Testing styles, techniques & strategies

[James Mead]

This is probably related to James Adam's question about the benefits of
Cucumber's natural language and Murray's question about the trend
towards only testing with Cucumber, but I'd be interested in learning
more generally about the pros & cons of different testing styles,
techniques & strategies.

Echoing one of James Adam's comments, I'm not interested in hearing
about the features of the latest shiny testing framework, but more in
hearing about specific problems that occur in testing and approaches to
solve these problems.

Nat Pryce & Steve Freeman, the authors of jMock, recently published a
book called "Growing Object-Oriented Software, Guided by Tests" [1]. I
haven't read it all yet, but I did read bits of it as it was published
incrementally on-line [2]. It describes a very specific style of
test-driven development. Although all the examples use Java, they say in
the preface that the book is really about a set of techniques that are
applicable to any OO language including Ruby. I'd be interested in
hearing about how well this style of TDD and these techniques fit in
with Ruby and more specifically with Rails.

I'd also be interested in hearing about the advantages and disadvantages
of using Test Spies [3] (versus other types of Test Double) which is a
topic that has come up quite a bit on the Mocha mailing list.

Cheers, James.

[1] http://www.growing-object-oriented-software.com/
[2] http://www.mockobjects.com/book/
[3] http://xunitpatterns.com/Test%20Spy.html

[Roland Swingler]

> I'm not interested in hearing
> about the features of the latest shiny testing framework, but more in
> hearing about specific problems that occur in testing and approaches to
> solve these problems.

I would be interested in hearing about testing approaches also.

Just to clarify - what I was most interested about in cucumber were
two implementation details which really have nothing to do with
testing at all: 1) Ragel, and how to use it and 2) How to go about
designing a protocol. Cucumber just seemed like a good concrete
example.

Cheers,
Roland

[Murray Steele]

2009/11/3 James Mead <ja...@floehopper.org>:

> I'd also be interested in hearing about the advantages and disadvantages
> of using Test Spies [3] (versus other types of Test Double) which is a
> topic that has come up quite a bit on the Mocha mailing list.

I wonder if you might actually be best placed to give this talk? I
know that you're looking to be convinced about Test Spies for
inclusion in Mocha, but maybe you could use the talk as a way to open
up the debate with the attendees of the Manor after presenting them
with the facts?

Can we convince you to turn this topic suggestion into a talk proposal?

Murray

[James Mead]

Hi Murray,

Hmm. I should've seen that coming ;-)

I only really added that bit about Test Spies as a bit of an after-thought. It's the first bit that I'm more interested in.

Also, given that I'm a bit of a sceptic, I'm not sure I could do justice to "the facts" about Test Spies. I had hoped that someone out there would care passionately enough about them to persuade me of their usefulness.

Cheers, James.