From: Douglas Philips <dgou@ma...> - 2003-08-22 20:36:35
% cat test3.lisp
(defclass t1 () ( (foo :accessor foo :initform :foo)))
(defun test-me ()
(let ((a (make-instance 't1)))
(if (typep a 't1)
(format t "worked~%")
(format t "failed!~%"))))
% clisp test3
% clisp -C test3
Man, I just hate chasing these kinds of bugs. No more -C for me for a
From: Sam Steingold <sds@gn...> - 2003-08-22 20:39:28
> * In message <6E3D596F-D4B5-11D7-AEAF-000393030214@...>
> * On the subject of "CLisp or Me, take three?"
> * Sent on Fri, 22 Aug 2003 11:29:33 -0400
> * Honorable Douglas Philips <dgou@...> writes:
> % cat test3.lisp
> (defclass t1 () ( (foo :accessor foo :initform :foo)))
> (defun test-me ()
> (let ((a (make-instance 't1)))
> (if (typep a 't1)
> (format t "worked~%")
> (format t "failed!~%"))))
> % clisp test3
> % clisp -C test3
thank you very much for your rigorous testing and concise test cases.
please try the appended patch.
> Man, I just hate chasing these kinds of bugs. No more -C for me for a
this just means that the bugs will be discovered later (if ever) and
you will have to deal with them in a release.
Sam Steingold (http://www.podval.org/~sds) running w2k
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
Parachute for sale, used once, never opened, small stain.
RCS file: /cvsroot/clisp/clisp/src/compiler.lisp,v
retrieving revision 1.133
diff -u -w -b -u -b -w -i -B -r1.133 compiler.lisp
--- compiler.lisp 20 Aug 2003 22:56:59 -0000 1.133
+++ compiler.lisp 22 Aug 2003 20:27:01 -0000
@@ -4250,7 +4250,7 @@
:seclass '(NIL . NIL)
:code (if *for-value*
- `((CONST ,(if *compiling-from-file*
+ `((CONST ,(if *fasoutput-stream*
(if (and (symbolp form) (c-constantp form))
(make-const :horizon ':all :form form
:value (c-constant-value form))
From: Douglas Philips <dgou@ma...> - 2003-08-23 21:48:16
On Friday, Aug 22, 2003, at 16:39 US/Eastern, Sam Steingold wrote:
> thank you very much for your rigorous testing and concise test cases.
> please try the appended patch.
The cvs HEAD that I checked out this AM seems to have that patch!
It looks as if it solves test1, test2 and test3 issues.
Thanks for such fast turn around!
>> Man, I just hate chasing these kinds of bugs. No more -C for me for a
> this just means that the bugs will be discovered later (if ever) and
> you will have to deal with them in a release.
I really hope that the two branch source will make that a lot less
painful if/when it happens.
Sorry for being so terse/harsh. So far the test cases have been
somewhat easy to identify. I went looking at compiler.lisp and am
puzzled by the dichotomy of tests, and when testing *fasoutput-stream*
is done and when testing *compiling-from-file* (and what about plain
ol' *compiling* ??)
I'll keep playing with -C, but only after I believe the code works
interpreted and as compiled as a file. ;-)
From: amit phalgune <phalgune@cs...> - 2003-08-23 21:53:44
I am getting the following error .
Does anyone have an idea what this error is. and How I can solve it.
*** - DEFGENERIC DESTROY: The only valid method combination is STANDARD :