[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
KCL CALL-NEXT-METHOD bug
- To: gregor.pa@Xerox.COM
- Subject: KCL CALL-NEXT-METHOD bug
- From: Jim Kempf <kempf%hplabsc@hplabs.HP.COM>
- Date: Tue, 17 Feb 87 14:37:24 pst
- Cc: commonloops.pa@Xerox.COM
KCL currently does not handle &ENVIRONMENT parameters correctly. For
example, the following test
(defmacro test (&environment e)
(format T "~S~%" e)
( (a 3) )
(flet ( (b (x) a) )
causes NIL to be printed, even the lexical environment passed to
the macro is considerably richer (e.g., both lexical variables
and a lexically defined function). This also is true of MACROLET.
Anyway, this causes the loading sequence for high.l to break
as soon as it tries to expand CALL-NEXT-METHOD, since the expansion
for CALL-NEXT-METHOD is heavily dependent on the lexically defined
macro defined during method generation.
While I think this is a problem with KCL (and *not* PCL), are there
any quick, short term workarounds suggested?
Jim Kempf email@example.com
PS: The backtrace during loading of high.l is: