Andreas Fuchs <asf@...> writes:
> On 2003-05-19, Andreas Fuchs <asf@...> wrote:
>> This patch ought to fix this, and add a few more simple-error
>> diagnostics for error cases which might pop up.
> And to continue in my glorious reply-to-a-patch message tradition, here
> is a patch which adds another check (and not, as before, fixes a stupid
> bug in the previous patch (-;), namely that for an &allow-other-keys
> without a &key keyword.
Thank you, I've merged this into sbcl-0.8.0.11.
As a general rule, I'd like to encourage patch submitters at least to
run their patched version through the test suite. In this case, it
caught one malformed specialized lambda list that wasn't flagged as
such (in tests/clos.impure.lisp); it didn't cause me too much worry,
because the problem was fairly obvious, but at least from my point of
view having a cogent argument as to why the test in the test suite is
wrong would be a good thing.
Added extra bonus points are available if submitters include tests to
go with their patches; in this case, working on the tests would have
revealed a problem -- namely, that the compiler was relying on
PARSE-SPECIALIZED-LAMBDA-LIST not to throw an error when going through
to find the relevant source context to print a condition. This
problem has now been fixed, too. :-)
All that notwithstanding, thank you again for the patch.
http://www-jcsu.jesus.cam.ac.uk/~csr21/ +44 1223 510 299/+44 7729 383 757
(set-pprint-dispatch 'number (lambda (s o) (declare (special b)) (format s b)))
(defvar b "~&Just another Lisp hacker~%") (pprint #36rJesusCollegeCambridge)