[rspec-devel] Are rspec code actually "driven" out from the spec incrementally?

David Chelimsky dchelimsky at gmail.com
Thu Jun 21 23:17:04 EDT 2007


On 6/21/07, Huang Liang <exceedhl at gmail.com> wrote:
> :-)
>
> I did not see that it is impossible. It's exciting for me to know that a
> tool can be actually driven out by itself because I have not done such kind
> of thing before. Many tools like CruiseControl eat their own dog food, but
> that happens after the tool has been in a minimal usable state.
>
> What I guess is if I want to use rspec to drive out the rspec, a spec might
> look like:
>
> describe Something, " ..." do
>       it "..." do
>         # setup fixture
>         # action and verify using should...
>       end
> end
>
> but in order to make this pass, actually we have to let rspec support
> several base functionalities like describe, it, should, which might be too
> much for just finishing one test. And when you are trying to fullfill this
> spec(which is like a test), you are not focusing on the functionality
> described by the spec, you have to work on some other functionality to make
> this spec work first and see the test result.
>
> Moreover, if we are using stories to manage the requirements of rspec, we
> may found the work contained in one story is not a vertical slice of the
> whole functionality. The story contains much other functionalities that
> spreads horizontally across different parts, which makes the story not
> small, very testable and estimable and describing a very clear business
> value.
>
> The above is just my guess. I'd like to know how the dev team actually drive
> out this great tool.

I misunderstood your initial question. Forgive me. The first versions
of rspec were developed in small steps using test/unit. Eventually we
got to a point where we decided it was dog food time. At that point we
converted our test/unit tests to rspec examples, and since then any
new developments have been driven directly with rspec.

Sorry if that's not as exciting as you had imagined. I wonder how
things would have been different if we had started with rspec
self-hosting from day 1.

Cheers,
David

>
>
> On 6/22/07, David Chelimsky < dchelimsky at gmail.com> wrote:
> > On 6/21/07, Huang Liang < exceedhl at gmail.com> wrote:
> > > The spec in the source code package are very good examples and
> description
> > > for rspec. When I am looking at those spec, I wonder if the rspec are
> > > actually driven out by those spec step by step or we just add those spec
> > > after the core functionality of rspec has been finished?
> >
> > What do you think? :)
> >
> > >
> > > _______________________________________________
> > > rspec-devel mailing list
> > > rspec-devel at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/rspec-devel
> > >
> > _______________________________________________
> > rspec-devel mailing list
> > rspec-devel at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-devel
> >
>
>
> _______________________________________________
> rspec-devel mailing list
> rspec-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-devel
>


More information about the rspec-devel mailing list