jsynthlib-devel — Mailing list for developers and users interested in improving the JSynthLib.

[Jsynthlib-devel] XML Drivers

[Rib R. <ri...@gm...>]

Well, I committed what I have so far.  Once again eclipse gave errors
during the commit, so let me know if it looks like something's
missing.  I've added two jar files which you will need to put in your
classpath.  These are for the Groovy scripting language, which has
most of the features of languages like Ruby and Python, but is
designed specifically to integrate with Java.
This stuff will still be pretty useless and untested until I add
editor support and sending and receiving patches. Once that's done (it
shouldn't be too hard, but I'm going to have less time to work on it
right now), EditorBuilder needs to be modified to remove sun's code so
it can be GPL'd, and updated to work with XML drivers. If I recall
correctly, the only part that needs the sun license is the method that
draws the arrows between widgets. Oh, and I'm going to have to
document this so people can start using it! ;-)

[Jsynthlib-devel] Converters Not Working

[Jeff W. <jww...@ya...>]

I started working on a Converter for the Bass Pod and
I immediately ran into a problem. After I added the
converter to my config, I tried opening a library
window and selecting Get... (making sure the library
window has the focus). The Get Sysex Data dialog does
not appear and the following exception is generated:
java.lang.NullPointerException
	at
core.SysexGetDialog$DriverActionListener.actionPerformed(SysexGetDialog.java:231)
	at
javax.swing.JComboBox.fireActionEvent(JComboBox.java:1197)
	at
javax.swing.JComboBox.contentsChanged(JComboBox.java:1268)
	at
javax.swing.AbstractListModel.fireContentsChanged(AbstractListModel.java:100)
	at
javax.swing.DefaultComboBoxModel.setSelectedItem(DefaultComboBoxModel.java:88)
	at
javax.swing.DefaultComboBoxModel.addElement(DefaultComboBoxModel.java:126)
	at javax.swing.JComboBox.addItem(JComboBox.java:690)
	at
core.SysexGetDialog$DeviceActionListener.actionPerformed(SysexGetDialog.java:201)
	at
javax.swing.JComboBox.fireActionEvent(JComboBox.java:1197)
	at
javax.swing.JComboBox.contentsChanged(JComboBox.java:1268)
	at
javax.swing.AbstractListModel.fireContentsChanged(AbstractListModel.java:100)
	at
javax.swing.DefaultComboBoxModel.setSelectedItem(DefaultComboBoxModel.java:88)
	at
javax.swing.DefaultComboBoxModel.addElement(DefaultComboBoxModel.java:126)
	at javax.swing.JComboBox.addItem(JComboBox.java:690)
	at core.SysexGetDialog.<init>(SysexGetDialog.java:75)
	at
core.Actions$GetAction.actionPerformed(Actions.java:896)
	at
javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:1819)
	at
javax.swing.AbstractButton$ForwardActionEvents.actionPerformed(AbstractButton.java:1872)
	at
javax.swing.DefaultButtonModel.fireActionPerformed(DefaultButtonModel.java:420)
	at
javax.swing.DefaultButtonModel.setPressed(DefaultButtonModel.java:258)
	at
javax.swing.AbstractButton.doClick(AbstractButton.java:321)
	at
apple.laf.ScreenMenuItem.actionPerformed(ScreenMenuItem.java:83)
	at
java.awt.MenuItem.processActionEvent(MenuItem.java:589)
	at java.awt.MenuItem.processEvent(MenuItem.java:548)
	at
java.awt.MenuComponent.dispatchEventImpl(MenuComponent.java:285)
	at
java.awt.MenuComponent.dispatchEvent(MenuComponent.java:273)
	at
java.awt.EventQueue.dispatchEvent(EventQueue.java:458)
	at
java.awt.EventDispatchThread.pumpOneEventForHierarchy(EventDispatchThread.java:234)
	at
java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:184)
	at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:178)
	at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:170)
	at
java.awt.EventDispatchThread.run(EventDispatchThread.java:100)

I thought it was something with my Bass Pod Converter
so I removed it from my config (leaving the other Bass
Pod drivers in place) and repeated the test. This time
the Get Sysex Data dialog displays and everything
appears to work normally. I put the converter back in
and the problem comes back. I decided to remove the
Bass Pod from the config entirely and test with other
device/drivers that use a Converter. I cleaned the
config out completely and tested with one
device/driver at a time. The same NullPointerException
occurs on all of the ones I tested. I tested the
following devices, quitting and restarting the program
and cleaning out the config between each test:
Yamaha TG100
Kawai K4
Yamaha DX7
Korg X3 (this one has two converters)
Yamaha FS1R

I tried using my debugger to place breakpoints around
where the exception is happening but for some reason,
my breakpoints aren't working. Anybody have any ideas?
Could somebody run this test and see if they get the
same result?

Thanks,
Jeff


		
_______________________________
Do you Yahoo!?
Win 1 of 4,000 free domain names from Yahoo! Enter now.
http://promotions.yahoo.com/goldrush

Re: [Jsynthlib-devel] SysexWidget.IParamModel/ISender interface

[Hiroo H. <hir...@co...>]

OK, take your time.

I have some questions.  (You don't have to answer.  I can wait to seeing
your code.)

Why some method in Parameter has 'IPatch p' argument.  I guess the
IPatch object is passed to Parameter object via constructor.

How does IPatchDriver.sendParameter(IPatch, Parameter) works.  Parameter
object does not have information about SysexMessage data format.  I
guess what we need is 'void Parameter.send()'.

I must be misunderstanding something.

On Sun, 29 Aug 2004 20:11:50 -0700
Rib Rdb <ri...@gm...> wrote:

Rib> On Sun, 29 Aug 2004 13:22:51 -0500, Hiroo Hayashi
Rib> <hir...@co...> wrote:
Rib> > Rib, are you changing these files for your XML driver?  If so, I'll
Rib> > check-in my fix after your check-in.
Rib> 
Rib> Yes.  What I've done is changed the sysex widgets to support both the
Rib> current param model/sysex senders, as well as a new api: an interface
Rib> Parameter, and a method IPatchDriver.sendParameter(IPatch, Parameter).
Rib>  Here is the parameter interface:
Rib> public interface Parameter {
Rib>     public String getName();
Rib> 
Rib>     /* For numeric parameters */
Rib>     public int getMin();
Rib>     public int getMax();
Rib>     public int get(IPatch p);
Rib>     public void set(IPatch p, int val);
Rib>     
Rib>     /* For list parameters */
Rib>     public String[] getValues();
Rib>     
Rib>     /* For String parameters */
Rib>     public String getString(IPatch p);
Rib>     public int getLength();
Rib>     public void set(IPatch p, String stringval);
Rib> }
Rib> 
Rib> This way all the sysex sender objects aren't necessary, and the
Rib> parameter objects can be shared between instances.  Also, sysex
Rib> widgets can be constructed from the parameter objects instead of
Rib> hardcoding the values to the constructor like is currently done.  This
Rib> is meant to help with editor builder.
Rib> 
Rib> Speaking of the XML driver: I have finished the parser and the plugin
Rib> support.  I merely have to finish implementing the XMLPatch class.
Rib> Hopefully I'll be able to get it done in the next couple days.

-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] Preferences Problem?

[Hiroo H. <hir...@co...>]

Jeff,

Jeff> One question--How do you clear your preferences?

Jeff> > If this is OK, try clear preference.  See FAQ in
Jeff> > documentation.html to
Jeff> > see where the preference is saved.  If you are using
Jeff> > Windows, you can
Jeff> > delete the preference by using regedit.

I need more specific question.  

# 'regedit' (Registory Editor) is a standard Windows application.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] Preferences Problem?

[Jeff W. <jww...@ya...>]

Hiroo,

I think I know what the problem is. I wanted to run
JSynthLib without my drivers installed. I checked out
a new version and compiled without my drivers and ran
the program. The synthdrivers.properties file did not
have my drivers listed but my preferences still did.
As soon as I added my drivers, everything worked
again.

One question--How do you clear your preferences?

Jeff


--- Hiroo Hayashi <hir...@co...> wrote:

> I made a change for AppConfig.  I made all methods
> in AppConfig static. 
> But it should not change the behavior of JSynthLib. 
> Of course I have no
> problem.
> 
> Double check if all *.java files are up-to-date and
> all *.class file is
> updated.
> If this is OK, try clear preference.  See FAQ in
> documentation.html to
> see where the preference is saved.  If you are using
> Windows, you can
> delete the preference by using regedit.
> 
> I've cleared my preference and rerun JSynthLib, but
> I don't see stack
> trace.
> 
> Hope this helps.
> 
> Jeff> Has something changed with the preferences? I
> tried
> Jeff> running a version of the code which I checked
> out Sun.
> Jeff> 8/29. Every time I start the program it throws
> a
> Jeff> NullPonterException:
> Jeff> java.lang.NullPointerException
> Jeff> 	at
> Jeff>
>
core.DevicesConfig.createDevice(DevicesConfig.java:162)
> Jeff> 	at
> core.AppConfig.addDevice(AppConfig.java:317)
> Jeff> 	at
> core.AppConfig.loadPrefs(AppConfig.java:67)
> Jeff> 	at core.PatchEdit.<init>(PatchEdit.java:35)
> Jeff> 	at JSynthLib.main(JSynthLib.java:31)
> Jeff> 
> Jeff> Even after I configure drivers and relaunch,
> the same
> Jeff> thing happens and none of the configs get
> saved.
> Jeff> 
> Jeff> Has anyone else noticed this or is it just me?
> 
> -- 
> Hiroo Hayashi
> 
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic
> Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1
> today.
>
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
>
https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
> 


	
		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail 

Re: [Jsynthlib-devel] Converter Question

[Hiroo H. <hir...@co...>]

Jeff,

You can use Converter for your purpose.

Do you mean;
        your converter worked fine a week ago.
        it does not work with current CVS head.
?

I made a change on this area, but it should not change the behavior. 
(Of course I might make a mistake.)

Make sure your driver does not have any reference to new interfaces,
IDriver, IPatchDriver, IConverter, IPatch, ... etc.  They don't have to
be referred by a driver which extends Driver class.  And their APIs are
unstable, your driver would be unstable if your driver refers them.

On Sun, 29 Aug 2004 13:47:03 -0700 (PDT)
Jeff Weber <jww...@ya...> wrote:

Jeff> I had my Line6 POD drivers working with a version of
Jeff> the code from about a week ago but this weekend when I
Jeff> downloaded the latest version I'm ran into some
Jeff> problems.
Jeff> 
Jeff> The PODs have some special issues that need to be
Jeff> dealt with. Whenever you request any kind of dump,
Jeff> they send a lot of garbage data along with the
Jeff> requested patch. The patch can appear anywhere within
Jeff> the garbage data and there are different lengths to
Jeff> the patches, depending upon what type of patch you
Jeff> have requested (bank, program, or edit buffer). The
Jeff> behavior is unpredictable. It goes beyond what
Jeff> Driver.trimSysex can deal with.
Jeff> 
Jeff> I want to use a Converter to filter out the garbage
Jeff> but judging by recent posts it sounds like the purpose
Jeff> of the Converter has shifted a little bit. I just want
Jeff> to make sure I'm on the right track. Is this still an
Jeff> appropriate use for a Converter?
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] Preferences Problem?

[Hiroo H. <hir...@co...>]

I made a change for AppConfig.  I made all methods in AppConfig static. 
But it should not change the behavior of JSynthLib.  Of course I have no
problem.

Double check if all *.java files are up-to-date and all *.class file is
updated.
If this is OK, try clear preference.  See FAQ in documentation.html to
see where the preference is saved.  If you are using Windows, you can
delete the preference by using regedit.

I've cleared my preference and rerun JSynthLib, but I don't see stack
trace.

Hope this helps.

Jeff> Has something changed with the preferences? I tried
Jeff> running a version of the code which I checked out Sun.
Jeff> 8/29. Every time I start the program it throws a
Jeff> NullPonterException:
Jeff> java.lang.NullPointerException
Jeff> 	at
Jeff> core.DevicesConfig.createDevice(DevicesConfig.java:162)
Jeff> 	at core.AppConfig.addDevice(AppConfig.java:317)
Jeff> 	at core.AppConfig.loadPrefs(AppConfig.java:67)
Jeff> 	at core.PatchEdit.<init>(PatchEdit.java:35)
Jeff> 	at JSynthLib.main(JSynthLib.java:31)
Jeff> 
Jeff> Even after I configure drivers and relaunch, the same
Jeff> thing happens and none of the configs get saved.
Jeff> 
Jeff> Has anyone else noticed this or is it just me?

-- 
Hiroo Hayashi

[Jsynthlib-devel] Preferences Problem?

[Jeff W. <jww...@ya...>]

Has something changed with the preferences? I tried
running a version of the code which I checked out Sun.
8/29. Every time I start the program it throws a
NullPonterException:
java.lang.NullPointerException
	at
core.DevicesConfig.createDevice(DevicesConfig.java:162)
	at core.AppConfig.addDevice(AppConfig.java:317)
	at core.AppConfig.loadPrefs(AppConfig.java:67)
	at core.PatchEdit.<init>(PatchEdit.java:35)
	at JSynthLib.main(JSynthLib.java:31)

Even after I configure drivers and relaunch, the same
thing happens and none of the configs get saved.

Has anyone else noticed this or is it just me?



	
		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail 

[Jsynthlib-devel] Converter Question

[Jeff W. <jww...@ya...>]

I had my Line6 POD drivers working with a version of
the code from about a week ago but this weekend when I
downloaded the latest version I'm ran into some
problems.

The PODs have some special issues that need to be
dealt with. Whenever you request any kind of dump,
they send a lot of garbage data along with the
requested patch. The patch can appear anywhere within
the garbage data and there are different lengths to
the patches, depending upon what type of patch you
have requested (bank, program, or edit buffer). The
behavior is unpredictable. It goes beyond what
Driver.trimSysex can deal with.

I want to use a Converter to filter out the garbage
but judging by recent posts it sounds like the purpose
of the Converter has shifted a little bit. I just want
to make sure I'm on the right track. Is this still an
appropriate use for a Converter?



		
__________________________________
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

Re: [Jsynthlib-devel] SysexWidget.IParamModel/ISender interface

[Jeff W. <jww...@ya...>]

Hi Hiroo,

That should solve my problem. All I would have to do
is add the override to send(IPatchDriver driver, int
value) to my sender.

Jeff
--- Hiroo Hayashi <hir...@co...> wrote:

> I'd like to propose two inner interfaces for
> SysexWidget.
> 
> 1. IParamModel
>     public interface IParamModel {
>         void set(int value);
>         int get();
>     }
> 
> The current ParamModel class depends on Patch class
> which is an
> implementation of IPatch.  It cannot be used for
> other implementation of
> IPatch.  By introducing this interface, SysexWidget
> and its subclasses
> can support any implementation of IPatch.
> 
> 2. ISender
>     public interface ISender {
>         void send(IPatchDriver driver, int value);
>     }
> 
> Now SysexSender.generate() returns a byte array and
> SysexWidget.sendSysex() converts the byte array into
> a SysexMessage and
> send it to MIDI port.  After my proposal
> SysexWidget.sendSysex() will
> just call ISender.send() method and send() method
> does all of the job.
> 
> The main job of SysexSender class becomes sending
> Sysex Message.  Easy
> to understand, isn't it?  More importantly a sender
> has more flexibility. 
> It can send any message, short message, multiple
> message, and mix of
> sysex and short message.
> 
> For example
> WaldorfPulseSingleEditor.WaldorfPulseSender (the
> only sender
> which sends short message, I could find.) becomes as
> follows;
> 
>     class WaldorfPulseSender implements
> SysexWidget.ISender {
>         private int param;
>         private int mult;
> 
>         public WaldorfPulseSender(int param) {
>             this(param, 1);
>         }
> 
>         public WaldorfPulseSender(int param, int
> multiplier) {
>             this.param = param;
>             this.mult = multiplier;
>         }
> 
>         public void send(IPatchDriver driver, int
> value) {
>             ShortMessage msg = new ShortMessage();
>             try {
>                
> msg.setMessage(ShortMessage.CONTROL_CHANGE,
>                         driver.getChannel() - 1,
> param, value * mult);
>                 driver.send(msg);
>             } catch (InvalidMidiDataException e) {
>                 ErrorMsg.reportStatus(e);
>             }
>         }
>     }
> 
> This proposal requires little changes of
> SysexWidget, its subclasses in
> core, SysexSender, and ParamModel.  No change for
> synthdrivers.  I'm
> ready to commit them.
> 
> Rib, are you changing these files for your XML
> driver?  If so, I'll
> check-in my fix after your check-in.
> -- 
> Hiroo Hayashi
> 
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic
> Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1
> today.
>
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
>
https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
> 


		
__________________________________
Do you Yahoo!?
Y! Messenger - Communicate in real time. Download now. 
http://messenger.yahoo.com

[Jsynthlib-devel] SysexWidget.IParamModel/ISender interface

[Hiroo H. <hir...@co...>]

I'd like to propose two inner interfaces for SysexWidget.

1. IParamModel
    public interface IParamModel {
        void set(int value);
        int get();
    }

The current ParamModel class depends on Patch class which is an
implementation of IPatch.  It cannot be used for other implementation of
IPatch.  By introducing this interface, SysexWidget and its subclasses
can support any implementation of IPatch.

2. ISender
    public interface ISender {
        void send(IPatchDriver driver, int value);
    }

Now SysexSender.generate() returns a byte array and
SysexWidget.sendSysex() converts the byte array into a SysexMessage and
send it to MIDI port.  After my proposal SysexWidget.sendSysex() will
just call ISender.send() method and send() method does all of the job.

The main job of SysexSender class becomes sending Sysex Message.  Easy
to understand, isn't it?  More importantly a sender has more flexibility. 
It can send any message, short message, multiple message, and mix of
sysex and short message.

For example WaldorfPulseSingleEditor.WaldorfPulseSender (the only sender
which sends short message, I could find.) becomes as follows;

    class WaldorfPulseSender implements SysexWidget.ISender {
        private int param;
        private int mult;

        public WaldorfPulseSender(int param) {
            this(param, 1);
        }

        public WaldorfPulseSender(int param, int multiplier) {
            this.param = param;
            this.mult = multiplier;
        }

        public void send(IPatchDriver driver, int value) {
            ShortMessage msg = new ShortMessage();
            try {
                msg.setMessage(ShortMessage.CONTROL_CHANGE,
                        driver.getChannel() - 1, param, value * mult);
                driver.send(msg);
            } catch (InvalidMidiDataException e) {
                ErrorMsg.reportStatus(e);
            }
        }
    }

This proposal requires little changes of SysexWidget, its subclasses in
core, SysexSender, and ParamModel.  No change for synthdrivers.  I'm
ready to commit them.

Rib, are you changing these files for your XML driver?  If so, I'll
check-in my fix after your check-in.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] Sequencer support works under Linux with JDK1.5

[Hiroo H. <hir...@co...>]

Hi Torsten,

Torsten> > When I click 'Play', it sends Sysex message but does not send MIDI
Torsten> > note messages.

Torsten> If I'm not wrong these SysexMessages are only the Patch which
Torsten> is send to your synth.  I think the NoteOn/NoteOff messages
Torsten> are send bye the sequencer and not catched by our
Torsten> midi-monitor.
I'm using an external MIDI monitor program (MIDI-OX).

Torsten> But alas the MIDI events of the Midi file are never send to
Torsten> this Receiver, only to Java's default synthesizer.

I guess default Synthesizer (returned by MidSystem.getSysnthsizer()) on
my system is software synth of Windows.  But I hear nothing.

Torsten> It seems that the MIDI Sequencer in the Linux 1.4 Java
Torsten> environment doesn't concern about the new assigned Receiver
Torsten> "outRcvr".  It only plays the midi file over Java's default
Torsten> synthesizer, which means you can hear the song over the "Line
Torsten> Out" output of your soundcard. Is this also correct under
Torsten> Windows?.

It seems my XP system is worse than Linux.

Torsten> Can you make a quick test with jdk1.5?
Torsten> If my sequencer part is working properly the MIDI file should
Torsten> be played over your synthesizer.  If this is the case, I
Torsten> think the wrong behaviour is a Java 1.4 issue and I don't
Torsten> activate this feature until jdk1.5 will become a requirement
Torsten> and my code reached a "stable" state, of course.

I have not decided to install 1.5 yet.
I don't prefer to install beta version.  And I'm afraid having two
version of SDK will cause trouble for me.

You can use core.Utility.getJavaVersion() and disable sequencer
without waiting for the time when 1.5 will be requirement.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] [Fwd: [YamahaDX] Re: about JSynthlib]

[<tt_...@gm...>]

Hi Hiroo,

Hiroo Hayashi wrote:
> All what I did is set PATH and JAVA_HOME for J2SE 1.4.2 on Windows XP.
>  
> (herbie) hiroo[503] echo $JAVA_HOME
> D:\Program Files\j2sdk1.4.2
> 
> I just followed the installation manual.

Since I'm just using Linux I don't concern about this problem by myself.
But I've read about some trouble to get JSynthLib running under Windows in several DX mailinglists.
I've talked again and again to these people to share our mailinglist to report their problems, but alas without success.

Now I found this comment in a DX mailinglist and forward it to our list.
.
I took a look into our documentation.html file under "Installing JSynthLib - Step 3: Installing and Invoking JSynthLib" in the current
CVS version, but can't find a comment about the need to set environment variables like PATH, CLASSPATH or JAVA_HOME.
Maybe it's not common for Windows user to handle environment variables or commandline tools and maybe it could be helpful to add some
comments about this issue.

This was all what I intended.

Bye
Torsten

Re: [Jsynthlib-devel] Sequencer support works under Linux with JDK1.5

[<tt_...@gm...>]

Hi Hiroo,

Hiroo Hayashi wrote:
> Torsten,
> 
> I've tried your sequencer today.
> 
> When I click 'Play', it sends Sysex message but does not send MIDI
> note messages.

If I'm not wrong these SysexMessages are only the Patch which is send to your synth.
I think the NoteOn/NoteOff messages are send bye the sequencer and not catched by our midi-monitor.
I don't see NoteOn/NoteOff messages too, but under jdk1.5 the midi file is played correctly about my synth.

> ------------------------------------------------------------------------
> DBG>JSynthLib: 0.20-alpha, Java: 1.4.2, OS: Windows XP, 5.1
> ...
> core.MidiUtil.startSequencer: seqTrans = com.sun.media.sound.AbstractPlayer$PlayerOutTransmitter@1885088
> core.MidiUtil.startSequencer: output MidiDevice = com.sun.media.sound.MidiOutDevice@5f387c
> core.MidiUtil.startSequencer: outRcvr = com.sun.media.sound.MidiOutDevice$MidiOutReceiver@1756db3
> ...
> ------------------------------------------------------------------------

This looks like my own output and tells that the right Receiver is connected.
But alas the MIDI events of the Midi file are never send to this Receiver, only to Java's default synthesizer.

> These are MIDI related descriptions in Bugs fixed in J2SETM 1.5.0
> Beta1 (http://java.sun.com/j2se/1.5.0/fixedbugs/fixedbugs.html).
> 
> ------------------------------------------------------------------------
>  4773012 java         classes_sound   36 votes   RFE: Implement a new stand-alone sequencer
>  4851018 java         classes_sound   3 votes    MidiMessage.getLength and .getData return wrong values
>  4493775 java         classes_sound   2 votes    Sequencer method, setTickPosition(long) doesnot set the Tick position
>  4782924 java         classes_sound   2 votes    RFE: Linux: Implement MIDI device I/O
>  4204105 java         classes_sound   1 vote     RFE: add loop() method(s) to Sequencer
>  4782900 java         classes_sound   1 vote     RFE: Linux: Implement Ports
> ------------------------------------------------------------------------
> 
> Does this mean the MIDI sequencer is not implemented in 1.4?

I'm suspecting that the MIDI sequencer isn't implemented properly in 1.4.

It seems that the MIDI Sequencer in the Linux 1.4 Java environment doesn't concern about the new assigned Receiver "outRcvr".
It only plays the midi file over Java's default synthesizer, which means you can hear the song over the "Line Out" output of your
soundcard. Is this also correct under Windows?.

Can you make a quick test with jdk1.5?
If my sequencer part is working properly the MIDI file should be played over your synthesizer.
If this is the case, I think the wrong behaviour is a Java 1.4 issue and I don't activate this feature until jdk1.5 will become a
requirement and my code reached a "stable" state, of course.

Bye
Torsten

Re: [Jsynthlib-devel] Patch Structure.

[<tt_...@gm...>]

Hi Hiroo,

Hiroo Hayashi wrote:
> Torsten> Perhaps this could make a future localization (translation
> Torsten> into other languages as english) easier.
> 
> Why making package for dialog helps localization?  I think it is
> opposite.  But the difference would be very little.

I got just the idea of localization as an future task and in my imagination it's easier if all dialogs are placed at one place.
That's all.

But I have to admit that I have no experience in this questions.

Bye
Torsten

Re: [Jsynthlib-devel] Sequencer support works under Linux with JDK1.5

[Hiroo H. <hir...@co...>]

Torsten,

I've tried your sequencer today.

When I click 'Play', it sends Sysex message but does not send MIDI
note messages.

------------------------------------------------------------------------
DBG>JSynthLib: 0.20-alpha, Java: 1.4.2, OS: Windows XP, 5.1
=2E..
core.MidiUtil.startSequencer: seqTrans =3D com.sun.media.sound.AbstractPlay=
er$PlayerOutTransmitter@1885088
core.MidiUtil.startSequencer: output MidiDevice =3D com.sun.media.sound.Mid=
iOutDevice@5f387c
core.MidiUtil.startSequencer: outRcvr =3D com.sun.media.sound.MidiOutDevice=
$MidiOutReceiver@1756db3
=2E..
------------------------------------------------------------------------


These are MIDI related descriptions in Bugs fixed in J2SETM 1.5.0
Beta1 (http://java.sun.com/j2se/1.5.0/fixedbugs/fixedbugs.html).

------------------------------------------------------------------------
 4773012 java         classes_sound   36 votes   RFE: Implement a new stand=
-alone sequencer
 4851018 java         classes_sound   3 votes    MidiMessage.getLength and .=
getData return wrong values
 4493775 java         classes_sound   2 votes    Sequencer method, setTickP=
osition(long) doesnot set the Tick position
 4782924 java         classes_sound   2 votes    RFE: Linux: Implement MIDI=
 device I/O
 4204105 java         classes_sound   1 vote     RFE: add loop() method(s) =
to Sequencer
 4782900 java         classes_sound   1 vote     RFE: Linux: Implement Port=
s
------------------------------------------------------------------------

Does this mean the MIDI sequencer is not implemented in 1.4?

Torsten> But I like to get some feedback, if the sequencer works also
Torsten> under Windows resp. MacOSX.

Torsten> If someone want to take a look to this feature he has to set
Torsten> "useSequencer" in core.NoteChooserConfigPanel to "true",
Torsten> recompile and make the corresponding settings in the config
Torsten> panel.

Torsten> BTW, the function isn't reentrant save and you can't stop the
Torsten> sequencer. So don't play a very long song.  (some of the
Torsten> things I have to do).

Torsten> Maybe the javasound part of the Java 1.4 version isn't
Torsten> generally able to make what I want.
Torsten> But maybe only the Linux port is buggy.
Torsten> This point is very interesting for me.

--=20
Hiroo Hayashi

Re: [Jsynthlib-devel] Patch Structure.

[Hiroo H. <hir...@co...>]

Rib and Torsten,

Rib > Second, I have one more comment on the package structure (although it
Rib > does not conflict with my proposed changes).  Since there are so many
Rib > dialogs, would it not be a good idea to have a package for them, and
Rib > perhaps also a package for the config panels?

I'm not sure which is better.

The number of files under jsynthlib (current core) will be about half
of the current one.  I would say the current core has too many files.
Half of files may be enough to be managed.

On the other hand, to access members in a different package we need
more public members (methods and fields).  A dialog class does not
access members of other dialogs, since there are independent each
other in general.  The package will be a collection of independent
classes, and jsynthlib (core) package needs to have more public
members.

If I have to choose one of the selection, I would take not to create a
package for dialogs.  But this is not strong preference.  Please
create it if think it is better.

Torsten> Perhaps this could make a future localization (translation
Torsten> into other languages as english) easier.

Why making package for dialog helps localization?  I think it is
opposite.  But the difference would be very little.
-- 
Hiroo Hayashi

[Jsynthlib-devel] [Fwd: [YamahaDX] Re: about JSynthlib]

[<tt_...@gm...>]

Hi all Windows XP developer,

I found this message in an DX mailinglist.

I don't now the circumstances why JSyntLib wasn't able to run on these systems, but maybe this helps you to find a bug or to extend the
documentation.

Bye
Torsten

-------- Original Message --------
Subject: [YamahaDX] Re: about JSynthlib
Date: Sat, 28 Aug 2004 06:42:03 -0000
From: warrick_wilson <war...@av...>
Reply-To: Yam...@ya...
To: Yam...@ya...

--- In Yam...@ya..., a33e <a33e@w...> wrote:
> Hi there,
> > 
> > I downloaded and unpacked JSynthlib as per instructions but there
> > appears to be a file missing. The only "JSynthlib" file I can 
find in
> > the zip is a "JSynthlib.class" file. I have the latest JRE 
installed
> > on a winXP system. Am I missing something?
> 
> Exactly same here. I just dont get it how to start this prog.
> I followed instructions carefully without results!!!!!!!!!!!!

I had to come back here looking for messages on JSynthLib. I just
downloaded it for Windows XP, and was having trouble.

The trick is to set the CLASSPATH environment variable to
include "."  (that is, the current directory). Under XP, you can get
at it by doing Start | Control Panel | System | Advanced  and then
clicking the Environment Variables button. Easy, huh?  Then you can
find the CLASSPATH variable and add  ".;" to the front (without the
double quotes, of course).

If you have a command prompt open to the directory (or folder) you
installed JSynthLib into, you can try this out first by entering the
following command at the prompt:

set CLASSPATH=.;%CLASSPATH%

and pressing Enter.

That's what I did and now JSynthLib is working.

You might also want to examine your PATH environment variable to
make sure you're pointing to the proper Java Runtime directory. I
found I was pointing to the 1.4.2_03 directory, and I'd upgraded to
1.4.2_05 a while ago.

Hope this helps someone. I figured I'd post something to try and put
some sort of extra help into the threads so that anyone searching
would actually find some answers.



------------------------ Yahoo! Groups Sponsor --------------------~-->
Over 1 billion served! The most music videos on the web.
Click to Watch now!
http://us.click.yahoo.com/Xkrq7C/IARHAA/n1hLAA/QnLolB/TM
--------------------------------------------------------------------~->


Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/YamahaDX/

<*> To unsubscribe from this group, send an email to:
     Yam...@ya...

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/

[Jsynthlib-devel] Sequencer support works under Linux with JDK1.5

[<tt_...@gm...>]

Hi,

I recently upgraded my Linux JAVA SDK to 1.5 (or 5.0?) and my sequencer function works as expected (with some minor blemishs, of course).

Now I will make some further improvements.
But I like to get some feedback, if the sequencer works also under Windows resp. MacOSX.

If someone want to take a look to this feature he has to set "useSequencer" in core.NoteChooserConfigPanel to "true", recompile and 
make the corresponding settings in the config panel.
BTW, the function isn't reentrant save and you can't stop the sequencer. So don't play a very long song.
(some of the things I have to do).

Maybe the javasound part of the Java 1.4 version isn't generally able to make what I want.
But maybe only the Linux port is buggy.
This point is very interesting for me.

Bye
Torsten

Re: [Jsynthlib-devel] Patch Structure.

[<tt_...@gm...>]

Hi Ryan,

Rib Rdb wrote:
> Second, I have one more comment on the package structure (although it
> does not conflict with my proposed changes).  Since there are so many
> dialogs, would it not be a good idea to have a package for them, and
> perhaps also a package for the config panels?

Perhaps this could make a future localization (translation into other languages as english) easier.
Of course, only if we want to do this.

Bye
Torsten

Re: [Jsynthlib-devel] Re: SysexSender not sending CC Messages

[<tt_...@gm...>]

Hi Jeff,

Jeff Weber wrote:
> If you have any suggestions let me know, or else feel
> free to make changes yourself if you feel so inclined.
> I don't have write access to the CVS repository yet (I
> just sent you another email requesting it), so if you
> want to go ahead and patch this in yourself, feel
> free. Otherwise I can put in a commit later.

If you want to become a developer I can change your state also.
This task doesn't need to be done by Hiroo.

Bye
Torsten

[Jsynthlib-devel] Re: SysexSender not sending CC Messages

[Jeff W. <jww...@ya...>]

Hiroo,

After taking care of family business last night I was
finally able to find some time to work on a simple fix
for the SysexWiget.sendSysex method. I'm guessing this
is probably pretty close to what you had in mind. The
SysexMessage handles sysex only and ShortMessage
handles everything else. Also--the current version
sends 0xF0 0xF7 even if the InvalidMidiDataException
occurs. I assumed that it shouldn't send anything when
the exception occurs so I moved driver.send(m) inside
the try block. Now nothing will get sent on an
exception.

    protected void sendSysex(SysexSender s, int v) {
        if (s != null) { // do it only if there is a
sysex-sender available
			s.channel = (byte) device.getDeviceID();
			byte[] sysex = s.generate(v);
			try {
				MidiMessage m;
				if (sysex[0] == (byte)0xF0) {
					m = new SysexMessage();
					((SysexMessage)m).setMessage(sysex,
sysex.length);
				} else {
					m = new ShortMessage();
					((ShortMessage)m).setMessage(sysex[0], sysex[1],
sysex[2]);
				}
				driver.send(m);
			} catch (InvalidMidiDataException e) {
				ErrorMsg.reportStatus(e);
			}
		}
	}

I haven't done extensive testing with it but I used my
Midi Monitor program to watch the MIDI datastream. I
tested with the Bass Pod and I also loaded up the
Yamaha TX81z and tested with that one. The Pod sends
CC messages and the Yamaha sends sysex, just like
they're supposed to.

If you have any suggestions let me know, or else feel
free to make changes yourself if you feel so inclined.
I don't have write access to the CVS repository yet (I
just sent you another email requesting it), so if you
want to go ahead and patch this in yourself, feel
free. Otherwise I can put in a commit later.

Thanks,
Jeff



		
_______________________________
Do you Yahoo!?
Express yourself with Y! Messenger! Free. Download now. 
http://messenger.yahoo.com

[Jsynthlib-devel] Patch Structure.

[Rib R. <ri...@gm...>]

(I accidentally sent this message to the yahoo groups list.  Please
ignore that one, as it contains an error)

I'm about ready to commit the first part of XML single bank support,
so I want to make sure I put the new files in the right package. After
reviewing the previous thread, here's what it looks like we've suggested so
far (all under org.jsynthlib):

.midi - MIDI support classes
.widgets - Sysex widgets
.desktop - JSL* classes
.editorbuilder - GUI editorbuilder
.utils - MacUtils, ErrorMsg
.jsynthlib - current core
.jsynthlib.synthdrivers - all (xml and java) drivers
.jsynthlib.xml - XML driver and support files

So, first of all, when I commit my changes I propose to create four package=
s:
org.jsynthlib.utils - I will place several generic xml utilities here
                           (it would be confusing to have both org.jsynthli=
b.xml
                            and org.jsynthlib.jsynthlib.xml)
org.jsynthlib.jsynthlib.xml - XML driver's parser, factories, etc.
org.jsynthilb.jsynthlib.plugins - checksum and decoding algorithms for
XML drivers
org.jsynthlib.jsynthlib.synthdrivers.yamaha.motif - YamahaMotif.xml
and NormalVoice.xml

Second, I have one more comment on the package structure (although it
does not conflict with my proposed changes). =A0Since there are so many
dialogs, would it not be a good idea to have a package for them, and
perhaps also a package for the config panels?

Re: [Jsynthlib-devel] Interfaces

[Hiroo H. <hir...@co...>]

Rib,

I still don't understand.  Sorry for my poor English.

Rib> > Rib> Also, I was thinking it might be better to have
Rib> > Rib> IPatchDriver.supportsMessage so that each message can be checked
Rib> > Rib> instead of just whatever's at the beginning.  Then it would probably
Rib> >
Rib> > Sorry, I don't understand the purpose of this method.  Does this check
Rib> > if IPatch.getDriver() returned correct value?  (I guess this is wrong.)
Rib> 
Rib> I was thinking of using this instead of supportsPatch(String, byte[])
Rib> in the patch factory method.  If there are multiple messages it can
Rib> check if there is a single driver that supports all the messages. If
Rib> so, it will use that driver; otherwise it can send messages
Rib> individually to drivers that support them.

I understand you mean multiple patch (IPatch) by multiple message
instead of multiple sysex message (SysexMessage).  Am I correct?

If I'm correct, this is the only job of Converter.  Converter's
createPatch() divides the input message into multiple patch and assigns
a proper driver onto each patch by using IDriver.supportsPatch().  The
process does not care if all of the patch are supported by a single
driver or not.

And now a converter can override createPatch() method directly instead
of Converter.extractPatch().  Converter has more flexibility than before.

Am I on the same page with you?
-- 
Hiroo Hayashi

[Jsynthlib-devel] SysexSender not sending CC Messages

[Jeff W. <jww...@ya...>]

Hiroo,

Somehow I lost your last message and I could not find
it in the mailing list archive either.  If I'm
understanding correctly, I believe what you were
suggesting was to use Driver.send to send the CC
message directly from my SysexSender subclass. Trouble
is, Driver.send is an instance method and I don't have
a reference to the driver object (or the Device object
for that matter) from within my SysexSender. Secondly,
even if I could get it to work, SysexSender.sendSysex
will still continue to send 0xF0 0xF7 in between all
of my CC messages. That probably wouldn't hurt
anything but it's probably not what we really want.

All of my SysexSender subclasses were already tested
under version 0.18 anyway so it's not something I need
to have addressed right away.  It sounds like you guys
already have enough on your plate right now as it is.

In the mean time if you'd like, I'd be happy to
volunteer to muck around with the
SysexSender.sendSysex method and see if I can come up
with a solution. It doesn't seem to me like it should
be that difficult. I won't commit anything until I've
given you a chance to review it. What do you think?

Thanks,
Jeff


		
__________________________________
Do you Yahoo!?
Y! Messenger - Communicate in real time. Download now. 
http://messenger.yahoo.com