freenode/#clasp - IRC Chatlog
Search
12:16:14
frgo
::notify drmeister Q: I do have an extension now almost getting through the compiler... I now get: In file included from ../../src/gctools/gc_interface.cc:815:
12:16:14
frgo
./generated/initClassesAndMethods_inc.h:5865:46: error: incomplete type 'clasp_ext_rti_dds::RTIDDSExposer_O' named in nested name specifier
12:16:14
frgo
- This is due to the fact that gc_interface.cc does not have the include file that contains the new class definition of the extension. I could add this manually, but the problem is here that we don't have a clear separation of extension code to clasp core code. I'd rather find a way of generating an include file with a well defined name containing all extension include files. That single include file could then be a
12:41:36
frgo
::notify drmeister I opened a new issue regardimg doing #define free in clasp header file as this causes problems. https://github.com/drmeister/clasp/issues/403
16:52:24
Colleen
drmeister: frgo said 4 hours, 36 minutes ago: Q: I do have an extension now almost getting through the compiler... I now get: In file included from ../../src/gctools/gc_interface.cc:815:
16:52:24
Colleen
drmeister: frgo said 4 hours, 10 minutes ago: I opened a new issue regardimg doing #define free in clasp header file as this causes problems. https://github.com/drmeister/clasp/issues/403
16:57:16
drmeister
I may have had that in there for the concurrent queue, which I am not using anymore.
16:58:01
drmeister
Bike: You mentioned some Common Lisp that uses a stub class to represent what gets generated by DEFCLASS at compile time - could you remind me what that was?
16:58:57
Bike
i've been wondering about implementing something like that for a while, but it would break everything without sandboxes
17:02:48
Bike
Yes, what I meant is that I know this because I wrote some code to see what classes exist at compile time.
17:06:56
Bike
something like (defclass foo () ((bar :initarg :bar :accessor bar))) (eval-when (:compile-toplevel) (print (find-class 'foo))), i think.