[Python-Dev] Directions for reproducing the coredump (original) (raw)
M.-A. Lemburg mal@lemburg.com
Sat, 12 Aug 2000 13:36:14 +0200
- Previous message: [Python-Dev] Directions for reproducing the coredump
- Next message: [Python-Dev] Directions for reproducing the coredump
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
"Eric S. Raymond" wrote:
Ka-Ping Yee <ping@lfw.org>: > One thing i ran into as a result of trying to run it under the > debugger, though: turning on cursesmodule was slightly nontrivial. > There's no cursesmodule.c; it's cursesmodule.c instead; but > Modules/Setup says "#curses cursesmodule.c". Taking out the "#" > wasn't sufficient; i had to edit and insert the underscores by hand > to get curses to work. Your Setup is out of date. But this reminds me. There's way too much hand-hacking in the Setup mechanism. It wouldn't be hard to enhance the Setup format to support #if/#endif so that config.c generation could take advantage of configure tests. That way, Setup could have constructs in it like this: #if defined(CURSES) #if defined(linux) curses cursesmodule.c -lncurses #else curses cursesmodule.c -lcurses -ltermcap #endif #endif I'm willing to do and test this.
This would be a cool thing to have :-)
Definitely +1 from me if it's done in a portable way.
(Not sure how you would get this to run without the C preprocessor though -- and Python's Makefile doesn't provide any information on how to call it in a platform independent way. It's probably cpp on most platforms, but you never know...)
-- Marc-Andre Lemburg
Business: http://www.lemburg.com/ Python Pages: http://www.lemburg.com/python/
- Previous message: [Python-Dev] Directions for reproducing the coredump
- Next message: [Python-Dev] Directions for reproducing the coredump
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]