Csound Csound-dev Csound-tekno Search About

[CSOUND-DEV:3233] more added to cs.h, fix in Opcodes/scansynx.c

Date2003-10-28 23:10
Fromstevenyi
Subject[CSOUND-DEV:3233] more added to cs.h, fix in Opcodes/scansynx.c
Hi all,

I added to csound5 cvs function unquote and ldmemfile to cs.h, csdl.h,
and oload.c.  Also, I've uncommented out a line in Opcodes/scansynx.c as
it was using an extern to unquote that wasn't necessary with it being
defined in the other files above. 

That helped me get a bit further for loading in the scanned synthesis
opcode libraries, but I haven't figured out what to do now as the
libraries are looking for cglob.exitjmp to longjmp to.  Any help here
appreciated!

BTW: regarding the long thread about return values, this is probably a
case where we should start using them, yes?  It seems a bit shaky to me
that an external object should be longjmp'ing anywhere (that is, if I
understand what's happening correctly). 

Thanks,
steven

Date2003-10-29 07:26
FromJohn ffitch
Subject[CSOUND-DEV:3234] Re: more added to cs.h, fix in Opcodes/scansynx.c
On Tue, 28 Oct 2003, stevenyi wrote:

> opcode libraries, but I haven't figured out what to do now as the
> libraries are looking for cglob.exitjmp to longjmp to.  Any help here
> appreciated!
> 

cglob.exitjmp -->> pcglob->exitjmp

==John

Date2003-10-29 07:28
FromJohn ffitch
Subject[CSOUND-DEV:3235] Re: more added to cs.h, fix in Opcodes/scansynx.c
On Tue, 28 Oct 2003, stevenyi wrote:

> BTW: regarding the long thread about return values, this is probably a
> case where we should start using them, yes?  It seems a bit shaky to me
> that an external object should be longjmp'ing anywhere (that is, if I
> understand what's happening correctly). 
> 

All opcodes now return a value; 0 for success, and an error code otherwise 
(usually 1 at present).  This value is not strongly used yet, but the 
basic code is in place.  Simplifies perferror and initerror.  I tink the 
result is user in the perf loop.

==John