A simpler test framework?

[Ben Finney]

Jonathan Lange <jml at mumak.net> writes:

> On Tue, Jul 14, 2009 at 5:47 PM, Martin Pool<mbp at sourcefrog.net> wrote:
> > Doctest has a bit of a bad reputation at least around Canonical for
> >causing some testing antipatterns
> 
> For the record, many people at Canonical think doctests are actually
> pretty good.
> 
> Of course, they are wrong.

I think doctests are pretty good — for testing narrative examples
interspersed within documentation (“narrative tests”). That's a good
purpose, and doctest does it simply and well.

But narrative tests are horrible for unit testing.

-- 
 \          “Those who write software only for pay should go hurt some |
  `\                 other field.” —Erik Naggum, in _gnu.misc.discuss_ |
_o__)                                                                  |
Ben Finney