Hi!
On Thu, May 10, 2001 at 09:56:34PM -0400, Jeff Tranter wrote:
> It looks good to me. I could add some Doxygen comments to cslsample.c if
> you want -- should I do that now or would it be better to wait?
> I'll add the two new files to the Doxygen config file now.
Well, I did a few changes about the error handling part today, and it works
now, so I don't think too much will happen anymore, but there might be still
some changes (Tim hasn't reviewed it yet).
> The example you mailed is much simpler than the one
> in csl/cslplay.c. Do you have to write the samples
> using csl_sample_write or will you be able to just
> specify an existing file?
You have the choice:
* either you create the sample with csl_sample_new, fill it with data with
csl_sample_write and finish it with csl_sample_finish
* or you use csl_sample_new_from_file instead, which will do this for you
In any case, you can only use samples after they have been completely
initialized.
> Will samples only be cached if the user explicitly calls
> csl_sample_cache_add() or is it just a hint to the underlying
> backend driver ?
As long as you actually use the sample (i.e. between csl_sample_new(_from_file)
and csl_sample_free) they will be cached (on the machine where the sound server
runs). After that, they will be deleted from there again, unless you call csl_
sample_cache_add.
So basically, you can use csl_sample_cache_add/find/remove to explicitely store
files on the sound server, and keep them there, which will for instance lessen
the overhead if you have fifty applications sharing the same set of desktop
scheme sounds (as each sample will be stored at most once).
Cu... Stefan
--
-* Stefan Westerfeld, [EMAIL PROTECTED] (PGP!), Hamburg/Germany
KDE Developer, project infos at http://space.twc.de/~stefan/kde *-
