[plt-dev] random testing, we need to do more of this

Carl Eastlund <carl.eastlund at gmail.com> writes:

> On Mon, Jan 25, 2010 at 9:30 AM, Michael Sperber
> <sperber at deinprogramm.de> wrote:
>>
>> Noel Welsh <noelwelsh at gmail.com> writes:
>>
>>> On Mon, Jan 25, 2010 at 8:05 AM, Michael Sperber
>>> <sperber at deinprogramm.de> wrote:
>>>> I'll point out that a quite complete port of the original QuickCheck is
>>>> sitting in the deinprogramm/quickcheck collection.  I wrote it for our
>>>> teaching languages (where feedback indicates it's a success), but it's
>>>> by no means restricted to that.
>>>
>>> For the sake of completeness, there is also fasttest on Planet.
>>
>> Right.  But note that it's different from QuickCheck: It features a
>> bunch of procedures that generate random values of various types,
>> whereas QuickCheck is a combinator library for generators.  That comes
>> into play with compound generators - such as generators for procedures.
>
> FastTest is also a combinator library.  For instance, it has higher
> order generators for lists and s-expressions.  If there's a useful set
> of generators that it's missing, I'm happy to take feature requests.

Sorry if I mischaracterized - FastTest does have higher-order
generators, but no generator datatype, so compositionality is limited.
The most useful set of generators that's missing is what I mentioned -
generators for procedures.  (And these, as far as I know, can't be
constructed from simple generator procedures, as used by FastTest.)

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla