Removal of should_be(), Hamcrest vs. FEST

[Eric Lefevre-Ardant]

Hi everyone,

Just upgraded to Narrative 1.0.

I wanted to mention that I rather liked the flow of reading made available by 'should_be' (but I do appreciate having less verbose code, too).

So my test went from

  Then.the(quant).expects_that(the_number_of_barcharts_for(CAC_40, EUR, Daily)).should_be(equalTo(1));

to

  Then.the(quant).expects_that(the_number_of_barcharts_for(CAC_40, EUR, Daily), equalTo(1));

An interesting side effect is that the signature of the 2-parameter expects_that() is now closer to the signature of the assertThat() method from JUnit 4. Probably a good thing when using Hamcrest matchers.

A natural conclusion is that, supposing that Narrative was to support FEST (which I might work on... some day), it could lead to a signature like this:

  Then.the(quant).expects_that(the_number_of_barcharts_for(CAC_40, EUR, Daily)).isEqualTo(1);

which would be closer to how FEST is often being used.

Of course, we (at Algodeal) also have our own wrappers around FEST. Integrating them with Narrative is probably quite hard.

Eric

[Douglas Squirrel]

Thanks Eric. By FEST did you mean http://code.google.com/p/fest/?
________________________________________
From: narrati...@googlegroups.com [narrati...@googlegroups.com] On Behalf Of Eric Lefevre-Ardant [er...@ericlefevre.net]
Sent: 29 November 2010 09:52
To: narrati...@googlegroups.com
Subject: Removal of should_be(), Hamcrest vs. FEST

[Eric Lefevre-Ardant]

yes. We only use the Fluent Assertions part.