In the course of attempting to make Csound into a JavaSound thing and a
VST-2 plugin, I am finding it helpful to place all the variables and
function declarations that are required for this purpose into a single .h
and .c file. When this work is completed, I imagine that most of the Csound
internal variables and global functions will end up in these files. At that
point, it would be possible to make them members of a Csound C++ class.
After that, it would be possible to run more than one instance of Csound at
the same time out of the same library.

The other thing I am considering doing (but have not actually done) is to
define abstract device driver classes for MIDI input and output, line
events, and audio input and output. There already are abstract functions for
audio input and output, so this is not so awful. The real mess is the
soundfile input and output and the MIDI input and output.

Unfortunately, JavaSound's limitations are showing up. It does not use
DirectX for audio input or output, and it does not have any official MIDI
input. I am hoping these limitations are resolved soon, either by Sun or by
independent developers. If they are not resolved, the JavaSound version of
Csound will not be usable as a live performance instrument.

-----Original Message-----
From: jpff@maths.bath.ac.uk 
To: csound@maths.ex.ac.uk 
Date: Thursday, August 05, 1999 4:33 PM
Subject: Re: random csound questions


>Message written at 5 Aug 1999 13:27:03 +0100
>--- Copy of mail to jasonf@ivy.the-i.net ---
>In-reply-to: 
> (jasonf@ivy.the-i.net)
>References:  
>
>>>>>> "jasonf" == jasonf   writes:
>
> jasonf> Anyway, when I compile I get the following error:
> jasonf> pgcc -O6 -DRTAUDIO -DSFIRCAM -DLINUX -DPIPES -DMACROS -c rdscor.c
> jasonf> rdscor.c:4: initializer element is not constant
> jasonf> make: *** [rdscor.o] Error 1
>
> jasonf> Any ideas?
>
>The line in question is
>FILE   *oscfp = stdout;     /* default Cscore outfile */
>
>and that assumes that stdout is a constant or the compiler is really a
>C++ one.  On many systems it is not the case, and so in reality the
>code is wrong.
>
>I think the simple fix is to replace this line by
>FILE   *oscfp = NULL;     /* default Cscore outfile */
>
>I have not checked thsi yet but I think it is safe
>
> jasonf> Also, I've been playing around with midi input the last few
> jasonf> days. How do I set the bend range? I see info on the subject
> jasonf> in the manual, but I'm more of a learn by example kind of
> jasonf> person. If you don't set the bend range,
>
>There is an optional parameter in the opcode to read the bend which
>scales it.
>
>==John ffitch