If you run the shell comand 'nm' you can see what symbols are exported. Just do:
nm (pdlib whatever that may be) | grep canvas_load and look at the man page of nm to find out what the result means. It could also be that whatever is looking for that symbol is looking in the wrong path. I believe the full path will show up in the console or system error log. On OSX it's in /Apps/Utils/Console.
wes
On 9/22/07, Claude Heiland-Allen claudiusmaximus@goto10.org wrote:
Hello GEM-dev!
IOhannes m zmoelnig wrote:
nevertheless, ask claude :-)
I just subscribed to this list, I'm reading the recent archives, so if I miss something in particular then let me know. Apologies for breaking your threading mail clients ;)
IOhannes m zmoelnig wrote:
the example works, but pdlua gives a lot of warnings and you cannot yet create a full-fledged (that passes on the GemList) gem-object in lua (both things because of the way pointers are handled)
True. I'm not sure what to do about pointers, on the one hand pd internal objects seem to do some kind of mystic voodoo for pointers, on the other many (all?) external objects just do whatever they like. I guess I'll follow the crowd and handle pointers the way other externals do.
marius schebella wrote:
when I create a [lua] object in pd i get an error
I don't have any Mac OS X machines to test with, so it's probably a bug in the pdlua Makefile. I'll look at some other Makefiles to see what special voodoo OS X requires (I think the bug is that lua.pd_darwin should be able to see symbols in the pd executable, but can't for some reason).
Moreover, creating a [lua] object isn't supposed to work yet (it's intended to be used with -lib lua) - but I will add that feature soon.
Also, I just want to note that pdlua is very fresh/new/alpha, and will probably change incompatibly at various points until something vaguely useable emerges.
Thanks,
Claude
http://claudiusmaximus.goto10.org
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev