I mean the API calls Csound::Perform (usually from a separate thread). Then, for example, the user decides he or she has heard enough, so the user clicks on a Stop button which calls Csound::Stop() and the Perform() call returns immediately. I have code for this in CppSound::perform and CppSound::stop, but I want to push this code, or some other code that does the same thing, down into the CSOUND object.

Regards,
Mike

-----Original Message-----
From: Istvan Varga 
Sent: Oct 28, 2005 11:18 AM
To: csound-devel@lists.sourceforge.net
Subject: Re: [Cs-dev] interfaces lib building in windows with MSVC

Michael Gogins wrote:

> I also am not sure where to put the file building code and what to do about
 > interrupting performances from the API.
> 
> I think I will push the performance interrupting stuff down into the "C" API anyway.

What do you exactly mean by interrupting performance from the API ?


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel





-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Michael Gogins wrote:

> I mean the API calls Csound::Perform (usually from a separate thread).
 > Then, for example, the user decides he or she has heard enough, so the
 > user clicks on a Stop button which calls Csound::Stop() and the Perform()
 > call returns immediately. I have code for this in CppSound::perform and
 > CppSound::stop, but I want to push this code, or some other code that
 > does the same thing, down into the CSOUND object.

Is a function that takes an integer argument (0: play, 1: pause, 2: stop)
suitable ? What name should be used ?


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thinking more about this issue, I am now not so sure if running
csoundPerform() in a separate thread is really so useful, perhaps
with the exception of the specific application where you have a
GUI that allows stopping (and maybe pausing) the performance, but
nothing more. As Csound was not designed at all for allowing access
to the same CSOUND* instance from multiple threads at the same time,
any of the more advanced functionality such as setting the score
offset in real time, triggering events, using the software bus, etc.,
would be unsafe. Unlike in the case of csoundPerformKsmps() and other
functions that only perform for a small amount of time, with
csoundPerform() the host application cannot implement reliable thread
synchronization.

Istvan Varga wrote:

> Michael Gogins wrote:
> 
>> I mean the API calls Csound::Perform (usually from a separate thread).
> 
>  > Then, for example, the user decides he or she has heard enough, so the
>  > user clicks on a Stop button which calls Csound::Stop() and the 
> Perform()
>  > call returns immediately. I have code for this in CppSound::perform and
>  > CppSound::stop, but I want to push this code, or some other code that
>  > does the same thing, down into the CSOUND object.
> 
> Is a function that takes an integer argument (0: play, 1: pause, 2: stop)
> suitable ? What name should be used ?



-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This should be done, I think, using the csoundYield callback.

The MacOS 9 front end already has a "transport" for pausing, stopping, and
resuming a performance.  It sets a yield callback to check for events and
when a click on the pause button is detected, it does not return to Csound
until the user clicks Play or Stop.  If the user clicks Stop, the yield
callback returns an error code which causes csoundPerform to end the
performance.

This all works fairly well (and has worked for nearly ten years).

Anthony Kozar
anthonykozar AT sbcglobal DOT net
http://akozar.spymac.net/

Michael Gogins wrote on 10/28/05 11:43 AM:

> I mean the API calls Csound::Perform (usually from a separate thread). Then,
> for example, the user decides he or she has heard enough, so the user clicks
> on a Stop button which calls Csound::Stop() and the Perform() call returns
> immediately. I have code for this in CppSound::perform and CppSound::stop, but
> I want to push this code, or some other code that does the same thing, down
> into the CSOUND object.



-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Anthony Kozar wrote:

> This should be done, I think, using the csoundYield callback.

I think the problem with the yield callback would be that the
interface is expected to be usable from high level languages,
and you may have problems setting callback functions that are
expected to have a "C" interface from a scripting language
like Python.


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net