rspec users - Oct 2008 - Expectation ordering

Hi

Can I still rely on this RSpec behaviour?

I''ve got a spec for a TCP socket client:

       it "should do things in a sane order" do
         @socket.should_receive(:write) do
           @socket.should_receive(:read) do
             @socket.should_receive(:close)
           end
         end
         @client.update_news_feeds
       end

But this passes, where I was expecting it to fail:

     def update_news_feeds
       socket = TCPSocket.new(@server, @port)
       socket.read
       socket.write("UPDATE\n")
       socket.close
     end

Has this form of expectation ordering been removed?

Thanks
Ashley

-- 
http://www.patchspace.co.uk/
http://aviewfromafar.net/

On Tue, Oct 7, 2008 at 7:59 AM, Ashley Moran
<ashley.moran at patchspace.co.uk> wrote:
> Hi
>
> Can I still rely on this RSpec behaviour?
>
> I''ve got a spec for a TCP socket client:
>
>      it "should do things in a sane order" do
>        @socket.should_receive(:write) do
>          @socket.should_receive(:read) do
>            @socket.should_receive(:close)
>          end
>        end
>        @client.update_news_feeds
>      end
>
> But this passes, where I was expecting it to fail:
>
>    def update_news_feeds
>      socket = TCPSocket.new(@server, @port)
>      socket.read
>      socket.write("UPDATE\n")
>      socket.close
>    end
>
> Has this form of expectation ordering been removed?
As far as I know this form of expectation ordering was never
supported. If you''ve seen documentation that suggests it should be,
please point me to it so I can resolve the discrepancy.

Ordering in rspec mocks is managed with the #ordered method:

@socket.should_receive(:write).ordered
@socket.should_receive(:read).ordered
@socket.should_receive(:close).ordered

See http://rspec.info/documentation/mocks/message_expectations.html -
almost all the way down the page.

Cheers,
David

>
> Thanks
> Ashley
>
> --
> http://www.patchspace.co.uk/
> http://aviewfromafar.net/
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 7 Oct 2008, at 15:15, David Chelimsky wrote:
> As far as I know this form of expectation ordering was never
> supported. If you''ve seen documentation that suggests it should
be,
> please point me to it so I can resolve the discrepancy.
Aslak told me himself... on this list :o)

> Ordering in rspec mocks is managed with the #ordered method:
>
> @socket.should_receive(:write).ordered
> @socket.should_receive(:read).ordered
> @socket.should_receive(:close).ordered
I forgot about that.  When I asked on the list before it was because I  
needed a way to do inter-mock ordering.  Thanks for the link, it  
solved the problem in the example I gave.  Unfortunately looks like  
I''ve lost the ability to do this:

       it "should migrate the database (after initialising it)" do
         DataMapper.should_receive(:setup) do
           Database::Migrator.should_receive(:migrate_up!)
         end
         @server.start
       end

Maybe I was the only person taking advantage of this?

Ashley

-- 
http://www.patchspace.co.uk/
http://aviewfromafar.net/

On Tue, Oct 7, 2008 at 9:40 AM, Ashley Moran
<ashley.moran at patchspace.co.uk> wrote:
>
> On 7 Oct 2008, at 15:15, David Chelimsky wrote:
>
>> As far as I know this form of expectation ordering was never
>> supported. If you''ve seen documentation that suggests it
should be,
>> please point me to it so I can resolve the discrepancy.
>
> Aslak told me himself... on this list :o)
>
>
>> Ordering in rspec mocks is managed with the #ordered method:
>>
>> @socket.should_receive(:write).ordered
>> @socket.should_receive(:read).ordered
>> @socket.should_receive(:close).ordered
>
> I forgot about that.  When I asked on the list before it was because I
> needed a way to do inter-mock ordering.  Thanks for the link, it solved the
> problem in the example I gave.  Unfortunately looks like I''ve lost
the
> ability to do this:
>
>      it "should migrate the database (after initialising it)" do
>        DataMapper.should_receive(:setup) do
>          Database::Migrator.should_receive(:migrate_up!)
>        end
>        @server.start
>      end
>
> Maybe I was the only person taking advantage of this?
Is the block no longer being executed? If so, then we have a problem
:) Please report this at http://rspec.lighthouseapp.com.

Thanks,
David
>
> Ashley
>
> --
> http://www.patchspace.co.uk/
> http://aviewfromafar.net/
>
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 7 Oct 2008, at 15:55, David Chelimsky wrote:
> Is the block no longer being executed? If so, then we have a problem
> :) Please report this at http://rspec.lighthouseapp.com.
Bizarrely, my little test...

   require ''spec''

   describe "Blocks" do
     it "should be called" do
       b = mock(Object)

       b.should_receive(:this) do
         b.should_receive(:that)
       end

       b.that
       b.this
     end
   end

Works as expected (the above fails)

I''ll have to investigate further...

Ashley

-- 
http://www.patchspace.co.uk/
http://aviewfromafar.net/