Re: Some comments on the ansi test suite
Raymond Toy <[email protected]>
| Newsgroups | gmane.lisp.clocc.devel |
|---|---|
| Message-ID | <[email protected]> |
>>>>> "Sam" == Sam Steingold <[email protected]> writes: >> * In message <[email protected]> >> * On the subject of "Re: Some comments on the ansi test suite" >> * Sent on 09 Jul 2002 12:59:43 -0400 >> * Honorable Raymond Toy <[email protected]> writes: >> >> There are also a few other bugs that I forgot to mention: >> >> IOFKTS-LEGACY-881 says all 10 characters are used up. I think this is >> wrong because vector-push-extend is supposed to be used. Sam> but if the supplied string is not actually expandable, an error should Sam> be signaled, as is clearly indicated by the writeup in issue 365. Sam> please revert the patch. Sam> the problem here is that implementations are not required to create Sam> non-expandable strings, so there is certain latitude here. Sam> I suggest that we condition the results on the implementations, Sam> assuming that the "default" implementation would create a Sam> non-expandable string and thus signal an error. I think the test should check if the array a is adjustable or not and have the result depend on that as well as the actual string, and not on the implementation. (Assuming you meant read-time conditionals.) Perhaps something like (check-for-bug :iofkts-legacy-881 (let ((a (make-array 10 :element-type 'character :fill-pointer 0))) (with-output-to-string (s a) (princ 123 s)) (with-output-to-string (s a) (princ 4567 s)) (with-output-to-string (s a) (princ 890 s)) (with-output-to-string (s a) (princ (quote a) s)) (if (adjustable-array-p a) (equalp a "0123456789A") (equalp a "0123456789"))) T) I think it's bad for this test to say some implementation fails just because the "default" thinks arrays with fill pointers are not adjustable. Ray ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf