doc/rspec.html and rcov report  
in doc/coverage/.

Jonathan
--
Tron Jonathan
http://jonathan.tron.name



From aslak.hellesoy at gmail.com  Fri Jan  5 06:01:32 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Fri, 5 Jan 2007 12:01:32 +0100
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
In-Reply-To: <E96C0CFF-9E38-44DA-80B0-C99CC065AA6C@gmail.com>
References: <CAD71583-F29B-4AE1-9F36-87D489A8D270@osesm.com>
	<E96C0CFF-9E38-44DA-80B0-C99CC065AA6C@gmail.com>
Message-ID: <8d961d900701050301j59b40e15pfd89748297cef5a5@mail.gmail.com>

In the next release (or rspec+spec/rails on trunk) you'll also be able
to leave the built-in spec:* tasks by editing a spec/spec.opts file.

On 1/5/07, Jonathan Tron <jonathan.tron.mailings at gmail.com> wrote:
> Le 5 janv. 07 ? 03:05, Bryan Liles a ?crit :
>
> > I seem to be missing something because I am at a loss on how to get
> > rcov, RSpec, and Rails working together.  Would someone mind dropping
> > some hints?
>
> I don't know what you want to achieve, but if your goal is to
> generate some reports here's what I added to lib/tasks/
> rspec_additions.rake :
>
> namespace :spec do
>    desc "Spec with rspec/rcov reports"
>    Spec::Rake::SpecTask.new('reports') do |t|
>      t.spec_files = FileList['spec/**/*.rb']
>      t.spec_opts = ["--format", "html", "--diff"]
>      t.out = 'doc/rspec.html'
>      t.rcov = true
>      t.rcov_dir = 'doc/coverage'
>
> # Optional :
> # Prevent rcov to generate report on specs, environment.rb and boot.rb
> #    t.rcov_opts = ['--exclude', 'spec,config\/environment.rb,config\/
> boot.rb']
>
>      t.fail_on_error = false
>      t.failure_message = "The specs failed. Check for the problem at
> http://projects.tron.name/flash007/rspec.html"
>    end
> end
>
> Run it with rake : spec:reports
> It will generate rspec.html report in doc/rspec.html and rcov report
> in doc/coverage/.
>
> Jonathan
> --
> Tron Jonathan
> http://jonathan.tron.name
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From mlangenberg at gmail.com  Fri Jan  5 09:09:08 2007
From: mlangenberg at gmail.com (Matthijs Langenberg)
Date: Fri, 5 Jan 2007 15:09:08 +0100
Subject: [rspec-users] Using RESTful routes in controller tests
Message-ID: <27c0ac6d0701050609p26603001j7127b1ffe663606e@mail.gmail.com>

Why can't I use the RESTful route helpers in my specs?
In my controller I'm doing a redirect: redirect_to list_path(:id =>
@item.list_id)
Now when I write: controller.should_redirect_to list_path(:id => 2)
I'm getting the following error:

NoMethodError in 'POST on /lists/2/items should redirect to index on
succesful POST'
You have a nil object when you didn't expect it!
The error occurred while evaluating nil.rewrite
(eval):19:in `list_path'
./spec/controllers/items_controller_spec.rb:36:

Finished in 0.114317 seconds

Writing "controller.should_redirect_to :controller => 'lists', :action
=> 'show', :id => 2" works though.

Am I forgetting something?

From aslak.hellesoy at gmail.com  Fri Jan  5 16:15:38 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Fri, 5 Jan 2007 22:15:38 +0100
Subject: [rspec-users] Using RESTful routes in controller tests
In-Reply-To: <27c0ac6d0701050609p26603001j7127b1ffe663606e@mail.gmail.com>
References: <27c0ac6d0701050609p26603001j7127b1ffe663606e@mail.gmail.com>
Message-ID: <8d961d900701051315y1d93404q69623fa4b9485d68@mail.gmail.com>

Sounds like a bug/missing feature. Please submit a feature request at
Rubyforge, with a failing spec if you can.

On 1/5/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
> Why can't I use the RESTful route helpers in my specs?
> In my controller I'm doing a redirect: redirect_to list_path(:id =>
> @item.list_id)
> Now when I write: controller.should_redirect_to list_path(:id => 2)
> I'm getting the following error:
>
> NoMethodError in 'POST on /lists/2/items should redirect to index on
> succesful POST'
> You have a nil object when you didn't expect it!
> The error occurred while evaluating nil.rewrite
> (eval):19:in `list_path'
> ./spec/controllers/items_controller_spec.rb:36:
>
> Finished in 0.114317 seconds
>
> Writing "controller.should_redirect_to :controller => 'lists', :action
> => 'show', :id => 2" works though.
>
> Am I forgetting something?
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From cwdinfo at gmail.com  Fri Jan  5 16:30:33 2007
From: cwdinfo at gmail.com (s.ross)
Date: Fri, 5 Jan 2007 13:30:33 -0800
Subject: [rspec-users] How To Spec Controllers with Finders
Message-ID: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>

Given this code (which renders rjs), I'm faced with the fixture- 
driven way of spec-ing or mocking. How the heck to you mock this so  
the code at line (2) and (4) work right? I'm still struggling with  
mocks but it seems like this can be done. Forgive the naivety of this  
question.

1.  def change_quantity
2.    @line_item = LineItem.find_by_id(params[:id])
3.    unless @line_item.nil?
4.      @line_item.update_attribute(:quantity, params[:quantity])
5.      @extension_id = "extension#{params[:id]}"
6.    end
7.  end

Thanks

From court3nay at gmail.com  Fri Jan  5 17:14:09 2007
From: court3nay at gmail.com (Courtenay)
Date: Fri, 5 Jan 2007 14:14:09 -0800
Subject: [rspec-users] How To Spec Controllers with Finders
In-Reply-To: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>
References: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>
Message-ID: <4b430c8f0701051414yd6e5a3r48551df4b818e739@mail.gmail.com>

try this:


@line_item = mock("line_item")
LineItem.should_receive(:find_by_id).with(3).and_return(@line_item)
@line_item.should_receive(:update_attribute).with(:quantity, 1)

get :change_quantity, :id => 3, :quantity => 1


On 1/5/07, s.ross <cwdinfo at gmail.com> wrote:
> Given this code (which renders rjs), I'm faced with the fixture-
> driven way of spec-ing or mocking. How the heck to you mock this so
> the code at line (2) and (4) work right? I'm still struggling with
> mocks but it seems like this can be done. Forgive the naivety of this
> question.
>
> 1.  def change_quantity
> 2.    @line_item = LineItem.find_by_id(params[:id])
> 3.    unless @line_item.nil?
> 4.      @line_item.update_attribute(:quantity, params[:quantity])
> 5.      @extension_id = "extension#{params[:id]}"
> 6.    end
> 7.  end
>
> Thanks
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From brian.takita at gmail.com  Fri Jan  5 20:39:26 2007
From: brian.takita at gmail.com (Brian Takita)
Date: Fri, 5 Jan 2007 17:39:26 -0800
Subject: [rspec-users] spec command and windows
Message-ID: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>

Hello,

Today I ran into some frustration with spec and windows.

We have a series of plugins that have either test suites and/or spec suites.
To avoid environment collisions, we have a "master" test suite that spawns
new process for all of the test suites and spec suites.
Every works great on posix. Unfortunately on windows, spec was not found.

The only way I figured to ensure we get to spec in a platform independent
way is to call:

success = true
dir = File.dirname(__FILE__)
spec_cmd = (RUBY_PLATFORM =~ /[^r]win/) ? 'C:\\ruby\\bin\\spec.cmd' : 'spec'
success &&= system("#{spec_cmd} --format specdoc --diff unified
#{dir}/project_dir/spec_suite.rb")
# ...
exit success

Lame!
Does somebody out there have a better solution? Could rspec be run from ruby
instead?

Thank you,
Brian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070105/6f6a5e28/attachment.html 

From waynev at gmail.com  Fri Jan  5 22:55:06 2007
From: waynev at gmail.com (Wayne Vucenic)
Date: Fri, 5 Jan 2007 19:55:06 -0800
Subject: [rspec-users] spec command and windows
In-Reply-To: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>
References: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>
Message-ID: <88c9ce410701051955y6b0079bav9c7b7537a00e7504@mail.gmail.com>

Hi Brian,

It looks like c:\ruby\bin isn't in your PATH environment variable on
Windows.  (I've had problems with the OneClickInstaller not adding
this directory to the PATH).  If this is true, adding it to PATH
should solve your problem.

> Could rspec be run from ruby instead?

This is similar to a question I was going to ask on this list.  I'm
trying to figure out how to run a xyz_spec.rb file from within my IDE,
rather than having to use spec on the command line.   Can anyone tell
me how to do this?

Thanks,

Wayne

---

Wayne Vucenic
No Bugs Software
Ruby, C#, and Erlang Agile Contract Programming in Silicon Valley

From dchelimsky at gmail.com  Fri Jan  5 23:04:21 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Fri, 5 Jan 2007 23:04:21 -0500
Subject: [rspec-users] spec command and windows
In-Reply-To: <88c9ce410701051955y6b0079bav9c7b7537a00e7504@mail.gmail.com>
References: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>
	<88c9ce410701051955y6b0079bav9c7b7537a00e7504@mail.gmail.com>
Message-ID: <57c63afe0701052004x39000224m9562f766069a9ddc@mail.gmail.com>

On 1/5/07, Wayne Vucenic <waynev at gmail.com> wrote:
> Hi Brian,
>
> It looks like c:\ruby\bin isn't in your PATH environment variable on
> Windows.  (I've had problems with the OneClickInstaller not adding
> this directory to the PATH).  If this is true, adding it to PATH
> should solve your problem.
>
> > Could rspec be run from ruby instead?
>
> This is similar to a question I was going to ask on this list.  I'm
> trying to figure out how to run a xyz_spec.rb file from within my IDE,
> rather than having to use spec on the command line.   Can anyone tell
> me how to do this?

Just require 'spec' in the file or a file it includes and go:

ruby path/to/the/file.rb

OR (if you MUST)

ruby path\to\the\file.rb

Cheers,
David

>
> Thanks,
>
> Wayne
>
> ---
>
> Wayne Vucenic
> No Bugs Software
> Ruby, C#, and Erlang Agile Contract Programming in Silicon Valley
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From waynev at gmail.com  Sat Jan  6 00:30:39 2007
From: waynev at gmail.com (Wayne Vucenic)
Date: Fri, 5 Jan 2007 21:30:39 -0800
Subject: [rspec-users] spec command and windows
In-Reply-To: <57c63afe0701052004x39000224m9562f766069a9ddc@mail.gmail.com>
References: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>
	<88c9ce410701051955y6b0079bav9c7b7537a00e7504@mail.gmail.com>
	<57c63afe0701052004x39000224m9562f766069a9ddc@mail.gmail.com>
Message-ID: <88c9ce410701052130u63cf2557oe4afa7d762fd9d07@mail.gmail.com>

Hi David,

Thanks for your amazingly prompt reply!

On 1/5/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> Just require 'spec' in the file or a file it includes and go:
>
> ruby path/to/the/file.rb
>
> OR (if you MUST)
>
> ruby path\to\the\file.rb

Works like a charm!  Now is there any way to get the same output as
with "-f s" ?

And I completely agree with you on / rather than \ in file paths.  The
use of \ was one of the most annoying things MS-DOS and then Windows
ever did.  When I switched from programming on the Mac to Windows NT
3.51 I was glad that all the Win32 APIs accepted either \ or / in
paths,  so I always preferred /.

Thanks,

Wayne

From cwdinfo at gmail.com  Sat Jan  6 01:35:22 2007
From: cwdinfo at gmail.com (s.ross)
Date: Fri, 5 Jan 2007 22:35:22 -0800
Subject: [rspec-users] How To Spec Controllers with Finders
In-Reply-To: <4b430c8f0701051414yd6e5a3r48551df4b818e739@mail.gmail.com>
References: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>
	<4b430c8f0701051414yd6e5a3r48551df4b818e739@mail.gmail.com>
Message-ID: <CD863B29-C7D1-4B35-8880-E171E327F456@gmail.com>

What I came up with was:

context "The cart controller" do
     controller_name "carts"
     integrate_views

     setup do
         @cart = mock("cart")
         @cart.stub!(:new_record).and_return(false)
         @cart.stub!(:new).and_return(@cart)
         @cart.stub!(:total_price).and_return(35.00)
         @cart.stub!(:total_quantity).and_return(1)
         @cart.stub!(:empty?).and_return(false)

         @line_item = mock("line_item")
         LineItem.stub!(:find_by_id).and_return(@line_item)
         @line_item.should_receive(:update_attribute).with(:quantity,  
"1").and_return(true)
         @line_item.stub!(:extension).and_return(25.00)
     end

     specify "should allow change of quantity" do
         post :change_quantity, {:id => 1, :quantity => 1}
         response.should_be_success
         assigns[:extension_id].should == "extension1"
         assigns[:line_item].should_be(@line_item)
         controller.should_render_rjs :replace_html, 'cartstatus',  
"(1 item, $35.00)"
     end
end

but somehow it seems a bit brittle. This is really what should happen  
if a user changes quantity, but the mocks seem like they won't flex  
or bend easily for further reuse. Any thoughts on this?

On Jan 5, 2007, at 2:14 PM, Courtenay wrote:

> try this:
>
>
> @line_item = mock("line_item")
> LineItem.should_receive(:find_by_id).with(3).and_return(@line_item)
> @line_item.should_receive(:update_attribute).with(:quantity, 1)
>
> get :change_quantity, :id => 3, :quantity => 1
>
>
> On 1/5/07, s.ross <cwdinfo at gmail.com> wrote:
>> Given this code (which renders rjs), I'm faced with the fixture-
>> driven way of spec-ing or mocking. How the heck to you mock this so
>> the code at line (2) and (4) work right? I'm still struggling with
>> mocks but it seems like this can be done. Forgive the naivety of this
>> question.
>>
>> 1.  def change_quantity
>> 2.    @line_item = LineItem.find_by_id(params[:id])
>> 3.    unless @line_item.nil?
>> 4.      @line_item.update_attribute(:quantity, params[:quantity])
>> 5.      @extension_id = "extension#{params[:id]}"
>> 6.    end
>> 7.  end
>>
>> Thanks
>> _______________________________________________
>> 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


From dchelimsky at gmail.com  Sat Jan  6 08:20:51 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sat, 6 Jan 2007 08:20:51 -0500
Subject: [rspec-users] How To Spec Controllers with Finders
In-Reply-To: <CD863B29-C7D1-4B35-8880-E171E327F456@gmail.com>
References: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>
	<4b430c8f0701051414yd6e5a3r48551df4b818e739@mail.gmail.com>
	<CD863B29-C7D1-4B35-8880-E171E327F456@gmail.com>
Message-ID: <57c63afe0701060520h50c2a556jbfc8f1856cc78cb4@mail.gmail.com>

On 1/6/07, s.ross <cwdinfo at gmail.com> wrote:
> What I came up with was:
>
> context "The cart controller" do
>      controller_name "carts"
>      integrate_views
>
>      setup do
>          @cart = mock("cart")
>          @cart.stub!(:new_record).and_return(false)
>          @cart.stub!(:new).and_return(@cart)
>          @cart.stub!(:total_price).and_return(35.00)
>          @cart.stub!(:total_quantity).and_return(1)
>          @cart.stub!(:empty?).and_return(false)
>
>          @line_item = mock("line_item")
>          LineItem.stub!(:find_by_id).and_return(@line_item)
>          @line_item.should_receive(:update_attribute).with(:quantity,
> "1").and_return(true)

I try to avoid using should_receive in setup, so I'd do this here:

@line_item.stub!(:update_attribute).and_return(true)

and then override the method stub w/ a mock expectation in the specify
block. The idea is that you put all of the stuff in setup that must be
there for the code to execute, and put the details in the specs that
are interesting for that spec.

>          @line_item.stub!(:extension).and_return(25.00)
>      end
>
>      specify "should allow change of quantity" do
>          post :change_quantity, {:id => 1, :quantity => 1}
>          response.should_be_success
>          assigns[:extension_id].should == "extension1"
>          assigns[:line_item].should_be(@line_item)
>          controller.should_render_rjs :replace_html, 'cartstatus',
> "(1 item, $35.00)"
>      end
> end
>
> but somehow it seems a bit brittle. This is really what should happen
> if a user changes quantity, but the mocks seem like they won't flex
> or bend easily for further reuse. Any thoughts on this?

Within this context they absolutely will flex or bend by overriding
the method stubs (functioning as defaults) w/ mock expectations.

Also, I might approach this differently. The behaviour here that you
want to spec is that when changing quantity some things should happen.
So....

context "Given a request to change quantity, the CartController should" do
  controller_name "carts"
  integrate_views

  setup do
    @cart = mock("cart")
    @cart.stub!(:new_record).and_return(false)
    @cart.stub!(:total_price).and_return(35.00)
    @cart.stub!(:total_quantity).and_return(1)
    @cart.stub!(:empty?).and_return(false)
    Cart.stub!(:new).and_return(@cart)

    @line_item = mock("line_item")
    @line_item.stub!(:update_attribute).and_return(true)
    @line_item.stub!(:extension).and_return(25.00)
    LineItem.stub!(:find_by_id).and_return(@line_item)

    post :change_quantity, {:id => 1, :quantity => 1}
  end

  specify "respond w/ success" do
    response.should_be_success
  end

  specify "assign exension_id" do
    assigns[:extension_id].should == "extension1"
  end

  specify "assign line_item" do
    assigns[:line_item].should_be(@line_item)
  end

  specify "update the html (using rjs)" do
    controller.should_render_rjs :replace_html,
      'cartstatus',
      "(1 item, $35.00)"
  end
end

This would then read:

Given a request to change quantity, the CartController should
- respond w/ success
- assign exension_id
- assign line_item
- update the html (using rjs)

If you're concerned about reuse of the stubs, you can do this:

module Stubs
  def new_stub_cart
    cart = mock("cart")
    cart.stub!(:new_record).and_return(false)
    cart.stub!(:total_price).and_return(0.0)
    cart.stub!(:total_quantity).and_return(0)
    cart.stub!(:empty?).and_return(false)
    cart
  end

  def new_stub_line_item
    line_item = mock("line_item")
    line_item.stub!(:update_attribute).and_return(true)
    line_item.stub!(:extension).and_return(0)
    line_item
  end

  def stub_cart
    @stub_cart ||= new_stub_cart
  end

  def stub_line_item
    @stub_line_item ||= new_stub_line_item
  end
end

context "Given a request to change quantity, the CartController should" do
  include Stubs
  controller_name "carts"
  integrate_views

  setup do
    Cart.stub!(:new).and_return(stub_cart)
    LineItem.stub!(:find_by_id).and_return(stub_line_item)
  end
  ...
end

Now any new spec that requires specific values can use
stub_cart.should_receive(:total_price).and_return(25.00), for example.


>
> On Jan 5, 2007, at 2:14 PM, Courtenay wrote:
>
> > try this:
> >
> >
> > @line_item = mock("line_item")
> > LineItem.should_receive(:find_by_id).with(3).and_return(@line_item)
> > @line_item.should_receive(:update_attribute).with(:quantity, 1)
> >
> > get :change_quantity, :id => 3, :quantity => 1
> >
> >
> > On 1/5/07, s.ross <cwdinfo at gmail.com> wrote:
> >> Given this code (which renders rjs), I'm faced with the fixture-
> >> driven way of spec-ing or mocking. How the heck to you mock this so
> >> the code at line (2) and (4) work right? I'm still struggling with
> >> mocks but it seems like this can be done. Forgive the naivety of this
> >> question.
> >>
> >> 1.  def change_quantity
> >> 2.    @line_item = LineItem.find_by_id(params[:id])
> >> 3.    unless @line_item.nil?
> >> 4.      @line_item.update_attribute(:quantity, params[:quantity])
> >> 5.      @extension_id = "extension#{params[:id]}"
> >> 6.    end
> >> 7.  end
> >>
> >> Thanks
> >> _______________________________________________
> >> 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
>

From bkeepers at gmail.com  Sat Jan  6 10:28:24 2007
From: bkeepers at gmail.com (Brandon Keepers)
Date: Sat, 6 Jan 2007 10:28:24 -0500
Subject: [rspec-users] Using RESTful routes in controller tests
In-Reply-To: <8d961d900701051315y1d93404q69623fa4b9485d68@mail.gmail.com>
References: <27c0ac6d0701050609p26603001j7127b1ffe663606e@mail.gmail.com>
	<8d961d900701051315y1d93404q69623fa4b9485d68@mail.gmail.com>
Message-ID: <6D1F570C-3C74-44C2-9660-818A4E25AB27@gmail.com>

There already is one:

https://rubyforge.org/tracker/index.php? 
func=detail&aid=6541&group_id=797&atid=3149

Brandon

On Jan 5, 2007, at 4:15 PM, aslak hellesoy wrote:

> Sounds like a bug/missing feature. Please submit a feature request at
> Rubyforge, with a failing spec if you can.
>
> On 1/5/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
>> Why can't I use the RESTful route helpers in my specs?
>> In my controller I'm doing a redirect: redirect_to list_path(:id =>
>> @item.list_id)
>> Now when I write: controller.should_redirect_to list_path(:id => 2)
>> I'm getting the following error:
>>
>> NoMethodError in 'POST on /lists/2/items should redirect to index on
>> succesful POST'
>> You have a nil object when you didn't expect it!
>> The error occurred while evaluating nil.rewrite
>> (eval):19:in `list_path'
>> ./spec/controllers/items_controller_spec.rb:36:
>>
>> Finished in 0.114317 seconds
>>
>> Writing "controller.should_redirect_to :controller =>  
>> 'lists', :action
>> => 'show', :id => 2" works though.
>>
>> Am I forgetting something?
>> _______________________________________________
>> 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


From cwdinfo at gmail.com  Sat Jan  6 12:09:10 2007
From: cwdinfo at gmail.com (s.ross)
Date: Sat, 6 Jan 2007 09:09:10 -0800
Subject: [rspec-users] How To Spec Controllers with Finders
In-Reply-To: <57c63afe0701060520h50c2a556jbfc8f1856cc78cb4@mail.gmail.com>
References: <A2A16278-54DF-43A5-A587-90F0230C02EB@gmail.com>
	<4b430c8f0701051414yd6e5a3r48551df4b818e739@mail.gmail.com>
	<CD863B29-C7D1-4B35-8880-E171E327F456@gmail.com>
	<57c63afe0701060520h50c2a556jbfc8f1856cc78cb4@mail.gmail.com>
Message-ID: <AB5C8957-FE12-4A3B-890B-F78A38F5B303@gmail.com>

Wow! Thanks. I'll look at this to see how it works for me.

Steve

On Jan 6, 2007, at 5:20 AM, David Chelimsky wrote:

> On 1/6/07, s.ross <cwdinfo at gmail.com> wrote:
>> What I came up with was:
>>
>> context "The cart controller" do
>>      controller_name "carts"
>>      integrate_views
>>
>>      setup do
>>          @cart = mock("cart")
>>          @cart.stub!(:new_record).and_return(false)
>>          @cart.stub!(:new).and_return(@cart)
>>          @cart.stub!(:total_price).and_return(35.00)
>>          @cart.stub!(:total_quantity).and_return(1)
>>          @cart.stub!(:empty?).and_return(false)
>>
>>          @line_item = mock("line_item")
>>          LineItem.stub!(:find_by_id).and_return(@line_item)
>>          @line_item.should_receive(:update_attribute).with(:quantity,
>> "1").and_return(true)
>
> I try to avoid using should_receive in setup, so I'd do this here:
>
> @line_item.stub!(:update_attribute).and_return(true)
>
> and then override the method stub w/ a mock expectation in the specify
> block. The idea is that you put all of the stuff in setup that must be
> there for the code to execute, and put the details in the specs that
> are interesting for that spec.
>
>>          @line_item.stub!(:extension).and_return(25.00)
>>      end
>>
>>      specify "should allow change of quantity" do
>>          post :change_quantity, {:id => 1, :quantity => 1}
>>          response.should_be_success
>>          assigns[:extension_id].should == "extension1"
>>          assigns[:line_item].should_be(@line_item)
>>          controller.should_render_rjs :replace_html, 'cartstatus',
>> "(1 item, $35.00)"
>>      end
>> end
>>
>> but somehow it seems a bit brittle. This is really what should happen
>> if a user changes quantity, but the mocks seem like they won't flex
>> or bend easily for further reuse. Any thoughts on this?
>
> Within this context they absolutely will flex or bend by overriding
> the method stubs (functioning as defaults) w/ mock expectations.
>
> Also, I might approach this differently. The behaviour here that you
> want to spec is that when changing quantity some things should happen.
> So....
>
> context "Given a request to change quantity, the CartController  
> should" do
>   controller_name "carts"
>   integrate_views
>
>   setup do
>     @cart = mock("cart")
>     @cart.stub!(:new_record).and_return(false)
>     @cart.stub!(:total_price).and_return(35.00)
>     @cart.stub!(:total_quantity).and_return(1)
>     @cart.stub!(:empty?).and_return(false)
>     Cart.stub!(:new).and_return(@cart)
>
>     @line_item = mock("line_item")
>     @line_item.stub!(:update_attribute).and_return(true)
>     @line_item.stub!(:extension).and_return(25.00)
>     LineItem.stub!(:find_by_id).and_return(@line_item)
>
>     post :change_quantity, {:id => 1, :quantity => 1}
>   end
>
>   specify "respond w/ success" do
>     response.should_be_success
>   end
>
>   specify "assign exension_id" do
>     assigns[:extension_id].should == "extension1"
>   end
>
>   specify "assign line_item" do
>     assigns[:line_item].should_be(@line_item)
>   end
>
>   specify "update the html (using rjs)" do
>     controller.should_render_rjs :replace_html,
>       'cartstatus',
>       "(1 item, $35.00)"
>   end
> end
>
> This would then read:
>
> Given a request to change quantity, the CartController should
> - respond w/ success
> - assign exension_id
> - assign line_item
> - update the html (using rjs)
>
> If you're concerned about reuse of the stubs, you can do this:
>
> module Stubs
>   def new_stub_cart
>     cart = mock("cart")
>     cart.stub!(:new_record).and_return(false)
>     cart.stub!(:total_price).and_return(0.0)
>     cart.stub!(:total_quantity).and_return(0)
>     cart.stub!(:empty?).and_return(false)
>     cart
>   end
>
>   def new_stub_line_item
>     line_item = mock("line_item")
>     line_item.stub!(:update_attribute).and_return(true)
>     line_item.stub!(:extension).and_return(0)
>     line_item
>   end
>
>   def stub_cart
>     @stub_cart ||= new_stub_cart
>   end
>
>   def stub_line_item
>     @stub_line_item ||= new_stub_line_item
>   end
> end
>
> context "Given a request to change quantity, the CartController  
> should" do
>   include Stubs
>   controller_name "carts"
>   integrate_views
>
>   setup do
>     Cart.stub!(:new).and_return(stub_cart)
>     LineItem.stub!(:find_by_id).and_return(stub_line_item)
>   end
>   ...
> end
>
> Now any new spec that requires specific values can use
> stub_cart.should_receive(:total_price).and_return(25.00), for example.
>
>
>>
>> On Jan 5, 2007, at 2:14 PM, Courtenay wrote:
>>
>>> try this:
>>>
>>>
>>> @line_item = mock("line_item")
>>> LineItem.should_receive(:find_by_id).with(3).and_return(@line_item)
>>> @line_item.should_receive(:update_attribute).with(:quantity, 1)
>>>
>>> get :change_quantity, :id => 3, :quantity => 1
>>>
>>>
>>> On 1/5/07, s.ross <cwdinfo at gmail.com> wrote:
>>>> Given this code (which renders rjs), I'm faced with the fixture-
>>>> driven way of spec-ing or mocking. How the heck to you mock this so
>>>> the code at line (2) and (4) work right? I'm still struggling with
>>>> mocks but it seems like this can be done. Forgive the naivety of  
>>>> this
>>>> question.
>>>>
>>>> 1.  def change_quantity
>>>> 2.    @line_item = LineItem.find_by_id(params[:id])
>>>> 3.    unless @line_item.nil?
>>>> 4.      @line_item.update_attribute(:quantity, params[:quantity])
>>>> 5.      @extension_id = "extension#{params[:id]}"
>>>> 6.    end
>>>> 7.  end
>>>>
>>>> Thanks
>>>> _______________________________________________
>>>> 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


From waynev at gmail.com  Sat Jan  6 15:19:49 2007
From: waynev at gmail.com (Wayne Vucenic)
Date: Sat, 6 Jan 2007 12:19:49 -0800
Subject: [rspec-users] spec command and windows
In-Reply-To: <88c9ce410701052130u63cf2557oe4afa7d762fd9d07@mail.gmail.com>
References: <1d7ddd110701051739j7285b552r7a785afca6b0bb0@mail.gmail.com>
	<88c9ce410701051955y6b0079bav9c7b7537a00e7504@mail.gmail.com>
	<57c63afe0701052004x39000224m9562f766069a9ddc@mail.gmail.com>
	<88c9ce410701052130u63cf2557oe4afa7d762fd9d07@mail.gmail.com>
Message-ID: <88c9ce410701061219q78466c0kdfc841b92dcc7080@mail.gmail.com>

> Now is there any way to get the same output as with "-f s" ?

I took a look at the source, and answered my own question:

require 'spec'
ARGV.unshift("-f", "s")

Not the most elegant approach, but it seems to work so far.

Wayne

From jchris at mfdz.com  Mon Jan  8 00:55:01 2007
From: jchris at mfdz.com (Chris Anderson)
Date: Sun, 7 Jan 2007 21:55:01 -0800
Subject: [rspec-users] thoughts on mocks and specs
Message-ID: <e282921e0701072155t468a49e2wd6ae5cdb734a8c2d@mail.gmail.com>

I spent the last couple of days getting my sea legs with Rails and
RSpec. I'd been waiting til things seemed calmer before jumping in,
and I'm overall very happy with my experience so far.

My only real annoyance so far has been forgetting to call "do_post" or
"do_create" from my specify blocks. My mocks don't get the calls they
want, and it usually isn't until I've mucked around in my application
code for a bit that I realize the real problem.

Wouldn't it be nice to be able to define a callback that gets run
after every specify block? Maybe it would look like this:

context "A leaner syntaxed context" do

  setup do
    @obj = mock('obj')
  end

  body do
    @obj.whatever
  end

  specify "should run the body block after every specification" do
    @obj.should_receive(:whatever)
  end

end

I know sometimes one might want to put some verification code after
the body. In that case, you could allow an explicit call to body from
within the specify block, which would preclude it from being called
again at the end.

It seems to me that this feature isn't just a fight for DRY, but also
encourages good spec design. Situations diverse enough to need
separate body code could use different contexts anyway.

If there is a call for this feature, I'd be happy to make a go at
patching it into RSpec.

Reactions?

-- 
Chris Anderson
http://jchris.mfdz.com

From pergesu at gmail.com  Mon Jan  8 01:19:50 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Sun, 7 Jan 2007 23:19:50 -0700
Subject: [rspec-users] thoughts on mocks and specs
In-Reply-To: <e282921e0701072155t468a49e2wd6ae5cdb734a8c2d@mail.gmail.com>
References: <e282921e0701072155t468a49e2wd6ae5cdb734a8c2d@mail.gmail.com>
Message-ID: <810a540e0701072219t2d4c6b65n94d18741f20d99ae@mail.gmail.com>

On 1/7/07, Chris Anderson <jchris at mfdz.com> wrote:
> I spent the last couple of days getting my sea legs with Rails and
> RSpec. I'd been waiting til things seemed calmer before jumping in,
> and I'm overall very happy with my experience so far.
>
> My only real annoyance so far has been forgetting to call "do_post" or
> "do_create" from my specify blocks. My mocks don't get the calls they
> want, and it usually isn't until I've mucked around in my application
> code for a bit that I realize the real problem.
>
> Wouldn't it be nice to be able to define a callback that gets run
> after every specify block? Maybe it would look like this:
>
> context "A leaner syntaxed context" do
>
>   setup do
>     @obj = mock('obj')
>   end
>
>   body do
>     @obj.whatever
>   end
>
>   specify "should run the body block after every specification" do
>     @obj.should_receive(:whatever)
>   end
>
> end
>
> I know sometimes one might want to put some verification code after
> the body. In that case, you could allow an explicit call to body from
> within the specify block, which would preclude it from being called
> again at the end.
>
> It seems to me that this feature isn't just a fight for DRY, but also
> encourages good spec design. Situations diverse enough to need
> separate body code could use different contexts anyway.
>
> If there is a call for this feature, I'd be happy to make a go at
> patching it into RSpec.
>
> Reactions?
>
> --
> Chris Anderson
> http://jchris.mfdz.com
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

You can do that using teardown.

Pat

From pergesu at gmail.com  Mon Jan  8 06:19:18 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Mon, 8 Jan 2007 04:19:18 -0700
Subject: [rspec-users] Controller actions do a bunch of javascript
	stuff...can I use spec that?
Message-ID: <810a540e0701080319n50286a5fu2ff78497ce7802a7@mail.gmail.com>

I've got a form that isn't doing a regular POST...what actually
happens is that when onClick gets called, a bunch of javascript stuff
happens.  It changes the form's action to be a different URL.  Then it
polls for progress on another server, and when it's 100% reverts back
to the initial URL.

I know that sounds really ugly (and I agree it is), but that's the way
it has to be at this point.  Working with a "legacy" system.  Is there
any way I can spec that behavior with rspec?  I sort of assume not,
since it's highly dependent on JS...probably need to use jsunit for
that instead.  Still I've love to hear any ideas, as well as how I
could move more of the behavior into Ruby instead of JS.

Pat

From dchelimsky at gmail.com  Mon Jan  8 08:40:23 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 8 Jan 2007 08:40:23 -0500
Subject: [rspec-users] Controller actions do a bunch of javascript
	stuff...can I use spec that?
In-Reply-To: <810a540e0701080319n50286a5fu2ff78497ce7802a7@mail.gmail.com>
References: <810a540e0701080319n50286a5fu2ff78497ce7802a7@mail.gmail.com>
Message-ID: <57c63afe0701080540p4acf2f3csff8126e646caab4c@mail.gmail.com>

On 1/8/07, Pat Maddox <pergesu at gmail.com> wrote:
> I've got a form that isn't doing a regular POST...what actually
> happens is that when onClick gets called, a bunch of javascript stuff
> happens.  It changes the form's action to be a different URL.  Then it
> polls for progress on another server, and when it's 100% reverts back
> to the initial URL.
>
> I know that sounds really ugly (and I agree it is), but that's the way
> it has to be at this point.  Working with a "legacy" system.  Is there
> any way I can spec that behavior with rspec?  I sort of assume not,
> since it's highly dependent on JS...probably need to use jsunit for
> that instead.  Still I've love to hear any ideas, as well as how I
> could move more of the behavior into Ruby instead of JS.

I would test stuff like that in the browser using selenium or watir.
There are existing hooks, like Selenium Remote Control, that allow you
to drive these browser-based testing frameworks from ruby files. We're
adding more integrated hooks into rspec for the next release as well.

As for pushing the behaviour down to ruby, I can imagine a couple of
options. One would be to have the onClick remote-link to an action
that replaces the form tag and introduces a new javascript block that
remote-links back to another action. The second action responds by
polling for progress on the new server and updating the page w/ the
same remote link (calling itself again) until the server responds
100%, at which point the second action restores the form tag.

Another approach might be to have the onClick remote-link to an action
that sets a variable in the session like
session[:server_update_complete] = false, invokes the server update,
starts polling for completion, and then sets
session[:server_update_complete] = true when it gets back 100%. The
form action would never change on the page, but the action would be
written to check for that session variable and redirect to the
appropriate action.

I'm sure both of these have pitfalls that I'm not thinking of. Just a
couple of ideas to bat around.

Cheers,
David


>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Mon Jan  8 10:02:12 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 8 Jan 2007 10:02:12 -0500
Subject: [rspec-users] thoughts on mocks and specs
In-Reply-To: <810a540e0701072219t2d4c6b65n94d18741f20d99ae@mail.gmail.com>
References: <e282921e0701072155t468a49e2wd6ae5cdb734a8c2d@mail.gmail.com>
	<810a540e0701072219t2d4c6b65n94d18741f20d99ae@mail.gmail.com>
Message-ID: <57c63afe0701080702m7466b62ew6c728f60f0153c16@mail.gmail.com>

On 1/8/07, Pat Maddox <pergesu at gmail.com> wrote:
> On 1/7/07, Chris Anderson <jchris at mfdz.com> wrote:
> > I spent the last couple of days getting my sea legs with Rails and
> > RSpec. I'd been waiting til things seemed calmer before jumping in,
> > and I'm overall very happy with my experience so far.
> >
> > My only real annoyance so far has been forgetting to call "do_post" or
> > "do_create" from my specify blocks. My mocks don't get the calls they
> > want, and it usually isn't until I've mucked around in my application
> > code for a bit that I realize the real problem.
> >
> > Wouldn't it be nice to be able to define a callback that gets run
> > after every specify block? Maybe it would look like this:
> >
> > context "A leaner syntaxed context" do
> >
> >   setup do
> >     @obj = mock('obj')
> >   end
> >
> >   body do
> >     @obj.whatever
> >   end
> >
> >   specify "should run the body block after every specification" do
> >     @obj.should_receive(:whatever)
> >   end
> >
> > end
> >
> > I know sometimes one might want to put some verification code after
> > the body. In that case, you could allow an explicit call to body from
> > within the specify block, which would preclude it from being called
> > again at the end.
> >
> > It seems to me that this feature isn't just a fight for DRY, but also
> > encourages good spec design. Situations diverse enough to need
> > separate body code could use different contexts anyway.
> >
> > If there is a call for this feature, I'd be happy to make a go at
> > patching it into RSpec.
> >
> > Reactions?
> >
> > --
> > Chris Anderson
> > http://jchris.mfdz.com
>
> You can do that using teardown.

You could, yes, but that means that part of your spec is in teardown.
To me, setup and teardown are for setting up and tearing down an
environment in which you'll execute the specs. Each spec should tell
it's whole story. You shouldn't have to look at setup or teardown to
understand what is being specified, though you may need to look there
to understand failures.

Also, sometimes the do_blah needs to come after the expectations (when
you're using mocks), and sometimes before, so we need some way to say
"execute the request before OR after each specify block". Seems to me
we're opening up a can of worms here. Simpler to just put it in the
spec block.

>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dylanegan at gmail.com  Tue Jan  9 05:37:10 2007
From: dylanegan at gmail.com (Dylan Egan)
Date: Tue, 9 Jan 2007 21:37:10 +1100
Subject: [rspec-users] rcov seg fault
Message-ID: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>

Hi

Trying to get rcov going (has been working), but with the following context
and specs it fails.

context '/account POST with invalid attendee' do
  controller_name :account
  setup do
    Attendee.stub!(:create!).and_raise(ActiveRecord::RecordInvalid.new(
Attendee.new))
  end

  specify 'should raise on create' do
    Attendee.should_receive(:create!).with({ 'name' => 'Attendee'
}).and_raise(ActiveRecord::RecordInvalid.new(Attendee.new))
    post :create, :attendee => { :name => 'Attendee' }
  end

  specify 'should render new' do
    controller.should_render :action => 'new'
    post :create
  end
end

All I get is /usr/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/mocks/error_generator.rb:53:
[BUG] Segmentation fault

I tried variations on the specs and setup, but it fails when only running
the should render new spec and everything else commented out.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070109/039ad785/attachment.html 

From jonathan.tron.mailings at gmail.com  Tue Jan  9 05:58:18 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Tue, 9 Jan 2007 11:58:18 +0100
Subject: [rspec-users] rcov seg fault
In-Reply-To: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
Message-ID: <514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>

Le 9 janv. 07 ? 11:37, Dylan Egan a ?crit :

> Hi
>
> Trying to get rcov going (has been working), but with the following  
> context and specs it fails.
...
> All I get is /usr/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/mocks/ 
> error_generator.rb:53: [BUG] Segmentation fault

I get the same error but only on my MacbookPro and only when using  
mock_model (cf: http://www.dzone.com/rsslinks/ 
making_a_mockery_of_activerecord.html).
The same code running on a debian server (ruby 1.8.5-4, rspec 0.7.5,  
rcov 0.7.0.1, Rails 1.2RC) don't produce the error... But I get  
failed specs I did not get on my MacbookPro (I'm using Cerberus for CI)

Really weird...

Jonathan
--
Tron Jonathan
http://jonathan.tron.name




From dchelimsky at gmail.com  Tue Jan  9 09:36:33 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 9 Jan 2007 08:36:33 -0600
Subject: [rspec-users] rcov seg fault
In-Reply-To: <514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
	<514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
Message-ID: <57c63afe0701090636p50166481oa069ad6b17a7da2@mail.gmail.com>

On 1/9/07, Jonathan Tron <jonathan.tron.mailings at gmail.com> wrote:
> Le 9 janv. 07 ? 11:37, Dylan Egan a ?crit :
>
> > Hi
> >
> > Trying to get rcov going (has been working), but with the following
> > context and specs it fails.
> ...
> > All I get is /usr/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/mocks/
> > error_generator.rb:53: [BUG] Segmentation fault
>
> I get the same error but only on my MacbookPro and only when using
> mock_model (cf: http://www.dzone.com/rsslinks/
> making_a_mockery_of_activerecord.html).
> The same code running on a debian server (ruby 1.8.5-4, rspec 0.7.5,
> rcov 0.7.0.1, Rails 1.2RC) don't produce the error... But I get
> failed specs I did not get on my MacbookPro (I'm using Cerberus for CI)
>
> Really weird...

I was having similar problems on an MBP when I was using sqlite and
rails. Since I started always using mysql, no problem.

What was really weird was that I could get rid of the error by simply
adding a few blank lines to any file that was involved in the spec
run. VERY weird.


>
> Jonathan
> --
> Tron Jonathan
> http://jonathan.tron.name
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From jonathan.tron.mailings at gmail.com  Tue Jan  9 09:52:27 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Tue, 9 Jan 2007 15:52:27 +0100
Subject: [rspec-users] rcov seg fault
In-Reply-To: <57c63afe0701090636p50166481oa069ad6b17a7da2@mail.gmail.com>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
	<514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
	<57c63afe0701090636p50166481oa069ad6b17a7da2@mail.gmail.com>
Message-ID: <69119765-D7BC-496A-AC08-F081E217D98F@gmail.com>

Le 9 janv. 07 ? 15:36, David Chelimsky a ?crit :

> I was having similar problems on an MBP when I was using sqlite and
> rails. Since I started always using mysql, no problem.
>
> What was really weird was that I could get rid of the error by simply
> adding a few blank lines to any file that was involved in the spec
> run. VERY weird.

That's exactly what I did after replying, I switched to MySQL (then  
to PostgreSQL ;) ) and all errors vanished.
Besides these errors, the true (compared to sqlite's one) foreign key  
implementation of both Mysql/PostgreSQL forced me to refactor my spec  
on vanilla model separating more appropriatly what should be tested  
with and without fixtures loaded and what should hit the db and what  
should not :)... (mock with randomly generated id is BAD!!!!! when  
your db check for foreign keys infringement).

Jonathan
--
Tron Jonathan
http://jonathan.tron.name


From jonathan at home.tron.name  Tue Jan  9 06:51:02 2007
From: jonathan at home.tron.name (Jonathan Tron)
Date: Tue, 9 Jan 2007 12:51:02 +0100
Subject: [rspec-users] rcov seg fault
In-Reply-To: <514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
	<514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
Message-ID: <B9F5E91C-904D-41AC-8FD0-C5B10227695B@home.tron.name>

Le 9 janv. 07 ? 11:58, Jonathan Tron a ?crit :

> Le 9 janv. 07 ? 11:37, Dylan Egan a ?crit :
>
>> Hi
>>
>> Trying to get rcov going (has been working), but with the following
>> context and specs it fails.
> ...
>> All I get is /usr/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/mocks/
>> error_generator.rb:53: [BUG] Segmentation fault
>
> I get the same error but only on my MacbookPro and only when using
> mock_model (cf: http://www.dzone.com/rsslinks/
> making_a_mockery_of_activerecord.html).
> The same code running on a debian server (ruby 1.8.5-4, rspec 0.7.5,
> rcov 0.7.0.1, Rails 1.2RC) don't produce the error... But I get
> failed specs I did not get on my MacbookPro (I'm using Cerberus for  
> CI)

I just find the problem with different rspec results between my two  
configuration... I use sqlite3 as test database and for a reason I  
ignore, the rake task db:test:prepare seems to cause ActiveRecord to  
pre-populate some of my models with quotes (''). My specs on  
presence_of validations then just failed.
If I do a db:migrate / db:test:clone_structure and then run specs the  
errors vanished...

Jonathan
--
Tron Jonathan
http://jonathan.tron.name



From brian at yamabe.net  Tue Jan  9 15:48:01 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Tue, 9 Jan 2007 12:48:01 -0800
Subject: [rspec-users] spec not running in TextMate
Message-ID: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>

I tried running a spec with command-r in TextMate and got:

/Library/Application Support/TextMate/Bundles/RSpec.tmbundle/Support/ 
lib/spec_mate.rb:1:in `require': No such file to load -- rubygems  
(LoadError) from /Library/Application Support/TextMate/Bundles/ 
RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/ 
temp_textmate.GplhPr:3:in `require' from /tmp/temp_textmate.GplhPr:3

The spec runs fine from the command line.

Any ideas?

Thanks,
Brian

From dylanegan at gmail.com  Tue Jan  9 18:07:48 2007
From: dylanegan at gmail.com (Dylan Egan)
Date: Wed, 10 Jan 2007 10:07:48 +1100
Subject: [rspec-users] rcov seg fault
In-Reply-To: <B9F5E91C-904D-41AC-8FD0-C5B10227695B@home.tron.name>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
	<514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
	<B9F5E91C-904D-41AC-8FD0-C5B10227695B@home.tron.name>
Message-ID: <cbbc1640701091507p4e392209kf59eddcd3b6ee20e@mail.gmail.com>

Tried both suggestions. I moved from sqlite to postgresql.

I also added some blank lines to the spec file, but without any luck.

Will try and get it working somehow.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070110/52a9d0d2/attachment.html 

From cwdinfo at gmail.com  Tue Jan  9 18:59:24 2007
From: cwdinfo at gmail.com (s.ross)
Date: Tue, 9 Jan 2007 15:59:24 -0800
Subject: [rspec-users] Date Approximation in Specs
Message-ID: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>

The floating-point expectations allow for an error tolerance. Is  
there any similar facility for dates? For example, say I have a  
custom class that handles date/time spans and I want to spec it:

context "A DateRange span" do
   specify "should know when a week ago is :)" do
     d = DateRange.new
     d.last_week.should_be_close_to(1.week.ago, 24*60*60)
   end
end

The idea of should_be_close_to is to provide a tolerance. My class  
only is required to be accurate to the same day, and I want to  
express that in terms of "close_to" uncertainty. Is there a great  
baked-in way I'm missing?

I have code to do this in Test::Unit that I could port over if need  
be but I'd rather not if something better already exists.

Thanks,

Steve

From dchelimsky at gmail.com  Tue Jan  9 19:45:08 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 9 Jan 2007 18:45:08 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
Message-ID: <57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>

On 1/9/07, s.ross <cwdinfo at gmail.com> wrote:
> The floating-point expectations allow for an error tolerance. Is
> there any similar facility for dates? For example, say I have a
> custom class that handles date/time spans and I want to spec it:
>
> context "A DateRange span" do
>    specify "should know when a week ago is :)" do
>      d = DateRange.new
>      d.last_week.should_be_close_to(1.week.ago, 24*60*60)
>    end
> end
>
> The idea of should_be_close_to is to provide a tolerance. My class
> only is required to be accurate to the same day, and I want to
> express that in terms of "close_to" uncertainty. Is there a great
> baked-in way I'm missing?
>
> I have code to do this in Test::Unit that I could port over if need
> be but I'd rather not if something better already exists.

Doesn't exist yet. Feel free to add a feature request:

http://rubyforge.org/tracker/?group_id=797

> Thanks,
>
> Steve
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Tue Jan  9 23:26:42 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 9 Jan 2007 22:26:42 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
Message-ID: <57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>

On 1/9/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/9/07, s.ross <cwdinfo at gmail.com> wrote:
> > The floating-point expectations allow for an error tolerance. Is
> > there any similar facility for dates? For example, say I have a
> > custom class that handles date/time spans and I want to spec it:
> >
> > context "A DateRange span" do
> >    specify "should know when a week ago is :)" do
> >      d = DateRange.new
> >      d.last_week.should_be_close_to(1.week.ago, 24*60*60)
> >    end
> > end
> >
> > The idea of should_be_close_to is to provide a tolerance. My class
> > only is required to be accurate to the same day, and I want to
> > express that in terms of "close_to" uncertainty. Is there a great
> > baked-in way I'm missing?
> >
> > I have code to do this in Test::Unit that I could port over if need
> > be but I'd rather not if something better already exists.
>
> Doesn't exist yet. Feel free to add a feature request:
>
> http://rubyforge.org/tracker/?group_id=797

Hey Steve - I just committed a small enhancement to rspec core that is
intended to pave the way for custom expectations. Are you working w/
0.7.5 or the trunk? If you're working w/ trunk, grab the latest trunk
and do this:

module DateExpectations
  class BeCloseTo
    def initialize(target_date, tolerance)
      @target_date = target_date
      @tolerance = tolerance
    end

    def met_by?(target)
      #return true if target is a date
      #that meets the expectation
    end

    def failure_message
      #return an appropriate failure message
    end
  end

  def should_be_close_to(target_date, tolerance)
    return BeCloseTo.new(target_date, tolerance)
  end
end

context "My package" do
  include DateExpectations
  specify "should be delivered no later than 2 days after my order" do
    delivery_date.should be_close_to("1/9/2007", :days => 2)
  end
end

This is only experimental today, but will very likely become the entry
way for custom expectations. Let me know how it works out for you.

Cheers,
David

>
> > Thanks,
> >
> > Steve
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
>

From bhelmkamp at gmail.com  Wed Jan 10 01:47:12 2007
From: bhelmkamp at gmail.com (Bryan Helmkamp)
Date: Wed, 10 Jan 2007 01:47:12 -0500
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
Message-ID: <c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>

I don't have any help to offer, but I can confirm that I'm
experiencing this issue right now, as well as some other people I have
spoken with.

-Bryan

On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
> I tried running a spec with command-r in TextMate and got:
>
> /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/Support/
> lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
> (LoadError) from /Library/Application Support/TextMate/Bundles/
> RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
> temp_textmate.GplhPr:3:in `require' from /tmp/temp_textmate.GplhPr:3
>
> The spec runs fine from the command line.
>
> Any ideas?
>
> Thanks,
> Brian
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From cwdinfo at gmail.com  Wed Jan 10 02:26:42 2007
From: cwdinfo at gmail.com (s.ross)
Date: Tue, 9 Jan 2007 23:26:42 -0800
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
Message-ID: <9C454716-9EB6-472C-BCED-B1F1CE94E932@gmail.com>

I'm running 0.7.5. I guess I should learn how to install rspec from  
svn, huh? Basically, each time I change my version of rspec, I have  
to dive into 5 or 6 projects to update their Rails plugins and so  
on... Any shortcuts you use when doing this?

Thanks for doing this so quickly!

Steve

On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:

> On 1/9/07, David Chelimsky <dchelimsky at gmail.com> wrote:
>> On 1/9/07, s.ross <cwdinfo at gmail.com> wrote:
>>> The floating-point expectations allow for an error tolerance. Is
>>> there any similar facility for dates? For example, say I have a
>>> custom class that handles date/time spans and I want to spec it:
>>>
>>> context "A DateRange span" do
>>>    specify "should know when a week ago is :)" do
>>>      d = DateRange.new
>>>      d.last_week.should_be_close_to(1.week.ago, 24*60*60)
>>>    end
>>> end
>>>
>>> The idea of should_be_close_to is to provide a tolerance. My class
>>> only is required to be accurate to the same day, and I want to
>>> express that in terms of "close_to" uncertainty. Is there a great
>>> baked-in way I'm missing?
>>>
>>> I have code to do this in Test::Unit that I could port over if need
>>> be but I'd rather not if something better already exists.
>>
>> Doesn't exist yet. Feel free to add a feature request:
>>
>> http://rubyforge.org/tracker/?group_id=797
>
> Hey Steve - I just committed a small enhancement to rspec core that is
> intended to pave the way for custom expectations. Are you working w/
> 0.7.5 or the trunk? If you're working w/ trunk, grab the latest trunk
> and do this:
>
> module DateExpectations
>   class BeCloseTo
>     def initialize(target_date, tolerance)
>       @target_date = target_date
>       @tolerance = tolerance
>     end
>
>     def met_by?(target)
>       #return true if target is a date
>       #that meets the expectation
>     end
>
>     def failure_message
>       #return an appropriate failure message
>     end
>   end
>
>   def should_be_close_to(target_date, tolerance)
>     return BeCloseTo.new(target_date, tolerance)
>   end
> end
>
> context "My package" do
>   include DateExpectations
>   specify "should be delivered no later than 2 days after my order" do
>     delivery_date.should be_close_to("1/9/2007", :days => 2)
>   end
> end
>
> This is only experimental today, but will very likely become the entry
> way for custom expectations. Let me know how it works out for you.
>
> Cheers,
> David
>
>>
>>> Thanks,
>>>
>>> Steve
>>> _______________________________________________
>>> 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


From jonathan.tron.mailings at gmail.com  Wed Jan 10 03:25:01 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Wed, 10 Jan 2007 09:25:01 +0100
Subject: [rspec-users] rcov seg fault
In-Reply-To: <cbbc1640701091507p4e392209kf59eddcd3b6ee20e@mail.gmail.com>
References: <cbbc1640701090237v7b733007t995ec11312614648@mail.gmail.com>
	<514DD5CA-2A2D-484E-824F-F2AF3F2B06C1@gmail.com>
	<B9F5E91C-904D-41AC-8FD0-C5B10227695B@home.tron.name>
	<cbbc1640701091507p4e392209kf59eddcd3b6ee20e@mail.gmail.com>
Message-ID: <A0C78B0A-B1C5-4308-B5AD-30A39DCA3546@gmail.com>

Le 10 janv. 07 ? 00:07, Dylan Egan a ?crit :

> Tried both suggestions. I moved from sqlite to postgresql.
>
> I also added some blank lines to the spec file, but without any luck.
>
> Will try and get it working somehow.

Sorry about the confusion, mysql/postgresql solutions was related to  
the fact that some specs doesn't work on sqlite3 when used in a  
different system.

My problem of seg fault when using rcov and mock_model remains  
unsolved (on my MacbookPro), I think it's a bug in ruby or rcov but I  
don't really have time to track it down.

Jonathan
--
Tron Jonathan
http://jonathan.tron.name




From aslak.hellesoy at gmail.com  Wed Jan 10 04:49:15 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Wed, 10 Jan 2007 10:49:15 +0100
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
	<c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
Message-ID: <8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>

Sounds like TM is running a Ruby interpreter that doesn't have
Rubygems installed.

Tell TM to run a different Ruby - it's documented somewhere, but I
don't remember where.

Aslak

On 1/10/07, Bryan Helmkamp <bhelmkamp at gmail.com> wrote:
> I don't have any help to offer, but I can confirm that I'm
> experiencing this issue right now, as well as some other people I have
> spoken with.
>
> -Bryan
>
> On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
> > I tried running a spec with command-r in TextMate and got:
> >
> > /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/Support/
> > lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
> > (LoadError) from /Library/Application Support/TextMate/Bundles/
> > RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
> > temp_textmate.GplhPr:3:in `require' from /tmp/temp_textmate.GplhPr:3
> >
> > The spec runs fine from the command line.
> >
> > Any ideas?
> >
> > Thanks,
> > Brian
> > _______________________________________________
> > 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
>

From brian at yamabe.net  Wed Jan 10 07:21:12 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Wed, 10 Jan 2007 04:21:12 -0800
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
	<c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
	<8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>
Message-ID: <DA7394F9-999B-489E-8732-545E21BDD093@yamabe.net>

I do have TM_RUBY set to the same Ruby interpreter as the one I use  
from the terminal.

---Brian

On Jan 10, 2007, at 1:49 AM, aslak hellesoy wrote:

> Sounds like TM is running a Ruby interpreter that doesn't have
> Rubygems installed.
>
> Tell TM to run a different Ruby - it's documented somewhere, but I
> don't remember where.
>
> Aslak
>
> On 1/10/07, Bryan Helmkamp <bhelmkamp at gmail.com> wrote:
>> I don't have any help to offer, but I can confirm that I'm
>> experiencing this issue right now, as well as some other people I  
>> have
>> spoken with.
>>
>> -Bryan
>>
>> On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
>>> I tried running a spec with command-r in TextMate and got:
>>>
>>> /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/ 
>>> Support/
>>> lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
>>> (LoadError) from /Library/Application Support/TextMate/Bundles/
>>> RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
>>> temp_textmate.GplhPr:3:in `require' from /tmp/temp_textmate.GplhPr:3
>>>
>>> The spec runs fine from the command line.
>>>
>>> Any ideas?
>>>
>>> Thanks,
>>> Brian
>>> _______________________________________________
>>> 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


From aslak.hellesoy at gmail.com  Wed Jan 10 10:00:54 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Wed, 10 Jan 2007 16:00:54 +0100
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <DA7394F9-999B-489E-8732-545E21BDD093@yamabe.net>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
	<c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
	<8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>
	<DA7394F9-999B-489E-8732-545E21BDD093@yamabe.net>
Message-ID: <8d961d900701100700j30c674e7pf216ad9e42039eb@mail.gmail.com>

Let's narrow it down a little to see if this is a problem with the
bundle or your environment.

Try this:

1) Create a file called foo.rb
2) Add one line to it: require 'rubygems'
3) Open it in TextMate
4) Run it from TextMate (Cmd-R - Run)

Do you get the same error? If so, your environment is bad.

Aslak

On 1/10/07, Brian Yamabe <brian at yamabe.net> wrote:
> I do have TM_RUBY set to the same Ruby interpreter as the one I use
> from the terminal.
>
> ---Brian
>
> On Jan 10, 2007, at 1:49 AM, aslak hellesoy wrote:
>
> > Sounds like TM is running a Ruby interpreter that doesn't have
> > Rubygems installed.
> >
> > Tell TM to run a different Ruby - it's documented somewhere, but I
> > don't remember where.
> >
> > Aslak
> >
> > On 1/10/07, Bryan Helmkamp <bhelmkamp at gmail.com> wrote:
> >> I don't have any help to offer, but I can confirm that I'm
> >> experiencing this issue right now, as well as some other people I
> >> have
> >> spoken with.
> >>
> >> -Bryan
> >>
> >> On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
> >>> I tried running a spec with command-r in TextMate and got:
> >>>
> >>> /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/
> >>> Support/
> >>> lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
> >>> (LoadError) from /Library/Application Support/TextMate/Bundles/
> >>> RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
> >>> temp_textmate.GplhPr:3:in `require' from /tmp/temp_textmate.GplhPr:3
> >>>
> >>> The spec runs fine from the command line.
> >>>
> >>> Any ideas?
> >>>
> >>> Thanks,
> >>> Brian
> >>> _______________________________________________
> >>> 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
>

From brian at yamabe.net  Wed Jan 10 10:56:11 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Wed, 10 Jan 2007 07:56:11 -0800
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <8d961d900701100700j30c674e7pf216ad9e42039eb@mail.gmail.com>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
	<c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
	<8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>
	<DA7394F9-999B-489E-8732-545E21BDD093@yamabe.net>
	<8d961d900701100700j30c674e7pf216ad9e42039eb@mail.gmail.com>
Message-ID: <647AD79E-646B-4075-8259-DE2A6FC54C6E@yamabe.net>

Tried it and it worked without a problem, so it looks like a bundle  
problem.
I don't know much about bundles, but I'd be happy to work with someone
who has suggestions for how to fix this.

Thank, Aslak. You and David do a great job on this list.

---Brian

On Jan 10, 2007, at 7:00 AM, aslak hellesoy wrote:

> Let's narrow it down a little to see if this is a problem with the
> bundle or your environment.
>
> Try this:
>
> 1) Create a file called foo.rb
> 2) Add one line to it: require 'rubygems'
> 3) Open it in TextMate
> 4) Run it from TextMate (Cmd-R - Run)
>
> Do you get the same error? If so, your environment is bad.
>
> Aslak
>
> On 1/10/07, Brian Yamabe <brian at yamabe.net> wrote:
>> I do have TM_RUBY set to the same Ruby interpreter as the one I use
>> from the terminal.
>>
>> ---Brian
>>
>> On Jan 10, 2007, at 1:49 AM, aslak hellesoy wrote:
>>
>>> Sounds like TM is running a Ruby interpreter that doesn't have
>>> Rubygems installed.
>>>
>>> Tell TM to run a different Ruby - it's documented somewhere, but I
>>> don't remember where.
>>>
>>> Aslak
>>>
>>> On 1/10/07, Bryan Helmkamp <bhelmkamp at gmail.com> wrote:
>>>> I don't have any help to offer, but I can confirm that I'm
>>>> experiencing this issue right now, as well as some other people I
>>>> have
>>>> spoken with.
>>>>
>>>> -Bryan
>>>>
>>>> On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
>>>>> I tried running a spec with command-r in TextMate and got:
>>>>>
>>>>> /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/
>>>>> Support/
>>>>> lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
>>>>> (LoadError) from /Library/Application Support/TextMate/Bundles/
>>>>> RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
>>>>> temp_textmate.GplhPr:3:in `require' from /tmp/ 
>>>>> temp_textmate.GplhPr:3
>>>>>
>>>>> The spec runs fine from the command line.
>>>>>
>>>>> Any ideas?
>>>>>
>>>>> Thanks,
>>>>> Brian
>>>>> _______________________________________________
>>>>> 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


From cwdinfo at gmail.com  Wed Jan 10 14:01:57 2007
From: cwdinfo at gmail.com (s.ross)
Date: Wed, 10 Jan 2007 11:01:57 -0800
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
Message-ID: <EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>

I upgraded to trunk and changed the code as follows:

module DateExpectations
     HOUR_TOLERANCE = 60 * 60
     DAY_TOLERANCE = HOUR_TOLERANCE * 24
     WEEK_TOLERANCE = DAY_TOLERANCE * 7

   class BeCloseTo
     def initialize(target_date, tolerance)
       @target_date = Time.parse(target_date)
       @tolerance = tolerance
     end

     def met_by?(target)
       compare_date = target - @target_date
       adjustment = @tolerance.inject do |item|
           case item
           when :days
               adjustment += DAY_TOLERANCE
           when :weeks
               adjustment += WEEK_TOLERANCE
           end
       end
       return true if compare_date.abs < adjustment
     end

     def failure_message
       return "#{@target_date} does not fall within tolerance."  # s/ 
b better
     end
   end

   def should_be_close_to(target_date, tolerance)
     return BeCloseTo.new(target_date, tolerance)
   end
end

context "My package" do
   include DateExpectations
   specify "should be delivered no later than 2 days after my order" do
     1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <- line 40
   end
end

The result is:

1)
NoMethodError in 'My package should be delivered no later than 2 days  
after my order'
undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
./spec/helpers/date_range_spec.rb:40:


Any thoughts as to why? (Note: You had written the expectation in  
your sample as:
"should<space>be_close_to" -- was that intentional?).

Steve


On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:

> module DateExpectations
>   class BeCloseTo
>     def initialize(target_date, tolerance)
>       @target_date = target_date
>       @tolerance = tolerance
>     end
>
>     def met_by?(target)
>       #return true if target is a date
>       #that meets the expectation
>     end
>
>     def failure_message
>       #return an appropriate failure message
>     end
>   end
>
>   def should_be_close_to(target_date, tolerance)
>     return BeCloseTo.new(target_date, tolerance)
>   end
> end
>
> context "My package" do
>   include DateExpectations
>   specify "should be delivered no later than 2 days after my order" do
>     delivery_date.should be_close_to("1/9/2007", :days => 2)
>   end
> end


From rob at muhlestein.net  Wed Jan 10 13:31:24 2007
From: rob at muhlestein.net (Rob Muhlestein)
Date: Wed, 10 Jan 2007 13:31:24 -0500
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec file name
	convention)
Message-ID: <1168453884.4245.53.camel@dads>

Is it too late to suggest some filename conventions for example rspec
files--especially when bundling with gems? 

I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
around. Would having a foo.rspec be worth talking about? Or is the
convention more or less to have 'spec' in the file name?

I have to confess the only real motivation I have at the moment is
syntax highlighting without adding RSpec DSL tags to ruby.vim (which I
went ahead and did anyway.) If acceptable as a published convention, we
could then recommend to editor projects.

Here's one link suggesting how to incorporate RSpec into rake and gems:

http://townx.org/running_rspec_as_a_rake_task

Personally I'd like so see this convention distilled enough to be
mentioned in the pickax "Creating Your Own Gems - Package Layout"
section (not to mention an entire additional section dedicated to
BDD to go with or replace the old asserts stuff).

Loving RSpec. You actually get work done while writing that spec doc any
PHB can understand, even if the application isn't written in Ruby.
Looking forward to using RSpec from JRuby to apply BDD to the legacy
Java stuff.

Regards,

-- 
Rob Muhlestein
http://rob.muhlestein.net


From shane.duan at gmail.com  Wed Jan 10 14:14:41 2007
From: shane.duan at gmail.com (Shane Duan)
Date: Wed, 10 Jan 2007 11:14:41 -0800
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec file
	name convention)
In-Reply-To: <1168453884.4245.53.camel@dads>
References: <1168453884.4245.53.camel@dads>
Message-ID: <bd7896da0701101114t43285a49kb4b7d1a3f751bd24@mail.gmail.com>

Coming up with new extension involves buy-ins from all the editors
(intelliJ, eclipse, vi, emacs), and major operation systems.  So I
really doubt it will take off.  Having the naming convention feels
like a good way to go.

As for the link provided, you just meant the naming convention, right?
 I learned to write it just following rspec's document.  Worked out
great for me, especially the rcov and rspec HTML report.

On 1/10/07, Rob Muhlestein <rob at muhlestein.net> wrote:
> Is it too late to suggest some filename conventions for example rspec
> files--especially when bundling with gems?
>
> I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
> around. Would having a foo.rspec be worth talking about? Or is the
> convention more or less to have 'spec' in the file name?
>
> I have to confess the only real motivation I have at the moment is
> syntax highlighting without adding RSpec DSL tags to ruby.vim (which I
> went ahead and did anyway.) If acceptable as a published convention, we
> could then recommend to editor projects.
>
> Here's one link suggesting how to incorporate RSpec into rake and gems:
>
> http://townx.org/running_rspec_as_a_rake_task
>
> Personally I'd like so see this convention distilled enough to be
> mentioned in the pickax "Creating Your Own Gems - Package Layout"
> section (not to mention an entire additional section dedicated to
> BDD to go with or replace the old asserts stuff).
>
> Loving RSpec. You actually get work done while writing that spec doc any
> PHB can understand, even if the application isn't written in Ruby.
> Looking forward to using RSpec from JRuby to apply BDD to the legacy
> Java stuff.
>
> Regards,
>
> --
> Rob Muhlestein
> http://rob.muhlestein.net
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>


-- 
Shane
http://www.shaneduan.com

From pergesu at gmail.com  Wed Jan 10 14:16:52 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Wed, 10 Jan 2007 12:16:52 -0700
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec file
	name convention)
In-Reply-To: <1168453884.4245.53.camel@dads>
References: <1168453884.4245.53.camel@dads>
Message-ID: <810a540e0701101116t289a4302p2a6a70ea3415bf33@mail.gmail.com>

On 1/10/07, Rob Muhlestein <rob at muhlestein.net> wrote:
> Is it too late to suggest some filename conventions for example rspec
> files--especially when bundling with gems?
>
> I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
> around. Would having a foo.rspec be worth talking about? Or is the
> convention more or less to have 'spec' in the file name?

I like foo_spec.rb

Pat

From bryan at osesm.com  Wed Jan 10 14:18:58 2007
From: bryan at osesm.com (Bryan Liles)
Date: Wed, 10 Jan 2007 14:18:58 -0500
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec file
	name convention)
In-Reply-To: <1168453884.4245.53.camel@dads>
References: <1168453884.4245.53.camel@dads>
Message-ID: <15D6D9AD-CEBD-43C3-A754-E79538E77D1F@osesm.com>


On Jan 10, 2007, at 1:31 PM, Rob Muhlestein wrote:

> Is it too late to suggest some filename conventions for example rspec
> files--especially when bundling with gems?
>
> I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
> around. Would having a foo.rspec be worth talking about? Or is the
> convention more or less to have 'spec' in the file name?
>

I would vote against this (even if my vote even mattered).  I like to
think of my rspecs of ruby that actually verifies my code is doing
what I think it is doing.



From dchelimsky at gmail.com  Wed Jan 10 14:22:01 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 10 Jan 2007 13:22:01 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
	<EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
Message-ID: <57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>

On 1/10/07, s.ross <cwdinfo at gmail.com> wrote:
> I upgraded to trunk and changed the code as follows:
>
> module DateExpectations
>      HOUR_TOLERANCE = 60 * 60
>      DAY_TOLERANCE = HOUR_TOLERANCE * 24
>      WEEK_TOLERANCE = DAY_TOLERANCE * 7
>
>    class BeCloseTo
>      def initialize(target_date, tolerance)
>        @target_date = Time.parse(target_date)
>        @tolerance = tolerance
>      end
>
>      def met_by?(target)
>        compare_date = target - @target_date
>        adjustment = @tolerance.inject do |item|
>            case item
>            when :days
>                adjustment += DAY_TOLERANCE
>            when :weeks
>                adjustment += WEEK_TOLERANCE
>            end
>        end
>        return true if compare_date.abs < adjustment
>      end
>
>      def failure_message
>        return "#{@target_date} does not fall within tolerance."  # s/
> b better
>      end
>    end
>
>    def should_be_close_to(target_date, tolerance)
>      return BeCloseTo.new(target_date, tolerance)
>    end
> end
>
> context "My package" do
>    include DateExpectations
>    specify "should be delivered no later than 2 days after my order" do
>      1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <- line 40
>    end
> end
>
> The result is:
>
> 1)
> NoMethodError in 'My package should be delivered no later than 2 days
> after my order'
> undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
> ./spec/helpers/date_range_spec.rb:40:
>
>
> Any thoughts as to why? (Note: You had written the expectation in
> your sample as:
> "should<space>be_close_to" -- was that intentional?).

Yes!!!! That's the whole point. ;)

should be_close_to

and your method should be named be_close_to

Let me know how it works out.

Thanks for giving this a shot - you're the guinea pig!

David

>
> Steve
>
>
> On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:
>
> > module DateExpectations
> >   class BeCloseTo
> >     def initialize(target_date, tolerance)
> >       @target_date = target_date
> >       @tolerance = tolerance
> >     end
> >
> >     def met_by?(target)
> >       #return true if target is a date
> >       #that meets the expectation
> >     end
> >
> >     def failure_message
> >       #return an appropriate failure message
> >     end
> >   end
> >
> >   def should_be_close_to(target_date, tolerance)
> >     return BeCloseTo.new(target_date, tolerance)
> >   end
> > end
> >
> > context "My package" do
> >   include DateExpectations
> >   specify "should be delivered no later than 2 days after my order" do
> >     delivery_date.should be_close_to("1/9/2007", :days => 2)
> >   end
> > end
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From shane.duan at gmail.com  Wed Jan 10 14:26:21 2007
From: shane.duan at gmail.com (Shane Duan)
Date: Wed, 10 Jan 2007 11:26:21 -0800
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec file
	name convention)
In-Reply-To: <810a540e0701101116t289a4302p2a6a70ea3415bf33@mail.gmail.com>
References: <1168453884.4245.53.camel@dads>
	<810a540e0701101116t289a4302p2a6a70ea3415bf33@mail.gmail.com>
Message-ID: <bd7896da0701101126m3b7d32e5h7964333ee3a1e3b3@mail.gmail.com>

The history from the Java world has shown that there will never be one
standard.  People choose different naming for different preferences.
I even have 'tc_' prefix in the project because I never bothered the
trouble after I converted them from ruby unit.

On 1/10/07, Pat Maddox <pergesu at gmail.com> wrote:
> On 1/10/07, Rob Muhlestein <rob at muhlestein.net> wrote:
> > Is it too late to suggest some filename conventions for example rspec
> > files--especially when bundling with gems?
> >
> > I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
> > around. Would having a foo.rspec be worth talking about? Or is the
> > convention more or less to have 'spec' in the file name?
>
> I like foo_spec.rb
>
> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>


-- 
Shane
http://www.shaneduan.com

From dchelimsky at gmail.com  Wed Jan 10 14:28:55 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 10 Jan 2007 13:28:55 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
	<EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
	<57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
Message-ID: <57c63afe0701101128t6f81d2f9y2896c928c5e6d6e1@mail.gmail.com>

On 1/10/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/10/07, s.ross <cwdinfo at gmail.com> wrote:
> > I upgraded to trunk and changed the code as follows:
> >
> > module DateExpectations
> >      HOUR_TOLERANCE = 60 * 60
> >      DAY_TOLERANCE = HOUR_TOLERANCE * 24
> >      WEEK_TOLERANCE = DAY_TOLERANCE * 7
> >
> >    class BeCloseTo
> >      def initialize(target_date, tolerance)
> >        @target_date = Time.parse(target_date)
> >        @tolerance = tolerance
> >      end
> >
> >      def met_by?(target)
> >        compare_date = target - @target_date
> >        adjustment = @tolerance.inject do |item|
> >            case item
> >            when :days
> >                adjustment += DAY_TOLERANCE
> >            when :weeks
> >                adjustment += WEEK_TOLERANCE
> >            end
> >        end
> >        return true if compare_date.abs < adjustment
> >      end
> >
> >      def failure_message
> >        return "#{@target_date} does not fall within tolerance."  # s/
> > b better
> >      end
> >    end
> >
> >    def should_be_close_to(target_date, tolerance)
> >      return BeCloseTo.new(target_date, tolerance)
> >    end
> > end
> >
> > context "My package" do
> >    include DateExpectations
> >    specify "should be delivered no later than 2 days after my order" do
> >      1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <- line 40
> >    end
> > end
> >
> > The result is:
> >
> > 1)
> > NoMethodError in 'My package should be delivered no later than 2 days
> > after my order'
> > undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
> > ./spec/helpers/date_range_spec.rb:40:
> >
> >
> > Any thoughts as to why? (Note: You had written the expectation in
> > your sample as:
> > "should<space>be_close_to" -- was that intentional?).
>
> Yes!!!! That's the whole point. ;)

I should clarify that a bit. The idea is to decouple expectations from
Object so that we can more easily write, maintain and extend them
while simultaneously minimizing the risk of other frameworks (like
rails) screwing w/ our use of method missing.

If you look at the initialize method in should.rb, it now takes an
optional expectation argument. If it gets one, it processes it. What
you are writing is an expectation that gets passed to should, which
then passes the target to the expectation for verification.

Make sense?

David

>
> should be_close_to
>
> and your method should be named be_close_to
>
> Let me know how it works out.
>
> Thanks for giving this a shot - you're the guinea pig!
>
> David
>
> >
> > Steve
> >
> >
> > On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:
> >
> > > module DateExpectations
> > >   class BeCloseTo
> > >     def initialize(target_date, tolerance)
> > >       @target_date = target_date
> > >       @tolerance = tolerance
> > >     end
> > >
> > >     def met_by?(target)
> > >       #return true if target is a date
> > >       #that meets the expectation
> > >     end
> > >
> > >     def failure_message
> > >       #return an appropriate failure message
> > >     end
> > >   end
> > >
> > >   def should_be_close_to(target_date, tolerance)
> > >     return BeCloseTo.new(target_date, tolerance)
> > >   end
> > > end
> > >
> > > context "My package" do
> > >   include DateExpectations
> > >   specify "should be delivered no later than 2 days after my order" do
> > >     delivery_date.should be_close_to("1/9/2007", :days => 2)
> > >   end
> > > end
> >
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
>

From dchelimsky at gmail.com  Wed Jan 10 14:40:27 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 10 Jan 2007 13:40:27 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701101128t6f81d2f9y2896c928c5e6d6e1@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
	<EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
	<57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
	<57c63afe0701101128t6f81d2f9y2896c928c5e6d6e1@mail.gmail.com>
Message-ID: <57c63afe0701101140i72e41457v7ed86b6e1d24f2c8@mail.gmail.com>

On 1/10/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/10/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 1/10/07, s.ross <cwdinfo at gmail.com> wrote:
> > > I upgraded to trunk and changed the code as follows:
> > >
> > > module DateExpectations
> > >      HOUR_TOLERANCE = 60 * 60
> > >      DAY_TOLERANCE = HOUR_TOLERANCE * 24
> > >      WEEK_TOLERANCE = DAY_TOLERANCE * 7
> > >
> > >    class BeCloseTo
> > >      def initialize(target_date, tolerance)
> > >        @target_date = Time.parse(target_date)
> > >        @tolerance = tolerance
> > >      end
> > >
> > >      def met_by?(target)
> > >        compare_date = target - @target_date
> > >        adjustment = @tolerance.inject do |item|
> > >            case item
> > >            when :days
> > >                adjustment += DAY_TOLERANCE
> > >            when :weeks
> > >                adjustment += WEEK_TOLERANCE
> > >            end
> > >        end
> > >        return true if compare_date.abs < adjustment
> > >      end
> > >
> > >      def failure_message
> > >        return "#{@target_date} does not fall within tolerance."  # s/
> > > b better
> > >      end
> > >    end
> > >
> > >    def should_be_close_to(target_date, tolerance)
> > >      return BeCloseTo.new(target_date, tolerance)
> > >    end
> > > end
> > >
> > > context "My package" do
> > >    include DateExpectations
> > >    specify "should be delivered no later than 2 days after my order" do
> > >      1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <- line 40
> > >    end
> > > end
> > >
> > > The result is:
> > >
> > > 1)
> > > NoMethodError in 'My package should be delivered no later than 2 days
> > > after my order'
> > > undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
> > > ./spec/helpers/date_range_spec.rb:40:
> > >
> > >
> > > Any thoughts as to why? (Note: You had written the expectation in
> > > your sample as:
> > > "should<space>be_close_to" -- was that intentional?).
> >
> > Yes!!!! That's the whole point. ;)
>
> I should clarify that a bit. The idea is to decouple expectations from
> Object so that we can more easily write, maintain and extend them
> while simultaneously minimizing the risk of other frameworks (like
> rails) screwing w/ our use of method missing.
>
> If you look at the initialize method in should.rb, it now takes an
> optional expectation argument. If it gets one, it processes it. What
> you are writing is an expectation that gets passed to should, which
> then passes the target to the expectation for verification.
>
> Make sense?

I blogged about this here:
http://blog.davidchelimsky.net/articles/2007/01/10/rspec-should-use_a_little_less_magic

Comments welcome and appreciated.

Thanks,
David


>
> David
>
> >
> > should be_close_to
> >
> > and your method should be named be_close_to
> >
> > Let me know how it works out.
> >
> > Thanks for giving this a shot - you're the guinea pig!
> >
> > David
> >
> > >
> > > Steve
> > >
> > >
> > > On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:
> > >
> > > > module DateExpectations
> > > >   class BeCloseTo
> > > >     def initialize(target_date, tolerance)
> > > >       @target_date = target_date
> > > >       @tolerance = tolerance
> > > >     end
> > > >
> > > >     def met_by?(target)
> > > >       #return true if target is a date
> > > >       #that meets the expectation
> > > >     end
> > > >
> > > >     def failure_message
> > > >       #return an appropriate failure message
> > > >     end
> > > >   end
> > > >
> > > >   def should_be_close_to(target_date, tolerance)
> > > >     return BeCloseTo.new(target_date, tolerance)
> > > >   end
> > > > end
> > > >
> > > > context "My package" do
> > > >   include DateExpectations
> > > >   specify "should be delivered no later than 2 days after my order" do
> > > >     delivery_date.should be_close_to("1/9/2007", :days => 2)
> > > >   end
> > > > end
> > >
> > > _______________________________________________
> > > rspec-users mailing list
> > > rspec-users at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/rspec-users
> > >
> >
>

From rob at muhlestein.net  Wed Jan 10 15:44:53 2007
From: rob at muhlestein.net (Rob Muhlestein)
Date: Wed, 10 Jan 2007 15:44:53 -0500
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec
	file	name convention)
In-Reply-To: <bd7896da0701101126m3b7d32e5h7964333ee3a1e3b3@mail.gmail.com>
References: <1168453884.4245.53.camel@dads>
	<810a540e0701101116t289a4302p2a6a70ea3415bf33@mail.gmail.com>
	<bd7896da0701101126m3b7d32e5h7964333ee3a1e3b3@mail.gmail.com>
Message-ID: <1168461893.4245.64.camel@dads>

On Wed, 2007-01-10 at 11:26 -0800, Shane Duan wrote:

> On 1/10/07, Pat Maddox <pergesu at gmail.com> wrote:
> > On 1/10/07, Rob Muhlestein <rob at muhlestein.net> wrote:
> > > Is it too late to suggest some filename conventions for example rspec
> > > files--especially when bundling with gems?
> > >
> > > I see spec_foo.rb and foo_spec.rb around. Also found some foo_ex.rb
> > > around. Would having a foo.rspec be worth talking about? Or is the
> > > convention more or less to have 'spec' in the file name?
> >
> > I like foo_spec.rb
>  [... corrected top-posting ...]
> The history from the Java world has shown that there will never be one
> standard.  People choose different naming for different preferences.
> I even have 'tc_' prefix in the project because I never bothered the
> trouble after I converted them from ruby unit.

Agreed, but convention is different than standard. Perl users starting
using .t for their test scripts and it stuck.

After a full reread of the http://rspec.rubyforge.org site (a good one I
might add) it is very clear the intent is to use foo_spec.rb and from
under the 'spec/' directory in rails. It makes sense, then, that for
gems these should go in 'spec/' next to 'lib/' for example although I'm
sure we'll find specs in 'test/' for many years to come.

I'll go ahead and submit a patch to ruby-vim adding the rspec DSL tags
(context, setup, etc). Is there even one other vim user on this list
that understands my point?

Thanks

-- 
Rob Muhlestein
http://rob.muhlestein.net


From rob at muhlestein.net  Wed Jan 10 15:56:19 2007
From: rob at muhlestein.net (Rob Muhlestein)
Date: Wed, 10 Jan 2007 15:56:19 -0500
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed RSpec
	file	name convention)
In-Reply-To: <bd7896da0701101114t43285a49kb4b7d1a3f751bd24@mail.gmail.com>
References: <1168453884.4245.53.camel@dads>
	<bd7896da0701101114t43285a49kb4b7d1a3f751bd24@mail.gmail.com>
Message-ID: <1168462579.4245.75.camel@dads>

On Wed, 2007-01-10 at 11:14 -0800, Shane Duan wrote:
> As for the link provided, you just meant the naming convention, right?
>  I learned to write it just following rspec's document.  Worked out
> great for me, especially the rcov and rspec HTML report.

The only part I was missing was that you can replace use of the 'spec'
command with the following to make your spec a standalone ruby script:

  require 'rubygems'
  require_gem 'rspec'

Obviously this is needed to define the 'context', 'setup', and the rest
of the RSpec DSL. This can be derived from cracking open the 'spec'
command, but is not immediately obvious to newbies, which I still
consider myself to be.

By the way, this isn't mentioned anywhere on the
http://rspec.rubyforge.org web site and was not immediately obvious to
me. This might be for a very good reason that I just don't see yet, but
if not, I'd like to humble suggest a note go into the example or
tutorial to cover it.

-- 
Rob Muhlestein
http://rob.muhlestein.net


From rob at muhlestein.net  Wed Jan 10 16:15:35 2007
From: rob at muhlestein.net (Rob Muhlestein)
Date: Wed, 10 Jan 2007 16:15:35 -0500
Subject: [rspec-users] foo_spec.rb -> foo.rspec (proposed
	RSpec	file	name convention)
In-Reply-To: <1168461893.4245.64.camel@dads>
References: <1168453884.4245.53.camel@dads>
	<810a540e0701101116t289a4302p2a6a70ea3415bf33@mail.gmail.com>
	<bd7896da0701101126m3b7d32e5h7964333ee3a1e3b3@mail.gmail.com>
	<1168461893.4245.64.camel@dads>
Message-ID: <1168463735.4245.83.camel@dads>

On Wed, 2007-01-10 at 15:44 -0500, Rob Muhlestein wrote:
> I'll go ahead and submit a patch to ruby-vim adding the rspec DSL tags
> (context, setup, etc). 

Here are all the keywords from the RSpec DSL that I can gather from the
website:

context
context_setup
setup
specify
context_teardown
teardown

Am I missing any? I've got them and few others (require_gem) I'm going
to submit as a ruby.vim patch to ruby-vim. Perhaps updates to the other
editors might be nice as well.

-- 
Rob Muhlestein
http://rob.muhlestein.net


From cwdinfo at gmail.com  Thu Jan 11 02:18:11 2007
From: cwdinfo at gmail.com (s.ross)
Date: Wed, 10 Jan 2007 23:18:11 -0800
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
	<EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
	<57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
Message-ID: <BFEBA93D-76A8-4D9D-BA4B-1BD3FA385984@gmail.com>

My net connection's gone spotty over the last couple of days, but I  
wanted to follow up and tell you that your new code works great.  
Here's just a proof of concept implementation (complete with hard  
coding!):

module DateExpectations
     HOUR_TOLERANCE = 60 * 60
     DAY_TOLERANCE = HOUR_TOLERANCE * 24
     WEEK_TOLERANCE = DAY_TOLERANCE * 7

   class BeCloseTo
     def initialize(target_date, tolerance)
       @target_date = Time.parse(target_date)
       @tolerance = tolerance
     end

     def met_by?(target)
       @compare_date = target
       compare_span = target - @target_date
       adjustment = 0
       @tolerance.each do |key, amount|
           case key
           when :hours
               adjustment += HOUR_TOLERANCE * amount
           when :days
               adjustment += DAY_TOLERANCE * amount
           when :weeks
               adjustment += WEEK_TOLERANCE * amount
           end
       end
       return true if compare_span.abs < adjustment
     end

     def failure_message
       return "#{@target_date} does not fall within tolerance (# 
{@tolerance.collect{|k,v| "#{k.to_s} => #{v}"}.join(", ")}) from # 
{@compare_date}."
     end
   end

   def be_close_to(target_date, tolerance)
     return BeCloseTo.new(target_date, tolerance)
   end
end

context "The DateRange helper" do
   include DateExpectations
   specify "should be able to approximate within two days" do
     1.day.ago.should be_close_to("1/9/2007", :days => 2)
   end
end


On Jan 10, 2007, at 11:22 AM, David Chelimsky wrote:

> On 1/10/07, s.ross <cwdinfo at gmail.com> wrote:
>> I upgraded to trunk and changed the code as follows:
>>
>> module DateExpectations
>>      HOUR_TOLERANCE = 60 * 60
>>      DAY_TOLERANCE = HOUR_TOLERANCE * 24
>>      WEEK_TOLERANCE = DAY_TOLERANCE * 7
>>
>>    class BeCloseTo
>>      def initialize(target_date, tolerance)
>>        @target_date = Time.parse(target_date)
>>        @tolerance = tolerance
>>      end
>>
>>      def met_by?(target)
>>        compare_date = target - @target_date
>>        adjustment = @tolerance.inject do |item|
>>            case item
>>            when :days
>>                adjustment += DAY_TOLERANCE
>>            when :weeks
>>                adjustment += WEEK_TOLERANCE
>>            end
>>        end
>>        return true if compare_date.abs < adjustment
>>      end
>>
>>      def failure_message
>>        return "#{@target_date} does not fall within tolerance."  # s/
>> b better
>>      end
>>    end
>>
>>    def should_be_close_to(target_date, tolerance)
>>      return BeCloseTo.new(target_date, tolerance)
>>    end
>> end
>>
>> context "My package" do
>>    include DateExpectations
>>    specify "should be delivered no later than 2 days after my  
>> order" do
>>      1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <-  
>> line 40
>>    end
>> end
>>
>> The result is:
>>
>> 1)
>> NoMethodError in 'My package should be delivered no later than 2 days
>> after my order'
>> undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
>> ./spec/helpers/date_range_spec.rb:40:
>>
>>
>> Any thoughts as to why? (Note: You had written the expectation in
>> your sample as:
>> "should<space>be_close_to" -- was that intentional?).
>
> Yes!!!! That's the whole point. ;)
>
> should be_close_to
>
> and your method should be named be_close_to
>
> Let me know how it works out.
>
> Thanks for giving this a shot - you're the guinea pig!
>
> David
>
>>
>> Steve
>>
>>
>> On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:
>>
>>> module DateExpectations
>>>   class BeCloseTo
>>>     def initialize(target_date, tolerance)
>>>       @target_date = target_date
>>>       @tolerance = tolerance
>>>     end
>>>
>>>     def met_by?(target)
>>>       #return true if target is a date
>>>       #that meets the expectation
>>>     end
>>>
>>>     def failure_message
>>>       #return an appropriate failure message
>>>     end
>>>   end
>>>
>>>   def should_be_close_to(target_date, tolerance)
>>>     return BeCloseTo.new(target_date, tolerance)
>>>   end
>>> end
>>>
>>> context "My package" do
>>>   include DateExpectations
>>>   specify "should be delivered no later than 2 days after my  
>>> order" do
>>>     delivery_date.should be_close_to("1/9/2007", :days => 2)
>>>   end
>>> end
>>
>> _______________________________________________
>> 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


From aslak.hellesoy at gmail.com  Thu Jan 11 02:55:09 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Thu, 11 Jan 2007 08:55:09 +0100
Subject: [rspec-users] spec not running in TextMate
In-Reply-To: <647AD79E-646B-4075-8259-DE2A6FC54C6E@yamabe.net>
References: <FBE2CCBA-875D-4AC5-9A6E-234F2BB3E0E8@yamabe.net>
	<c0b8beca0701092247k641dcbb4j6583c7547e4855bb@mail.gmail.com>
	<8d961d900701100149k1bcd48d1q7de9791cd0bb9c72@mail.gmail.com>
	<DA7394F9-999B-489E-8732-545E21BDD093@yamabe.net>
	<8d961d900701100700j30c674e7pf216ad9e42039eb@mail.gmail.com>
	<647AD79E-646B-4075-8259-DE2A6FC54C6E@yamabe.net>
Message-ID: <8d961d900701102355w5f75bee9t102e89677ad54056@mail.gmail.com>

Please submit a bug report at Rubyforge for this. I'm currently
without Internet at home so it will take a while before I get to it.

On 1/10/07, Brian Yamabe <brian at yamabe.net> wrote:
> Tried it and it worked without a problem, so it looks like a bundle
> problem.
> I don't know much about bundles, but I'd be happy to work with someone
> who has suggestions for how to fix this.
>
> Thank, Aslak. You and David do a great job on this list.
>
> ---Brian
>
> On Jan 10, 2007, at 7:00 AM, aslak hellesoy wrote:
>
> > Let's narrow it down a little to see if this is a problem with the
> > bundle or your environment.
> >
> > Try this:
> >
> > 1) Create a file called foo.rb
> > 2) Add one line to it: require 'rubygems'
> > 3) Open it in TextMate
> > 4) Run it from TextMate (Cmd-R - Run)
> >
> > Do you get the same error? If so, your environment is bad.
> >
> > Aslak
> >
> > On 1/10/07, Brian Yamabe <brian at yamabe.net> wrote:
> >> I do have TM_RUBY set to the same Ruby interpreter as the one I use
> >> from the terminal.
> >>
> >> ---Brian
> >>
> >> On Jan 10, 2007, at 1:49 AM, aslak hellesoy wrote:
> >>
> >>> Sounds like TM is running a Ruby interpreter that doesn't have
> >>> Rubygems installed.
> >>>
> >>> Tell TM to run a different Ruby - it's documented somewhere, but I
> >>> don't remember where.
> >>>
> >>> Aslak
> >>>
> >>> On 1/10/07, Bryan Helmkamp <bhelmkamp at gmail.com> wrote:
> >>>> I don't have any help to offer, but I can confirm that I'm
> >>>> experiencing this issue right now, as well as some other people I
> >>>> have
> >>>> spoken with.
> >>>>
> >>>> -Bryan
> >>>>
> >>>> On 1/9/07, Brian Yamabe <brian at yamabe.net> wrote:
> >>>>> I tried running a spec with command-r in TextMate and got:
> >>>>>
> >>>>> /Library/Application Support/TextMate/Bundles/RSpec.tmbundle/
> >>>>> Support/
> >>>>> lib/spec_mate.rb:1:in `require': No such file to load -- rubygems
> >>>>> (LoadError) from /Library/Application Support/TextMate/Bundles/
> >>>>> RSpec.tmbundle/Support/lib/spec_mate.rb:1 from /tmp/
> >>>>> temp_textmate.GplhPr:3:in `require' from /tmp/
> >>>>> temp_textmate.GplhPr:3
> >>>>>
> >>>>> The spec runs fine from the command line.
> >>>>>
> >>>>> Any ideas?
> >>>>>
> >>>>> Thanks,
> >>>>> Brian
> >>>>> _______________________________________________
> >>>>> 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
>

From dchelimsky at gmail.com  Thu Jan 11 10:44:50 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 11 Jan 2007 09:44:50 -0600
Subject: [rspec-users] Date Approximation in Specs
In-Reply-To: <BFEBA93D-76A8-4D9D-BA4B-1BD3FA385984@gmail.com>
References: <9B13D352-1F62-4433-83BC-68A1682EB05A@gmail.com>
	<57c63afe0701091645i41c3f2dehaa3bacd4ebbb3b13@mail.gmail.com>
	<57c63afe0701092026y45b5f66fjeadb5b109b85448a@mail.gmail.com>
	<EBC7915C-3B7F-498F-BA8C-DFC333DB1F2C@gmail.com>
	<57c63afe0701101122t281a11ebr4b542a885081accf@mail.gmail.com>
	<BFEBA93D-76A8-4D9D-BA4B-1BD3FA385984@gmail.com>
Message-ID: <57c63afe0701110744o79892a9ck787ba999689b328d@mail.gmail.com>

On 1/11/07, s.ross <cwdinfo at gmail.com> wrote:
> My net connection's gone spotty over the last couple of days, but I
> wanted to follow up and tell you that your new code works great.
> Here's just a proof of concept implementation (complete with hard
> coding!):
>
> module DateExpectations
>      HOUR_TOLERANCE = 60 * 60
>      DAY_TOLERANCE = HOUR_TOLERANCE * 24
>      WEEK_TOLERANCE = DAY_TOLERANCE * 7
>
>    class BeCloseTo
>      def initialize(target_date, tolerance)
>        @target_date = Time.parse(target_date)
>        @tolerance = tolerance
>      end
>
>      def met_by?(target)
>        @compare_date = target
>        compare_span = target - @target_date
>        adjustment = 0
>        @tolerance.each do |key, amount|
>            case key
>            when :hours
>                adjustment += HOUR_TOLERANCE * amount
>            when :days
>                adjustment += DAY_TOLERANCE * amount
>            when :weeks
>                adjustment += WEEK_TOLERANCE * amount
>            end
>        end
>        return true if compare_span.abs < adjustment
>      end
>
>      def failure_message
>        return "#{@target_date} does not fall within tolerance (#
> {@tolerance.collect{|k,v| "#{k.to_s} => #{v}"}.join(", ")}) from #
> {@compare_date}."
>      end
>    end
>
>    def be_close_to(target_date, tolerance)
>      return BeCloseTo.new(target_date, tolerance)
>    end
> end
>
> context "The DateRange helper" do
>    include DateExpectations
>    specify "should be able to approximate within two days" do
>      1.day.ago.should be_close_to("1/9/2007", :days => 2)
>    end
> end

Great to hear it's working for you.

I just committed support for negative expectations as well. You just
need to supply a negative_failure_message in addition to
failure_message. Then  you can say (given your extension):

3.days.ago.should_not be_close_to("1/10/2007", :days => 2)

I am totally psyched about this. In addition to opening the door to
custom expectations it seems to be leading towards a leaner
implementation in rspec.

Cheers,
David

>
>
> On Jan 10, 2007, at 11:22 AM, David Chelimsky wrote:
>
> > On 1/10/07, s.ross <cwdinfo at gmail.com> wrote:
> >> I upgraded to trunk and changed the code as follows:
> >>
> >> module DateExpectations
> >>      HOUR_TOLERANCE = 60 * 60
> >>      DAY_TOLERANCE = HOUR_TOLERANCE * 24
> >>      WEEK_TOLERANCE = DAY_TOLERANCE * 7
> >>
> >>    class BeCloseTo
> >>      def initialize(target_date, tolerance)
> >>        @target_date = Time.parse(target_date)
> >>        @tolerance = tolerance
> >>      end
> >>
> >>      def met_by?(target)
> >>        compare_date = target - @target_date
> >>        adjustment = @tolerance.inject do |item|
> >>            case item
> >>            when :days
> >>                adjustment += DAY_TOLERANCE
> >>            when :weeks
> >>                adjustment += WEEK_TOLERANCE
> >>            end
> >>        end
> >>        return true if compare_date.abs < adjustment
> >>      end
> >>
> >>      def failure_message
> >>        return "#{@target_date} does not fall within tolerance."  # s/
> >> b better
> >>      end
> >>    end
> >>
> >>    def should_be_close_to(target_date, tolerance)
> >>      return BeCloseTo.new(target_date, tolerance)
> >>    end
> >> end
> >>
> >> context "My package" do
> >>    include DateExpectations
> >>    specify "should be delivered no later than 2 days after my
> >> order" do
> >>      1.day.ago.should_be_close_to("1/9/2007", :days => 2)   # <-
> >> line 40
> >>    end
> >> end
> >>
> >> The result is:
> >>
> >> 1)
> >> NoMethodError in 'My package should be delivered no later than 2 days
> >> after my order'
> >> undefined method `close_to?' for Tue Jan 09 10:59:31 -0800 2007:Time
> >> ./spec/helpers/date_range_spec.rb:40:
> >>
> >>
> >> Any thoughts as to why? (Note: You had written the expectation in
> >> your sample as:
> >> "should<space>be_close_to" -- was that intentional?).
> >
> > Yes!!!! That's the whole point. ;)
> >
> > should be_close_to
> >
> > and your method should be named be_close_to
> >
> > Let me know how it works out.
> >
> > Thanks for giving this a shot - you're the guinea pig!
> >
> > David
> >
> >>
> >> Steve
> >>
> >>
> >> On Jan 9, 2007, at 8:26 PM, David Chelimsky wrote:
> >>
> >>> module DateExpectations
> >>>   class BeCloseTo
> >>>     def initialize(target_date, tolerance)
> >>>       @target_date = target_date
> >>>       @tolerance = tolerance
> >>>     end
> >>>
> >>>     def met_by?(target)
> >>>       #return true if target is a date
> >>>       #that meets the expectation
> >>>     end
> >>>
> >>>     def failure_message
> >>>       #return an appropriate failure message
> >>>     end
> >>>   end
> >>>
> >>>   def should_be_close_to(target_date, tolerance)
> >>>     return BeCloseTo.new(target_date, tolerance)
> >>>   end
> >>> end
> >>>
> >>> context "My package" do
> >>>   include DateExpectations
> >>>   specify "should be delivered no later than 2 days after my
> >>> order" do
> >>>     delivery_date.should be_close_to("1/9/2007", :days => 2)
> >>>   end
> >>> end
> >>
> >> _______________________________________________
> >> 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
>

From ed.howland at gmail.com  Thu Jan 11 19:32:34 2007
From: ed.howland at gmail.com (Ed Howland)
Date: Thu, 11 Jan 2007 18:32:34 -0600
Subject: [rspec-users] After upgrade to 0.7.5,
	specs are now using the development, not the test DB
Message-ID: <3df642dd0701111632p1ef98f0exaf9d4e94192722f0@mail.gmail.com>

Hi,

After I upgraded RSpec from 0.7.4 to 0.7.5, first the gem, then the
Rails plugin, I noticed that when I ran my spec/models, the fixtures
were overwriting the data in the development environment DB, not the
test one. The test environment was being used prior to the upgrade.

I am using SQLite3 and an in-memory db for the test environment.

Any help would be appreciated.

Thanks
Ed
-- 
Ed Howland
http://greenprogrammer.blogspot.com

From listaccount at e-tobi.net  Thu Jan 11 19:46:15 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Fri, 12 Jan 2007 01:46:15 +0100
Subject: [rspec-users] First steps with RSpec on Rails
Message-ID: <45A6DA57.9060704@e-tobi.net>

Hi!

If anyone is interested, these are my first tiny steps with RSpec on Rails:

https://www.heise.de:444/svn/ctvdr/ctvdrwebadmin/trunk/

The "ctvdrwebadmin" application should become a web administration tool
for a Debian based distribution that is specialized to VDR [1] - the
Linux video disc recoder by Klaus Schmidinger.  It shall provide an
interface for the installation / uninstallation of software packages,
configuration handling and several status information. It doesn't use a
database - instead it reads data from the file system or grabs it from
the OS and other external tools. For now this is still in the "proof of
concept" phase, giving me the opportunity to play around with Rails and
RSpec.

"ctvdrwebadmin" is tightly coupled to Debian and has a German-only UI,
so most of you probably won't be able to run it. But the interesting
part is the source code, especially the specifications - all English of
course.

Any suggestions for improvements are welcome! I still have a problem
with finding the "right" phrases for specifications, so especially on
this topic any feedback is welcome.

The most interesting part of the application in it's current state
probably is the channel editor. One of the use cases of the channel
editor is, that it should be possible to rearrange the order of
channels. With plain old http get/post this quickly becomes a usability
nightmare, so I decided to throw in some Ajax/JavaScript to implement a
nice drag 'n drop UI for rearranging the list of channels. With
sortable_element() and Rails this was much easier than I initially
thought. It just became a little bit more tricky, when I realized, that
with 1000 channels the Javascripts became damn slow. And several hundred
channels (TV and radio) are not unusual when you use DVB-S. So I needed
to paginate the list of channels. This required a little bit more Ajax.
And because I now needed a way to move a channel from the last page to
the first, I decided to introduce a clipboard where the user can
temporarily store channels when switching between pages.

So finally I had this: Two lists where the user can drag 'n drop
channels within and between the lists. The clipboard list is static, the
channel list is paginated. Each time the user moves a channel, an Ajax
request triggers the rearrangement in the underlying controller.

And these are the specifications for this controller:

https://www.heise.de:444/svn/ctvdr/ctvdrwebadmin/branches/snapshot_for_rspec_ml/spec/controllers/channel_list_spec.rb

Some of the things that still puzzle me there:

spec "should result in an empty clipboard when pulling the last channel
to the order list" - when a channel is moved from the clipboard to the
list, two Ajax requests are triggered. One that passes the new list of
items in the clipboard and one with the items in the channel list. For
this specification I'm only interested in the request from the
clipboard. The question is, if I then can skip the request from the
channel list? In a real use of the application both events always occur
together, when dragging between the two lists. Or should I just rename
the specification to  "should result in an empty clipboard when removing
the last channel from it" and skip the request from the channel list?
(But what if I had a bug in the code that handles the requests from the
channel list?)

What about the contexts? Should I split them into more contexts? Or
should I merge some? The first two contexts only have one specification.
They could easily be merged, but the setup of the second context is a
little bit more complex which is not needed to perform the specification
in the first context.

So far I think that doing it the RSpec-way gave me a little bit more
freedom to evolve the implementation and the overall design than doing
it with "plain old" unit tests. But on the other hand I have the
feeling, that the specifications don't test the code as deeply as unit
test would have done. But that's only a vague feeling at the moment.

bye,

Tobias

PS: David - I followed your suggestion and moved the "logged-in"-code to
a set_logged_in method, which I simply put into spec_helper.rb, so I
don't need to explicitly include a helper module. Thanks!

[1]: http://www.cadsoft.de/vdr/


From dchelimsky at gmail.com  Thu Jan 11 22:57:53 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 11 Jan 2007 21:57:53 -0600
Subject: [rspec-users] After upgrade to 0.7.5,
	specs are now using the development, not the test DB
In-Reply-To: <3df642dd0701111632p1ef98f0exaf9d4e94192722f0@mail.gmail.com>
References: <3df642dd0701111632p1ef98f0exaf9d4e94192722f0@mail.gmail.com>
Message-ID: <57c63afe0701111957w7e53e8e5w441e0d45ec529799@mail.gmail.com>

Add this to the top of spec_helper.rb:

ENV["RAILS_ENV"] = "test"

This will not be a problem in the next release.

David

On 1/11/07, Ed Howland <ed.howland at gmail.com> wrote:
> Hi,
>
> After I upgraded RSpec from 0.7.4 to 0.7.5, first the gem, then the
> Rails plugin, I noticed that when I ran my spec/models, the fixtures
> were overwriting the data in the development environment DB, not the
> test one. The test environment was being used prior to the upgrade.
>
> I am using SQLite3 and an in-memory db for the test environment.
>
> Any help would be appreciated.
>
> Thanks
> Ed
> --
> Ed Howland
> http://greenprogrammer.blogspot.com
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From ed.howland at gmail.com  Fri Jan 12 09:23:03 2007
From: ed.howland at gmail.com (Ed Howland)
Date: Fri, 12 Jan 2007 08:23:03 -0600
Subject: [rspec-users] After upgrade to 0.7.5,
	specs are now using the development, not the test DB
In-Reply-To: <57c63afe0701111957w7e53e8e5w441e0d45ec529799@mail.gmail.com>
References: <3df642dd0701111632p1ef98f0exaf9d4e94192722f0@mail.gmail.com>
	<57c63afe0701111957w7e53e8e5w441e0d45ec529799@mail.gmail.com>
Message-ID: <3df642dd0701120623s246def36n24c49a0a61f543fa@mail.gmail.com>

On 1/11/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> Add this to the top of spec_helper.rb:
>
> ENV["RAILS_ENV"] = "test"
>
> This will not be a problem in the next release.
>
> David

Thanks, that did the trick.

Ed

From brian at yamabe.net  Fri Jan 12 11:20:23 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Fri, 12 Jan 2007 08:20:23 -0800
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
Message-ID: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>

I'm evaluating what to use for integration testing:

SafariWatir
- integrates nicely with RSpec
- doesn't work with Ajax because of Safari event firing issues

Selenium
- Works with Ajax
- doesn't integrate with RSpec

What is the best guess timeframe for RSpec integration testing?
Will it be able to test Ajax application in the early iterations?

Thanks,
Brian Yamabe

From shane.duan at gmail.com  Fri Jan 12 11:34:26 2007
From: shane.duan at gmail.com (Shane Duan)
Date: Fri, 12 Jan 2007 08:34:26 -0800
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
In-Reply-To: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
References: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
Message-ID: <bd7896da0701120834u7e61066cp460d2c5992e70265@mail.gmail.com>

Hi,

What did you mean by Selenium integration with rspec?

You can install the selenium gem.  Then you can control the start/stop
of selenium server and use the driver to drive the browser,

http://selenium.rubyforge.org

Cheers
Shane

On 1/12/07, Brian Yamabe <brian at yamabe.net> wrote:
> I'm evaluating what to use for integration testing:
>
> SafariWatir
> - integrates nicely with RSpec
> - doesn't work with Ajax because of Safari event firing issues
>
> Selenium
> - Works with Ajax
> - doesn't integrate with RSpec
>
> What is the best guess timeframe for RSpec integration testing?
> Will it be able to test Ajax application in the early iterations?
>
> Thanks,
> Brian Yamabe
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>


-- 
Shane
http://www.shaneduan.com

From aslak.hellesoy at gmail.com  Fri Jan 12 11:38:10 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Fri, 12 Jan 2007 17:38:10 +0100
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
In-Reply-To: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
References: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
Message-ID: <8d961d900701120838s49b6bc18vec3a4bd095c48208@mail.gmail.com>

Rspec and selemnium (via selenium-rc) do indeed play nicely together.
See the spec-ui directory in svn trunk.

On 1/12/07, Brian Yamabe <brian at yamabe.net> wrote:
> I'm evaluating what to use for integration testing:
>
> SafariWatir
> - integrates nicely with RSpec
> - doesn't work with Ajax because of Safari event firing issues
>
> Selenium
> - Works with Ajax
> - doesn't integrate with RSpec
>
> What is the best guess timeframe for RSpec integration testing?
> Will it be able to test Ajax application in the early iterations?
>
> Thanks,
> Brian Yamabe
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From brian at yamabe.net  Fri Jan 12 12:05:42 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Fri, 12 Jan 2007 09:05:42 -0800
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
In-Reply-To: <bd7896da0701120834u7e61066cp460d2c5992e70265@mail.gmail.com>
References: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
	<bd7896da0701120834u7e61066cp460d2c5992e70265@mail.gmail.com>
Message-ID: <01C57737-1697-4297-9D52-1C43548C7308@yamabe.net>

Thanks for both the replies.  I saw Selenium on Rails, but missed the  
Selenium gem.

---Brian

On Jan 12, 2007, at 8:34 AM, Shane Duan wrote:

> Hi,
>
> What did you mean by Selenium integration with rspec?
>
> You can install the selenium gem.  Then you can control the start/stop
> of selenium server and use the driver to drive the browser,
>
> http://selenium.rubyforge.org
>
> Cheers
> Shane
>
> On 1/12/07, Brian Yamabe <brian at yamabe.net> wrote:
>> I'm evaluating what to use for integration testing:
>>
>> SafariWatir
>> - integrates nicely with RSpec
>> - doesn't work with Ajax because of Safari event firing issues
>>
>> Selenium
>> - Works with Ajax
>> - doesn't integrate with RSpec
>>
>> What is the best guess timeframe for RSpec integration testing?
>> Will it be able to test Ajax application in the early iterations?
>>
>> Thanks,
>> Brian Yamabe
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
>
>
> -- 
> Shane
> http://www.shaneduan.com
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users


From shane.duan at gmail.com  Fri Jan 12 14:16:16 2007
From: shane.duan at gmail.com (Shane Duan)
Date: Fri, 12 Jan 2007 11:16:16 -0800
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
In-Reply-To: <01C57737-1697-4297-9D52-1C43548C7308@yamabe.net>
References: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
	<bd7896da0701120834u7e61066cp460d2c5992e70265@mail.gmail.com>
	<01C57737-1697-4297-9D52-1C43548C7308@yamabe.net>
Message-ID: <bd7896da0701121116l510af351vf9470269714a65f1@mail.gmail.com>

Selenium on rails uses the original selenium HTML driver and provides
API for generating the HTML

Selenium Ruby bundles the selenium server and the Selenium ruby RC as
well as providing additional API for domain based web testing
(http://selenium.rubyforge.org/doc/domain-based-web-testing.html)

I have contacted them to see if two projects can work together in some way.

Cheers
Shane

On 1/12/07, Brian Yamabe <brian at yamabe.net> wrote:
> Thanks for both the replies.  I saw Selenium on Rails, but missed the
> Selenium gem.
>
> ---Brian
>
> On Jan 12, 2007, at 8:34 AM, Shane Duan wrote:
>
> > Hi,
> >
> > What did you mean by Selenium integration with rspec?
> >
> > You can install the selenium gem.  Then you can control the start/stop
> > of selenium server and use the driver to drive the browser,
> >
> > http://selenium.rubyforge.org
> >
> > Cheers
> > Shane
> >
> > On 1/12/07, Brian Yamabe <brian at yamabe.net> wrote:
> >> I'm evaluating what to use for integration testing:
> >>
> >> SafariWatir
> >> - integrates nicely with RSpec
> >> - doesn't work with Ajax because of Safari event firing issues
> >>
> >> Selenium
> >> - Works with Ajax
> >> - doesn't integrate with RSpec
> >>
> >> What is the best guess timeframe for RSpec integration testing?
> >> Will it be able to test Ajax application in the early iterations?
> >>
> >> Thanks,
> >> Brian Yamabe
> >> _______________________________________________
> >> rspec-users mailing list
> >> rspec-users at rubyforge.org
> >> http://rubyforge.org/mailman/listinfo/rspec-users
> >>
> >
> >
> > --
> > Shane
> > http://www.shaneduan.com
> > _______________________________________________
> > 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
>


-- 
Shane
http://www.shaneduan.com

From work at ashleymoran.me.uk  Fri Jan 12 14:36:34 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Fri, 12 Jan 2007 19:36:34 +0000
Subject: [rspec-users] Integration Testing - Wait, Watir, or Selenium
In-Reply-To: <bd7896da0701121116l510af351vf9470269714a65f1@mail.gmail.com>
References: <E9396A4C-B66E-45A4-816A-09098C04327B@yamabe.net>
	<bd7896da0701120834u7e61066cp460d2c5992e70265@mail.gmail.com>
	<01C57737-1697-4297-9D52-1C43548C7308@yamabe.net>
	<bd7896da0701121116l510af351vf9470269714a65f1@mail.gmail.com>
Message-ID: <AA7CCEBA-FA69-452E-BDB4-F3059C3E09F1@ashleymoran.me.uk>


On Jan 12, 2007, at 7:16 pm, Shane Duan wrote:

>>>> SafariWatir
>>>> - integrates nicely with RSpec
>>>> - doesn't work with Ajax because of Safari event firing issues

I e-mailed the author about this and he fixed it over Christmas.  Try  
version 0.2.1, it works for me.

Ashley	

From brian at yamabe.net  Fri Jan 12 18:00:14 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Fri, 12 Jan 2007 15:00:14 -0800
Subject: [rspec-users] spec_ui problems
Message-ID: <ADAB7E40-D308-4AF5-B220-11FDDD052D93@yamabe.net>

While looking into spec_ui, I decided to run the examples.

The watir example works a little, but always chokes on the 'better  
than fudge' spec (failure output below).  Also, is there any command  
to pause the "browser"?  If my connection slows, the test gets out of  
whack.

The selenium example fails right away (output below).  I do have the  
0.9.0 selenium-rc server running.

I executed rake from inside each of the specific example directories.

Thanks for any help.

---Brian Yamabe

watir failure
----------
/opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/expectations/ 
sugar.rb:13:in `call': undefined local variable or method  
`current_spec_number' for #<Spec::Ui::WebappFormatter:0x10083d0>  
(NameError)
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
expectations/sugar.rb:13:in `_method_missing'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
expectations/sugar.rb:9:in `method_missing'
         from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb: 
5:in `extra_failure_content'
         from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb: 
11:in `extra_failure_content'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/formatter/html_formatter.rb:57:in `spec_failed'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/reporter.rb:78:in `spec_failed'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/reporter.rb:27:in `spec_finished'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/specification.rb:37:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context.rb:57:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context.rb:54:in `each'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context.rb:54:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context_runner.rb:23:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context_runner.rb:22:in `each'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/context_runner.rb:22:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:26:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4
rake aborted!

selenium failure
----------
/opt/local/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in  
`gem_original_require': no such file to load -- ./spec/selenium  
(LoadError)
         from /opt/local/lib/ruby/site_ruby/1.8/rubygems/ 
custom_require.rb:27:in `require'
         from ./spec/spec_helper.rb:7
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/option_parser.rb:107:in `require'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/option_parser.rb:107:in `parse'
         from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `call'
         from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `order!'
         from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `catch'
         from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `order!'
         from /opt/local/lib/ruby/1.8/optparse.rb:1279:in `permute!'
         from /opt/local/lib/ruby/1.8/optparse.rb:1300:in `parse!'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/option_parser.rb:141:in `parse'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/option_parser.rb:13:in `create_context_runner'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/ 
runner/command_line.rb:12:in `run'
         from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4

From cwdinfo at gmail.com  Sat Jan 13 14:18:38 2007
From: cwdinfo at gmail.com (s.ross)
Date: Sat, 13 Jan 2007 11:18:38 -0800
Subject: [rspec-users] svn reorganized?
Message-ID: <A7D88DAB-7C1E-45BF-B247-4AC03A17CF54@gmail.com>

It appears the repository has been rearranged when compared to the  
documentation. For the Rails plugin, the correct script/plugin  
install now seems to be:

script/plugin install svn://rubyforge.org/var/svn/rspec/trunk/ 
rspec_on_rails

Is this correct?

From dchelimsky at gmail.com  Sat Jan 13 14:51:54 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sat, 13 Jan 2007 13:51:54 -0600
Subject: [rspec-users] svn reorganized?
In-Reply-To: <A7D88DAB-7C1E-45BF-B247-4AC03A17CF54@gmail.com>
References: <A7D88DAB-7C1E-45BF-B247-4AC03A17CF54@gmail.com>
Message-ID: <57c63afe0701131151h66016f6cpd14024e1fb1a16b3@mail.gmail.com>

Where are you seeing the wrong doc? I'm looking at this:

http://rspec.rubyforge.org/documentation/rails/install.html

and I see the correct path for tagged versions (0.7.5 and later)

svn://rubyforge.org/var/svn/rspec/tags/REL_X_Y_Z/rspec_on_rails/vendor/plugins/rspec_on_rails

and the current trunk:

svn://rubyforge.org/var/svn/rspec/trunk/rspec_on_rails/vendor/plugins/rspec_on_rails

Is there another page that you are seeing the wrong information?

Thanks
David

On 1/13/07, s.ross <cwdinfo at gmail.com> wrote:
> It appears the repository has been rearranged when compared to the
> documentation. For the Rails plugin, the correct script/plugin
> install now seems to be:
>
> script/plugin install svn://rubyforge.org/var/svn/rspec/trunk/
> rspec_on_rails
>
> Is this correct?
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From cwdinfo at gmail.com  Sat Jan 13 16:03:42 2007
From: cwdinfo at gmail.com (s.ross)
Date: Sat, 13 Jan 2007 13:03:42 -0800
Subject: [rspec-users] svn reorganized?
In-Reply-To: <57c63afe0701131151h66016f6cpd14024e1fb1a16b3@mail.gmail.com>
References: <A7D88DAB-7C1E-45BF-B247-4AC03A17CF54@gmail.com>
	<57c63afe0701131151h66016f6cpd14024e1fb1a16b3@mail.gmail.com>
Message-ID: <FFE75306-53A1-49BB-94D7-2CE9DEF50EB1@gmail.com>

When installing from trunk, the location (apparently) is:

svn://rubyforge.org/var/svn/rspec/trunk/rspec_on_rails

This page (http://rspec.rubyforge.org/tools/rails.html) mentions  
trunk as being:

script/plugin install svn://rubyforge.org/var/svn/rspec/trunk/vendor/ 
rspec_on_rails/vendor/plugins/rspec

It is tagged as being for rSpec 0.6.3

This page (http://rspec.rubyforge.org/documentation/rails/index.html)  
mentions trunk as being:

script/plugin install svn://rubyforge.org/var/svn/rspec/trunk/ 
rspec_on_rails/vendor/plugins/rspec_on_rails

The point is that the former is wrong with respect to trunk and (IMO)  
should be revised to point to the 0.7.5 and later trunk installation.

Steve



On Jan 13, 2007, at 11:51 AM, David Chelimsky wrote:

> Where are you seeing the wrong doc? I'm looking at this:
>
> http://rspec.rubyforge.org/documentation/rails/install.html
>
> and I see the correct path for tagged versions (0.7.5 and later)
>
> svn://rubyforge.org/var/svn/rspec/tags/REL_X_Y_Z/rspec_on_rails/ 
> vendor/plugins/rspec_on_rails
>
> and the current trunk:
>
> svn://rubyforge.org/var/svn/rspec/trunk/rspec_on_rails/vendor/ 
> plugins/rspec_on_rails
>
> Is there another page that you are seeing the wrong information?
>
> Thanks
> David
>
> On 1/13/07, s.ross <cwdinfo at gmail.com> wrote:
>> It appears the repository has been rearranged when compared to the
>> documentation. For the Rails plugin, the correct script/plugin
>> install now seems to be:
>>
>> script/plugin install svn://rubyforge.org/var/svn/rspec/trunk/
>> rspec_on_rails
>>
>> Is this correct?
>> _______________________________________________
>> 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


From richard at infoarts.info  Sat Jan 13 21:51:09 2007
From: richard at infoarts.info (Richard Sandilands)
Date: Sun, 14 Jan 2007 13:51:09 +1100
Subject: [rspec-users] rspec and set_table_name?
Message-ID: <3961F8F2-D53C-4F3F-A116-49D4CB8AAA09@infoarts.info>

Hi there

I have a Value class in my Rails app; however 'values' is a reserved  
word in MySQL so in my value.rb file, I've declared set_table_name as  
'vals'.

However, in setting up my specs for this model, I'm hitting errors  
that seem to be resulting from rspec not seeing the set_table_name  
declaration in my model file.

For instance, rspec is throwing a MySQL error as it is attempting to  
delete records from a table named 'values' which of course does not  
exist - the table is 'vals' but the model is Value.

Is this an issue with rspec? If so, is there a way around this? I  
want to keep using rspec and am prepared to rename my class so as to  
not need the set_table_name declaration but want to be clear that  
this is the source of the problem.

Any clues would be appreciated.

Richard



From matt at iseekgolf.com  Sun Jan 14 02:40:49 2007
From: matt at iseekgolf.com (Matt Allen)
Date: Sun, 14 Jan 2007 18:40:49 +1100
Subject: [rspec-users] rspec and set_table_name?
In-Reply-To: <3961F8F2-D53C-4F3F-A116-49D4CB8AAA09@infoarts.info>
References: <3961F8F2-D53C-4F3F-A116-49D4CB8AAA09@infoarts.info>
Message-ID: <45A9DE81.2010508@iseekgolf.com>

Hi Richard;

I use set_table_name too.

Here is my set-up.

product.rb
class Product < ActiveRecord::Base
 set_table_name "cart_products"
end

product_spec.rb
context "A product with fixtures loaded" do
 set_fixture_class :cart_products => 'Product'
 fixtures :cart_products
end

Your fixtures should be the same name as you table, so in my case 
cart_products.

HTH,
Matta

Richard Sandilands did write the following on 14/01/2007 1:51 PM:
> Hi there
>
> I have a Value class in my Rails app; however 'values' is a reserved  
> word in MySQL so in my value.rb file, I've declared set_table_name as  
> 'vals'.
>
> However, in setting up my specs for this model, I'm hitting errors  
> that seem to be resulting from rspec not seeing the set_table_name  
> declaration in my model file.
>
> For instance, rspec is throwing a MySQL error as it is attempting to  
> delete records from a table named 'values' which of course does not  
> exist - the table is 'vals' but the model is Value.
>
> Is this an issue with rspec? If so, is there a way around this? I  
> want to keep using rspec and am prepared to rename my class so as to  
> not need the set_table_name declaration but want to be clear that  
> this is the source of the problem.
>
> Any clues would be appreciated.
>
> Richard
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>
> .
>

-- 
Matt Allen
iseekgolf.com
1300 663 813

From richard at infoarts.info  Sun Jan 14 15:27:37 2007
From: richard at infoarts.info (Richard Sandilands)
Date: Mon, 15 Jan 2007 07:27:37 +1100
Subject: [rspec-users] rspec and set_table_name?
In-Reply-To: <45A9DE81.2010508@iseekgolf.com>
References: <3961F8F2-D53C-4F3F-A116-49D4CB8AAA09@infoarts.info>
	<45A9DE81.2010508@iseekgolf.com>
Message-ID: <0C4AD1BF-75C2-4204-B39E-30B3017C2049@infoarts.info>


On 14/01/2007, at 6:40 PM, Matt Allen wrote:

> set_fixture_class :cart_products => 'Product'

Ah, that's the missing piece there. It's all working nicely now -  
thanks for your help.

-- 
Richard



From eastmedianyc at gmail.com  Mon Jan 15 00:04:56 2007
From: eastmedianyc at gmail.com (Matt Pelletier)
Date: Mon, 15 Jan 2007 00:04:56 -0500
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
Message-ID: <96af85420701142104u640e0ceav875ae4fabd0bd4b3@mail.gmail.com>

> Le 5 janv. 07 ? 03:05, Bryan Liles a ?crit :
>
> > I seem to be missing something because I am at a loss on how to get
> > rcov, RSpec, and Rails working together.  Would someone mind dropping
> > some hints?
>
> I don't know what you want to achieve, but if your goal is to
> generate some reports here's what I added to lib/tasks/
> rspec_additions.rake :
>
> namespace :spec do
>    desc "Spec with rspec/rcov reports"
>    Spec::Rake::SpecTask.new('reports') do |t|
>      t.spec_files = FileList['spec/**/*.rb']
>      t.spec_opts = ["--format", "html", "--diff"]
>      t.out = 'doc/rspec.html'
>      t.rcov = true
>      t.rcov_dir = 'doc/coverage'
>
> # Optional :
> # Prevent rcov to generate report on specs, environment.rb and boot.rb
> #    t.rcov_opts = ['--exclude', 'spec,config\/environment.rb,config\/
> boot.rb']
>
>      t.fail_on_error = false
>      t.failure_message = "The specs failed. Check for the problem at
> http://projects.tron.name/flash007/rspec.html"
>    end
> end
>
> Run it with rake : spec:reports
> It will generate rspec.html report in doc/rspec.html and rcov report
> in doc/coverage/.

This results in a bus error in OSX 10.4.8 (I had a similar task that
did the same thing):

<snip path>/config/../vendor/rails/activerecord/lib/active_record/associations/has_many_association.rb:13:
[BUG] Bus Error
ruby 1.8.5 (2006-08-25) [i686-darwin8.8.1]

I'm running Rails 1.2 in vendor/rails

Anyone else getting this? Any idea why? Thanks.

Matt

>
> Jonathan
> --
> Tron Jonathan
> http://jonathan.tron.name
>

From jchris at mfdz.com  Mon Jan 15 04:29:50 2007
From: jchris at mfdz.com (Chris Anderson)
Date: Mon, 15 Jan 2007 01:29:50 -0800
Subject: [rspec-users] heckle and rspec on rails
Message-ID: <e282921e0701150129m924b4e3rbcc30589bf7bfd4c@mail.gmail.com>

Now that I'm developing a big suite of specs, I really want to run
Heckle and see how well I've done.
Using r1359, when I run:

spec spec/models/metadata_report_spec.rb --heckle MetadataReport

the result is just to run the specs once and then exit, as though I
hadn't mentioned Heckle. I started looking though rspec to figure out
why, but it's late, so I should sleep instead.

I guess the question is, has anyone had luck using Heckle with Rails?
I've got it working for regular ruby specs, and it's a lot of fun
(although I do hit infinite loops sometimes).

Thanks!
-- 
Chris Anderson
http://jchris.mfdz.com

From cwdinfo at gmail.com  Mon Jan 15 18:55:28 2007
From: cwdinfo at gmail.com (s.ross)
Date: Mon, 15 Jan 2007 15:55:28 -0800
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
In-Reply-To: <96af85420701142104u640e0ceav875ae4fabd0bd4b3@mail.gmail.com>
References: <96af85420701142104u640e0ceav875ae4fabd0bd4b3@mail.gmail.com>
Message-ID: <3FCA8364-BDAA-4C26-9EF4-F38F7A5A004E@gmail.com>

I'm running the same system configuration as you, except my Rails  
install is 5623. The bus error does not occur on my system when  
running this task. Note: I'm running trunk rSpec. What version of  
rSpec did you say you were running?

--steve


On Jan 14, 2007, at 9:04 PM, Matt Pelletier wrote:

>> Le 5 janv. 07 ? 03:05, Bryan Liles a ?crit :
>>
>>> I seem to be missing something because I am at a loss on how to get
>>> rcov, RSpec, and Rails working together.  Would someone mind  
>>> dropping
>>> some hints?
>>
>> I don't know what you want to achieve, but if your goal is to
>> generate some reports here's what I added to lib/tasks/
>> rspec_additions.rake :
>>
>> namespace :spec do
>>    desc "Spec with rspec/rcov reports"
>>    Spec::Rake::SpecTask.new('reports') do |t|
>>      t.spec_files = FileList['spec/**/*.rb']
>>      t.spec_opts = ["--format", "html", "--diff"]
>>      t.out = 'doc/rspec.html'
>>      t.rcov = true
>>      t.rcov_dir = 'doc/coverage'
>>
>> # Optional :
>> # Prevent rcov to generate report on specs, environment.rb and  
>> boot.rb
>> #    t.rcov_opts = ['--exclude', 'spec,config\/ 
>> environment.rb,config\/
>> boot.rb']
>>
>>      t.fail_on_error = false
>>      t.failure_message = "The specs failed. Check for the problem at
>> http://projects.tron.name/flash007/rspec.html"
>>    end
>> end
>>
>> Run it with rake : spec:reports
>> It will generate rspec.html report in doc/rspec.html and rcov report
>> in doc/coverage/.
>
> This results in a bus error in OSX 10.4.8 (I had a similar task that
> did the same thing):
>
> <snip path>/config/../vendor/rails/activerecord/lib/active_record/ 
> associations/has_many_association.rb:13:
> [BUG] Bus Error
> ruby 1.8.5 (2006-08-25) [i686-darwin8.8.1]
>
> I'm running Rails 1.2 in vendor/rails
>
> Anyone else getting this? Any idea why? Thanks.
>
> Matt
>
>>
>> Jonathan
>> --
>> Tron Jonathan
>> http://jonathan.tron.name
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users


From jonathan.tron.mailings at gmail.com  Tue Jan 16 03:35:42 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Tue, 16 Jan 2007 09:35:42 +0100
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
In-Reply-To: <3FCA8364-BDAA-4C26-9EF4-F38F7A5A004E@gmail.com>
References: <96af85420701142104u640e0ceav875ae4fabd0bd4b3@mail.gmail.com>
	<3FCA8364-BDAA-4C26-9EF4-F38F7A5A004E@gmail.com>
Message-ID: <C0F836EB-F4D3-4FFF-B7C4-D3601A3EA8A5@gmail.com>

Le 16 janv. 07 ? 00:55, s.ross a ?crit :

> I'm running the same system configuration as you, except my Rails
> install is 5623. The bus error does not occur on my system when
> running this task. Note: I'm running trunk rSpec. What version of
> rSpec did you say you were running?

As I said on a different thread, I had the problem of "Bus Error"  
too, but only when I'm using mock_model from (http://metaclass.org/ 
2006/12/22/making-a-mockery-of-activerecord). I run specs on two  
different config and only the local one segfault.

My local config is : MacOS X.4.8 / Ruby 1.8.5 (from MacPorts) / Rails  
edge 1.2-branch / RSpec 0.7.5 (gem) / RspecOnRails 0.7.5 (from  
trunk at 1331) / rcov 0.7.0.1 (gem)
My remote config is : Debian Testing / Ruby 1.8.5-4 / Rails edge 1.2- 
branch / RSpec 0.7.5 (gem) / RspecOnRails 0.7.5 (from trunk at 1331) /  
rcov 0.7.0.1 (gem)


Jonathan
--
Tron Jonathan
http://jonathan.tron.name




From aslak.hellesoy at gmail.com  Tue Jan 16 04:41:23 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Tue, 16 Jan 2007 10:41:23 +0100
Subject: [rspec-users] RSpec 0.7.5 with Rails and rcov
In-Reply-To: <C0F836EB-F4D3-4FFF-B7C4-D3601A3EA8A5@gmail.com>
References: <96af85420701142104u640e0ceav875ae4fabd0bd4b3@mail.gmail.com>
	<3FCA8364-BDAA-4C26-9EF4-F38F7A5A004E@gmail.com>
	<C0F836EB-F4D3-4FFF-B7C4-D3601A3EA8A5@gmail.com>
Message-ID: <8d961d900701160141v6747f6bfq9ddba4acaad4fd04@mail.gmail.com>

Is anyone telling Mauricio about this? We're not going to fix it - bus
error is a ruby/rcov problem

Aslak

On 1/16/07, Jonathan Tron <jonathan.tron.mailings at gmail.com> wrote:
> Le 16 janv. 07 ? 00:55, s.ross a ?crit :
>
> > I'm running the same system configuration as you, except my Rails
> > install is 5623. The bus error does not occur on my system when
> > running this task. Note: I'm running trunk rSpec. What version of
> > rSpec did you say you were running?
>
> As I said on a different thread, I had the problem of "Bus Error"
> too, but only when I'm using mock_model from (http://metaclass.org/
> 2006/12/22/making-a-mockery-of-activerecord). I run specs on two
> different config and only the local one segfault.
>
> My local config is : MacOS X.4.8 / Ruby 1.8.5 (from MacPorts) / Rails
> edge 1.2-branch / RSpec 0.7.5 (gem) / RspecOnRails 0.7.5 (from
> trunk at 1331) / rcov 0.7.0.1 (gem)
> My remote config is : Debian Testing / Ruby 1.8.5-4 / Rails edge 1.2-
> branch / RSpec 0.7.5 (gem) / RspecOnRails 0.7.5 (from trunk at 1331) /
> rcov 0.7.0.1 (gem)
>
>
> Jonathan
> --
> Tron Jonathan
> http://jonathan.tron.name
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From work at ashleymoran.me.uk  Tue Jan 16 08:02:43 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Tue, 16 Jan 2007 13:02:43 +0000
Subject: [rspec-users] Links in view specs
Message-ID: <79EAA51F-6DEF-49F5-85BF-F62DA9BD4883@ashleymoran.me.uk>

Hi

I have this simple view (used by the Gap controller):

   <h1>Gap#index</h1>
   <%= link_to "Get a GAP quote", :action => "get_quote" %>

And this spec:

   context "A rendered gap/index" do
     setup do
       render 'gap/index'
     end

     specify "should have a link to the get_quote page" do
       response.should_have_tag 'a', :attributes => { :href => /\/gap 
\/get_quote/ }
     end
   end

But is fails because it actually links like this:
    <a href="/view_spec/get_quote">
not
   <a href="/gap/get_quote">

What's the cleanest way to test links without integrating it into the  
controller?  For now I'm doing...
   response.should_have_tag 'a', :attributes => { :href => /\/ 
get_quote$/ }

  Sorry if this is a FAQ, can't see any posts on the list about it.

Ta
Ashley

From dchelimsky at gmail.com  Tue Jan 16 08:21:51 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 16 Jan 2007 07:21:51 -0600
Subject: [rspec-users] Links in view specs
In-Reply-To: <79EAA51F-6DEF-49F5-85BF-F62DA9BD4883@ashleymoran.me.uk>
References: <79EAA51F-6DEF-49F5-85BF-F62DA9BD4883@ashleymoran.me.uk>
Message-ID: <57c63afe0701160521j4c2ac5den77870f6948050f78@mail.gmail.com>

On 1/16/07, Ashley Moran <work at ashleymoran.me.uk> wrote:
> Hi
>
> I have this simple view (used by the Gap controller):
>
>    <h1>Gap#index</h1>
>    <%= link_to "Get a GAP quote", :action => "get_quote" %>
>
> And this spec:
>
>    context "A rendered gap/index" do
>      setup do
>        render 'gap/index'
>      end
>
>      specify "should have a link to the get_quote page" do
>        response.should_have_tag 'a', :attributes => { :href => /\/gap
> \/get_quote/ }
>      end
>    end
>
> But is fails because it actually links like this:
>     <a href="/view_spec/get_quote">
> not
>    <a href="/gap/get_quote">
>
> What's the cleanest way to test links without integrating it into the
> controller?  For now I'm doing...
>    response.should_have_tag 'a', :attributes => { :href => /\/
> get_quote$/ }
>
>   Sorry if this is a FAQ, can't see any posts on the list about it.

Try checking the tracker at ruby forge too. This is a known bug:

http://rubyforge.org/tracker/index.php?func=detail&aid=7795&group_id=797&atid=3149

Cheers,
David

>
> Ta
> Ashley
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From work at ashleymoran.me.uk  Tue Jan 16 09:02:55 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Tue, 16 Jan 2007 14:02:55 +0000
Subject: [rspec-users] Links in view specs
In-Reply-To: <57c63afe0701160521j4c2ac5den77870f6948050f78@mail.gmail.com>
References: <79EAA51F-6DEF-49F5-85BF-F62DA9BD4883@ashleymoran.me.uk>
	<57c63afe0701160521j4c2ac5den77870f6948050f78@mail.gmail.com>
Message-ID: <F0D043B6-4E63-4052-8826-22639851D72C@ashleymoran.me.uk>


On 16 Jan 2007, at 13:21, David Chelimsky wrote:

> Try checking the tracker at ruby forge too. This is a known bug:
>
> http://rubyforge.org/tracker/index.php? 
> func=detail&aid=7795&group_id=797&atid=3149
>
> Cheers,
> David

Thanks David I'll just keep it half-specced for now

From work at ashleymoran.me.uk  Tue Jan 16 11:51:11 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Tue, 16 Jan 2007 16:51:11 +0000
Subject: [rspec-users] Specifying the contents of a date selector
Message-ID: <817850A8-00B6-4EB8-8347-0B8B8EE0E771@ashleymoran.me.uk>

Another n00b question (can  you tell I don't normally write view code  
yet???)...

I'm using the date_select form helper to give me three select popups,  
(day, month, year).  I want to specify that these exist and are set  
to today's date

I've tried this:

     response.should_have 'form > p > select 
[name="gap_quick_quote_parameters[purchase_date(3i)]"] > option',
                           :attributes => { :value => '16', :selected  
=> 'selected' },
                           :content => '16'

But it fails, saying:

     Invalid selector: "] > option

Is this a limitation of the CSS selector?  I can't think of a way of  
phrasing the spec without trying to select children of the select  
elements.  I want to avoid specifying that SOME select popup on the  
page has a value equal to today's day, month, or year.

Ashley

From dchelimsky at gmail.com  Tue Jan 16 12:27:57 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 16 Jan 2007 11:27:57 -0600
Subject: [rspec-users] Specifying the contents of a date selector
In-Reply-To: <817850A8-00B6-4EB8-8347-0B8B8EE0E771@ashleymoran.me.uk>
References: <817850A8-00B6-4EB8-8347-0B8B8EE0E771@ashleymoran.me.uk>
Message-ID: <57c63afe0701160927j6a407f7eu87c89ce3eaf698d6@mail.gmail.com>

On 1/16/07, Ashley Moran <work at ashleymoran.me.uk> wrote:
> Another n00b question (can  you tell I don't normally write view code
> yet???)...
>
> I'm using the date_select form helper to give me three select popups,
> (day, month, year).  I want to specify that these exist and are set
> to today's date
>
> I've tried this:
>
>      response.should_have 'form > p > select
> [name="gap_quick_quote_parameters[purchase_date(3i)]"] > option',
>                            :attributes => { :value => '16', :selected
> => 'selected' },
>                            :content => '16'
>
> But it fails, saying:
>
>      Invalid selector: "] > option
>
> Is this a limitation of the CSS selector?  I can't think of a way of
> phrasing the spec without trying to select children of the select
> elements.  I want to avoid specifying that SOME select popup on the
> page has a value equal to today's day, month, or year.

The syntax you are using invokes a preliminary port of assert_select
that is not officially supported yet.

That said, take a look at the tests for the selector. All the examples
that use double quotes have them escaped:

"form>p>select[name=\"gap_quick_quote_parameters[purchase_date(3i)]\"]>option"

I'm guessing you could also do this:

"form>p>select[name='gap_quick_quote_parameters[purchase_date(3i)]']>option"

or even this:

"form>p>select[name=gap_quick_quote_parameters[purchase_date(3i)]]>option"

Let us know what works, if any of those suggestions.

Cheers,
David

> Ashley
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From work at ashleymoran.me.uk  Tue Jan 16 12:52:04 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Tue, 16 Jan 2007 17:52:04 +0000
Subject: [rspec-users] Specifying the contents of a date selector
In-Reply-To: <57c63afe0701160927j6a407f7eu87c89ce3eaf698d6@mail.gmail.com>
References: <817850A8-00B6-4EB8-8347-0B8B8EE0E771@ashleymoran.me.uk>
	<57c63afe0701160927j6a407f7eu87c89ce3eaf698d6@mail.gmail.com>
Message-ID: <A48AEB18-F30D-40AF-95A8-DC1F233D0CF7@ashleymoran.me.uk>


On 16 Jan 2007, at 17:27, David Chelimsky wrote:

> "form>p>select[name='gap_quick_quote_parameters[purchase_date(3i)]'] 
> >option"


Hi David

Thanks for the pointer - that works fine.  I found this out playing  
around with it:

This fails (this is the original one I posted):
     response.should_have "'orm > p > select 
[name="gap_quick_quote_parameters[purchase_date(3i)]"] > option'

This succeeds (obviously the penultimate ]-character is interpreted  
as closing the *first* [-character):
     response.should_have 'form > p > select 
[name^="gap_quick_quote_parameters[purchase_date(3i)"] > option'

So apparently, using double quotes inside the CSS selector breaks  
bracket matching.  I wouldn't like to guess why it does that.

Anyway it's working now so I'm happy.

Cheers
Ashley

PS I edited those code lines by hand in case they contains errors

From aslak.hellesoy at gmail.com  Wed Jan 17 05:14:24 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Wed, 17 Jan 2007 11:14:24 +0100
Subject: [rspec-users] spec_ui problems
In-Reply-To: <ADAB7E40-D308-4AF5-B220-11FDDD052D93@yamabe.net>
References: <ADAB7E40-D308-4AF5-B220-11FDDD052D93@yamabe.net>
Message-ID: <8d961d900701170214v47e229dj438bdea6a4065c@mail.gmail.com>

On 1/13/07, Brian Yamabe <brian at yamabe.net> wrote:
> While looking into spec_ui, I decided to run the examples.
>
> The watir example works a little, but always chokes on the 'better
> than fudge' spec (failure output below).  Also, is there any command
> to pause the "browser"?  If my connection slows, the test gets out of
> whack.
>

For now, you have to build and install the RSpec gem from svn trunk to
make spec_ui work.
You can sprinkle some sleeps in your code (or monkey-patch Watir) to
pause things.

HTH,
Aslak

> The selenium example fails right away (output below).  I do have the
> 0.9.0 selenium-rc server running.
>
> I executed rake from inside each of the specific example directories.
>
> Thanks for any help.
>
> ---Brian Yamabe
>
> watir failure
> ----------
> /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/expectations/
> sugar.rb:13:in `call': undefined local variable or method
> `current_spec_number' for #<Spec::Ui::WebappFormatter:0x10083d0>
> (NameError)
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> expectations/sugar.rb:13:in `_method_missing'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> expectations/sugar.rb:9:in `method_missing'
>          from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb:
> 5:in `extra_failure_content'
>          from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb:
> 11:in `extra_failure_content'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/formatter/html_formatter.rb:57:in `spec_failed'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/reporter.rb:78:in `spec_failed'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/reporter.rb:27:in `spec_finished'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/specification.rb:37:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context.rb:57:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context.rb:54:in `each'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context.rb:54:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context_runner.rb:23:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context_runner.rb:22:in `each'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/context_runner.rb:22:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/command_line.rb:26:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4
> rake aborted!
>
> selenium failure
> ----------
> /opt/local/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> `gem_original_require': no such file to load -- ./spec/selenium
> (LoadError)
>          from /opt/local/lib/ruby/site_ruby/1.8/rubygems/
> custom_require.rb:27:in `require'
>          from ./spec/spec_helper.rb:7
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/option_parser.rb:107:in `require'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/option_parser.rb:107:in `parse'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `call'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `order!'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `catch'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `order!'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1279:in `permute!'
>          from /opt/local/lib/ruby/1.8/optparse.rb:1300:in `parse!'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/option_parser.rb:141:in `parse'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/option_parser.rb:13:in `create_context_runner'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> runner/command_line.rb:12:in `run'
>          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From aschur1 at telus.net  Wed Jan 17 19:26:17 2007
From: aschur1 at telus.net (aschur1 at telus.net)
Date: Wed, 17 Jan 2007 16:26:17 -0800
Subject: [rspec-users] spec_ui problems
In-Reply-To: <8d961d900701170214v47e229dj438bdea6a4065c@mail.gmail.com>
References: <ADAB7E40-D308-4AF5-B220-11FDDD052D93@yamabe.net>
	<8d961d900701170214v47e229dj438bdea6a4065c@mail.gmail.com>
Message-ID: <1169079977.45aebea9ad702@webmail.telus.net>

See patch #7685 to fix the undefined local variable message

http://rubyforge.org/tracker/index.php?func=detail&aid=7685&group_id=797&atid=3151

Alvin.

Quoting aslak hellesoy <aslak.hellesoy at gmail.com>:

> On 1/13/07, Brian Yamabe <brian at yamabe.net> wrote:
> > While looking into spec_ui, I decided to run the examples.
> >
> > The watir example works a little, but always chokes on the 'better
> > than fudge' spec (failure output below).  Also, is there any command
> > to pause the "browser"?  If my connection slows, the test gets out of
> > whack.
> >
> 
> For now, you have to build and install the RSpec gem from svn trunk to
> make spec_ui work.
> You can sprinkle some sleeps in your code (or monkey-patch Watir) to
> pause things.
> 
> HTH,
> Aslak
> 
> > The selenium example fails right away (output below).  I do have the
> > 0.9.0 selenium-rc server running.
> >
> > I executed rake from inside each of the specific example directories.
> >
> > Thanks for any help.
> >
> > ---Brian Yamabe
> >
> > watir failure
> > ----------
> > /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/expectations/
> > sugar.rb:13:in `call': undefined local variable or method
> > `current_spec_number' for #<Spec::Ui::WebappFormatter:0x10083d0>
> > (NameError)
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > expectations/sugar.rb:13:in `_method_missing'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > expectations/sugar.rb:9:in `method_missing'
> >          from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb:
> > 5:in `extra_failure_content'
> >          from /Users/byamabe/Desktop/spec_ui/lib/spec/ui/formatter.rb:
> > 11:in `extra_failure_content'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/formatter/html_formatter.rb:57:in `spec_failed'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/reporter.rb:78:in `spec_failed'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/reporter.rb:27:in `spec_finished'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/specification.rb:37:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context.rb:57:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context.rb:54:in `each'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context.rb:54:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context_runner.rb:23:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context_runner.rb:22:in `each'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/context_runner.rb:22:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/command_line.rb:26:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4
> > rake aborted!
> >
> > selenium failure
> > ----------
> > /opt/local/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:27:in
> > `gem_original_require': no such file to load -- ./spec/selenium
> > (LoadError)
> >          from /opt/local/lib/ruby/site_ruby/1.8/rubygems/
> > custom_require.rb:27:in `require'
> >          from ./spec/spec_helper.rb:7
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/option_parser.rb:107:in `require'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/option_parser.rb:107:in `parse'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `call'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1218:in `order!'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `catch'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1205:in `order!'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1279:in `permute!'
> >          from /opt/local/lib/ruby/1.8/optparse.rb:1300:in `parse!'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/option_parser.rb:141:in `parse'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/option_parser.rb:13:in `create_context_runner'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/lib/spec/
> > runner/command_line.rb:12:in `run'
> >          from /opt/local/lib/ruby/gems/1.8/gems/rspec-0.7.5/bin/spec:4
> > _______________________________________________
> > 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
> 





From jaydonnell at yahoo.com  Wed Jan 17 20:02:41 2007
From: jaydonnell at yahoo.com (Jay Donnell)
Date: Wed, 17 Jan 2007 17:02:41 -0800 (PST)
Subject: [rspec-users] problems testing a rails controller
Message-ID: <545907.19730.qm@web56014.mail.re3.yahoo.com>

I'm trying to test a rails controller, but my mock
isn't working. Everything looks right to me. Can
someone take a look at it and see if I'm missing
something obvious

http://pastie.caboo.se/33883

Domain.should_receive(:find_domain).once.and_return(domain)

That line in the spec is failing but I don't see why.
If I comment out that line the test passes so it is
running the controller method as far as I can tell.

Thanks,
Jay


 
____________________________________________________________________________________
Looking for earth-friendly autos? 
Browse Top Cars by "Green Rating" at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/

From work at ashleymoran.me.uk  Thu Jan 18 08:57:29 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Thu, 18 Jan 2007 13:57:29 +0000
Subject: [rspec-users] Best way to specify an HTML element inside another
Message-ID: <C2A316EA-303A-4225-8974-6A73A25407CA@ashleymoran.me.uk>

Hi

I know should_have is preliminary but I can't resist using it...

I have a form that contains a date select (Rails-generated).  I want  
to specify that the default value for each popup menu is correct

The best I can manage is this:
     response.should_have "form > p > select[name='quote_parameters 
[purchase_date(3i)]'] > option[selected='selected']", :text => '16'

I really want to specify two attributes for the <option> tag, like  
option[selected='selected',value='16'] - is there any syntax that  
allows this?

Ashley

From listaccount at e-tobi.net  Thu Jan 18 04:43:42 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Thu, 18 Jan 2007 10:43:42 +0100
Subject: [rspec-users] problems testing a rails controller
In-Reply-To: <545907.19730.qm@web56014.mail.re3.yahoo.com>
References: <545907.19730.qm@web56014.mail.re3.yahoo.com>
Message-ID: <45AF414E.40107@e-tobi.net>

Jay Donnell wrote:
> someone take a look at it and see if I'm missing
> something obvious
>   

> post 'search.rjs'

Mmm... You're posting to the action 'search.rjs', but what you want to
test is the action 'search'

bye,

Tobias

From bryan at osesm.com  Thu Jan 18 13:57:39 2007
From: bryan at osesm.com (Bryan Liles)
Date: Thu, 18 Jan 2007 13:57:39 -0500
Subject: [rspec-users] What do you think of this controller spec?
In-Reply-To: <810a540e0612260908k55c67a79wf281c58fa9fb7ce2@mail.gmail.com>
References: <810a540e0612241253gd5b327ex706cb07e3ccacc6e@mail.gmail.com>
	<57c63afe0612250834nf735f0cge4cb2346a03e1057@mail.gmail.com>
	<810a540e0612260908k55c67a79wf281c58fa9fb7ce2@mail.gmail.com>
Message-ID: <0FEDF8E5-9065-40E2-80BB-A26B6D033EBE@osesm.com>


On Dec 26, 2006, at 12:08 PM, Pat Maddox wrote:

> On 12/25/06, David Chelimsky <dchelimsky at gmail.com> wrote:
>> On 12/24/06, Pat Maddox <pergesu at gmail.com> wrote:
>>> Here's a controller spec I wrote, it's for a very simple RESTful
>>> controller (well, aren't all RESTful controllers simple? :)
>>>
>>> I've created a couple spec helper methods to refactor some of the
>>> common code...for example, require_login_and_correct_user creates  
>>> two
>>> specifications: one for when the user isn't logged in, and one when
>>> the user is logged in but requests someone else's book.  do_request
>>> automatically calls do_get/post/put/delete depending on what's
>>> implemented.
>
> Here's the code with the incorporated changes:
> spec_helper.rb: http://pastie.caboo.se/29577
> books_controller_spec.rb: http://pastie.caboo.se/29578
>

I'm trying to follow your example, but the include command isn't  
extending what ever class my context runs in. 
  

From pergesu at gmail.com  Thu Jan 18 14:51:33 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Thu, 18 Jan 2007 12:51:33 -0700
Subject: [rspec-users] What do you think of this controller spec?
In-Reply-To: <0FEDF8E5-9065-40E2-80BB-A26B6D033EBE@osesm.com>
References: <810a540e0612241253gd5b327ex706cb07e3ccacc6e@mail.gmail.com>
	<57c63afe0612250834nf735f0cge4cb2346a03e1057@mail.gmail.com>
	<810a540e0612260908k55c67a79wf281c58fa9fb7ce2@mail.gmail.com>
	<0FEDF8E5-9065-40E2-80BB-A26B6D033EBE@osesm.com>
Message-ID: <810a540e0701181151s5eceb5c2j71f3155b729400fe@mail.gmail.com>

On 1/18/07, Bryan Liles <bryan at osesm.com> wrote:
>
> On Dec 26, 2006, at 12:08 PM, Pat Maddox wrote:
>
> > On 12/25/06, David Chelimsky <dchelimsky at gmail.com> wrote:
> >> On 12/24/06, Pat Maddox <pergesu at gmail.com> wrote:
> >>> Here's a controller spec I wrote, it's for a very simple RESTful
> >>> controller (well, aren't all RESTful controllers simple? :)
> >>>
> >>> I've created a couple spec helper methods to refactor some of the
> >>> common code...for example, require_login_and_correct_user creates
> >>> two
> >>> specifications: one for when the user isn't logged in, and one when
> >>> the user is logged in but requests someone else's book.  do_request
> >>> automatically calls do_get/post/put/delete depending on what's
> >>> implemented.
> >
> > Here's the code with the incorporated changes:
> > spec_helper.rb: http://pastie.caboo.se/29577
> > books_controller_spec.rb: http://pastie.caboo.se/29578
> >
>
> I'm trying to follow your example, but the include command isn't
> extending what ever class my context runs in.

See http://rubyforge.org/tracker/index.php?func=detail&aid=7461&group_id=797&atid=3151

From cwdinfo at gmail.com  Thu Jan 18 16:44:38 2007
From: cwdinfo at gmail.com (s.ross)
Date: Thu, 18 Jan 2007 13:44:38 -0800
Subject: [rspec-users] spec'ing models with observers that send mail
Message-ID: <AFD1AF18-FC30-4A70-B567-7F6FB789F353@gmail.com>

I'm spec'ing a model with an observer that sends mail. Mailers behave  
similarly to controllers, so to make the whole thing work, I believe  
something like integrate_views has to be available. Has anyone dealt  
with observers and ActionMailer?

Thanks

From jaydonnell at yahoo.com  Thu Jan 18 17:51:30 2007
From: jaydonnell at yahoo.com (Jay Donnell)
Date: Thu, 18 Jan 2007 14:51:30 -0800 (PST)
Subject: [rspec-users] problems testing a rails controller
Message-ID: <245009.71018.qm@web56007.mail.re3.yahoo.com>

"Mmm... You're posting to the action 'search.rjs', but
what you want to
test is the action 'search'"

My understanding is that the new restful stuff in edge
rails (i.e. the respond_to) method will use the
extension to determine which output type to use. This
code runs fine outside of the tests and I am using
urls with extensions.

What is the usual way to test various respond_to
output types from rspec? 


 
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com

From dchelimsky at gmail.com  Thu Jan 18 18:09:53 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 18 Jan 2007 17:09:53 -0600
Subject: [rspec-users] problems testing a rails controller
In-Reply-To: <245009.71018.qm@web56007.mail.re3.yahoo.com>
References: <245009.71018.qm@web56007.mail.re3.yahoo.com>
Message-ID: <57c63afe0701181509y42d88a72o54d9acd1d6108b16@mail.gmail.com>

On 1/18/07, Jay Donnell <jaydonnell at yahoo.com> wrote:
> "Mmm... You're posting to the action 'search.rjs', but
> what you want to
> test is the action 'search'"
>
> My understanding is that the new restful stuff in edge
> rails (i.e. the respond_to) method will use the
> extension to determine which output type to use. This
> code runs fine outside of the tests and I am using
> urls with extensions.
>
> What is the usual way to test various respond_to
> output types from rspec?

This bit looks odd:

respond_to do |format|
  format.js { render :action => "search.rjs" }
end

I think you'd want to either render :template => 'seach.rjs' or just
say format.js (and the rjs template would be rendered implicitly, no?)

>
>
>
> ____________________________________________________________________________________
> Do you Yahoo!?
> Everyone is raving about the all-new Yahoo! Mail beta.
> http://new.mail.yahoo.com
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From jaydonnell at yahoo.com  Thu Jan 18 20:48:04 2007
From: jaydonnell at yahoo.com (Jay Donnell)
Date: Thu, 18 Jan 2007 17:48:04 -0800 (PST)
Subject: [rspec-users] problems testing a rails controller
In-Reply-To: <57c63afe0701181509y42d88a72o54d9acd1d6108b16@mail.gmail.com>
Message-ID: <917599.16539.qm@web56009.mail.re3.yahoo.com>

I'll try it without render :action as soon as I get a
chance. I don't know why that would be a problem
though.

 
--- David Chelimsky <dchelimsky at gmail.com> wrote:

> On 1/18/07, Jay Donnell <jaydonnell at yahoo.com>
> wrote:
> > "Mmm... You're posting to the action 'search.rjs',
> but
> > what you want to
> > test is the action 'search'"
> >
> > My understanding is that the new restful stuff in
> edge
> > rails (i.e. the respond_to) method will use the
> > extension to determine which output type to use.
> This
> > code runs fine outside of the tests and I am using
> > urls with extensions.
> >
> > What is the usual way to test various respond_to
> > output types from rspec?
> 
> This bit looks odd:
> 
> respond_to do |format|
>   format.js { render :action => "search.rjs" }
> end
> 
> I think you'd want to either render :template =>
> 'seach.rjs' or just
> say format.js (and the rjs template would be
> rendered implicitly, no?)
> 
> >
> >
> >
> >
>
____________________________________________________________________________________
> > Do you Yahoo!?
> > Everyone is raving about the all-new Yahoo! Mail
> beta.
> > http://new.mail.yahoo.com
> > _______________________________________________
> > 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
> 



 
____________________________________________________________________________________
Yahoo! Music Unlimited
Access over 1 million songs.
http://music.yahoo.com/unlimited

From dchelimsky at gmail.com  Thu Jan 18 21:23:13 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 18 Jan 2007 20:23:13 -0600
Subject: [rspec-users] [ANN] RSpec 0.7.5.1
Message-ID: <57c63afe0701181823ub9b7d1bq63d85f6627ffc3d4@mail.gmail.com>

Announcing the release of RSpec 0.7.5.1.

This release fixes a bug that prevents you from installing RSpec with
rubygems 0.9.1. It does not include any other fixes, so we do not
recommend that you upgrade to this if you are already using 0.7.5.

Thank you,
The RSpec Development Team

From jenny at likemindedgroup.com  Thu Jan 18 19:03:14 2007
From: jenny at likemindedgroup.com (jenny lawrence)
Date: Fri, 19 Jan 2007 00:03:14 -0000
Subject: [rspec-users] Error when following online tutorial
Message-ID: <20070119000958.895F552422C8@rubyforge.org>

Hi

 

I'm following the tutorial from the website.

 

I've got to the end, but I've been getting this failure:

 

'A stack with one item should return top when you send it top FAILED

"one item"  should equal "one item"

./stack_spec.rb:28:

./stack_spec.rb:27:in 'instance_eval'

 

3 specs 1 failure

 

I do not understand this.  Surely "one item" does equal "one item". I'm a
beginner and I'm just trying to copy exactly what it says on the tin.  What
is going on?

 

Can anyone help?

 

Thanks

 

Jason

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070119/56e826c4/attachment.html 

From dchelimsky at gmail.com  Thu Jan 18 22:12:57 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 18 Jan 2007 21:12:57 -0600
Subject: [rspec-users] Error when following online tutorial
In-Reply-To: <20070119000958.895F552422C8@rubyforge.org>
References: <20070119000958.895F552422C8@rubyforge.org>
Message-ID: <57c63afe0701181912i397f1fcby429dfc63de5bfab5@mail.gmail.com>

On 1/18/07, jenny lawrence <jenny at likemindedgroup.com> wrote:
>
>
>
>
> Hi
>
>
>
> I'm following the tutorial from the website.
>
>
>
> I've got to the end, but I've been getting this failure:
>
>
>
> 'A stack with one item should return top when you send it top FAILED
>
> "one item"  should equal "one item"
>
> ./stack_spec.rb:28:
>
> ./stack_spec.rb:27:in 'instance_eval'
>
>
>
> 3 specs 1 failure
>
>
>
> I do not understand this.  Surely "one item" does equal "one item". I'm a
> beginner and I'm just trying to copy exactly what it says on the tin.  What
> is going on?

Sorry Jenny - I haven't updated the tutorial since we made some
changes to rspec syntax.

The deal is that ruby evaluates actual.equal?(expected) using object
identity. Even though "one item" and "one item" look the same, they
are actually different objects in ruby. Instead, you want to use
either:

@stack.top.should_eql "one item"

OR

@stack.top.should == "one item"

Check out http://rspec.rubyforge.org/documentation/expectations.html
for more information.

I'll update the tutorial soon.

Cheers,
David
>
>
>
> Can anyone help?
>
>
>
> Thanks
>
>
>
> Jason
>
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>
>

From brian at yamabe.net  Thu Jan 18 22:18:29 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Thu, 18 Jan 2007 19:18:29 -0800
Subject: [rspec-users] rake spec:views
Message-ID: <2A173275-C63D-4F05-B92A-FCEB774E9D39@yamabe.net>

rake spec:views does not run the specs in my views folder.  I can run  
them with spec <path>.  Is there something I missed?

---Brian Yamabe

From dchelimsky at gmail.com  Thu Jan 18 22:34:04 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Thu, 18 Jan 2007 21:34:04 -0600
Subject: [rspec-users] rake spec:views
In-Reply-To: <2A173275-C63D-4F05-B92A-FCEB774E9D39@yamabe.net>
References: <2A173275-C63D-4F05-B92A-FCEB774E9D39@yamabe.net>
Message-ID: <57c63afe0701181934hc5b3c72x994dae5ec92f2e73@mail.gmail.com>

The files need be below spec/views/ and need be named ending in
_spec.rb. Is that the case?

On 1/18/07, Brian Yamabe <brian at yamabe.net> wrote:
> rake spec:views does not run the specs in my views folder.  I can run
> them with spec <path>.  Is there something I missed?
>
> ---Brian Yamabe
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From brian at yamabe.net  Thu Jan 18 22:41:19 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Thu, 18 Jan 2007 19:41:19 -0800
Subject: [rspec-users] rake spec:views
In-Reply-To: <57c63afe0701181934hc5b3c72x994dae5ec92f2e73@mail.gmail.com>
References: <2A173275-C63D-4F05-B92A-FCEB774E9D39@yamabe.net>
	<57c63afe0701181934hc5b3c72x994dae5ec92f2e73@mail.gmail.com>
Message-ID: <0A05D175-3DBE-47B2-AA36-FF4B414E2AA8@yamabe.net>

Thanks, can't believe I didn't notice the _spec. part.

---Brian Yamabe

On Jan 18, 2007, at 7:34 PM, David Chelimsky wrote:

> The files need be below spec/views/ and need be named ending in
> _spec.rb. Is that the case?
>
> On 1/18/07, Brian Yamabe <brian at yamabe.net> wrote:
>> rake spec:views does not run the specs in my views folder.  I can run
>> them with spec <path>.  Is there something I missed?
>>
>> ---Brian Yamabe
>> _______________________________________________
>> 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


From jonathan.tron.mailings at gmail.com  Fri Jan 19 04:09:50 2007
From: jonathan.tron.mailings at gmail.com (Jonathan Tron)
Date: Fri, 19 Jan 2007 10:09:50 +0100
Subject: [rspec-users] spec'ing models with observers that send mail
In-Reply-To: <AFD1AF18-FC30-4A70-B567-7F6FB789F353@gmail.com>
References: <AFD1AF18-FC30-4A70-B567-7F6FB789F353@gmail.com>
Message-ID: <C5388C0C-46B7-4D20-8D5C-B4E267F89E9A@gmail.com>

Le 18 janv. 07 ? 22:44, s.ross a ?crit :

> I'm spec'ing a model with an observer that sends mail. Mailers behave
> similarly to controllers, so to make the whole thing work, I believe
> something like integrate_views has to be available. Has anyone dealt
> with observers and ActionMailer?

I did it to add specs to restful_authentication plugin, it uses an  
observer to send mail on signup/activation.

For the observer, I choosed to get it by calling  
UserObserver.instance and then spec from that with a mock User  
object. (cf: http://jonathan.tron.name/files/user_observer_spec.v2.rb)

For the mailer, I choosed to spec based on the different mailer  
actions and what UserNotifier.create_my_action returns. (cf: http:// 
jonathan.tron.name/files/user_notifier_spec.v2.rb)

I'm not sure it's the right way to go, especially for the mailer  
part, so feel free to correct my mistakes.

Jonathan
--
Tron Jonathan
http://jonathan.tron.name




From pergesu at gmail.com  Fri Jan 19 05:05:01 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Fri, 19 Jan 2007 03:05:01 -0700
Subject: [rspec-users] Something cool I learned about rspec today
Message-ID: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>

I found an awesome feature today.  When you're setting expectations,
you can actually pattern match the parameters to methods.  Here's my
discussion and example:
http://evang.eli.st/blog/2007/1/19/parameter-matching-in-rspec

Pat

From work at ashleymoran.me.uk  Fri Jan 19 05:37:05 2007
From: work at ashleymoran.me.uk (Ashley Moran)
Date: Fri, 19 Jan 2007 10:37:05 +0000
Subject: [rspec-users] Something cool I learned about rspec today
In-Reply-To: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>
References: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>
Message-ID: <DE1E163E-07D0-4F8C-A55B-3BC4F120E0A3@ashleymoran.me.uk>


On 19 Jan 2007, at 10:05, Pat Maddox wrote:

> I found an awesome feature today.  When you're setting expectations,
> you can actually pattern match the parameters to methods.  Here's my
> discussion and example:
> http://evang.eli.st/blog/2007/1/19/parameter-matching-in-rspec

Just out of curiosity, what would happen if you needed to pass a  
regex as a message argument?

Ashley


From aslak.hellesoy at gmail.com  Fri Jan 19 05:39:40 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Fri, 19 Jan 2007 11:39:40 +0100
Subject: [rspec-users] Something cool I learned about rspec today
In-Reply-To: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>
References: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>
Message-ID: <8d961d900701190239k119b25b2pcbf56cfec3cbc740@mail.gmail.com>

On 1/19/07, Pat Maddox <pergesu at gmail.com> wrote:
> I found an awesome feature today.  When you're setting expectations,
> you can actually pattern match the parameters to methods.  Here's my
> discussion and example:
> http://evang.eli.st/blog/2007/1/19/parameter-matching-in-rspec
>

Nice writeup Pat!

> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From aslak.hellesoy at gmail.com  Fri Jan 19 05:44:57 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Fri, 19 Jan 2007 11:44:57 +0100
Subject: [rspec-users] Something cool I learned about rspec today
In-Reply-To: <DE1E163E-07D0-4F8C-A55B-3BC4F120E0A3@ashleymoran.me.uk>
References: <810a540e0701190205n17a24de5p632b0a2fef8a2c5@mail.gmail.com>
	<DE1E163E-07D0-4F8C-A55B-3BC4F120E0A3@ashleymoran.me.uk>
Message-ID: <8d961d900701190244l2ebee4f5vbdfd1213e4fe94a0@mail.gmail.com>

On 1/19/07, Ashley Moran <work at ashleymoran.me.uk> wrote:
>
> On 19 Jan 2007, at 10:05, Pat Maddox wrote:
>
> > I found an awesome feature today.  When you're setting expectations,
> > you can actually pattern match the parameters to methods.  Here's my
> > discussion and example:
> > http://evang.eli.st/blog/2007/1/19/parameter-matching-in-rspec
>
> Just out of curiosity, what would happen if you needed to pass a
> regex as a message argument?
>

Your mock expectation would look the same, and you'd have to call the
method with an eql regexp to make it pass. -Like the example in the
docs illustrates.

Aslak

> Ashley
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From justnothing at tiscali.co.uk  Fri Jan 19 07:07:16 2007
From: justnothing at tiscali.co.uk (David Green)
Date: Fri, 19 Jan 2007 12:07:16 -0000
Subject: [rspec-users] problems updating to 0.7.5.1
Message-ID: <000501c73bc2$5d6828f0$8801a8c0@DCQ95K1J>

Hi all

I've upgraded the rails plugin from 0.7.5 to 0.7.5.1 (plugin/remove 
rspec_on_rails followed by :
ruby script/plugin install 
svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
I've also updated the rspec gem and removed the older version. I'm running 
rubygems 0.9.1 on a windows box

however, when I try to run a spec (which was running fine under the old 
setup) I get the error message below.

here are the steps I took to update:

- ruby scipt/plugin/remove rspec_on_rails
- ruby script/plugin install 
svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
- ruby script/generate rspec (Y to over-write)

- gem update rspec
- gem uninstall rspec -v 0.7.5

- manually downloaded and installed rubygems 0.9.1. as per the instructions 
in this thread:
http://groups.google.com/group/rubyonrails-talk/browse_thread/thread/d7bc67cb9c7372d3/cb9af3d48cd63863

needless to say I'm a little stumped now and would appreciate some help
thanks

david

error message:

M:\devel\wd\phoss>drbspec spec/models/photo_spec.rb
(druby://localhost:8989) 
M:/devel/wd/phoss/vendor/plugins/rspec_on_rails/lib/spe
c/rails/version.rb:14:  (RuntimeError)
############################################################################
Your RSpec on Rails plugin is incompatible with your installed RSpec.

RSpec          : 0.7.5.1 (r1395)
RSpec on Rails : r1394

Make sure your RSpec on Rails plugin is compatible with your RSpec gem.
See http://rspec.rubyforge.org/documentation/rails/install.html for details.
############################################################################
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
stom_require.rb:27:in `gem_original_require'
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
stom_require.rb:27:in `require'
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
        from (druby://localhost:8989) 
M:/devel/wd/phoss/config/../vendor/plugins
/rspec_on_rails/lib/spec/rails.rb:15
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
stom_require.rb:27:in `gem_original_require'
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
stom_require.rb:27:in `require'
        from (druby://localhost:8989) 
c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
        from (druby://localhost:8989) ./spec/models/../spec_helper.rb:5
         ... 10 levels...
        from 
c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/lib/spec/runner/drb_co
mmand_line.rb:13:in `run'
        from c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/bin/drbspec:3
        from c:/ruby/bin/drbspec:16:in `load'
        from c:/ruby/bin/drbspec:16 


From dchelimsky at gmail.com  Fri Jan 19 08:07:17 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Fri, 19 Jan 2007 07:07:17 -0600
Subject: [rspec-users] problems updating to 0.7.5.1
In-Reply-To: <000501c73bc2$5d6828f0$8801a8c0@DCQ95K1J>
References: <000501c73bc2$5d6828f0$8801a8c0@DCQ95K1J>
Message-ID: <57c63afe0701190507k41a60035ob3df3aeed7f106a2@mail.gmail.com>

Working on this now. I'll follow up shortly....

On 1/19/07, David Green <justnothing at tiscali.co.uk> wrote:
> Hi all
>
> I've upgraded the rails plugin from 0.7.5 to 0.7.5.1 (plugin/remove
> rspec_on_rails followed by :
> ruby script/plugin install
> svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
> I've also updated the rspec gem and removed the older version. I'm running
> rubygems 0.9.1 on a windows box
>
> however, when I try to run a spec (which was running fine under the old
> setup) I get the error message below.
>
> here are the steps I took to update:
>
> - ruby scipt/plugin/remove rspec_on_rails
> - ruby script/plugin install
> svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
> - ruby script/generate rspec (Y to over-write)
>
> - gem update rspec
> - gem uninstall rspec -v 0.7.5
>
> - manually downloaded and installed rubygems 0.9.1. as per the instructions
> in this thread:
> http://groups.google.com/group/rubyonrails-talk/browse_thread/thread/d7bc67cb9c7372d3/cb9af3d48cd63863
>
> needless to say I'm a little stumped now and would appreciate some help
> thanks
>
> david
>
> error message:
>
> M:\devel\wd\phoss>drbspec spec/models/photo_spec.rb
> (druby://localhost:8989)
> M:/devel/wd/phoss/vendor/plugins/rspec_on_rails/lib/spe
> c/rails/version.rb:14:  (RuntimeError)
> ############################################################################
> Your RSpec on Rails plugin is incompatible with your installed RSpec.
>
> RSpec          : 0.7.5.1 (r1395)
> RSpec on Rails : r1394
>
> Make sure your RSpec on Rails plugin is compatible with your RSpec gem.
> See http://rspec.rubyforge.org/documentation/rails/install.html for details.
> ############################################################################
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> stom_require.rb:27:in `gem_original_require'
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> stom_require.rb:27:in `require'
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
> rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
>         from (druby://localhost:8989)
> M:/devel/wd/phoss/config/../vendor/plugins
> /rspec_on_rails/lib/spec/rails.rb:15
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> stom_require.rb:27:in `gem_original_require'
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> stom_require.rb:27:in `require'
>         from (druby://localhost:8989)
> c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
> rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
>         from (druby://localhost:8989) ./spec/models/../spec_helper.rb:5
>          ... 10 levels...
>         from
> c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/lib/spec/runner/drb_co
> mmand_line.rb:13:in `run'
>         from c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/bin/drbspec:3
>         from c:/ruby/bin/drbspec:16:in `load'
>         from c:/ruby/bin/drbspec:16
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Fri Jan 19 08:47:17 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Fri, 19 Jan 2007 07:47:17 -0600
Subject: [rspec-users] problems updating to 0.7.5.1
In-Reply-To: <57c63afe0701190507k41a60035ob3df3aeed7f106a2@mail.gmail.com>
References: <000501c73bc2$5d6828f0$8801a8c0@DCQ95K1J>
	<57c63afe0701190507k41a60035ob3df3aeed7f106a2@mail.gmail.com>
Message-ID: <57c63afe0701190547u303b835o8fcb93ac9b8209c6@mail.gmail.com>

OK. This is fixed now.

For those who are interested.... I'm not quite sure how it happened,
but when I did the release the gem and and the tag ended up w/
different revision numbers. We've got code in the plugin that enforces
a rule that the rspec and rspec_on_rails revision numbers must match.
I went ahead and hard-coded the revision number in the plugin to match
that of the gem. All should be working correctly now.

Thanks for your patience.

Cheers,
David

On 1/19/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> Working on this now. I'll follow up shortly....
>
> On 1/19/07, David Green <justnothing at tiscali.co.uk> wrote:
> > Hi all
> >
> > I've upgraded the rails plugin from 0.7.5 to 0.7.5.1 (plugin/remove
> > rspec_on_rails followed by :
> > ruby script/plugin install
> > svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
> > I've also updated the rspec gem and removed the older version. I'm running
> > rubygems 0.9.1 on a windows box
> >
> > however, when I try to run a spec (which was running fine under the old
> > setup) I get the error message below.
> >
> > here are the steps I took to update:
> >
> > - ruby scipt/plugin/remove rspec_on_rails
> > - ruby script/plugin install
> > svn://rubyforge.org/var/svn/rspec/tags/REL_0_7_5_1/rspec_on_rails/vendor/plugins/rspec_on_rails
> > - ruby script/generate rspec (Y to over-write)
> >
> > - gem update rspec
> > - gem uninstall rspec -v 0.7.5
> >
> > - manually downloaded and installed rubygems 0.9.1. as per the instructions
> > in this thread:
> > http://groups.google.com/group/rubyonrails-talk/browse_thread/thread/d7bc67cb9c7372d3/cb9af3d48cd63863
> >
> > needless to say I'm a little stumped now and would appreciate some help
> > thanks
> >
> > david
> >
> > error message:
> >
> > M:\devel\wd\phoss>drbspec spec/models/photo_spec.rb
> > (druby://localhost:8989)
> > M:/devel/wd/phoss/vendor/plugins/rspec_on_rails/lib/spe
> > c/rails/version.rb:14:  (RuntimeError)
> > ############################################################################
> > Your RSpec on Rails plugin is incompatible with your installed RSpec.
> >
> > RSpec          : 0.7.5.1 (r1395)
> > RSpec on Rails : r1394
> >
> > Make sure your RSpec on Rails plugin is compatible with your RSpec gem.
> > See http://rspec.rubyforge.org/documentation/rails/install.html for details.
> > ############################################################################
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> > stom_require.rb:27:in `gem_original_require'
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> > stom_require.rb:27:in `require'
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
> > rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
> >         from (druby://localhost:8989)
> > M:/devel/wd/phoss/config/../vendor/plugins
> > /rspec_on_rails/lib/spec/rails.rb:15
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> > stom_require.rb:27:in `gem_original_require'
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/site_ruby/1.8/rubygems/cu
> > stom_require.rb:27:in `require'
> >         from (druby://localhost:8989)
> > c:/ruby/lib/ruby/gems/1.8/gems/activesuppo
> > rt-1.3.1/lib/active_support/dependencies.rb:147:in `require'
> >         from (druby://localhost:8989) ./spec/models/../spec_helper.rb:5
> >          ... 10 levels...
> >         from
> > c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/lib/spec/runner/drb_co
> > mmand_line.rb:13:in `run'
> >         from c:/ruby/lib/ruby/gems/1.8/gems/rspec-0.7.5.1/bin/drbspec:3
> >         from c:/ruby/bin/drbspec:16:in `load'
> >         from c:/ruby/bin/drbspec:16
> >
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
>

From justnothing at tiscali.co.uk  Fri Jan 19 10:30:29 2007
From: justnothing at tiscali.co.uk (David Green)
Date: Fri, 19 Jan 2007 15:30:29 -0000
Subject: [rspec-users] problems updating to 0.7.5.1
Message-ID: <000f01c73bde$c1e53130$8801a8c0@DCQ95K1J>

> OK. This is fixed now.

thanks david! I appreciate you fixing it so quickly.

From lists-rspec at shopwatch.org  Fri Jan 19 22:51:05 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Fri, 19 Jan 2007 22:51:05 -0500
Subject: [rspec-users] heckle and rspec on rails
In-Reply-To: <e282921e0701150129m924b4e3rbcc30589bf7bfd4c@mail.gmail.com>
References: <e282921e0701150129m924b4e3rbcc30589bf7bfd4c@mail.gmail.com>
Message-ID: <45B191A9.8070209@rubyforge.org>

Chris Anderson wrote:
> I guess the question is, has anyone had luck using Heckle with Rails?
> I've got it working for regular ruby specs, and it's a lot of fun
> (although I do hit infinite loops sometimes).

If I heckle an ActiveRecord model, I get all sorts of noise about the 
various methods ActiveRecord adds.  However, if I heckle a regular old 
class I wrote, I get the results you're seeing - single run, no 
heckling.  Not sure what to go look at from here.

Jay Levitt


From lists-rspec at shopwatch.org  Sat Jan 20 11:02:36 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sat, 20 Jan 2007 11:02:36 -0500
Subject: [rspec-users] heckle and rspec on rails
In-Reply-To: <45B191A9.8070209@rubyforge.org>
References: <e282921e0701150129m924b4e3rbcc30589bf7bfd4c@mail.gmail.com>
	<45B191A9.8070209@rubyforge.org>
Message-ID: <45B23D1C.3050908@rubyforge.org>

Jay Levitt wrote:
> If I heckle an ActiveRecord model, I get all sorts of noise about the 
> various methods ActiveRecord adds.  However, if I heckle a regular old 
> class I wrote, I get the results you're seeing - single run, no 
> heckling.  Not sure what to go look at from here.

No, ignore that - I was mistyping my own class name!  When I heckle the 
class, it works as expected.

Jay



From listaccount at e-tobi.net  Sat Jan 20 20:25:00 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Sun, 21 Jan 2007 02:25:00 +0100
Subject: [rspec-users] A Thread / Mock question
Message-ID: <45B2C0EC.1030609@e-tobi.net>

Hi!

What I'm trying at the moment is to build specifications for some
background thread code. What I want is a class Task and a class
TaskRunner. The TaskRunner should run a task in a background thread. It
should also indicate the status of the running task when calling the
running? method.

That's basically, what I came up with so far:

class TaskRunner
    def initialize
        @is_running = false
    end
   
    def running?
        return @is_running
    end
   
    def execute(task)
        @is_running = true
        Thread.new do
            task.run
            @is_running = false
        end
    end
end

class SampleTask
    def run
    end
end

context "The TaskRunner" do
    setup do
        Thread.stub!(:new).and_return do |block|
            block.call
        end
        @runner = TaskRunner.new
    end

    specify "should be running when a task is executed" do
        task = SampleTask.new
        task.should_receive(:run) do
            @runner.should_be_running
        end
        @runner.execute(task)
    end

    specify "should not be running after the task has been executed" do
        @runner.execute(SampleTask.new)
        @runner.should_not_be_running
    end
end

I don't want to have real threads in the specs, so I mocked the Thread
class to just execute the passed block within the main thread. In the
first specification I also mock the run method of the task to check
within that method, if the TaskRunner correctly indicates it's running
status. So far this works, but maybe there is a better solution. Any
suggestions?

Is there a way to use the above mocking of Thread.new and  pass a
parameter to the block executed by the thread, like this?:


    def execute(task)
        @is_running = true
        Thread.new(task) do |task_to_execute|
            task_to_execute.run
            @is_running = false
        end
    end


How can RSpec mock Thread.new in this case?

Maybe I should introduce some kind of ThreadFactory that creates
instances of TestableThread for the specifications.

bye,

Tobias


From francois.beausoleil at gmail.com  Sat Jan 20 23:45:17 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Sat, 20 Jan 2007 23:45:17 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
Message-ID: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>

Hi all !

I just started writing specs on a new project.  I would just like to
validate that this is the way you would write it.  I know about mocks,
stubs and expectations.  I don't think this is a problem for me.

My question really boils down to:

def index
  @projects = current_user.projects.active
end

My spec needs to return the proxy, no ?  Here's my code:

context "A logged in user visiting the index action" do
  controller_name :dashboard

  setup do
    @user = mock("user")
    controller.stub!(:current_user).and_return(@user)

    @user.should_receive(:projects).and_return(projects_proxy = Object.new)
    projects_proxy.should_receive(:active).and_return([:a, :b])
    get :index
  end

  specify "should have a list of projects available" do
    assigns[:projects].should == [:a, :b]
  end

  specify "should render the index view" do
    controller.should_render :template => "dashboard/index"
  end

  specify "should be a successful call" do
    response.should_be_success
  end
end

Thanks !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From listaccount at e-tobi.net  Sun Jan 21 06:25:30 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Sun, 21 Jan 2007 12:25:30 +0100
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
Message-ID: <45B34DAA.7040004@e-tobi.net>

Hi!

Francois Beausoleil wrote:
> My question really boils down to:
>
> def index
>   @projects = current_user.projects.active
> end
>
> My spec needs to return the proxy, no ?  Here's my code:
>   
>From a RSpec-newbie's point of view:

Looks like current_user is a method of the controller, that selects the
user from the request params, right? Mocking this means, current_user()
will not be covered by the specs. Normally I avoid mocking/stubbing
members of a class under test or private and protected members of any
class. You could instead mock the User class directly to return the User
mock.

But why not use fixtures and let it run with real data instead of
mocking the database away?

bye,

Tobias



From dchelimsky at gmail.com  Sun Jan 21 06:37:59 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 05:37:59 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
Message-ID: <57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>

On 1/20/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> Hi all !
>
> I just started writing specs on a new project.  I would just like to
> validate that this is the way you would write it.  I know about mocks,
> stubs and expectations.  I don't think this is a problem for me.
>
> My question really boils down to:
>
> def index
>   @projects = current_user.projects.active
> end
>
> My spec needs to return the proxy, no ?  Here's my code:
>
> context "A logged in user visiting the index action" do
>   controller_name :dashboard
>
>   setup do
>     @user = mock("user")
>     controller.stub!(:current_user).and_return(@user)
>
>     @user.should_receive(:projects).and_return(projects_proxy = Object.new)
>     projects_proxy.should_receive(:active).and_return([:a, :b])
>     get :index
>   end
>
>   specify "should have a list of projects available" do
>     assigns[:projects].should == [:a, :b]
>   end
>
>   specify "should render the index view" do
>     controller.should_render :template => "dashboard/index"
>   end
>
>   specify "should be a successful call" do
>     response.should_be_success
>   end
> end

I would approach this quite differently. The first difference being I
would have started with a spec, letting the spec help me to discover
the interface that I want on User.

context "A logged in user visiting the index action" do
  controller_name :dashboard

  setup do
    @user = mock("user")
    controller.stub!(:current_user).and_return(@user)
    @user.should_receive(:active_projects).and_return([:a, :b])

    get :index
  end

  specify "should have a list of projects available" do
    assigns[:projects].should == [:a, :b]
  end

  specify "should render the index view" do
    response.should_render :template => "dashboard/index"
  end

  specify "should be a successful call" do
    response.should_be_success
  end
end

This would lead me to this implementation ...

def index
  @projects = current_user.active_projects
end

... which would, in turn, lead me to spec out a method named
active_projects on User.

The other thing that I've been doing lately is making an effort to
separate out the noise (in setup) from the interesting bits in each
specification. I do that by using stub! in setup and should_receive in
specify, using should_receive to override stub! when appropriate. This
leads to a more verbose spec, but it makes each specify block complete
(so you don't have to look back at setup).

context "A logged in user visiting the index action" do
  controller_name :dashboard

  setup do
    @user = mock("user")
    @user.stub!(:active_projects).and_return([])
    controller.stub!(:current_user).and_return(@user)
  end

  specify "should have a list of projects available" do
    #given
    @user.should_receive(:active_projects).and_return([:a, :b])

    #when
    get :index

    #then
    assigns[:projects].should == [:a, :b]
  end

  specify "should render the index view" do
    #given - nothing unusual

    #when
    get :index

    #then
    response.should_be_success
    response.should_render :template => "dashboard/index"
  end
end

Here I've identified in each spec the Given/When/Then with comments.
Note that I abhor comments like this in production code, but I've come
to view specs as different from code. The result is that each spec is
easy to read and understand in isolation. You don't have to look up at
setup to understand a problem.

Note also that I've used two expectations in one spec. For me, having
a separate spec for should_be_success is not that meaningful when
we're spec'ing the template. From a developer standpoint it might be
helpful to have the statement there when trying to understand a
problem, but from the outside we know it was successful if the right
template shows up, no?

A little more than 2 cents. Hope this is all helpful.

Cheers,
David


>
> Thanks !
> --
> Fran?ois Beausoleil
> http://blog.teksol.info/
> http://piston.rubyforge.org/
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

From dchelimsky at gmail.com  Sun Jan 21 06:46:13 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 05:46:13 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <45B34DAA.7040004@e-tobi.net>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B34DAA.7040004@e-tobi.net>
Message-ID: <57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>

On 1/21/07, Tobias Grimm <listaccount at e-tobi.net> wrote:
> Hi!
>
> Francois Beausoleil wrote:
> > My question really boils down to:
> >
> > def index
> >   @projects = current_user.projects.active
> > end
> >
> > My spec needs to return the proxy, no ?  Here's my code:
> >
> >From a RSpec-newbie's point of view:
>
> Looks like current_user is a method of the controller, that selects the
> user from the request params, right? Mocking this means, current_user()
> will not be covered by the specs. Normally I avoid mocking/stubbing
> members of a class under test or private and protected members of any
> class. You could instead mock the User class directly to return the User
> mock.

If the current_user is actually coming from request params I agree w/
this. If it's a user stored in the session, I don't agree that it
should come from the User class. I would probably stick the mock user
in the session though, rather than stub the method.

> But why not use fixtures and let it run with real data instead of
> mocking the database away?

Because specs run faster when you don't rely on the database. You're
going to be spec'ing all the same stuff in your model specs, so why
hit the database more than you need to?

Cheers,
David

>
> bye,
>
> Tobias
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From listaccount at e-tobi.net  Sun Jan 21 07:24:07 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Sun, 21 Jan 2007 13:24:07 +0100
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>
	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>
Message-ID: <45B35B67.2070402@e-tobi.net>

David Chelimsky wrote:
> this. If it's a user stored in the session, I don't agree that it
> should come from the User class. I would probably stick the mock user
> in the session though, rather than stub the method.
>   

Agreed - if it's in the session, this should be the preferred way.

> Because specs run faster when you don't rely on the database. You're
> going to be spec'ing all the same stuff in your model specs, so why
> hit the database more than you need to?
>   

I don't have any experience with RSpec/Rails using a database - my
current Rails project is DB-less :-) But on a recent C# project (with
NUnit) I've chosen to just use a database in my tests. It wasn't even an
in-memory database (embedded Firebird) but it was still fast enough with
several hundred tests hitting the database. Rails with SQLite in memory
should even be faster. So speed isn't really much of an issue in my
experience.

But you're right that stubbing database access should be preferred.
Sometimes it's just easier to put complex data into a fixture and
retrieve it from the database. Maybe that's a design smell - I'll keep
this in mind when I run into this situation the next time.


Tobias


From dchelimsky at gmail.com  Sun Jan 21 07:36:01 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 06:36:01 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <45B35B67.2070402@e-tobi.net>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B34DAA.7040004@e-tobi.net>
	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>
	<45B35B67.2070402@e-tobi.net>
Message-ID: <57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>

On 1/21/07, Tobias Grimm <listaccount at e-tobi.net> wrote:
> David Chelimsky wrote:
> > this. If it's a user stored in the session, I don't agree that it
> > should come from the User class. I would probably stick the mock user
> > in the session though, rather than stub the method.
> >
>
> Agreed - if it's in the session, this should be the preferred way.
>
> > Because specs run faster when you don't rely on the database. You're
> > going to be spec'ing all the same stuff in your model specs, so why
> > hit the database more than you need to?
> >
>
> I don't have any experience with RSpec/Rails using a database - my
> current Rails project is DB-less :-) But on a recent C# project (with
> NUnit) I've chosen to just use a database in my tests. It wasn't even an
> in-memory database (embedded Firebird) but it was still fast enough with
> several hundred tests hitting the database.

What's fast enough?

When doing BDD or TDD, the goal is to run the specs (tests) as often
as possible, after every little change. I run my specs dozens (if not
hundreds) of times a day, so a difference of a few seconds adds up to
a big difference.

>Rails with SQLite in memory
> should even be faster. So speed isn't really much of an issue in my
> experience.
>
> But you're right that stubbing database access should be preferred.
> Sometimes it's just easier to put complex data into a fixture and
> retrieve it from the database. Maybe that's a design smell - I'll keep
> this in mind when I run into this situation the next time.
>
>
> Tobias
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From listaccount at e-tobi.net  Sun Jan 21 07:58:42 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Sun, 21 Jan 2007 13:58:42 +0100
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>
	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
Message-ID: <45B36382.2030503@e-tobi.net>

David Chelimsky wrote:
> What's fast enough?
>   

It's fast enough, if it doesn't break my "flow" - which most likely is
slower than your's :-)

Tobias


From francois.beausoleil at gmail.com  Sun Jan 21 15:42:56 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Sun, 21 Jan 2007 15:42:56 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
Message-ID: <41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>

Hello David, all,

2007/1/21, David Chelimsky <dchelimsky at gmail.com>:
> On 1/20/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> > def index
> >   @projects = current_user.projects.active
> > end
>
> I would approach this quite differently. The first difference being I
> would have started with a spec, letting the spec help me to discover
> the interface that I want on User.
>
> def index
>   @projects = current_user.active_projects
> end
>
> ... which would, in turn, lead me to spec out a method named
> active_projects on User.

I agree with you, David, but Rails has this nice facility for allowing
developers to add methods on collection proxies.  The end goal is that
I get only active / inactive projects, right ?  Whatever
implementation floats the boat is good.

Adding #active_projects on User leads to a kind of namespace
pollution, no ?  Really, who knows what projects are active ?  The
Project class, or the User ?

has_many :projects do
  def active
    find(:all, :conditions => ['active = 1'])
  end
end

has_many :projects
def active_projects

end

We're going off-topic here, but these are good design questions anyway.

Thank you for enlightning me !

Bye !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From francois.beausoleil at gmail.com  Sun Jan 21 15:44:38 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Sun, 21 Jan 2007 15:44:38 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
Message-ID: <41d5fadf0701211244t71ba8ceci495a73aaa0ed4f15@mail.gmail.com>

Shoot!  Message went away too fast.

has_many :projects
def active_projects
  self.projects.find(:all, :conditions => ['active = 1'])
end

Or another alternative:
class Project
  def self.active
    with_scope(:find => ['active = 1']) do
      yield
    end
  end
end

In the end, it boils down to preferences, no ?

Thanks !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From lists-rspec at shopwatch.org  Sun Jan 21 15:46:23 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sun, 21 Jan 2007 15:46:23 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B36382.2030503@e-tobi.net>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
	<45B36382.2030503@e-tobi.net>
Message-ID: <45B3D11F.90703@rubyforge.org>

I've got a question similar to Francois's.  I'm writing a to-do-like 
application, where each Task has a number of Events, consisting of 
event.date and event.status.  I want to see how long it's been since the 
Task was last completed.  So:

context "A task completed once on 2001-01-02" do
   setup do
     @task = Task.new
     mock_event = Struct.new(:date, :status)
     @task.stub!(:events).and_return(
       [mock_event.new(Date.parse("2001-01-02"), "completed")])
   end

   specify "was completed 0 days ago on 2001-01-02" do
     @task.completed_days_ago(Date.parse("2001-01-02")).should_eql 0
   end
end

task.rb looks like:

   def cache_completions
     @completions = {}
     events.each do |event|
       @completions[event.date] = event.status
     end
   end

   def completed_days_ago(todays_date)
     cache_completions unless @completions
     date = todays_date
     while date > todays_date - 7 do
       return (todays_date - date).to_i if @completions[date] == "completed"
       date = date - 1
     end

     "never"
   end


This smells a bit.  I could separate part of completed_days_ago into 
last_completed_on(date), but no matter how I slice it I still end up 
mocking Task.cells for some routine.  Suggestions?

Jay



From cwdinfo at gmail.com  Sun Jan 21 15:52:09 2007
From: cwdinfo at gmail.com (s.ross)
Date: Sun, 21 Jan 2007 12:52:09 -0800
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
Message-ID: <A09E1267-DD7F-4693-B57A-D267EAB3048F@gmail.com>

Do partial mocks solve your problem?

http://rspec.rubyforge.org/documentation/mocks/partial_mocks.html

I run into a lot of these issues when spec'ing, and it seems writing  
mocks that cover as many cases as I expect AR to cough up can be  
difficult. Others might have different experiences. I found partial  
mocks intriguing, although I've not yet explored them.

Hope I'm not barking up the wrong tree.

--steve

On Jan 21, 2007, at 12:42 PM, Francois Beausoleil wrote:

> Hello David, all,
>
> 2007/1/21, David Chelimsky <dchelimsky at gmail.com>:
>> On 1/20/07, Francois Beausoleil <francois.beausoleil at gmail.com>  
>> wrote:
>>> def index
>>>   @projects = current_user.projects.active
>>> end
>>
>> I would approach this quite differently. The first difference being I
>> would have started with a spec, letting the spec help me to discover
>> the interface that I want on User.
>>
>> def index
>>   @projects = current_user.active_projects
>> end
>>
>> ... which would, in turn, lead me to spec out a method named
>> active_projects on User.
>
> I agree with you, David, but Rails has this nice facility for allowing
> developers to add methods on collection proxies.  The end goal is that
> I get only active / inactive projects, right ?  Whatever
> implementation floats the boat is good.
>
> Adding #active_projects on User leads to a kind of namespace
> pollution, no ?  Really, who knows what projects are active ?  The
> Project class, or the User ?
>
> has_many :projects do
>   def active
>     find(:all, :conditions => ['active = 1'])
>   end
> end
>
> has_many :projects
> def active_projects
>
> end
>
> We're going off-topic here, but these are good design questions  
> anyway.
>
> Thank you for enlightning me !
>
> Bye !
> -- 
> Fran?ois Beausoleil
> http://blog.teksol.info/
> http://piston.rubyforge.org/
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users


From lists-rspec at shopwatch.org  Sun Jan 21 15:53:43 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sun, 21 Jan 2007 15:53:43 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
Message-ID: <45B3D2D7.5050400@rubyforge.org>

Francois Beausoleil wrote:

> 
> Adding #active_projects on User leads to a kind of namespace
> pollution, no ?  Really, who knows what projects are active ?  The
> Project class, or the User ?

This seems to be a basic conflict between "Tell, don't ask" and 
domain-driven design.  DDD, as I understand it, says the Project class 
should encapsulate all knowledge about projects; TDA says we shouldn't 
ask for the project proxy.  Yes?

Jay


From dchelimsky at gmail.com  Sun Jan 21 16:59:27 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 15:59:27 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
Message-ID: <57c63afe0701211359l1238f200v3dde8cf7ad447fc3@mail.gmail.com>

On 1/21/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> Hello David, all,
>
> 2007/1/21, David Chelimsky <dchelimsky at gmail.com>:
> > On 1/20/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> > > def index
> > >   @projects = current_user.projects.active
> > > end
> >
> > I would approach this quite differently. The first difference being I
> > would have started with a spec, letting the spec help me to discover
> > the interface that I want on User.
> >
> > def index
> >   @projects = current_user.active_projects
> > end
> >
> > ... which would, in turn, lead me to spec out a method named
> > active_projects on User.
>
> I agree with you, David, but Rails has this nice facility for allowing
> developers to add methods on collection proxies.  The end goal is that
> I get only active / inactive projects, right ?  Whatever
> implementation floats the boat is good.
>
> Adding #active_projects on User leads to a kind of namespace
> pollution, no ?  Really, who knows what projects are active ?  The
> Project class, or the User ?
>
> has_many :projects do
>   def active
>     find(:all, :conditions => ['active = 1'])
>   end
> end
>
> has_many :projects
> def active_projects
>
> end
>
> We're going off-topic here, but these are good design questions anyway.

RSpec is ALL ABOUT design, so these questions are certainly on topic
(in my view).

>
> Thank you for enlightning me !
>
> Bye !
> --
> Fran?ois Beausoleil
> http://blog.teksol.info/
> http://piston.rubyforge.org/
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

From brian at yamabe.net  Sun Jan 21 17:02:01 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Sun, 21 Jan 2007 14:02:01 -0800
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <45B3D2D7.5050400@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
Message-ID: <40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>

Sorry for continuing the divergence, but the design thoughts are very  
interesting.

Couldn't DDD and TDA be satisfied by:
Project.find_active_projects_for_user(:current_user)

---Brian Yamabe

On Jan 21, 2007, at 12:53 PM, Jay Levitt wrote:

> Francois Beausoleil wrote:
>
>>
>> Adding #active_projects on User leads to a kind of namespace
>> pollution, no ?  Really, who knows what projects are active ?  The
>> Project class, or the User ?
>
> This seems to be a basic conflict between "Tell, don't ask" and
> domain-driven design.  DDD, as I understand it, says the Project class
> should encapsulate all knowledge about projects; TDA says we shouldn't
> ask for the project proxy.  Yes?
>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users


From dchelimsky at gmail.com  Sun Jan 21 17:14:37 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 16:14:37 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <45B3D2D7.5050400@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
Message-ID: <57c63afe0701211414t189c0df1md288cce6885c0a56@mail.gmail.com>

On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> Francois Beausoleil wrote:
>
> >
> > Adding #active_projects on User leads to a kind of namespace
> > pollution, no ?  Really, who knows what projects are active ?  The
> > Project class, or the User ?
>
> This seems to be a basic conflict between "Tell, don't ask" and
> domain-driven design.  DDD, as I understand it, says the Project class
> should encapsulate all knowledge about projects; TDA says we shouldn't
> ask for the project proxy.  Yes?

TDA says we shouldn't even ask the User for its Projects. It's the Law
of Demeter that says we shouldn't ask the User for it's Projects and
THEN ask the Projects whether they are active.

Just because rails makes it easy doesn't make it a good design choice.
It really depends on context. If your app is never going to change
requirements, then the problems associated with violations of OO
design principles don't really matter. They are all about writing
maintainable software.

I'm serious here - I don't mean to patronize. There are times when the
application you're writing will have simple requirements and a short
potential life-span. In those cases, the costs of heeding to design
principles might outweigh the benefits of a flexible design.

That said, imagine that somewhere down the line the requirements
change such that User knows its Projects through some middle man, like
a ProjectManager. Now you'd have to change all the places in the code
that say "user.projects.active" would have to change to
"user.project_manager.projects.active".

Of course at that point you could change #projects to delegate to
project_manager.projects, so clients of User wouldn't have to change,
but you see the point.

So I'd prefer to not advertise how the User keeps its Projects.

But that's me. 2 cents worth.

>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Sun Jan 21 17:23:51 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 16:23:51 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
Message-ID: <57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>

On 1/21/07, Brian Yamabe <brian at yamabe.net> wrote:
> Sorry for continuing the divergence, but the design thoughts are very
> interesting.
>
> Couldn't DDD and TDA be satisfied by:
> Project.find_active_projects_for_user(:current_user)

How would that be implemented?

def self.find_active_projects_for_user(user)
  return user.projects.active_projects
end

This bi-directional dependency seems problematic to me. I realize that
rails makes it easy to do these, but again, just because its easy
doesn't make it a good decision.

This is one of those questions that can be enlightened by client need.
Do clients of Projects ask for their users? Do clients of User ask for
its Projects? If both answers are yes, then maybe the bi-directional
deal makes sense. If only one is needed, then only one should be
implemented.

OK. I'll shut up now. Go Bears!

David


> ---Brian Yamabe
>
> On Jan 21, 2007, at 12:53 PM, Jay Levitt wrote:
>
> > Francois Beausoleil wrote:
> >
> >>
> >> Adding #active_projects on User leads to a kind of namespace
> >> pollution, no ?  Really, who knows what projects are active ?  The
> >> Project class, or the User ?
> >
> > This seems to be a basic conflict between "Tell, don't ask" and
> > domain-driven design.  DDD, as I understand it, says the Project class
> > should encapsulate all knowledge about projects; TDA says we shouldn't
> > ask for the project proxy.  Yes?
> >
> > Jay
> >
> > _______________________________________________
> > 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
>

From dchelimsky at gmail.com  Sun Jan 21 17:28:44 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 16:28:44 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B3D11F.90703@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B34DAA.7040004@e-tobi.net>
	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>
	<45B35B67.2070402@e-tobi.net>
	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
	<45B36382.2030503@e-tobi.net> <45B3D11F.90703@rubyforge.org>
Message-ID: <57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>

On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> I've got a question similar to Francois's.  I'm writing a to-do-like
> application, where each Task has a number of Events, consisting of
> event.date and event.status.  I want to see how long it's been since the
> Task was last completed.  So:
>
> context "A task completed once on 2001-01-02" do
>    setup do
>      @task = Task.new
>      mock_event = Struct.new(:date, :status)
>      @task.stub!(:events).and_return(
>        [mock_event.new(Date.parse("2001-01-02"), "completed")])
>    end
>
>    specify "was completed 0 days ago on 2001-01-02" do
>      @task.completed_days_ago(Date.parse("2001-01-02")).should_eql 0
>    end
> end
>
> task.rb looks like:
>
>    def cache_completions
>      @completions = {}
>      events.each do |event|
>        @completions[event.date] = event.status
>      end
>    end
>
>    def completed_days_ago(todays_date)
>      cache_completions unless @completions
>      date = todays_date
>      while date > todays_date - 7 do
>        return (todays_date - date).to_i if @completions[date] == "completed"
>        date = date - 1
>      end
>
>      "never"
>    end
>
>
> This smells a bit.  I could separate part of completed_days_ago into
> last_completed_on(date), but no matter how I slice it I still end up
> mocking Task.cells for some routine.  Suggestions?

Ah, how I love these questions. Thanks Jay!

So first of all, lets think about what we REALLY want the interface to
Task to be. What you have there is:

  @task.completed_days_ago(date)

In your description, you say "how long it's been since the Task was
last completed". That leads me to a method name like this:

  @task.days_since_completed(date)

The next question is, why are we passing in the date? We want to how
many days have passed since it was completed, not how many days
between completion and some arbitrary date. Now this gets tricky
because the method becomes more testable if you can supply the date,
but supplying the date doesn't "feel" right in the context of
"days_since_completed".

>From here, I see a couple of ways to go. One would be to stub Time.new
(or local, or whatever method you wish to use to generate current
time) and leave the method with no argument:

context "A task completed once on 2001-02-03" do
  setup do
    event_stub = Struct.new(:date,
:status).new(Time.parse("2001-02-03"), "completed")
    Time.stub!(:new).and_return(Time.parse("2001-02-03"))

    @task = Task.new
    @task.stub!(:events).and_return([event_stub])
  end

  specify "was completed 0 days ago on 2001-02-03" do
    @task.days_since_completed.should_eql 0
  end
end

Another would be to change the meaning of the method to something like:

  @task.days_from_completion_date_to(date)

No perfect answer here. Just food for thought.

Cheers,
David


>
> Jay
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From lists-rspec at shopwatch.org  Sun Jan 21 18:09:18 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sun, 21 Jan 2007 18:09:18 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>	<45B36382.2030503@e-tobi.net>
	<45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
Message-ID: <45B3F29E.1080009@rubyforge.org>

David Chelimsky wrote:
> On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
>> I've got a question similar to Francois's.  I'm writing a to-do-like
>> application, where each Task has a number of Events, consisting of
>> event.date and event.status.  I want to see how long it's been since the
>> Task was last completed.  So:

[snip]

>> This smells a bit.  I could separate part of completed_days_ago into
>> last_completed_on(date), but no matter how I slice it I still end up
>> mocking Task.cells for some routine.  Suggestions?
> 
> The next question is, why are we passing in the date? We want to how
> many days have passed since it was completed, not how many days
> between completion and some arbitrary date. 

Actually, we want the latter - but of course you couldn't tell that from 
my method name!  The tasks are repeating, and this app fills in a 
spreadsheet-like series of cells, one per day, with various colors, 
depending on how often you're completing a task.  So we actually need to 
check the task's status on any given day.

[..]
>     event_stub = Struct.new(:date,
> :status).new(Time.parse("2001-02-03"), "completed")
>     @task.stub!(:events).and_return([event_stub])
[..]

You don't mind stubbing the association proxy as a struct?  That was my 
real question; it seems to lead down a messy road of trying to stub 
every ActiveRecord method.  On the one hand, it seems like BDD says I 
should stub or mock any object other than the one under test.  On the 
other, since I'm still in the models realm, maybe it's okay to use 
fixtures to represent the model relations.  Or maybe there's a third way.

Jay


From lists-rspec at shopwatch.org  Sun Jan 21 18:13:46 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sun, 21 Jan 2007 18:13:46 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701211414t189c0df1md288cce6885c0a56@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>	<45B3D2D7.5050400@rubyforge.org>
	<57c63afe0701211414t189c0df1md288cce6885c0a56@mail.gmail.com>
Message-ID: <45B3F3AA.1080606@rubyforge.org>

David Chelimsky wrote:
> TDA says we shouldn't even ask the User for its Projects. It's the Law
> of Demeter that says we shouldn't ask the User for it's Projects and
> THEN ask the Projects whether they are active.

Oops, my bad, then!  So is it my misunderstanding of domain-driven 
design - especially the AGGREGATE pattern - or is TDA diametrically 
opposed to DDD here?  Can we get Alec Sharp and Eric Evans in a boxing ring?

Jay


From brian at yamabe.net  Sun Jan 21 18:32:20 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Sun, 21 Jan 2007 15:32:20 -0800
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
	<57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>
Message-ID: <9DB465C9-119C-4577-8896-EDA12E744709@yamabe.net>

David, thanks for making me think. If the bi-directional dependency  
is needed, would it be cleaned up with a "has_many :through" (An  
Assignment class as an example)?

> def self.find_active_projects_for_user(user)
>   return user.projects.active_projects
> end
>
> This bi-directional dependency seems problematic to me. I realize that
> rails makes it easy to do these, but again, just because its easy
> doesn't make it a good decision.
>
> This is one of those questions that can be enlightened by client need.
> Do clients of Projects ask for their users? Do clients of User ask for
> its Projects? If both answers are yes, then maybe the bi-directional
> deal makes sense. If only one is needed, then only one should be
> implemented.
>
> OK. I'll shut up now. Go Bears!
>
> David
>
>
>> ---Brian Yamabe
>>
>> On Jan 21, 2007, at 12:53 PM, Jay Levitt wrote:
>>
>>> Francois Beausoleil wrote:
>>>
>>>>
>>>> Adding #active_projects on User leads to a kind of namespace
>>>> pollution, no ?  Really, who knows what projects are active ?  The
>>>> Project class, or the User ?
>>>
>>> This seems to be a basic conflict between "Tell, don't ask" and
>>> domain-driven design.  DDD, as I understand it, says the Project  
>>> class
>>> should encapsulate all knowledge about projects; TDA says we  
>>> shouldn't
>>> ask for the project proxy.  Yes?
>>>
>>> Jay
>>>
>>> _______________________________________________
>>> 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


From dchelimsky at gmail.com  Sun Jan 21 18:51:19 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 17:51:19 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <45B3F3AA.1080606@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<57c63afe0701211414t189c0df1md288cce6885c0a56@mail.gmail.com>
	<45B3F3AA.1080606@rubyforge.org>
Message-ID: <57c63afe0701211551x783afc8am640c389785a55e81@mail.gmail.com>

On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> David Chelimsky wrote:
> > TDA says we shouldn't even ask the User for its Projects. It's the Law
> > of Demeter that says we shouldn't ask the User for it's Projects and
> > THEN ask the Projects whether they are active.
>
> Oops, my bad, then!  So is it my misunderstanding of domain-driven
> design - especially the AGGREGATE pattern - or is TDA diametrically
> opposed to DDD here?  Can we get Alec Sharp and Eric Evans in a boxing ring?

No - I think that's right. My read is that aggregate looks like the
antithesis of LoD, BUT, there is an important distinction to be made.
In DDD, the goal is ubiquitous language built around the business
domain. If the business domain says that there are Leagues that have
Conferences that have Teams that have Players, then those
relationships are very unlikely to change, and therefore minimize the
risks assoicated with LoD violations.

The real problem w/ LoD is when you see stuff like:

  player = team.jersey_number_to_player_map[17]

as opposed to

  player = team.find_player_by_jersey_number[17]

Even though it's not obvious, the first is an LoD violation. Here are
the same statements in java:

  player = team.getJerseyNumberToPlayerMap().get("17")

as opposed to

  player = team.findPlayerByJerseyNumber("17")

The fact that team stores players in a map is none of anybody's
business. The fact that it stores players is arguably a domain concept
that makes sense.

Make sense?

David


>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Sun Jan 21 18:58:16 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 21 Jan 2007 17:58:16 -0600
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <9DB465C9-119C-4577-8896-EDA12E744709@yamabe.net>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
	<57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>
	<9DB465C9-119C-4577-8896-EDA12E744709@yamabe.net>
Message-ID: <57c63afe0701211558t2127490bw2f5146a4821ca5c1@mail.gmail.com>

On 1/21/07, Brian Yamabe <brian at yamabe.net> wrote:
> David, thanks for making me think.

Back atcha.

> If the bi-directional dependency
> is needed, would it be cleaned up with a "has_many :through" (An
> Assignment class as an example)?

That happens to work in a lot of cases because rails supports it. But
bear in mind that :through relationships weren't supported in 1.0.
They came later because the problem of a particular type of
relationship came up enough to do something about it in the framework.

I'm sure that sooner or later you're going to come up w/ a model that
isn't supported directly by rails, and then you need to think about
all this stuff.

<soapbox>
Also, while this may be far from most people's minds, letting all of
this AR goodness seep out past your model classes binds your
application to AR. There are other persistence frameworks, like OG,
that are promising. If you decided to give on a try, you'd have a much
easier time if you encapsulate all of that goodness inside your models
and expose methods that are framework neutral.

Not saying that you shouldn't exploit AR goodness. Just that
localizing the exploitation makes things more flexible in the long
run.
</soapbox>

>
> > def self.find_active_projects_for_user(user)
> >   return user.projects.active_projects
> > end
> >
> > This bi-directional dependency seems problematic to me. I realize that
> > rails makes it easy to do these, but again, just because its easy
> > doesn't make it a good decision.
> >
> > This is one of those questions that can be enlightened by client need.
> > Do clients of Projects ask for their users? Do clients of User ask for
> > its Projects? If both answers are yes, then maybe the bi-directional
> > deal makes sense. If only one is needed, then only one should be
> > implemented.
> >
> > OK. I'll shut up now. Go Bears!
> >
> > David
> >
> >
> >> ---Brian Yamabe
> >>
> >> On Jan 21, 2007, at 12:53 PM, Jay Levitt wrote:
> >>
> >>> Francois Beausoleil wrote:
> >>>
> >>>>
> >>>> Adding #active_projects on User leads to a kind of namespace
> >>>> pollution, no ?  Really, who knows what projects are active ?  The
> >>>> Project class, or the User ?
> >>>
> >>> This seems to be a basic conflict between "Tell, don't ask" and
> >>> domain-driven design.  DDD, as I understand it, says the Project
> >>> class
> >>> should encapsulate all knowledge about projects; TDA says we
> >>> shouldn't
> >>> ask for the project proxy.  Yes?
> >>>
> >>> Jay
> >>>
> >>> _______________________________________________
> >>> 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
>

From lists-rspec at shopwatch.org  Sun Jan 21 19:00:43 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Sun, 21 Jan 2007 19:00:43 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701211551x783afc8am640c389785a55e81@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>	<45B3D2D7.5050400@rubyforge.org>	<57c63afe0701211414t189c0df1md288cce6885c0a56@mail.gmail.com>	<45B3F3AA.1080606@rubyforge.org>
	<57c63afe0701211551x783afc8am640c389785a55e81@mail.gmail.com>
Message-ID: <45B3FEAB.9060201@rubyforge.org>

David Chelimsky wrote:
> The fact that team stores players in a map is none of anybody's
> business. The fact that it stores players is arguably a domain concept
> that makes sense.

Great distinction... thanks.

Jay


From francois.beausoleil at gmail.com  Mon Jan 22 09:46:36 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Mon, 22 Jan 2007 09:46:36 -0500
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701211558t2127490bw2f5146a4821ca5c1@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
	<57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>
	<9DB465C9-119C-4577-8896-EDA12E744709@yamabe.net>
	<57c63afe0701211558t2127490bw2f5146a4821ca5c1@mail.gmail.com>
Message-ID: <41d5fadf0701220646y119fcf73wf7ffd72307ba1e22@mail.gmail.com>

Hi all,

2007/1/21, David Chelimsky <dchelimsky at gmail.com>:
> <soapbox>
> Also, while this may be far from most people's minds, letting all of
> this AR goodness seep out past your model classes binds your
> application to AR. There are other persistence frameworks, like OG,
> that are promising. If you decided to give on a try, you'd have a much
> easier time if you encapsulate all of that goodness inside your models
> and expose methods that are framework neutral.
>
> Not saying that you shouldn't exploit AR goodness. Just that
> localizing the exploitation makes things more flexible in the long
> run.
> </soapbox>

So in essence, you are arguing for current_user.active_projects
instead of anything else, right :)

In my application, projects exist only in relationship to users.  And
users access projects through a Role.  That wasn't relevant in the
initial discussions, so I didn't speak about it.

Thanks all !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From brian at yamabe.net  Mon Jan 22 10:58:44 2007
From: brian at yamabe.net (Brian Yamabe)
Date: Mon, 22 Jan 2007 07:58:44 -0800
Subject: [rspec-users] Collection proxies need to be stubbed ?
In-Reply-To: <57c63afe0701211558t2127490bw2f5146a4821ca5c1@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701210337h27416165sebb7fa9f23743a50@mail.gmail.com>
	<41d5fadf0701211242n68928cedq36329b5f23c112d6@mail.gmail.com>
	<45B3D2D7.5050400@rubyforge.org>
	<40DABC99-130D-48D8-946B-B82F9FD28D8F@yamabe.net>
	<57c63afe0701211423r5fde1438ib5ca73505391647d@mail.gmail.com>
	<9DB465C9-119C-4577-8896-EDA12E744709@yamabe.net>
	<57c63afe0701211558t2127490bw2f5146a4821ca5c1@mail.gmail.com>
Message-ID: <ADB43E3C-61CF-44F1-8897-012E54A8677E@yamabe.net>

> I'm sure that sooner or later you're going to come up w/ a model that
> isn't supported directly by rails, and then you need to think about
> all this stuff.
>
> <soapbox>
> Also, while this may be far from most people's minds, letting all of
> this AR goodness seep out past your model classes binds your
> application to AR. There are other persistence frameworks, like OG,
> that are promising. If you decided to give on a try, you'd have a much
> easier time if you encapsulate all of that goodness inside your models
> and expose methods that are framework neutral.
>
> Not saying that you shouldn't exploit AR goodness. Just that
> localizing the exploitation makes things more flexible in the long
> run.
> </soapbox>

Even before I got your <soapbox/> I was thinking about what to do if
this weren't using Rails. I still like the notion of an intermediary;  
a class
that manages the relationship between Users and Projects. This would
minimize any coupling between the User and Project classes and allow
for a richer set of relationships to be tracked (inactive project  
members,
inactive user projects, etc.). The intermediary class could either  
exploit the
"AR goodness" or not. I could hear someone crying, "YAGNI," but the
decoupling of User and Project is the main concern.  The flexibility  
comes
almost as a side effect.

---Brian Yamabe


From francois.beausoleil at gmail.com  Mon Jan 22 11:25:26 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Mon, 22 Jan 2007 11:25:26 -0500
Subject: [rspec-users] Observed models cause failures with DRBSpec ?
Message-ID: <41d5fadf0701220825v266a97c5l4a02c99cbefb34b8@mail.gmail.com>

Hi all !

I don't believe I am the only one using model observers, right ?
Well, here's a nice one:

$ drbspec spec

.

Finished in 0.125322 seconds

1 specification, 0 failures
$ drbspec spec

F

1)
'A user with an inactive and an active project should be able to
return active projects only' FAILED
[] should == [#<Spec::Mocks::Mock:0xb75590bc @name="active", @options={}>]
./spec/models/user_spec.rb:11:
script/rails_spec_server:17:in `run'
script/rails_spec_server:38:

Finished in 0.024455 seconds

1 specification, 1 failure



# spec/models/user_spec.rb
require File.dirname(__FILE__) + "/../spec_helper"

context "A user with an inactive and an active project" do
  setup do
    @user = User.new
  end

  specify "should be able to return active projects only" do
    @active = mock("active")
    Project.stub!(:active).and_return([@active])
    @user.active_projects.should == [@active]
  end
end


# app/models/user.rb
class User < ActiveRecord::Base
  has_many :projects, :foreign_key => "owner_id", :order => "projects.title"

  def active_projects(force=false)
    @active_projects = nil if force
    @active_projects ||= self.projects.active
  end
end

The observer is loaded in config/environment.rb like this:
Rails::Initializer.run do |config|
  config.active_record.observers = :user_observer
end

There's nothing very interesting about this, except that if I comment
the user observer in the configuration, I can run drbspec spec
multiple times in a row without failures.  If I add the observer, I
can run drbspec without failures once, then I get failures like the
one above for ever after.

Can anyone confirm / deny this behavior ?

Thanks !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From aslak.hellesoy at gmail.com  Mon Jan 22 11:45:23 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Mon, 22 Jan 2007 17:45:23 +0100
Subject: [rspec-users] Observed models cause failures with DRBSpec ?
In-Reply-To: <41d5fadf0701220825v266a97c5l4a02c99cbefb34b8@mail.gmail.com>
References: <41d5fadf0701220825v266a97c5l4a02c99cbefb34b8@mail.gmail.com>
Message-ID: <8d961d900701220845i7a6bc9e8v4516e12c87b37679@mail.gmail.com>

On 1/22/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> Hi all !
>
> I don't believe I am the only one using model observers, right ?
> Well, here's a nice one:
>
> $ drbspec spec
>
> .
>
> Finished in 0.125322 seconds
>
> 1 specification, 0 failures
> $ drbspec spec
>
> F
>
> 1)
> 'A user with an inactive and an active project should be able to
> return active projects only' FAILED
> [] should == [#<Spec::Mocks::Mock:0xb75590bc @name="active", @options={}>]
> ./spec/models/user_spec.rb:11:
> script/rails_spec_server:17:in `run'
> script/rails_spec_server:38:
>
> Finished in 0.024455 seconds
>
> 1 specification, 1 failure
>
>
>
> # spec/models/user_spec.rb
> require File.dirname(__FILE__) + "/../spec_helper"
>
> context "A user with an inactive and an active project" do
>   setup do
>     @user = User.new
>   end
>
>   specify "should be able to return active projects only" do
>     @active = mock("active")
>     Project.stub!(:active).and_return([@active])
>     @user.active_projects.should == [@active]
>   end
> end
>
>
> # app/models/user.rb
> class User < ActiveRecord::Base
>   has_many :projects, :foreign_key => "owner_id", :order => "projects.title"
>
>   def active_projects(force=false)
>     @active_projects = nil if force
>     @active_projects ||= self.projects.active
>   end
> end
>
> The observer is loaded in config/environment.rb like this:
> Rails::Initializer.run do |config|
>   config.active_record.observers = :user_observer
> end
>
> There's nothing very interesting about this, except that if I comment
> the user observer in the configuration, I can run drbspec spec
> multiple times in a row without failures.  If I add the observer, I
> can run drbspec without failures once, then I get failures like the
> one above for ever after.
>
> Can anyone confirm / deny this behavior ?
>

I haven't tried to reproduce this, but it sounds like
rails_spec_server is not reinitialising the rails environment
properly. Essentially, we're trying to make rails_spec_server behave
the same way as Rails in development mode - reload everything on each
HTTP request.

Take a peek inside rails_spec_server - perhaps we're not doing this
correctly? If it works with Rails in development mode (can you check
that please), it should work with rails_spec_server too.

If you can confirm that reloading happens ok in dev mode (hitting with
your browser), but not with rails_spec_server, then please submit a
bug report, along with useful versions/revision numbers etc.

By the way, drbspec will be replaced with spec --drb in the next RSpec release.

Not much help, but maybe a nudge in the right direction for more debugging.
Aslak

> Thanks !
> --
> Fran?ois Beausoleil
> http://blog.teksol.info/
> http://piston.rubyforge.org/
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

From pergesu at gmail.com  Tue Jan 23 21:24:51 2007
From: pergesu at gmail.com (Pat Maddox)
Date: Tue, 23 Jan 2007 19:24:51 -0700
Subject: [rspec-users] A spec where interaction-based testing breaks down...
	(at least for now)
Message-ID: <810a540e0701231824w35211fd1sf75aaacbda9d4c4f@mail.gmail.com>

Here are the specs:
context "Finding all the stylesheets available to a company" do
  setup do
    @mock_company = mock("company")
    @mock_company.stub!(:to_param).and_return "1"
    Stylesheet.stub!(:find)
  end

  def do_find
    Stylesheet.find_available_to @mock_company
  end

  specify "should convert the company into a parameter" do
    @mock_company.should_receive(:to_param).and_return "1"
    do_find
  end

  specify "should find stylesheets belonging to only the company or to
nobody" do
    Stylesheet.should_receive(:find).with(:all, :conditions =>
["company_id IS NULL OR company_id=?", "1"])
    do_find
  end
end

And here's the code:
class Stylesheet < ActiveRecord::Base
  def self.find_available_to(company)
    find :all, :conditions => ["company_id IS NULL OR company_id=?",
company.to_param]
  end
end

You can see that the implementation is duplicated.  In fact, I wrote
the implementation in the spec and then basically just copied it over.

That bugs me.

The obvious way to fix it would be to create a couple stylesheet
entries in the database, run the query, and see if it matches the
expected results.

The only benefit of the current approach (that I can see) is that the
behavior is explicit.  If you've just got the results, you have to
mentally link them to the query.

Anyway I'm just sort of stuck on this one.  I think there's a much
better way that is evading me.

Pat

From dchelimsky at gmail.com  Wed Jan 24 01:26:19 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 24 Jan 2007 00:26:19 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B3F29E.1080009@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B34DAA.7040004@e-tobi.net>
	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>
	<45B35B67.2070402@e-tobi.net>
	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
	<45B36382.2030503@e-tobi.net> <45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
Message-ID: <57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>

On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> David Chelimsky wrote:
> > On 1/21/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> >> I've got a question similar to Francois's.  I'm writing a to-do-like
> >> application, where each Task has a number of Events, consisting of
> >> event.date and event.status.  I want to see how long it's been since the
> >> Task was last completed.  So:
>
> [snip]
>
> >> This smells a bit.  I could separate part of completed_days_ago into
> >> last_completed_on(date), but no matter how I slice it I still end up
> >> mocking Task.cells for some routine.  Suggestions?
> >
> > The next question is, why are we passing in the date? We want to how
> > many days have passed since it was completed, not how many days
> > between completion and some arbitrary date.
>
> Actually, we want the latter - but of course you couldn't tell that from
> my method name!  The tasks are repeating, and this app fills in a
> spreadsheet-like series of cells, one per day, with various colors,
> depending on how often you're completing a task.  So we actually need to
> check the task's status on any given day.
>
> [..]
> >     event_stub = Struct.new(:date,
> > :status).new(Time.parse("2001-02-03"), "completed")
> >     @task.stub!(:events).and_return([event_stub])
> [..]
>
> You don't mind stubbing the association proxy as a struct?  That was my
> real question; it seems to lead down a messy road of trying to stub
> every ActiveRecord method.  On the one hand, it seems like BDD says I
> should stub or mock any object other than the one under test.  On the
> other, since I'm still in the models realm, maybe it's okay to use
> fixtures to represent the model relations.  Or maybe there's a third way.

I'm kind of on the fence on this topic. I've been of the mindset that
since models are so tightly bound to AR that it's more trouble than
its worth to try and break that dependency. Because I mock models in
all of my controller and view specs, I'm only hitting the db for the
things that are interesting about models.

Recently, I read a post from Jay Fields where he talks about what
amounts to mocking out the loading of a file. So if you want to spec
that User should validate presence of email, you do this:

User.should_receive(:validates_presence_of).with(:email)
load "${RAILS_ROOT}/app/models/user.rb"

Odd, huh? It wouldn't be if the API for AR was like this:

@model_registry.should_receive(:validate_presence_of).with(User, :email)
User.register(@model_registry)

Essentially, loading the file IS the entry point to the AR API.
Interesting point of view, no?

I like two things about this:
clear specs
fast specs

I can write all my validation specs in two lines each. No models to
create. No database to connect to. Pretty awesome.

So, no, I don't mind mocking the proxy in model specs, though I would
always try to avoid that seeping out to the controllers and views.

That helpful?

Cheers,
David


>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From lists-rspec at shopwatch.org  Wed Jan 24 08:45:55 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Wed, 24 Jan 2007 08:45:55 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>	<45B36382.2030503@e-tobi.net>
	<45B3D11F.90703@rubyforge.org>	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
Message-ID: <45B76313.4000305@rubyforge.org>

David Chelimsky wrote:
> I'm kind of on the fence on this topic. I've been of the mindset that
> since models are so tightly bound to AR that it's more trouble than
> its worth to try and break that dependency. Because I mock models in
> all of my controller and view specs, I'm only hitting the db for the
> things that are interesting about models.

I think I'm of the same mind; I just wanted to make sure I wasn't taking 
the easy way out.

I was thinking... a great addition to rSpec someday - though a major 
project in its own right - might be a pre-stubbed version of 
ActiveRecord with fixtures.  Then again, that might be just as complex 
as the real AR, and maybe the solution is to just use an in-memory 
database when testing.
> 
> Recently, I read a post from Jay Fields where he talks about what
> amounts to mocking out the loading of a file. So if you want to spec
> that User should validate presence of email, you do this:
> 
> User.should_receive(:validates_presence_of).with(:email)
> load "${RAILS_ROOT}/app/models/user.rb"
> 
> Odd, huh? It wouldn't be if the API for AR was like this:
> 
> @model_registry.should_receive(:validate_presence_of).with(User, :email)
> User.register(@model_registry)
> 
> Essentially, loading the file IS the entry point to the AR API.
> Interesting point of view, no?

Wow - I never thought of it like that.  I think (as a new convert to 
mocks) that type of mock is a little too specific and un-DRY for my 
taste, though; it feels like specifying that "line 6 of method foo 
should be an if statement".  There's double-entry bookkeeping, and then 
there's keeping two sets of books.

> So, no, I don't mind mocking the proxy in model specs, though I would
> always try to avoid that seeping out to the controllers and views.
> 
> That helpful?

Very!


From lists-rspec at shopwatch.org  Wed Jan 24 09:08:10 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Wed, 24 Jan 2007 09:08:10 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B76313.4000305@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B34DAA.7040004@e-tobi.net>	<57c63afe0701210346v69792d9bgdca54afef008efc6@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>	<45B36382.2030503@e-tobi.net>	<45B3D11F.90703@rubyforge.org>	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>	<45B3F29E.1080009@rubyforge.org>	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org>
Message-ID: <45B7684A.7030404@rubyforge.org>

Jay Levitt wrote:
> David Chelimsky wrote:
>> User.should_receive(:validates_presence_of).with(:email)
> Wow - I never thought of it like that.  I think (as a new convert to 
> mocks) that type of mock is a little too specific and un-DRY for my 
> taste, though; it feels like specifying that "line 6 of method foo 
> should be an if statement".  There's double-entry bookkeeping, and then 
> there's keeping two sets of books.

While I'm out on this limb, I'm gonna say that, given the way Rails 
validations are structured, it's better to use state-based testing than 
interaction-based testing!

validates_presence_of is a bad example, because the two methods are 
practically interchangeable.  But consider a validation that uses a 
regex to verify a legal IP address.  Do you want your specs to repeat 
the regex, or do you want to test various legal and illegal IP-address 
strings and see what breaks?  To me, it's the second one that's actually 
testing the behavior of the application.


Jay


From dchelimsky at gmail.com  Wed Jan 24 10:15:07 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 24 Jan 2007 09:15:07 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B7684A.7030404@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B35B67.2070402@e-tobi.net>
	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>
	<45B36382.2030503@e-tobi.net> <45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
Message-ID: <57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>

On 1/24/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> Jay Levitt wrote:
> > David Chelimsky wrote:
> >> User.should_receive(:validates_presence_of).with(:email)
> > Wow - I never thought of it like that.  I think (as a new convert to
> > mocks) that type of mock is a little too specific and un-DRY for my
> > taste, though; it feels like specifying that "line 6 of method foo
> > should be an if statement".  There's double-entry bookkeeping, and then
> > there's keeping two sets of books.
>
> While I'm out on this limb, I'm gonna say that, given the way Rails
> validations are structured, it's better to use state-based testing than
> interaction-based testing!
>
> validates_presence_of is a bad example, because the two methods are
> practically interchangeable.  But consider a validation that uses a
> regex to verify a legal IP address.  Do you want your specs to repeat
> the regex, or do you want to test various legal and illegal IP-address
> strings and see what breaks?  To me, it's the second one that's actually
> testing the behavior of the application.

Another view would be that the Regexp is a separate component that
you'd want to test separately from the use of that component. So the
test that your model validates_format_of using a Regexp uses the right
one and then have other tests just for that Regexp.

WDYT?

>
>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From lists-rspec at shopwatch.org  Wed Jan 24 10:54:18 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Wed, 24 Jan 2007 10:54:18 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>	<45B36382.2030503@e-tobi.net>
	<45B3D11F.90703@rubyforge.org>	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>	<45B3F29E.1080009@rubyforge.org>	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>	<45B76313.4000305@rubyforge.org>
	<45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
Message-ID: <45B7812A.1090506@rubyforge.org>

David Chelimsky wrote:
> On 1/24/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
>> validates_presence_of is a bad example, because the two methods are
>> practically interchangeable.  But consider a validation that uses a
>> regex to verify a legal IP address.  Do you want your specs to repeat
>> the regex, or do you want to test various legal and illegal IP-address
>> strings and see what breaks?  To me, it's the second one that's actually
>> testing the behavior of the application.
> 
> Another view would be that the Regexp is a separate component that
> you'd want to test separately from the use of that component. So the
> test that your model validates_format_of using a Regexp uses the right
> one and then have other tests just for that Regexp.
> 
> WDYT?

Hmm.. I can see your argument, but I'm not convinced.  Perhaps more 
experience will convince me :)

Jay


From dchelimsky at gmail.com  Wed Jan 24 11:08:52 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 24 Jan 2007 10:08:52 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45B7812A.1090506@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B36382.2030503@e-tobi.net> <45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
	<45B7812A.1090506@rubyforge.org>
Message-ID: <57c63afe0701240808q16fa0d3h59b48970ae6a559a@mail.gmail.com>

On 1/24/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> David Chelimsky wrote:
> > On 1/24/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> >> validates_presence_of is a bad example, because the two methods are
> >> practically interchangeable.  But consider a validation that uses a
> >> regex to verify a legal IP address.  Do you want your specs to repeat
> >> the regex, or do you want to test various legal and illegal IP-address
> >> strings and see what breaks?  To me, it's the second one that's actually
> >> testing the behavior of the application.
> >
> > Another view would be that the Regexp is a separate component that
> > you'd want to test separately from the use of that component. So the
> > test that your model validates_format_of using a Regexp uses the right
> > one and then have other tests just for that Regexp.
> >
> > WDYT?
>
> Hmm.. I can see your argument, but I'm not convinced.  Perhaps more
> experience will convince me :)

FWIW, I'm not convinced either. It was just an idea for the purpose of
discussion.


>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From jodi at nnovation.ca  Wed Jan 24 14:39:54 2007
From: jodi at nnovation.ca (Jodi Showers)
Date: Wed, 24 Jan 2007 14:39:54 -0500
Subject: [rspec-users] hello! first post + context_setup
Message-ID: <86659E25-EEAE-4B94-8FE7-9C126B004B86@nNovation.ca>

Greetings all.

I'm looking over a 'greenfield' pasture, and enjoying the TDD(BDD)  
process of creation.

I'm currently rspec'ing a finite state machine  
(acts_as_state_machine). As such, I need to persist the model through  
'specifies' within the context - from top to bottom.

I have bumped into a couple of probs:
	1. context_setup to define the doesn't permit access to fixtures.  
"You have a nil object when you didn't expect it!"
         2. transactions are defined around each specify? For this  
reason my specifies are very repetitive, encompassing all state  
movement for all previous contexts.
         3. since 2 seems to be the case, it probably doesn't matter  
- but is execution order  guaranteed for specifiy's ?

Thanx to the core team and contributors for rspec. The "diameter of  
my smile" is wide.

Cheers,
Jodi
General Partner
The nNovation Group inc.
www.nnovation.ca/blog



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070124/f3e7c934/attachment.html 

From dchelimsky at gmail.com  Wed Jan 24 15:03:26 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 24 Jan 2007 14:03:26 -0600
Subject: [rspec-users] hello! first post + context_setup
In-Reply-To: <86659E25-EEAE-4B94-8FE7-9C126B004B86@nNovation.ca>
References: <86659E25-EEAE-4B94-8FE7-9C126B004B86@nNovation.ca>
Message-ID: <57c63afe0701241203n668a22e9ud82c8163eab3a532@mail.gmail.com>

On 1/24/07, Jodi Showers <jodi at nnovation.ca> wrote:
>
> Greetings all.
> I'm looking over a 'greenfield' pasture, and enjoying the TDD(BDD) process
> of creation.
>

Consider yourself lucky!!!


> I'm currently rspec'ing a finite state machine (acts_as_state_machine). As
> such, I need to persist the model through 'specifies' within the context -
> from top to bottom.
>
> I have bumped into a couple of probs:
> 1. context_setup to define the doesn't permit access to fixtures. "You
> have a nil object when you didn't expect it!"
>

Please submit a bug on this to the tracker:

http://rubyforge.org/tracker/?group_id=797

        2. transactions are defined around each specify? For this reason my
> specifies are very repetitive, encompassing all state movement for all
> previous contexts.
>         3. since 2 seems to be the case, it probably doesn't matter - but
> is execution order  guaranteed for specifiy's ?
>

Order is not guaranteed as in a commitment from the RSpec team to keep it
that way. We believe that each specify block should be completely
independent from the next, so order should not be related.

That said, the specify method generates a Specification object which is
added to a Ruby Array held by the Context object (which is generated by the
context method). Given that implementation, the specs will run in order. And
it is HIGHLY unlikely that this would ever change. Depending on that,
however, is a risk that you would have to understand and choose to take.

Thinking about the particular problem at hand, imagine that we solved the
transaction problem and allowed you to declare the entire context to be
wrapped in a single transaction and guaranteed order. Now if spec 1 fails,
you can't really trust the outcome of spec 2 (because it depends on state
leftover from spec 1). This seems to me to be no different than having a
single specify block that takes you through all the states. In fact, it
strikes me as worse than the single spec block because you can't trust the
other outcomes.

So I'd recommend just writing one spec block.


Thanx to the core team and contributors for rspec. The "diameter of my
> smile" is wide.
>

Thanks for the feedback.
David



Cheers,
> Jodi
> General Partner
> The nNovation Group inc.
> www.nnovation.ca/blog
> [image: ranting o the technology and business of software]<http://feeds.feedburner.com/on-innovation>
>
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070124/fd12f76e/attachment.html 

From aslak.hellesoy at gmail.com  Wed Jan 24 16:47:49 2007
From: aslak.hellesoy at gmail.com (aslak hellesoy)
Date: Wed, 24 Jan 2007 22:47:49 +0100
Subject: [rspec-users] A spec where interaction-based testing breaks
	down... (at least for now)
In-Reply-To: <810a540e0701231824w35211fd1sf75aaacbda9d4c4f@mail.gmail.com>
References: <810a540e0701231824w35211fd1sf75aaacbda9d4c4f@mail.gmail.com>
Message-ID: <8d961d900701241347m79ed0a89waa59b1de77df8973@mail.gmail.com>

On 1/24/07, Pat Maddox <pergesu at gmail.com> wrote:
> Here are the specs:
> context "Finding all the stylesheets available to a company" do
>   setup do
>     @mock_company = mock("company")
>     @mock_company.stub!(:to_param).and_return "1"
>     Stylesheet.stub!(:find)
>   end
>
>   def do_find
>     Stylesheet.find_available_to @mock_company
>   end
>
>   specify "should convert the company into a parameter" do
>     @mock_company.should_receive(:to_param).and_return "1"
>     do_find
>   end
>
>   specify "should find stylesheets belonging to only the company or to
> nobody" do
>     Stylesheet.should_receive(:find).with(:all, :conditions =>
> ["company_id IS NULL OR company_id=?", "1"])
>     do_find
>   end
> end
>
> And here's the code:
> class Stylesheet < ActiveRecord::Base
>   def self.find_available_to(company)
>     find :all, :conditions => ["company_id IS NULL OR company_id=?",
> company.to_param]
>   end
> end
>
> You can see that the implementation is duplicated.  In fact, I wrote
> the implementation in the spec and then basically just copied it over.
>
> That bugs me.
>
> The obvious way to fix it would be to create a couple stylesheet
> entries in the database, run the query, and see if it matches the
> expected results.
>
> The only benefit of the current approach (that I can see) is that the
> behavior is explicit.  If you've just got the results, you have to
> mentally link them to the query.
>
> Anyway I'm just sort of stuck on this one.  I think there's a much
> better way that is evading me.
>

I wouldn't use mocks for model specs. I'd test it against the real database.

Aslak

> Pat
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From listaccount at e-tobi.net  Wed Jan 24 19:03:02 2007
From: listaccount at e-tobi.net (Tobias Grimm)
Date: Thu, 25 Jan 2007 01:03:02 +0100
Subject: [rspec-users] mocking methods that receive blocks
In-Reply-To: <45B2C0EC.1030609@e-tobi.net>
References: <45B2C0EC.1030609@e-tobi.net>
Message-ID: <45B7F3B6.80608@e-tobi.net>

Hi!

Maybe my last (but real life) example was too boring - what about this one?

see end of mail or http://pastie.caboo.se/35443

So basically what I want to do, is mock a method that receives
parameters and a block, but I still need to execute the passed block in
order to test that it does what it is expected to do.

With a different design, I could avoid this problem. Instead of passing
blocks around I could e.g. use some kind of Strategy class.
I was just wondering if this can be done somehow with RSpec's mocking
capabilities. If not, maybe RSpec can be extended to make it possible.

bye,

Tobias

Sorry, I hold no degree in Trek physics !!! :-)

class Antimatter
    def stabilize
    end
end

class TransWarpEngine
    def TransWarpEngine.initiate(antimatter)
       create_quantum_singularity
       yield antimatter
    end
end

class BorgCube
    def travel_with_light_speed(antimatter)
        TransWarpEngine.initiate(antimatter) do |antimatter|
            antimatter.stabilize
        end
    end
end

context "The Borg Cube, when travelling with light speed" do
    setup do
        @borg_cube = BorgCube.new
    end

    specify "should initiate the TransWarpEngine with the given
antimatter" do
        antimatter = mock(Antimatter)
        TransWarpEngine.should_receive(:initiate).with(antimatter)
        @borg_cube.travel_with_light_speed(antimatter)
    end

    specify "should tell the TransWarpEngine, that the antimatter should
be stabilized" do
        antimatter = mock(Antimatter)
        antimatter.should_receive(:stabilize)

        #
        # Put in here the part, that I don't know how to do :-)
        # I must mock or stub TransWarpEngine.initiate(), because I
can't risk
        # to create a quantum singularity when running RSpec.
        # But i also must test, that my little BorgCube receives the correct
        # instructions to stabilize the antimatter, otherwise my
customer won't
        # be very happy, when she tries out the BorgCube the first time
and it
        # simply detonates because of unstabilized antimatter.
        #

        @borg_cube.travel_with_light_speed(antimatter)
    end
end


From jodi at nnovation.ca  Wed Jan 24 20:52:28 2007
From: jodi at nnovation.ca (Jodi Showers)
Date: Wed, 24 Jan 2007 20:52:28 -0500
Subject: [rspec-users] hello! first post + context_setup
In-Reply-To: <57c63afe0701241203n668a22e9ud82c8163eab3a532@mail.gmail.com>
References: <86659E25-EEAE-4B94-8FE7-9C126B004B86@nNovation.ca>
	<57c63afe0701241203n668a22e9ud82c8163eab3a532@mail.gmail.com>
Message-ID: <A52446E0-F3CD-4176-8022-2193A7012844@nNovation.ca>

On 24-Jan-07, at 3:03 PM, David Chelimsky wrote:
> I have bumped into a couple of probs:
> 	1. context_setup to define the doesn't permit access to fixtures.  
> "You have a nil object when you didn't expect it!"
>
> Please submit a bug on this to the tracker:
>
> http://rubyforge.org/tracker/?group_id=797
>

>         2. transactions are defined around each specify? For this  
> reason my specifies are very repetitive, encompassing all state  
> movement for all previous contexts.
>         3. since 2 seems to be the case, it probably doesn't matter  
> - but is execution order  guaranteed for specifiy's ?
>
> Order is not guaranteed as in a commitment from the RSpec team to  
> keep it that way. We believe that each specify block should be  
> completely independent from the next, so order should not be related.
>
> That said, the specify method generates a Specification object  
> which is added to a Ruby Array held by the Context object (which is  
> generated by the context method). Given that implementation, the  
> specs will run in order. And it is HIGHLY unlikely that this would  
> ever change. Depending on that, however, is a risk that you would  
> have to understand and choose to take.
>
> Thinking about the particular problem at hand, imagine that we  
> solved the transaction problem and allowed you to declare the  
> entire context to be wrapped in a single transaction and guaranteed  
> order. Now if spec 1 fails, you can't really trust the outcome of  
> spec 2 (because it depends on state leftover from spec 1). This  
> seems to me to be no different than having a single specify block  
> that takes you through all the states. In fact, it strikes me as  
> worse than the single spec block because you can't trust the other  
> outcomes.
>
> So I'd recommend just writing one spec block.
>

I'm in agreement David.

I'm facing a bunch of code duplication (that doesn't feel safe), so  
I'd perhaps I'll work out a means to iterate over the events and  
expected states.

I'll file that bug tomorrow.

Jodi


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070124/19b46816/attachment-0001.html 

From listaccount at e-tobi.net  Thu Jan 25 06:36:16 2007
From: listaccount at e-tobi.net (Tobi)
Date: Thu, 25 Jan 2007 12:36:16 +0100
Subject: [rspec-users] mocking methods that receive blocks - back to mocking
	Thread again
Message-ID: <45B89630.7000701@e-tobi.net>

Hi again,

my last example wasn't good either. It really makes only sense in the 
context of using Thread.

I finally came up with this code, that should now test all important 
aspects of running a task in a background thread (to keep it simple, I 
left out the exception handling).

see http://pastie.caboo.se/35559 or end of mail.

It seems, TaskRunner is now very well covered by the specs - if you find 
a way to break TaskRunner without breaking the specs, please let me know.
Unfortunately the specifications seem to be a little bit hard to read 
now. Any suggestions for improvements?

bye,

Tobias

class TaskRunner
    def initialize
        @is_running = false
    end
  
    def running?
        return @is_running
    end
  
    def execute(task)
        @is_running = true
        Thread.new(task) do |myTask|
            myTask.run
            @is_running = false
        end
    end
end

context "The TaskRunner" do
    setup do
        @runner = TaskRunner.new
        @task = mock('SampleTask')
        @task.stub!(:run)
        Thread.stub!(:new)
    end

    specify "should not run a task in the main thread" do
        @task.should_not_receive(:run)
        @runner.execute(@task)
    end

    specify "should run a task in a new thread" do
        Thread.should_receive(:new).with(@task).and_yield(@task)
        @task.should_receive(:run)
        @runner.execute(@task)
    end
   
    specify "should be running while a task is beeing executed" do
        Thread.should_receive(:new).with(@task).and_yield(@task)
        @task.should_receive(:run) do
            @runner.should_be_running
        end
        @runner.execute(@task)
    end

    specify "should not be running after the task has been executed" do
        Thread.should_receive(:new).with(@task).and_return do |task, block|
            block.call(task)
            @runner.should_not_be_running
        end
        @runner.execute(@task)
    end
end


From papipo at gmail.com  Thu Jan 25 08:30:02 2007
From: papipo at gmail.com (=?ISO-8859-1?Q?Rodrigo_Alvarez_Fern=E1ndez?=)
Date: Thu, 25 Jan 2007 14:30:02 +0100
Subject: [rspec-users] Testing RJS select method
Message-ID: <6d2bdda0701250530p5838e132mbf38afc9c6a5a113@mail.gmail.com>

Hi

How can I test this kind of RJS code?

page.select '#id .class'.each do |element|
  element.removeClassName('whatever')
end

I know that should_have_rjs admits a block, but i don't know exactly
which is its purpose.
And I can test that the "select" part is beign called, but I don't
know hot to test the  "removeClassName" part.

Thanks in advance.

From yrashk at gmail.com  Thu Jan 25 12:52:04 2007
From: yrashk at gmail.com (Yurii Rashkovskii)
Date: Thu, 25 Jan 2007 19:52:04 +0200
Subject: [rspec-users] specify_negatively
Message-ID: <9A9FE63F-3735-4595-9B7C-B20F1F4ECB93@gmail.com>

Hello,

I have played a bit on a negative specification idea:

http://rashkovskii.com/articles/2007/1/25/rspec-specify_negatively

What do you think about it?

Yurii Rashkovskii
Technology Director
Railsware.com




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070125/12c4f473/attachment.html 

From martin.emde at gmail.com  Thu Jan 25 15:02:42 2007
From: martin.emde at gmail.com (Martin Emde)
Date: Thu, 25 Jan 2007 13:02:42 -0700
Subject: [rspec-users] A spec where interaction-based testing breaks
	down... (at least for now)
In-Reply-To: <8d961d900701241347m79ed0a89waa59b1de77df8973@mail.gmail.com>
References: <810a540e0701231824w35211fd1sf75aaacbda9d4c4f@mail.gmail.com>
	<8d961d900701241347m79ed0a89waa59b1de77df8973@mail.gmail.com>
Message-ID: <145359ad0701251202w4f6a67fbuee039cf61e347914@mail.gmail.com>

On 1/24/07, aslak hellesoy <aslak.hellesoy at gmail.com> wrote:
>
> On 1/24/07, Pat Maddox <pergesu at gmail.com> wrote:
> > Here are the specs:
> > context "Finding all the stylesheets available to a company" do
> >   setup do
> >     @mock_company = mock("company")
> >     @mock_company.stub!(:to_param).and_return "1"
> >     Stylesheet.stub!(:find)
> >   end
> >
> >   def do_find
> >     Stylesheet.find_available_to @mock_company
> >   end
> >
> >   specify "should convert the company into a parameter" do
> >     @mock_company.should_receive(:to_param).and_return "1"
> >     do_find
> >   end
> >
> >   specify "should find stylesheets belonging to only the company or to
> > nobody" do
> >     Stylesheet.should_receive(:find).with(:all, :conditions =>
> > ["company_id IS NULL OR company_id=?", "1"])
> >     do_find
> >   end
> > end
> >
> > And here's the code:
> > class Stylesheet < ActiveRecord::Base
> >   def self.find_available_to(company)
> >     find :all, :conditions => ["company_id IS NULL OR company_id=?",
> > company.to_param]
> >   end
> > end
> >
> > You can see that the implementation is duplicated.  In fact, I wrote
> > the implementation in the spec and then basically just copied it over.
> >
> > That bugs me.
> >
> > The obvious way to fix it would be to create a couple stylesheet
> > entries in the database, run the query, and see if it matches the
> > expected results.
> >
> > The only benefit of the current approach (that I can see) is that the
> > behavior is explicit.  If you've just got the results, you have to
> > mentally link them to the query.
> >
> > Anyway I'm just sort of stuck on this one.  I think there's a much
> > better way that is evading me.
> >
>
> I wouldn't use mocks for model specs. I'd test it against the real
> database.
>
> Aslak


In addition to actually finding real records from the database for the
Stylesheet model (don't stub find), I would also suggest that to_param is
not the right method to use in this case. You're stating that company_id
should equal the url formatted string for a company (which it is typical to
overload to handle things like 10-name_of_company). Clearly your database
naming says company.id so it's safer to just use company.id.

-Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rspec-users/attachments/20070125/32558d34/attachment.html 

From lists-rspec at shopwatch.org  Fri Jan 26 15:04:07 2007
From: lists-rspec at shopwatch.org (Jay Levitt)
Date: Fri, 26 Jan 2007 15:04:07 -0500
Subject: [rspec-users] More on collection proxies
In-Reply-To: <57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>	<45B35B67.2070402@e-tobi.net>	<57c63afe0701210436v28002fcds4dd6aa945171b8b4@mail.gmail.com>	<45B36382.2030503@e-tobi.net>
	<45B3D11F.90703@rubyforge.org>	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>	<45B3F29E.1080009@rubyforge.org>	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>	<45B76313.4000305@rubyforge.org>
	<45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
Message-ID: <45BA5EB7.9010508@rubyforge.org>

David Chelimsky wrote:
>> Jay Levitt wrote:
>>
>> validates_presence_of is a bad example, because the two methods are
>> practically interchangeable.  But consider a validation that uses a
>> regex to verify a legal IP address.  Do you want your specs to repeat
>> the regex, or do you want to test various legal and illegal IP-address
>> strings and see what breaks?  To me, it's the second one that's actually
>> testing the behavior of the application.
> 
> Another view would be that the Regexp is a separate component that
> you'd want to test separately from the use of that component. So the
> test that your model validates_format_of using a Regexp uses the right
> one and then have other tests just for that Regexp.

I'd counter that the Regexp is essentially a private method that 
shouldn't be tested.  It's never directly used by the outside world; 
only the validation is.

Jay


From mlangenberg at gmail.com  Sat Jan 27 06:59:36 2007
From: mlangenberg at gmail.com (Matthijs Langenberg)
Date: Sat, 27 Jan 2007 12:59:36 +0100
Subject: [rspec-users] More on collection proxies
In-Reply-To: <45BA5EB7.9010508@rubyforge.org>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B36382.2030503@e-tobi.net> <45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
	<45BA5EB7.9010508@rubyforge.org>
Message-ID: <27c0ac6d0701270359t5e57d9bh258d208e4bbc7664@mail.gmail.com>

Why shouldn't private methods be tested? I use Object#send all the
time to test private methods, otherwise I've got the idea the step to
implement the spec is too big, for example look at the specs of my
SMSer project (http://pastie.caboo.se/36044).

On 1/26/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> David Chelimsky wrote:
> >> Jay Levitt wrote:
> >>
> >> validates_presence_of is a bad example, because the two methods are
> >> practically interchangeable.  But consider a validation that uses a
> >> regex to verify a legal IP address.  Do you want your specs to repeat
> >> the regex, or do you want to test various legal and illegal IP-address
> >> strings and see what breaks?  To me, it's the second one that's actually
> >> testing the behavior of the application.
> >
> > Another view would be that the Regexp is a separate component that
> > you'd want to test separately from the use of that component. So the
> > test that your model validates_format_of using a Regexp uses the right
> > one and then have other tests just for that Regexp.
>
> I'd counter that the Regexp is essentially a private method that
> shouldn't be tested.  It's never directly used by the outside world;
> only the validation is.
>
> Jay
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From dchelimsky at gmail.com  Sat Jan 27 10:03:18 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sat, 27 Jan 2007 09:03:18 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <27c0ac6d0701270359t5e57d9bh258d208e4bbc7664@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B3D11F.90703@rubyforge.org>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
	<45BA5EB7.9010508@rubyforge.org>
	<27c0ac6d0701270359t5e57d9bh258d208e4bbc7664@mail.gmail.com>
Message-ID: <57c63afe0701270703q58b8af68n9c33f8856a122ff@mail.gmail.com>

On 1/27/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
> Why shouldn't private methods be tested? I use Object#send all the
> time to test private methods, otherwise I've got the idea the step to
> implement the spec is too big, for example look at the specs of my
> SMSer project (http://pastie.caboo.se/36044).

There is a school of thought that suggests that if you need to test
privates that you're probably doing too much in the one class and you
should consider breaking things out. Admittedly, this school of
thought comes from static languages in which you have to go through
crazy and error prone hoops in order to test privates. Since Ruby
makes it so simple, it seems less harmful and risky.

It should, however, be at least a red flag about your design. It
should make you ask "why am I testing privates?".

The questions I have are:

What is the impact of extracting the error code?
Does doing so change some other behaviour?
Isn't it enough to spec "should raise an exception if the webservice
returned an error"?

Getting back to my earlier statement about breaking things out, you
*could* (and I likely would) choose to make a new class SmsResponse
that gets initialized with the xml response text and exposes :code and
:message. Then you could eliminate the last two specs in your example
and have a separate spec just for this new class:
http://pastie.caboo.se/36071

The cost of this is an extra class. The benefit is good separation of
concerns (parsing XML vs processing a logical response).

Similarly, the product of :create_post_data must go somewhere, right?
What effect does that have on behavior? If it simply gets stored, then
you can mock out the storage to make sure it gets stored correctly. If
it changes the behaviour of the SMSer, then you can have examples that
show the different behaviours.

All of this said, these principles exist because people have
experienced pain from doing things. You may be absolutely fine w/ what
you have. The risk that you run is that business rules about SMSer and
the XML schema for the response change independently, and then you
have two reasons to change the same class (violation of the Single
Responsibility Principle), making it more prone to ripple effects.

An other thing to consider is whether there are other parts of your
system that have to parse the same XML. If that is the case, then its
definitely to your benefit to break out an SmsResponse class.

WDYT?

David




>
> On 1/26/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> > David Chelimsky wrote:
> > >> Jay Levitt wrote:
> > >>
> > >> validates_presence_of is a bad example, because the two methods are
> > >> practically interchangeable.  But consider a validation that uses a
> > >> regex to verify a legal IP address.  Do you want your specs to repeat
> > >> the regex, or do you want to test various legal and illegal IP-address
> > >> strings and see what breaks?  To me, it's the second one that's actually
> > >> testing the behavior of the application.
> > >
> > > Another view would be that the Regexp is a separate component that
> > > you'd want to test separately from the use of that component. So the
> > > test that your model validates_format_of using a Regexp uses the right
> > > one and then have other tests just for that Regexp.
> >
> > I'd counter that the Regexp is essentially a private method that
> > shouldn't be tested.  It's never directly used by the outside world;
> > only the validation is.
> >
> > Jay
> >
> > _______________________________________________
> > 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
>

From cwdinfo at gmail.com  Sat Jan 27 14:29:23 2007
From: cwdinfo at gmail.com (s.ross)
Date: Sat, 27 Jan 2007 11:29:23 -0800
Subject: [rspec-users] should expectation syntax has changed
Message-ID: <3EEEBBD1-B8EB-4BA4-8714-496052E39873@gmail.com>

This is just a note in case anyone else is using the (I believe  
experimental) custom expectations.

The custom expectation I had written used the custom 'met_by?'  
function, which now appears to have been replaced by the 'matches?'  
function to determine whether the expectation is met.

Steve


From dchelimsky at gmail.com  Sat Jan 27 14:55:19 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sat, 27 Jan 2007 13:55:19 -0600
Subject: [rspec-users] should expectation syntax has changed
In-Reply-To: <3EEEBBD1-B8EB-4BA4-8714-496052E39873@gmail.com>
References: <3EEEBBD1-B8EB-4BA4-8714-496052E39873@gmail.com>
Message-ID: <57c63afe0701271155x4152e139t32d5891274b639e0@mail.gmail.com>

To be clear, this ONLY applies to those of you using trunk and
experimenting with an un-released feature supporting custom
expectations.

To those of you who are, it is now pretty much how it will be
released, so if you use #matches?, failure_message and
negative_failure_message you should be OK.

Cheers,
David

On 1/27/07, s.ross <cwdinfo at gmail.com> wrote:
> This is just a note in case anyone else is using the (I believe
> experimental) custom expectations.
>
> The custom expectation I had written used the custom 'met_by?'
> function, which now appears to have been replaced by the 'matches?'
> function to determine whether the expectation is met.
>
> Steve
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From listaccount at e-tobi.net  Sun Jan 28 19:58:00 2007
From: listaccount at e-tobi.net (Tobi)
Date: Mon, 29 Jan 2007 01:58:00 +0100
Subject: [rspec-users] Bug in should_not_be - What else to use?
Message-ID: <45BD4698.5000909@e-tobi.net>

Hi!

I just stumbled over a possible bug in 0.7.5.1:

`1.should_not_be == 1` does not fail.

I took a look into the code and figured out, that this is caused by
Not.be() using :no_arg instead of :___no_arg. The expected argument of
the method be() in Not is passed from should_not_be() with :___no_arg.

Not.be() is marked with "Gone for 0.9", so I assume that "should_not_be
==" shouldn't be used at all, right? So what else to use, when testing
of "==" is required?

1.should_not == 1

...gives a "not so nice" failure message: "1 should not == 1 nil" - this
would read nicer as "1 should not be == 1"

Where is the "nil" in the failure message coming from anyway? I digged a
little bit deeper into the sources and figured that out too:

In the Not class, instead of passing the not expected value as the
second argument to default_message, it gets appended to the message text,
so the missing argument will then be added by default_message() as nil.
I've already added a bug report.

bye,

Tobias




From jchris at mfdz.com  Sun Jan 28 20:47:41 2007
From: jchris at mfdz.com (Chris Anderson)
Date: Sun, 28 Jan 2007 17:47:41 -0800
Subject: [rspec-users] Rails trunk breakages
Message-ID: <e282921e0701281747o17445311s38e06112dcaef927@mail.gmail.com>

Anyone else having an issue with route-generation in specs breaking as
of Rails trunk r6062? When I run my specs at r6061, they all pass, and
in 6062 I get a trace like this every time my controller generates a
route:

TypeError in 'Requesting /artists/1 using PUT should redirect to the artist'
can't convert Fixnum into String
(eval):19:in `artist_url'
/Users/jchris/ruby/rails/grabbit/config/../app/controllers/artists_controller.rb:59:in
`update'
/Users/jchris/ruby/rails/grabbit/config/../app/controllers/artists_controller.rb:56:in
`update'
./spec/controllers/artists_controller_spec.rb:267:in `do_update'
./spec/controllers/artists_controller_spec.rb:287:

For now I'm just using Rails r6061, but I wonder if people using a new
version of RSpec are not having trouble. For reference, I'm using
RSpec and RSpec on Rails from trunk as of revision 1359. (Even more
reference. I just took a five minute break and upgraded to RSpec 0.8.0
(r1440) and am having the same problem.

Parallel Test::Unit tests aren't showing any signs of breakage.

I've put this in the tracker, as well.
http://rubyforge.org/tracker/?func=detail&aid=8238&group_id=797&atid=3149

Best,
Chris


-- 
Chris Anderson
http://jchris.mfdz.com

From dchelimsky at gmail.com  Sun Jan 28 21:33:07 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 28 Jan 2007 20:33:07 -0600
Subject: [rspec-users] Bug in should_not_be - What else to use?
In-Reply-To: <45BD4698.5000909@e-tobi.net>
References: <45BD4698.5000909@e-tobi.net>
Message-ID: <57c63afe0701281833t1f191696x77ee553b3a356cd3@mail.gmail.com>

On 1/28/07, Tobi <listaccount at e-tobi.net> wrote:
> Hi!
>
> I just stumbled over a possible bug in 0.7.5.1:
>
> `1.should_not_be == 1` does not fail.

Try should_not == 1

>
> I took a look into the code and figured out, that this is caused by
> Not.be() using :no_arg instead of :___no_arg. The expected argument of
> the method be() in Not is passed from should_not_be() with :___no_arg.
>
> Not.be() is marked with "Gone for 0.9", so I assume that "should_not_be
> ==" shouldn't be used at all, right? So what else to use, when testing
> of "==" is required?
>
> 1.should_not == 1
>
> ...gives a "not so nice" failure message: "1 should not == 1 nil" - this
> would read nicer as "1 should not be == 1"
>
> Where is the "nil" in the failure message coming from anyway? I digged a
> little bit deeper into the sources and figured that out too:
>
> In the Not class, instead of passing the not expected value as the
> second argument to default_message, it gets appended to the message text,
> so the missing argument will then be added by default_message() as nil.
> I've already added a bug report.
>
> bye,
>
> Tobias
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From francois.beausoleil at gmail.com  Sun Jan 28 22:12:30 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Sun, 28 Jan 2007 22:12:30 -0500
Subject: [rspec-users] Cryptic error message when no controller_name
Message-ID: <41d5fadf0701281912u2fd2e98bk868ef5f73b0c541e@mail.gmail.com>

Hi !

In the following spec:

context "A project owner" do
  specify "can assign roles to other users" do
    # ...
  end
end

I get the following cryptic error message:

1)
NoMethodError in 'A project owner can assign roles to other users'
undefined method `session=' for #<Object:0xb741bed4>
/home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/context/controller.rb:134:in
`setup_extra'
/home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/functional_eval_context.rb:22:in
`setup'
/home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/context/controller.rb:151:in
`setup'

Line 134 of controller.rb is:

@controller.session = session

We are probably only missing a check to ensure that @controller is
somehow defined.  Failing that should simply say "You forgot to define
controller_name in your controller spec".

This is on 0.7.5.1.

Thanks !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From dchelimsky at gmail.com  Sun Jan 28 22:15:16 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Sun, 28 Jan 2007 21:15:16 -0600
Subject: [rspec-users] Cryptic error message when no controller_name
In-Reply-To: <41d5fadf0701281912u2fd2e98bk868ef5f73b0c541e@mail.gmail.com>
References: <41d5fadf0701281912u2fd2e98bk868ef5f73b0c541e@mail.gmail.com>
Message-ID: <57c63afe0701281915g12fdc81bybc30aa1b0862663a@mail.gmail.com>

Please put this in the tracker:
http://rubyforge.org/tracker/?group_id=797

Thanks,
David

On 1/28/07, Francois Beausoleil <francois.beausoleil at gmail.com> wrote:
> Hi !
>
> In the following spec:
>
> context "A project owner" do
>   specify "can assign roles to other users" do
>     # ...
>   end
> end
>
> I get the following cryptic error message:
>
> 1)
> NoMethodError in 'A project owner can assign roles to other users'
> undefined method `session=' for #<Object:0xb741bed4>
> /home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/context/controller.rb:134:in
> `setup_extra'
> /home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/functional_eval_context.rb:22:in
> `setup'
> /home/francois/src/smrty2/vendor/plugins/rspec_on_rails/lib/spec/rails/context/controller.rb:151:in
> `setup'
>
> Line 134 of controller.rb is:
>
> @controller.session = session
>
> We are probably only missing a check to ensure that @controller is
> somehow defined.  Failing that should simply say "You forgot to define
> controller_name in your controller spec".
>
> This is on 0.7.5.1.
>
> Thanks !
> --
> Fran?ois Beausoleil
> http://blog.teksol.info/
> http://piston.rubyforge.org/
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

From francois.beausoleil at gmail.com  Sun Jan 28 22:27:11 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Sun, 28 Jan 2007 22:27:11 -0500
Subject: [rspec-users] Cryptic error message when no controller_name
In-Reply-To: <57c63afe0701281915g12fdc81bybc30aa1b0862663a@mail.gmail.com>
References: <41d5fadf0701281912u2fd2e98bk868ef5f73b0c541e@mail.gmail.com>
	<57c63afe0701281915g12fdc81bybc30aa1b0862663a@mail.gmail.com>
Message-ID: <41d5fadf0701281927p1d8230cbkc3f85b31e3aba235@mail.gmail.com>

Hi !

2007/1/28, David Chelimsky <dchelimsky at gmail.com>:
> Please put this in the tracker:
> http://rubyforge.org/tracker/?group_id=797

Yes, of course.  I'm so sorry not to have done that right at first.

Here's the link:
http://rubyforge.org/tracker/index.php?func=detail&aid=8240&group_id=797&atid=3149

Bye !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From francois.beausoleil at gmail.com  Mon Jan 29 00:12:23 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Mon, 29 Jan 2007 00:12:23 -0500
Subject: [rspec-users] mock.with(Hash) expects an Array ?
Message-ID: <41d5fadf0701282112hf4ad5dctf97d1dcfebd5fd52@mail.gmail.com>

Hi all,

I looked at the open issues on the bug tracker, but couldn't find
anything related to this.  I also searched the archives.  And I can't
believe someone didn't hit on this before.

Here's my spec:
context "A project owner" do
  controller_name :roles

  setup do
    controller.stub!(:current_user).and_return(@user = mock("user"))
    controller.stub!(:current_project).and_return(@project = mock("project"))
    @user.stub!(:role_on).and_return(Role.owner)
  end

  specify "can change the role of a user" do
    @project.should_receive(:roles).and_return(roles_proxy =
mock("roles proxy"))
    roles_proxy.should_receive(:find).with("17").and_return(role = mock("role"))
    role.should_receive(:update_attributes!).with(:name =>
Role.developer).and_return(true)
    role.stub!(:name).and_return(Role.developer)

    post :update, :project_id => 41, :id => 17, :role => {:name =>
Role.developer}
    response.body.should == Role.developer
  end
end

And the failure:
'A project owner can change the role of a user' FAILED
Mock 'role' expected :update_attributes! with ([:name, "developer"])
but received it with ({"name"=>"developer"})
(eval):3:in `update_attributes!'
/home/francois/src/smrty2/config/../app/controllers/roles_controller.rb:11:in
`update'
./spec/controllers/roles_spec.rb:35:

Looks like the call to #with changed the Hash to an Array.

Can anyone else confirm this ?  This is on 0.7.5.1.

Thanks !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From listaccount at e-tobi.net  Mon Jan 29 04:41:01 2007
From: listaccount at e-tobi.net (Tobi)
Date: Mon, 29 Jan 2007 10:41:01 +0100
Subject: [rspec-users] Bug in should_not_be - What else to use?
In-Reply-To: <57c63afe0701281833t1f191696x77ee553b3a356cd3@mail.gmail.com>
References: <45BD4698.5000909@e-tobi.net>
	<57c63afe0701281833t1f191696x77ee553b3a356cd3@mail.gmail.com>
Message-ID: <45BDC12D.6000503@e-tobi.net>

David Chelimsky wrote:
> Try should_not == 1
>   

Ok. It just gives me a not so nice to read failure message:

"1 should not == 1"  compared to  "1 should not be == 1"

Maybe the "Gone for 0.9"-methods should print a deprecation warning in
one of the next releases.

Tobias


From justnothing at tiscali.co.uk  Mon Jan 29 04:30:13 2007
From: justnothing at tiscali.co.uk (David Green)
Date: Mon, 29 Jan 2007 01:30:13 -0800 (PST)
Subject: [rspec-users] rspec and cookies
Message-ID: <8686339.post@talk.nabble.com>


Hello

can anyone tell me how to test cookies using rspec? specifically, I'd like
to be able to set a cookie before a get/post request and also to test the
cookies which have been set by a get/post request

I know I can use cookies[:name] = 'value' to set a cookie but how would I
set expiry information on such a cookie? when i try to pass a hash {:value
=> 'value', :expires => 3.days.from_now} the entire hash becomes the value
of cookies[:name] 

thanks
david
-- 
View this message in context: http://www.nabble.com/rspec-and-cookies-tf3134883.html#a8686339
Sent from the rspec-users mailing list archive at Nabble.com.


From dchelimsky at gmail.com  Mon Jan 29 07:31:36 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Mon, 29 Jan 2007 06:31:36 -0600
Subject: [rspec-users] Bug in should_not_be - What else to use?
In-Reply-To: <45BDC12D.6000503@e-tobi.net>
References: <45BD4698.5000909@e-tobi.net>
	<57c63afe0701281833t1f191696x77ee553b3a356cd3@mail.gmail.com>
	<45BDC12D.6000503@e-tobi.net>
Message-ID: <57c63afe0701290431o45ab86d4g93a21f6b3032d417@mail.gmail.com>

On 1/29/07, Tobi <listaccount at e-tobi.net> wrote:
> David Chelimsky wrote:
> > Try should_not == 1
> >
>
> Ok. It just gives me a not so nice to read failure message:
>
> "1 should not == 1"  compared to  "1 should not be == 1"
>
> Maybe the "Gone for 0.9"-methods should print a deprecation warning in
> one of the next releases.

Already in the plan. It won't print warnings in 0.8.0, but it will be
shortly thereafter, and well before the deprecated methods are
removed.

>
> Tobias
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From francois.beausoleil at gmail.com  Mon Jan 29 15:03:40 2007
From: francois.beausoleil at gmail.com (Francois Beausoleil)
Date: Mon, 29 Jan 2007 15:03:40 -0500
Subject: [rspec-users] rspec and cookies
In-Reply-To: <8686339.post@talk.nabble.com>
References: <8686339.post@talk.nabble.com>
Message-ID: <41d5fadf0701291203y6b4fb291l234a32000594ede2@mail.gmail.com>

Hello David,

2007/1/29, David Green <justnothing at tiscali.co.uk>:
> I know I can use cookies[:name] = 'value' to set a cookie but how would I
> set expiry information on such a cookie? when i try to pass a hash {:value
> => 'value', :expires => 3.days.from_now} the entire hash becomes the value
> of cookies[:name]

When the cookie comes back through the browser, it never has any
expiration or options.  The browser handles that information.  That's
why you can't set those options on the requesting side.

When responding, however... that's another story.

Hope that helps !
-- 
Fran?ois Beausoleil
http://blog.teksol.info/
http://piston.rubyforge.org/

From deepak.jois at gmail.com  Tue Jan 30 03:45:13 2007
From: deepak.jois at gmail.com (Deepak Jois)
Date: Tue, 30 Jan 2007 16:45:13 +0800
Subject: [rspec-users] errors while testing resource controller using rpec
Message-ID: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>

I am testing a resource called venue in this piece of code (generated
using script/rspec_resource)

====================
context "Requesting /venues using POST" do
  controller_name :venues

  setup do
    @mock_venue = mock('Venue')
    @mock_venue.stub!(:save).and_return(true)
    @mock_venue.stub!(:to_param).and_return(1)
    Venue.stub!(:new).and_return(@mock_venue)
  end

  def do_post
    post :create, :venue => {:name => 'Venue'}
  end

  specify "should create a new venue" do
    Venue.should_receive(:new).with({'name' => 'Venue'}).and_return(@mock_venue)
    do_post
  end
end
==========================

and the test fails as
===============
TypeError in 'Requesting /venues using POST should create a new venue'
can't convert Fixnum into String
(eval):19:in `venue_url'
/home/deepak/personalcode/worlds/config/../app/controllers/venues_controller.rb:43:in
`create'
/home/deepak/personalcode/worlds/config/../app/controllers/venues_controller.rb:40:in
`create'
./spec/controllers/venues_controller_spec.rb:242:in `do_post'
./spec/controllers/venues_controller_spec.rb:247:
================

the relevant portion of the controller code is

================
 # POST /venues
  # POST /venues.xml
  def create
    @venue = Venue.new(params[:venue])

    respond_to do |format|
      if @venue.save
        flash[:notice] = 'Venue was successfully created.'
        format.html { redirect_to venue_url(@venue) }
        format.xml  { head :created, :location => venue_url(@venue) }
      else
        format.html { render :action => "new" }
        format.xml  { render :xml => @venue.errors.to_xml }
      end
    end
  end
==============

however, If I change the mock stub to return a string, the spec passes fine.

@mock_venue.stub!(:to_param).and_return("1")  # note the string

There is something happening in the generated code venue_url which I
am not aware of, which is causing the test to fail on the mock object.

Is this a bug in the generated code.. or am I doing something wrong?

Thanks
Deepak

From deepak.jois at gmail.com  Tue Jan 30 03:47:38 2007
From: deepak.jois at gmail.com (Deepak Jois)
Date: Tue, 30 Jan 2007 16:47:38 +0800
Subject: [rspec-users] errors while testing resource controller using
	rpec
In-Reply-To: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
References: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
Message-ID: <24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>

On 1/30/07, Deepak Jois <deepak.jois at gmail.com> wrote:
>
> however, If I change the mock stub to return a string, the spec passes fine.
>
> @mock_venue.stub!(:to_param).and_return("1")  # note the string
>
> There is something happening in the generated code venue_url which I
> am not aware of, which is causing the test to fail on the mock object.

According to http://api.rubyonrails.org/classes/ActiveRecord/Base.html#M001033

the to_param method sends a stringified ID. Hence this should be a bug.

Thanks
Deepak

From dchelimsky at gmail.com  Tue Jan 30 06:04:35 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 30 Jan 2007 05:04:35 -0600
Subject: [rspec-users] errors while testing resource controller using
	rpec
In-Reply-To: <24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>
References: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
	<24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>
Message-ID: <57c63afe0701300304h687fbb1cg3cd34db2c2f6caf@mail.gmail.com>

On 1/30/07, Deepak Jois <deepak.jois at gmail.com> wrote:
> On 1/30/07, Deepak Jois <deepak.jois at gmail.com> wrote:
> >
> > however, If I change the mock stub to return a string, the spec passes fine.
> >
> > @mock_venue.stub!(:to_param).and_return("1")  # note the string
> >
> > There is something happening in the generated code venue_url which I
> > am not aware of, which is causing the test to fail on the mock object.
>
> According to http://api.rubyonrails.org/classes/ActiveRecord/Base.html#M001033
>
> the to_param method sends a stringified ID. Hence this should be a bug.

Hence you should report this to the rspec tracker:

http://rspec.rubyforge.org/contribute.html
http://rubyforge.org/mail/?group_id=797

Cheers,
David

>
> Thanks
> Deepak
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From deepak.jois at gmail.com  Tue Jan 30 06:20:48 2007
From: deepak.jois at gmail.com (Deepak Jois)
Date: Tue, 30 Jan 2007 19:20:48 +0800
Subject: [rspec-users] errors while testing resource controller using
	rpec
In-Reply-To: <57c63afe0701300304h687fbb1cg3cd34db2c2f6caf@mail.gmail.com>
References: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
	<24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>
	<57c63afe0701300304h687fbb1cg3cd34db2c2f6caf@mail.gmail.com>
Message-ID: <24dc07380701300320le73cfb6r40cb89e3914dbf77@mail.gmail.com>

On 1/30/07, David Chelimsky <dchelimsky at gmail.com> wrote:
>
> Hence you should report this to the rspec tracker:
>
> http://rspec.rubyforge.org/contribute.html
> http://rubyforge.org/mail/?group_id=797
>
> Cheers,
> David
>

And thus arises my meta-question :).. Since I am quote new to this, I
dont know how to submit a patch for this with a spec

As far as I can see, it is a VERY trivial patch which involves adding
quotes around a number in the generator templates. Should I just go
ahead and do that? Or is there a spec that I can actually write for
such a trivial patch,

Thanks
Deepak

From dchelimsky at gmail.com  Tue Jan 30 06:33:26 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Tue, 30 Jan 2007 05:33:26 -0600
Subject: [rspec-users] errors while testing resource controller using
	rpec
In-Reply-To: <24dc07380701300320le73cfb6r40cb89e3914dbf77@mail.gmail.com>
References: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
	<24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>
	<57c63afe0701300304h687fbb1cg3cd34db2c2f6caf@mail.gmail.com>
	<24dc07380701300320le73cfb6r40cb89e3914dbf77@mail.gmail.com>
Message-ID: <57c63afe0701300333u634b1f73m17b8337023cd011a@mail.gmail.com>

On 1/30/07, Deepak Jois <deepak.jois at gmail.com> wrote:
> On 1/30/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> >
> > Hence you should report this to the rspec tracker:
> >
> > http://rspec.rubyforge.org/contribute.html
> > http://rubyforge.org/mail/?group_id=797
> >
> > Cheers,
> > David
> >
>
> And thus arises my meta-question :).. Since I am quote new to this, I
> dont know how to submit a patch for this with a spec
>
> As far as I can see, it is a VERY trivial patch which involves adding
> quotes around a number in the generator templates. Should I just go
> ahead and do that? Or is there a spec that I can actually write for
> such a trivial patch,

I'll put up better instructions re: patches in the next release of the docs.

In the mean time, since you've diagnosed the problem what you've
reported above would be just fine. Something along the lines of
"mocking :to_param should return a Fixnum" as the title and then
describe the situation. Go ahead and stick that in the Bugs tracker
(rather than a patch).

Cheers,
David

>
> Thanks
> Deepak
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

From tils-rspec at tils.net  Tue Jan 30 06:06:57 2007
From: tils-rspec at tils.net (Tilmann Singer)
Date: Tue, 30 Jan 2007 12:06:57 +0100
Subject: [rspec-users] Emacs spec parsing?
Message-ID: <20070130110657.GD25504@tils.net>

Hi,

Did anyone with better emacs lisp knowledge than me already come up
with a function like spec-at-point which finds the specify and context
declaration for the current cursor position and returns a string which
can be used to be passed to the --spec option of the spec runner?

Damn that would be useful ...

I am currently using a function that runs spec --line with the current
line number, the problem with this is however that 1.) the
specification string is not shown again on the commandline if the spec
passed, and 2.) the line number may change when editing the file, and
re-running the previous spec commandline (with the very nice emacs
command recompile) may hit a different spot.


tia, Til

From jchris at mfdz.com  Tue Jan 30 12:24:59 2007
From: jchris at mfdz.com (Chris Anderson)
Date: Tue, 30 Jan 2007 09:24:59 -0800
Subject: [rspec-users] errors while testing resource controller using
	rpec
In-Reply-To: <57c63afe0701300333u634b1f73m17b8337023cd011a@mail.gmail.com>
References: <24dc07380701300045x6aa80fbx3fbd2bad9855a5e3@mail.gmail.com>
	<24dc07380701300047g6a3cc02cv491bec66fc08ad96@mail.gmail.com>
	<57c63afe0701300304h687fbb1cg3cd34db2c2f6caf@mail.gmail.com>
	<24dc07380701300320le73cfb6r40cb89e3914dbf77@mail.gmail.com>
	<57c63afe0701300333u634b1f73m17b8337023cd011a@mail.gmail.com>
Message-ID: <e282921e0701300924y76286f5ekf0ab7230453612b1@mail.gmail.com>

I filed a ticket on this issue the other night. I didn't realize that
to_param ought to return a string, so the bug does seem to be with the
generated specs. But it is interesting to note that the issue does not
affect Rails prior to rev 6062.

I've invalidated my old ticket and submitted a patch that fixes the
issue. Here it is:

http://rubyforge.org/tracker/?func=detail&aid=8316&group_id=797&atid=3151


On 1/30/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/30/07, Deepak Jois <deepak.jois at gmail.com> wrote:
> > On 1/30/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > >
> > > Hence you should report this to the rspec tracker:
> > >
> > > http://rspec.rubyforge.org/contribute.html
> > > http://rubyforge.org/mail/?group_id=797
> > >
> > > Cheers,
> > > David
> > >
> >
> > And thus arises my meta-question :).. Since I am quote new to this, I
> > dont know how to submit a patch for this with a spec
> >
> > As far as I can see, it is a VERY trivial patch which involves adding
> > quotes around a number in the generator templates. Should I just go
> > ahead and do that? Or is there a spec that I can actually write for
> > such a trivial patch,
>
> I'll put up better instructions re: patches in the next release of the docs.
>
> In the mean time, since you've diagnosed the problem what you've
> reported above would be just fine. Something along the lines of
> "mocking :to_param should return a Fixnum" as the title and then
> describe the situation. Go ahead and stick that in the Bugs tracker
> (rather than a patch).
>
> Cheers,
> David
>
> >
> > Thanks
> > Deepak
> > _______________________________________________
> > 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
>


-- 
Chris Anderson
http://jchris.mfdz.com

From jchris at mfdz.com  Tue Jan 30 17:36:57 2007
From: jchris at mfdz.com (Chris Anderson)
Date: Tue, 30 Jan 2007 14:36:57 -0800
Subject: [rspec-users] nil object in mocks
Message-ID: <e282921e0701301436w4c102121yab19ad819e60e88@mail.gmail.com>

RSpecrs,

I've been running into some trouble lately when passing mock objects
into rich views. Either I have to obscure my specs by stubbing every
method the view would call on my object, or I have to resort to
trickery like stubbing ActionController#render, which is also less
than ideal.

The solution I've some up with is a variant on the :null_object option
that can be passed to the mock factory. Null objects return themselves
from any unstubbed method call. The option I've added changes that
return value to nil, which doesn't play so well with method chaining,
but does work nicely in most views.

The question for you all is: what should the syntax look like? We've
considered a couple of options but found them lacking. They are

m = mock('returns nil', :nil_object => true)
m = mock('returns nil', :null_object => :returns_nil)

Hopefully someone on the list can come up with something more compelling.

Another idea that just came to me (funny how lunch break has that effect):

m = mock('returns nil', :default => nil)

could be useful like this also

m = mock('returns 5', :default => 5)

and if you allow the default value to be a lambda that takes the mock
as an argument, you can make :null_object a special case:

m = mock('returns self', :default => lambda{|m| m})

but now I'm just getting silly.

--
Chris Anderson
http://jchris.mfdz.com

From snessnet at gmail.com  Tue Jan 30 22:27:36 2007
From: snessnet at gmail.com (sness)
Date: Tue, 30 Jan 2007 19:27:36 -0800
Subject: [rspec-users] rspec + CakePHP
Message-ID: <7f7f86440701301927s2363a006u87a88c2141d95d85@mail.gmail.com>

Hi,

I am in a situation where I am going to have to rewrite a working Ruby
on Rails application in CakePHP with PHP5.  I have a whole bunch of
specs written with rspec, and I was wondering if it would be possible
to somehow use these specs to test the Models and Controllers that I
am rewriting in CakePHP.  I am going to be basically just exactly
duplicating the Rails application, with identical class and method
names in PHP.

I know this is perverse, but is it possible?

Thanks,
sness.

From cwdinfo at gmail.com  Tue Jan 30 23:39:25 2007
From: cwdinfo at gmail.com (s.ross)
Date: Tue, 30 Jan 2007 20:39:25 -0800
Subject: [rspec-users] rspec + CakePHP
In-Reply-To: <7f7f86440701301927s2363a006u87a88c2141d95d85@mail.gmail.com>
References: <7f7f86440701301927s2363a006u87a88c2141d95d85@mail.gmail.com>
Message-ID: <2C9E17A6-09C0-410E-9E95-D4DF80AF4206@gmail.com>

I'll take a shot at answering this. rSpec doesn't passively observe  
the code under test, it actively interrogates it. So rSpec, which is  
written in Ruby will not be mappable into the PHP space. Consider  
this example:

assigns[:person].first_name.should == 'bob'

Here's a snippet of controller spec that is reaching into the  
controller and checking out the contents of @person. See why it won't  
work?

Have a look at this:

http://www.sitepoint.com/forums/showthread.php?t=439480

Still want to work in PHP?

Maybe?


On Jan 30, 2007, at 7:27 PM, sness wrote:

> Hi,
>
> I am in a situation where I am going to have to rewrite a working Ruby
> on Rails application in CakePHP with PHP5.  I have a whole bunch of
> specs written with rspec, and I was wondering if it would be possible
> to somehow use these specs to test the Models and Controllers that I
> am rewriting in CakePHP.  I am going to be basically just exactly
> duplicating the Rails application, with identical class and method
> names in PHP.
>
> I know this is perverse, but is it possible?
>
> Thanks,
> sness.
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users


From mlangenberg at gmail.com  Wed Jan 31 03:20:22 2007
From: mlangenberg at gmail.com (Matthijs Langenberg)
Date: Wed, 31 Jan 2007 09:20:22 +0100
Subject: [rspec-users] rspec + CakePHP
In-Reply-To: <2C9E17A6-09C0-410E-9E95-D4DF80AF4206@gmail.com>
References: <7f7f86440701301927s2363a006u87a88c2141d95d85@mail.gmail.com>
	<2C9E17A6-09C0-410E-9E95-D4DF80AF4206@gmail.com>
Message-ID: <27c0ac6d0701310020w3fe64911r7cdd8679f5ff0c70@mail.gmail.com>

Why actually doing a rewrite of a RoR application? Performance? I bet
there are other solutions.

On 1/31/07, s.ross <cwdinfo at gmail.com> wrote:
> I'll take a shot at answering this. rSpec doesn't passively observe
> the code under test, it actively interrogates it. So rSpec, which is
> written in Ruby will not be mappable into the PHP space. Consider
> this example:
>
> assigns[:person].first_name.should == 'bob'
>
> Here's a snippet of controller spec that is reaching into the
> controller and checking out the contents of @person. See why it won't
> work?
>
> Have a look at this:
>
> http://www.sitepoint.com/forums/showthread.php?t=439480
>
> Still want to work in PHP?
>
> Maybe?
>
>
> On Jan 30, 2007, at 7:27 PM, sness wrote:
>
> > Hi,
> >
> > I am in a situation where I am going to have to rewrite a working Ruby
> > on Rails application in CakePHP with PHP5.  I have a whole bunch of
> > specs written with rspec, and I was wondering if it would be possible
> > to somehow use these specs to test the Models and Controllers that I
> > am rewriting in CakePHP.  I am going to be basically just exactly
> > duplicating the Rails application, with identical class and method
> > names in PHP.
> >
> > I know this is perverse, but is it possible?
> >
> > Thanks,
> > sness.
> > _______________________________________________
> > 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
>

From mlangenberg at gmail.com  Wed Jan 31 04:20:21 2007
From: mlangenberg at gmail.com (Matthijs Langenberg)
Date: Wed, 31 Jan 2007 10:20:21 +0100
Subject: [rspec-users] More on collection proxies
In-Reply-To: <57c63afe0701270703q58b8af68n9c33f8856a122ff@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<57c63afe0701211428w263a3bebr4c84d454fd31db0b@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
	<45BA5EB7.9010508@rubyforge.org>
	<27c0ac6d0701270359t5e57d9bh258d208e4bbc7664@mail.gmail.com>
	<57c63afe0701270703q58b8af68n9c33f8856a122ff@mail.gmail.com>
Message-ID: <27c0ac6d0701310120s2147c930o6441ef60efb752ea@mail.gmail.com>

David, thanks for your reply. I've made some changes to my specs and
code, but I'm still not really sure where I should move the private
'create_post_data' method, should there be a seperate class for that
method to?
Same question for the 'send_http_post_request' method, although I
think it's a good idea to put that method in a different class /
context, allowing me to write a sample webservice to assert the http
post.
This is currently my spec and code: http://pastie.caboo.se/36900

- Matthijs

On 1/27/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> On 1/27/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
> > Why shouldn't private methods be tested? I use Object#send all the
> > time to test private methods, otherwise I've got the idea the step to
> > implement the spec is too big, for example look at the specs of my
> > SMSer project (http://pastie.caboo.se/36044).
>
> There is a school of thought that suggests that if you need to test
> privates that you're probably doing too much in the one class and you
> should consider breaking things out. Admittedly, this school of
> thought comes from static languages in which you have to go through
> crazy and error prone hoops in order to test privates. Since Ruby
> makes it so simple, it seems less harmful and risky.
>
> It should, however, be at least a red flag about your design. It
> should make you ask "why am I testing privates?".
>
> The questions I have are:
>
> What is the impact of extracting the error code?
> Does doing so change some other behaviour?
> Isn't it enough to spec "should raise an exception if the webservice
> returned an error"?
>
> Getting back to my earlier statement about breaking things out, you
> *could* (and I likely would) choose to make a new class SmsResponse
> that gets initialized with the xml response text and exposes :code and
> :message. Then you could eliminate the last two specs in your example
> and have a separate spec just for this new class:
> http://pastie.caboo.se/36071
>
> The cost of this is an extra class. The benefit is good separation of
> concerns (parsing XML vs processing a logical response).
>
> Similarly, the product of :create_post_data must go somewhere, right?
> What effect does that have on behavior? If it simply gets stored, then
> you can mock out the storage to make sure it gets stored correctly. If
> it changes the behaviour of the SMSer, then you can have examples that
> show the different behaviours.
>
> All of this said, these principles exist because people have
> experienced pain from doing things. You may be absolutely fine w/ what
> you have. The risk that you run is that business rules about SMSer and
> the XML schema for the response change independently, and then you
> have two reasons to change the same class (violation of the Single
> Responsibility Principle), making it more prone to ripple effects.
>
> An other thing to consider is whether there are other parts of your
> system that have to parse the same XML. If that is the case, then its
> definitely to your benefit to break out an SmsResponse class.
>
> WDYT?
>
> David
>
>
>
>
> >
> > On 1/26/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> > > David Chelimsky wrote:
> > > >> Jay Levitt wrote:
> > > >>
> > > >> validates_presence_of is a bad example, because the two methods are
> > > >> practically interchangeable.  But consider a validation that uses a
> > > >> regex to verify a legal IP address.  Do you want your specs to repeat
> > > >> the regex, or do you want to test various legal and illegal IP-address
> > > >> strings and see what breaks?  To me, it's the second one that's actually
> > > >> testing the behavior of the application.
> > > >
> > > > Another view would be that the Regexp is a separate component that
> > > > you'd want to test separately from the use of that component. So the
> > > > test that your model validates_format_of using a Regexp uses the right
> > > > one and then have other tests just for that Regexp.
> > >
> > > I'd counter that the Regexp is essentially a private method that
> > > shouldn't be tested.  It's never directly used by the outside world;
> > > only the validation is.
> > >
> > > Jay
> > >
> > > _______________________________________________
> > > 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
>

From dchelimsky at gmail.com  Wed Jan 31 08:07:26 2007
From: dchelimsky at gmail.com (David Chelimsky)
Date: Wed, 31 Jan 2007 07:07:26 -0600
Subject: [rspec-users] More on collection proxies
In-Reply-To: <27c0ac6d0701310120s2147c930o6441ef60efb752ea@mail.gmail.com>
References: <41d5fadf0701202045y208bdca0ub06dba767ad770c1@mail.gmail.com>
	<45B3F29E.1080009@rubyforge.org>
	<57c63afe0701232226u30653b12g6dad4e18a719ce9c@mail.gmail.com>
	<45B76313.4000305@rubyforge.org> <45B7684A.7030404@rubyforge.org>
	<57c63afe0701240715t47a30e3cs625624565dc6f583@mail.gmail.com>
	<45BA5EB7.9010508@rubyforge.org>
	<27c0ac6d0701270359t5e57d9bh258d208e4bbc7664@mail.gmail.com>
	<57c63afe0701270703q58b8af68n9c33f8856a122ff@mail.gmail.com>
	<27c0ac6d0701310120s2147c930o6441ef60efb752ea@mail.gmail.com>
Message-ID: <57c63afe0701310507o494bac2cw14bad21dcac8a0c2@mail.gmail.com>

On 1/31/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
> David, thanks for your reply. I've made some changes to my specs and
> code, but I'm still not really sure where I should move the private
> 'create_post_data' method, should there be a seperate class for that
> method to?
> Same question for the 'send_http_post_request' method, although I
> think it's a good idea to put that method in a different class /
> context, allowing me to write a sample webservice to assert the http
> post.
> This is currently my spec and code: http://pastie.caboo.se/36900

This line is the perfect place to introduce a new class:

@sms_response = SMSResponse.new(send_http_post_request(SERVICE_URI,
create_post_data(dutch_mobile, msg)))

How about this:

@sms_response = SMSRequest.new(dutch_mobile, msg).post(SERVICE_URI)

and then moving the creation of the SMSResponse to the SMSRequest?
This applies TDA and separates things nicely.

WDYT?

>
> - Matthijs
>
> On 1/27/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 1/27/07, Matthijs Langenberg <mlangenberg at gmail.com> wrote:
> > > Why shouldn't private methods be tested? I use Object#send all the
> > > time to test private methods, otherwise I've got the idea the step to
> > > implement the spec is too big, for example look at the specs of my
> > > SMSer project (http://pastie.caboo.se/36044).
> >
> > There is a school of thought that suggests that if you need to test
> > privates that you're probably doing too much in the one class and you
> > should consider breaking things out. Admittedly, this school of
> > thought comes from static languages in which you have to go through
> > crazy and error prone hoops in order to test privates. Since Ruby
> > makes it so simple, it seems less harmful and risky.
> >
> > It should, however, be at least a red flag about your design. It
> > should make you ask "why am I testing privates?".
> >
> > The questions I have are:
> >
> > What is the impact of extracting the error code?
> > Does doing so change some other behaviour?
> > Isn't it enough to spec "should raise an exception if the webservice
> > returned an error"?
> >
> > Getting back to my earlier statement about breaking things out, you
> > *could* (and I likely would) choose to make a new class SmsResponse
> > that gets initialized with the xml response text and exposes :code and
> > :message. Then you could eliminate the last two specs in your example
> > and have a separate spec just for this new class:
> > http://pastie.caboo.se/36071
> >
> > The cost of this is an extra class. The benefit is good separation of
> > concerns (parsing XML vs processing a logical response).
> >
> > Similarly, the product of :create_post_data must go somewhere, right?
> > What effect does that have on behavior? If it simply gets stored, then
> > you can mock out the storage to make sure it gets stored correctly. If
> > it changes the behaviour of the SMSer, then you can have examples that
> > show the different behaviours.
> >
> > All of this said, these principles exist because people have
> > experienced pain from doing things. You may be absolutely fine w/ what
> > you have. The risk that you run is that business rules about SMSer and
> > the XML schema for the response change independently, and then you
> > have two reasons to change the same class (violation of the Single
> > Responsibility Principle), making it more prone to ripple effects.
> >
> > An other thing to consider is whether there are other parts of your
> > system that have to parse the same XML. If that is the case, then its
> > definitely to your benefit to break out an SmsResponse class.
> >
> > WDYT?
> >
> > David
> >
> >
> >
> >
> > >
> > > On 1/26/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> > > > David Chelimsky wrote:
> > > > >> Jay Levitt wrote:
> > > > >>
> > > > >> validates_presence_of is a bad example, because the two methods are
> > > > >> practically interchangeable.  But consider a validation that uses a
> > > > >> regex to verify a legal IP address.  Do you want your specs to repeat
> > > > >> the regex, or do you want to test various legal and illegal IP-address
> > > > >> strings and see what breaks?  To me, it's the second one that's actually
> > > > >> testing the behavior of the application.
> > > > >
> > > > > Another view would be that the Regexp is a separate component that
> > > > > you'd want to test separately from the use of that component. So the
> > > > > test that your model validates_format_of using a Regexp uses the right
> > > > > one and then have other tests just for that Regexp.
> > > >
> > > > I'd counter that the Regexp is essentially a private method that
> > > > shouldn't be tested.  It's never directly used by the outside world;
> > > > only the validation is.
> > > >
> > > > Jay
> > > >
> > > > _______________________________________________
> > > > 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
>