[slime-devel] [PATCH] ABCL broken with contrib/swank-asdf.lisp
ABCL uses a very primitive version of ASDF because of the lack of an implementation of the long form of DEFINE-METHOD-COMBINATION. This causes Stas Boukarev's implementation on Tue Dec 15 of SLIME-RELOAD-SYSTEM to fail under ABCL with complaints of being unable to resolve the symbol ASDF:AROUND. The attached patch at least allows ABCL to load SLIME but makes the SLIME-RELOAD-SYSTEM still recompile the dependencies leading to slightly inconsistent behavior. I was already looking into how to implement the long form of DEFINE-METHOD-COMBINATION for ABCL, but even if I get that working, we will still have a fair number of ABCL implementations broken with SLIME HEAD until they update their implementations, so I would like to patch SLIME in some manner. I have commit rights to SLIME to add my kludge, but would like to solicit the comments of other maintainers before I add such macro conditionals to other's code. Comments please? -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."
Mark Evenson <evenson@panix.com> writes:
ABCL uses a very primitive version of ASDF because of the lack of an implementation of the long form of DEFINE-METHOD-COMBINATION. This causes Stas Boukarev's implementation on Tue Dec 15 of SLIME-RELOAD-SYSTEM to fail under ABCL with complaints of being unable to resolve the symbol ASDF:AROUND.
The attached patch at least allows ABCL to load SLIME but makes the SLIME-RELOAD-SYSTEM still recompile the dependencies leading to slightly inconsistent behavior.
I was already looking into how to implement the long form of DEFINE-METHOD-COMBINATION for ABCL, but even if I get that working, we will still have a fair number of ABCL implementations broken with SLIME HEAD until they update their implementations, so I would like to patch SLIME in some manner.
I have commit rights to SLIME to add my kludge, but would like to solicit the comments of other maintainers before I add such macro conditionals to other's code. Comments please? Fine by me. I've applied it. Looking forward to the long form of d-m-c.
-- With Best Regards, Stas.
participants (2)
-
Mark Evenson -
Stas Boukarev