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

[Jsynthlib-devel] AW: Eclipse: Indentation / Displayed tab width

[Joachim B. <jba...@pi...>]

Hi,

I searched in the old mails and found the following:

(written by Hiroo)
> The problem is the default "Displayed Tab Width"
> (Preferences->Java->Editor) is set to 4.  This is bad choice.  Most
> tools (ex. cat on console, CVS browser on JSynthLib=20
> SourceForge site, ....)
> assumes "Displayed Tab Width" is 8.  I see indentation of some code is
> broken in JSynthLib.  This default value may be one of the reason.

I don't think the setting of the "Displayed tab width" to 8 is a good =
solution
for that, as it just moves the layout issues to another place.

Best in my opinion would be the using of blank spaces for tabs
reset "Displayed Tab Width" to 4.

Regardless of that: The "Displayed tab width" problem should
be included in the "Programming.html" documentation.

Regards,
Joachim Backhaus




> -----Urspr=FCngliche Nachricht-----
> Von: Joachim Backhaus=20
> Gesendet: Dienstag, 1. Februar 2005 10:21
> An: jsy...@li...
> Betreff: Eclipse: Indentation / Displayed tab width
>=20
>=20
> Hi,
>=20
> I've started a discussion with Hiroo concerning
> the indentation and displayed tab width.
>=20
> In the "programming.html" the following can
> be found:
>=20
> "1. Preferences->Java->Editor->Displayed tab width must be=20
> set to '8'. "
>=20
> This seems to be the issue.
>=20
> Can anybody explain why the "Displayed tab width" should be set to 8
> when indentation level should be set to 4?
>=20
> I know that in most tools you can set up both differently but
> it usually doesn't make sense.
>=20
> In this case it leads to the point that there are two kinds=20
> of JSynthLib
> developers:
> A The one who use Eclipse with the setting above
> B Those who use an editor with the recommended indentation &=20
> displayed tab width
> setting of 4 (may also be Eclipse in my case).
>=20
> Of course, the code of Group B developers looks strange for Group A
> and vice versa.
>=20
> And this leads to the discussion we have now here.
>=20
> Regards,
> Joachim Backhaus
>=20

[Jsynthlib-devel] Eclipse: Indentation / Displayed tab width

[Joachim B. <jba...@pi...>]

Hi,

I've started a discussion with Hiroo concerning
the indentation and displayed tab width.

In the "programming.html" the following can
be found:

"1. Preferences->Java->Editor->Displayed tab width must be set to '8'. "

This seems to be the issue.

Can anybody explain why the "Displayed tab width" should be set to 8
when indentation level should be set to 4?

I know that in most tools you can set up both differently but
it usually doesn't make sense.

In this case it leads to the point that there are two kinds of JSynthLib
developers:
A The one who use Eclipse with the setting above
B Those who use an editor with the recommended indentation & displayed =
tab width
setting of 4 (may also be Eclipse in my case).

Of course, the code of Group B developers looks strange for Group A
and vice versa.

And this leads to the discussion we have now here.

Regards,
Joachim Backhaus

Re: AW: [Jsynthlib-devel] New Widget: MultiWidget

[Thorsten K. <Tho...@gm...>]

Hi,

Joachim Backhaus wrote:
> Is there an example in CVS? Does a synthdriver already use this?

the synth drivers of MIDIbox SID and MIDIbox FM are using 
models and senders for bit/bitfield operations

Best Regards, Thorsten.

AW: [Jsynthlib-devel] New synthdrivers: RolandJV80 & WaldorfMW2

[Joachim B. <jba...@pi...>]

Hi,

I accidently checked in the "Makefile" I changed by hand.

Don't know if it works as I don't use it.

Regards,
Joachim Backhaus

> -----Urspr=FCngliche Nachricht-----
> Von: jsy...@li...
> [mailto:jsy...@li...]Im Auftrag von
> Joachim Backhaus
> Gesendet: Montag, 31. Januar 2005 08:46
> An: JSynthLib Development
> Betreff: [Jsynthlib-devel] New synthdrivers: RolandJV80 & WaldorfMW2
>=20
>=20
> Hello,
>=20
> recently I added two new synthdrivers to CVS:
>=20
> One for the Roland JV80/880 (thanks to Sander Brandenburg)
> and one for the Waldorf Microwave 2/XT/XTK.
>=20
> Can someone please update the makefile and check it into CVS?
>=20
> Best regards,
> Joachim Backhaus
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive=20
> Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
> https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
>=20

AW: [Jsynthlib-devel] New Widget: MultiWidget

[Joachim B. <jba...@pi...>]

Hi,

> ParamModel can handle this kind of case. You need two ParamModels, one
> for check box and one for combobox.  One SysexSender can be shared by
> both check box and combobox.

Is there an example in CVS? Does a synthdriver already use this?

But I think the MultiWidget is more easy to handle than different
ParamModels (when I add some Javadoc comments that describe it
better).
And the developer of the synthdriver can use the already exisiting
ParamModel or the one he already needed to create.

Best regards,
Joachim Backhaus

> -----Urspr=FCngliche Nachricht-----
> Von: jsy...@li...
> [mailto:jsy...@li...]Im=20
> Auftrag von Hiroo
> Hayashi
> Gesendet: Dienstag, 1. Februar 2005 06:54
> An: JSynthLib Development
> Betreff: Re: [Jsynthlib-devel] New Widget: MultiWidget
>=20
>=20
> Joachim> They can be retrieved by using bitmasks.
> Joachim>=20
> Joachim> In this case I needed a check box for the switch and=20
> a combo box
> Joachim> for the "Arpeggiator resolution" but one SysexSender=20
> and one ParamModel
> Joachim> for both because they are saved in one byte and single bits
> Joachim> can't be handled.
>=20
> ParamModel can handle this kind of case. You need two ParamModels, one
> for check box and one for combobox.  One SysexSender can be shared by
> both check box and combobox.
>=20
> I took a look your code, but I could not understand well.
>=20
> Correct me, if I'm missing something.
> --=20
> Hiroo Hayashi
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive=20
> Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
> https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
>=20

AW: [Jsynthlib-devel] MidiScan fix

[Joachim B. <jba...@pi...>]

Hi,

I added a comment in MidiScan.java.

> And please fix indentation, too.

For the indentation issue: Please use a different editor
than Eclipse to compare (use an indenation level of 4 for a tab):
My changed code looks good in: Eclipse 3.0.1, UltraEdit, PSPad, SciTE

Best regards,
Joachim Backhaus

> -----Urspr=FCngliche Nachricht-----
> Von: jsy...@li...
> [mailto:jsy...@li...]Im=20
> Auftrag von Hiroo
> Hayashi
> Gesendet: Dienstag, 1. Februar 2005 06:19
> An: JSynthLib Development
> Betreff: [Jsynthlib-devel] MidiScan fix
>=20
>=20
> Hi, Joachim,
>=20
> I took a look at your recent check-in into core.  Thank you very much.
>=20
> jbackhaus> Modified Files:
> jbackhaus> 	MidiScan.java=20
> jbackhaus> Log Message:
> jbackhaus> Waldorf Microwave 2 is now recognized per Auto-Scan.
> jbackhaus>=20
> jbackhaus> It uses a different inquiry response header:
> jbackhaus> F7 7E 06 02 channel
> jbackhaus> instead of
> jbackhaus> F7 7E channel 06 02
>=20
> This means "Waldorf Microwave 2" does not obey the MIDI=20
> specification.=20
> It's not bad idea to put a hack for a particular synth. (Let's support
> more synths!)  But please put a comment.  Otherwise nobody cannot
> understand this code.
>=20
> And please fix indentation, too.
>=20
> Thank you, again.
> --=20
> Hiroo Hayashi
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive=20
> Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
> https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
>=20

Re: [Jsynthlib-devel] New Widget: MultiWidget

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

Joachim> They can be retrieved by using bitmasks.
Joachim> 
Joachim> In this case I needed a check box for the switch and a combo box
Joachim> for the "Arpeggiator resolution" but one SysexSender and one ParamModel
Joachim> for both because they are saved in one byte and single bits
Joachim> can't be handled.

ParamModel can handle this kind of case. You need two ParamModels, one
for check box and one for combobox.  One SysexSender can be shared by
both check box and combobox.

I took a look your code, but I could not understand well.

Correct me, if I'm missing something.
-- 
Hiroo Hayashi

[Jsynthlib-devel] MidiScan fix

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

Hi, Joachim,

I took a look at your recent check-in into core.  Thank you very much.

jbackhaus> Modified Files:
jbackhaus> 	MidiScan.java 
jbackhaus> Log Message:
jbackhaus> Waldorf Microwave 2 is now recognized per Auto-Scan.
jbackhaus> 
jbackhaus> It uses a different inquiry response header:
jbackhaus> F7 7E 06 02 channel
jbackhaus> instead of
jbackhaus> F7 7E channel 06 02

This means "Waldorf Microwave 2" does not obey the MIDI specification. 
It's not bad idea to put a hack for a particular synth. (Let's support
more synths!)  But please put a comment.  Otherwise nobody cannot
understand this code.

And please fix indentation, too.

Thank you, again.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] New Widget: MultiWidget

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

To All,

I'm just beginning work on drivers/editor for the
Behringer FCB1010 MIDI Foot Controller. Joachim's
message about the MultiWidget was particularly
interesting to me. This is definitely something I
could use for the FCB1010 and it's certainly a more
elegant solution that what I had in mind. Integrating
this into the Core package would certainly help me
out.

Thanks,
Jeff



--- Joachim Backhaus <jba...@pi...>
wrote:

> Hello,
> 
> I needed a widget for my QuasimidiQuasar editor
> which 
> could handle two or more parameters stored in one or
> more 
> bit of one byte.
> 
> Confused?
> 
> OK, I make an example:
> 
> There is a switch parameter "Arpeggiator on/off"
> stored in bit 0.
> The "Arpeggiator resolution" ist stored in bit 1-3
> (16 values).
> 
> They can be retrieved by using bitmasks.
> 
> In this case I needed a check box for the switch and
> a combo box
> for the "Arpeggiator resolution" but one SysexSender
> and one ParamModel
> for both because they are saved in one byte and
> single bits
> can't be handled.
> 
> In the Quasar there are three bytes that contains
> two or more
> parameters, so I needed two write three new Widgets.
> 
> That's why I decided to finally write this new
> widget because
> I need it also for other editors I'll write.
> 
> The new widget which I called "MultiWidget" can
> handle these
> cases and uses check boxes for switches and combo
> boxes
> for other parameters.
> 
> You can look at it in the CVS:
>
http://cvs.sourceforge.net/viewcvs.py/jsynthlib/JSynthLib/synthdrivers/QuasimidiQuasar/MultiWidget.java
> 
> To simplify the use of the widget I separated the
> parameters
> for each widget in there in a separate class:
>
http://cvs.sourceforge.net/viewcvs.py/jsynthlib/JSynthLib/synthdrivers/QuasimidiQuasar/MultiWidgetParams.java
> 
> Maybe it would be useful to integrate this into the
> core classes?
> 
> What do you think?
> 
> Best regards,
> Joachim Backhaus

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

[Jsynthlib-devel] Re: AlesisDM5 Inquiry ID

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

Joachim,

I committed the fix for the INQUIRY_ID for the Alesis
DM5.

Thanks for pointing this out.

Jeff


--- Joachim Backhaus <jba...@us...>
wrote:

> Message body follows:
> 
> Hello,
> 
> the Inquiry ID in the Constants.java of the
> AlesisDM5 driver
> was wrong:
> 
> static final String INQUIRY_ID = "F07E..0601F7";
> 
> It should be somewhat with "...0602..." as it is the
> response not the request.
> 
> I set it temporarily to null.
> 
> Can you fix that and set the correct ID?
> 
> Best regards,
> Joachim Backhaus



		
__________________________________ 
Do you Yahoo!? 
The all-new My Yahoo! - Get yours free! 
http://my.yahoo.com 
 

[Jsynthlib-devel] New synthdrivers: RolandJV80 & WaldorfMW2

[Joachim B. <jba...@pi...>]

Hello,

recently I added two new synthdrivers to CVS:

One for the Roland JV80/880 (thanks to Sander Brandenburg)
and one for the Waldorf Microwave 2/XT/XTK.

Can someone please update the makefile and check it into CVS?

Best regards,
Joachim Backhaus

[Jsynthlib-devel] New Widget: MultiWidget

[Joachim B. <jba...@pi...>]

Hello,

I needed a widget for my QuasimidiQuasar editor which=20
could handle two or more parameters stored in one or more=20
bit of one byte.

Confused?

OK, I make an example:

There is a switch parameter "Arpeggiator on/off" stored in bit 0.
The "Arpeggiator resolution" ist stored in bit 1-3 (16 values).

They can be retrieved by using bitmasks.

In this case I needed a check box for the switch and a combo box
for the "Arpeggiator resolution" but one SysexSender and one ParamModel
for both because they are saved in one byte and single bits
can't be handled.

In the Quasar there are three bytes that contains two or more
parameters, so I needed two write three new Widgets.

That's why I decided to finally write this new widget because
I need it also for other editors I'll write.

The new widget which I called "MultiWidget" can handle these
cases and uses check boxes for switches and combo boxes
for other parameters.

You can look at it in the CVS:
http://cvs.sourceforge.net/viewcvs.py/jsynthlib/JSynthLib/synthdrivers/Qu=
asimidiQuasar/MultiWidget.java

To simplify the use of the widget I separated the parameters
for each widget in there in a separate class:
http://cvs.sourceforge.net/viewcvs.py/jsynthlib/JSynthLib/synthdrivers/Qu=
asimidiQuasar/MultiWidgetParams.java

Maybe it would be useful to integrate this into the core classes?

What do you think?

Best regards,
Joachim Backhaus

[Jsynthlib-devel] JSLFrame and JSLDeskTop update phase II.

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

I checked in some changes for JSLFrame and JSLDesktop.  If you see any
problem, let me know.

1. JSLDesktop is not non-static.
  This makes JSLDesktop more compatible with standard Swing components.
  JSLDesktop.xxx were replaced by PatchEdit.getDesktop().xxx.

2. deprecate JSLFrame.show() method.
  show() is deprecated by J2SE 1.5.  setVisible() is preferred.
  Get rid of show() in patch editors.  Action.editPatch do the job.

3. move the code for setDefaultLocation() from JSLFrame to JSLDesktop.
  This reduced some methods for communication between the two classes.

And I also disable 'Save' and 'Save As...' menu entries when bank editor
is disabled to fix the bug reported last week.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] ConcurrentModificationException when bringing up editor window

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

Jeff,

I might introduce this bug in the last change of JSLJFrame.  I think I
already fixed last week this on my local directory.  I'm checking other
changes works.  Give me some more time.

Jeff> Has anyone else noticed this? I'm seeing
Jeff> ConcurrentModificationException sometimes when I try
Jeff> to bring up a patch editor window. Any ideas?
Jeff> 
Jeff> Here's a stack trace:
Jeff> apple.awt.EventQueueExceptionHandler Caught Throwable
Jeff> : java.util.ConcurrentModificationException
Jeff> java.util.ConcurrentModificationException
Jeff> 	at java.util.AbstractList$Itr.checkForComodification(AbstractList.java:448)
Jeff> 	at java.util.AbstractList$Itr.next(AbstractList.java:419)
Jeff> 	at core.JSLFrame$JSLJFrame.windowActivated(JSLFrame.java:378)

-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] NullPointerException on YAMAHA FS1R Performance Editor

[denis q. <dqu...@fr...>]

hi

> YAMAHA FS1R Performance Editor causes NullPointerException on both the
> latest release and the current CVS version.
> 
> How to reproduce the bug;
> 1. "create new patch" -> select "Yamaha FS1R Performance"
> 2. "edit" the patch

hmmm, a rapid response : the FS1R editor don't support to edit patch out of a 
bank. This is because a performance has to take single voices from the current 
bank 128 voices (or from ROM voices).

Anyway I investigate...

-- 
Denis

[Jsynthlib-devel] NullPointerException on YAMAHA FS1R Performance Editor

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

Hi Denis,

YAMAHA FS1R Performance Editor causes NullPointerException on both the
latest release and the current CVS version.

How to reproduce the bug;
1. "create new patch" -> select "Yamaha FS1R Performance"
2. "edit" the patch

(herbie) hiroo[518] java -jar JSynthLib-0.19-pre1.jar 2
DBG>Something selected "JavaSound"
DBG>Initializing driver:JavaSound
DBG>open inport: Yamaha PCI External MIDI, port: 8
DBG>open outport: Yamaha PCI FM Synthesizer, port: 10
DBG>"JSynthLib Tool Bar" activated.
=2E..
DBG>"JSynthLib Tool Bar" deactivated.
DBG>"aaa.patchlib" activated.
DBG>"aaa.patchlib" selected
DBG>open inport: MIDI Yoke NT:  1, port: 0
DBG>"aaa.patchlib" deactivated.
DBG>"Yamaha FS1R Performance Editor" activated.
java.lang.NullPointerException
        at synthdrivers.YamahaFS1R.YamahaFS1RPerformanceEditor$1.JSLFrameAc=
tivated(YamahaFS1RPerformanceEditor.java:115)
        at core.JSLFrame$JSLJFrame.windowActivated(JSLFrame.java:287)
        at java.awt.AWTEventMulticaster.windowActivated(AWTEventMulticaster.=
java:331)
=2E...

Here is the method which throws the exception.

	protected void frameOpened()
	{
		super.frameOpened();
		for (int oPart =3D 0; oPart < 4; oPart++)=20
		{
			// send part voice if bank is int
			if (mBankSelector[oPart].getValue() =3D=3D 1)// && mPartChannel[oPart].g=
etValue() !=3D 17)
			{
				Patch oPatch =3D null;
				if (mVoicesInEdit[oPart] !=3D null) {
					// voice currently in editing
					oPatch =3D (Patch)mVoicesInEdit[oPart].getPatch();
				}
				else=20
				{
					oPatch =3D (Patch)YamahaFS1RBankDriver.getInstance().getPatch(((Yamaha=
FS1RBankEditor)bankFrame).getBankPatch(), 128+mVoiceSelector[oPart].getValu=
e());
				}
				//System.out.println("SEND VOICE "+oPart+" "+mVoiceSelector[oPart].getV=
alue());
				YamahaFS1RVoiceDriver.getInstance().sendPatch (oPatch, oPart+1);
			}
		}
	}

Both "mBankSelector[oPart].getValue() =3D=3D 1" and "mVoicesInEdit[oPart] !=
=3D null"
are true, but "bankFrame" is null.  I'm guessing the below may be a fix,
but I don't understand this code well.

	if (mVoicesInEdit[oPart] !=3D null) {
		// voice currently in editing
		oPatch =3D (Patch)mVoicesInEdit[oPart].getPatch();
                YamahaFS1RVoiceDriver.getInstance().sendPatch (oPatch, oPar=
t+1);
	} else if (bankFrame !=3D null) {
		oPatch =3D (Patch)YamahaFS1RBankDriver.get Instance().getPatch(((YamahaFS=
1RBankEditor)bankFrame).getBankPatch(),128+mVoiceSelector[oPart].getValue()=
);
                YamahaFS1RVoiceDriver.getInstance().sendPatch (oPatch, oPar=
t+1);
	}

Is this correct?
--=20
Hiroo Hayashi

[Jsynthlib-devel] ConcurrentModificationException when bringing up editor window

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

Has anyone else noticed this? I'm seeing
ConcurrentModificationException sometimes when I try
to bring up a patch editor window. Any ideas?

Here's a stack trace:
apple.awt.EventQueueExceptionHandler Caught Throwable
: java.util.ConcurrentModificationException
java.util.ConcurrentModificationException
	at
java.util.AbstractList$Itr.checkForComodification(AbstractList.java:448)
	at
java.util.AbstractList$Itr.next(AbstractList.java:419)
	at
core.JSLFrame$JSLJFrame.windowActivated(JSLFrame.java:378)
	at
java.awt.AWTEventMulticaster.windowActivated(AWTEventMulticaster.java:331)
	at
java.awt.Window.processWindowEvent(Window.java:1133)
	at
javax.swing.JFrame.processWindowEvent(JFrame.java:266)
	at java.awt.Window.processEvent(Window.java:1079)
	at
java.awt.Component.dispatchEventImpl(Component.java:3681)
	at
java.awt.Container.dispatchEventImpl(Container.java:1671)
	at
java.awt.Window.dispatchEventImpl(Window.java:1606)
	at
java.awt.Component.dispatchEvent(Component.java:3543)
	at
java.awt.KeyboardFocusManager.redispatchEvent(KeyboardFocusManager.java:1713)
	at
java.awt.DefaultKeyboardFocusManager.typeAheadAssertions(DefaultKeyboardFocusManager.java:777)
	at
java.awt.DefaultKeyboardFocusManager.dispatchEvent(DefaultKeyboardFocusManager.java:365)
	at
java.awt.Component.dispatchEventImpl(Component.java:3572)
	at
java.awt.Container.dispatchEventImpl(Container.java:1671)
	at
java.awt.Window.dispatchEventImpl(Window.java:1606)
	at
java.awt.Component.dispatchEvent(Component.java:3543)
	at
java.awt.EventQueue.dispatchEvent(EventQueue.java:456)
	at java.awt.SentEvent.dispatch(SentEvent.java:50)
	at
java.awt.DefaultKeyboardFocusManager$DefaultKeyboardFocusManagerSentEvent.dispatch(DefaultKeyboardFocusManager.java:145)
	at
java.awt.DefaultKeyboardFocusManager.sendMessage(DefaultKeyboardFocusManager.java:171)
	at
java.awt.DefaultKeyboardFocusManager.dispatchEvent(DefaultKeyboardFocusManager.java:247)
	at
java.awt.Component.dispatchEventImpl(Component.java:3572)
	at
java.awt.Container.dispatchEventImpl(Container.java:1671)
	at
java.awt.Window.dispatchEventImpl(Window.java:1606)
	at
java.awt.Component.dispatchEvent(Component.java:3543)
	at
java.awt.EventQueue.dispatchEvent(EventQueue.java:456)
	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)



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250

[Jsynthlib-devel] SourceForge updates

[Joachim B. <jba...@pi...>]

Hello,

there will come some changes on the SourceForge site:
https://sourceforge.net/tracker/index.php?func=3Ddetail&aid=3D1109940&gro=
up_id=3D1&atid=3D350001

I don't know if any of this changes affects the JSynthLib
project.

Best regards,
Joachim Backhaus

Re: AW: [Jsynthlib-devel] Once again: Package structure

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

Hi,

I could not feel that you were offensive.  Yes, it might because my
English was not good, either:-).

I know nobody likes jsynthlib.jsynthlib.  Issue is that nobody can
propose a better application name than JSynthLib (that is a good name). 
This is an international project.  Is there a good name in your language?

Joachim> thanks Rib and Hiroo,
Joachim> now I understand it. :)
Joachim> 
Joachim> I hope I was not too offensive.
Joachim> Work was very stressful yesterday
Joachim> and my english is still not that good
Joachim> (Yes, I have two excuses ;) ).
-- 
Hiroo Hayashi

AW: [Jsynthlib-devel] Once again: Package structure

[Joachim B. <jba...@pi...>]

Hi,

thanks Rib and Hiroo,
now I understand it. :)

I hope I was not too offensive.
Work was very stressful yesterday
and my english is still not that good
(Yes, I have two excuses ;) ).

Best regards,
Joachim Backhaus

> -----Urspr=FCngliche Nachricht-----
> Von: jsy...@li...
> [mailto:jsy...@li...]Im=20
> Auftrag von Hiroo
> Hayashi
> Gesendet: Donnerstag, 27. Januar 2005 04:57
> An: jsy...@li...
> Betreff: Re: [Jsynthlib-devel] Once again: Package structure
>=20
>=20
> Hi,
>=20
> Joachim> I think it's hard to keep track of the suggested
> Joachim> changes in a mailing list.
> I agree with you.  Does anyone have a good idea?
>=20
> Joachim> So nobody read Dennis's mail or what?
> Joachim>=20
> Joachim> I agree with Dennis that an org.jsynthlib.jsynthlib package
> Joachim> is superfluous and I've never seen such a doubled structure
> Joachim> somewhere else.
>=20
> Rib answered to the mail.  (Both the original mail and the reply were
> sent to old yahoo mailing list.)
>=20
> --------------------------------------------------------------------
> The reasoning was to have the package name be tld.domain.application.=20
> The OS X Preferences class expects this, which is why there's
> currently the org.jsynthlib.jsynthlib.dummy class.
>=20
> Packages under org.jsynthlib are meant to be standalone. For example,
> EditorBuilder uses the widgets without running JSynthLib, and the
> desktop classes could be used in any java project. Since core,
> synthdrivers, and the xml driver don't really make sense outside of
> the Editor/Librarian, they all go into an application package.=20
> Unfortunately org.jsynthlib.jsynthlib is confusing, but I couldn't
> think of a better name.
> --------------------------------------------------------------------
>=20
> I agreed with Rib silently.
>=20
> org.jsynthlib is for JSynthLib project which may release=20
> other programs.
> org.jsynthlib.jsynthlib is for an application called 'JSynthLib'
> released by JSynthLib project.
> --=20
> Hiroo Hayashi
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive=20
> Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Jsynthlib-devel mailing list
> Jsy...@li...
> https://lists.sourceforge.net/lists/listinfo/jsynthlib-devel
>=20

Re: [Jsynthlib-devel] Once again: Package structure

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

Hi,

Joachim> I think it's hard to keep track of the suggested
Joachim> changes in a mailing list.
I agree with you.  Does anyone have a good idea?

Joachim> So nobody read Dennis's mail or what?
Joachim> 
Joachim> I agree with Dennis that an org.jsynthlib.jsynthlib package
Joachim> is superfluous and I've never seen such a doubled structure
Joachim> somewhere else.

Rib answered to the mail.  (Both the original mail and the reply were
sent to old yahoo mailing list.)

--------------------------------------------------------------------
The reasoning was to have the package name be tld.domain.application. 
The OS X Preferences class expects this, which is why there's
currently the org.jsynthlib.jsynthlib.dummy class.

Packages under org.jsynthlib are meant to be standalone. For example,
EditorBuilder uses the widgets without running JSynthLib, and the
desktop classes could be used in any java project. Since core,
synthdrivers, and the xml driver don't really make sense outside of
the Editor/Librarian, they all go into an application package. 
Unfortunately org.jsynthlib.jsynthlib is confusing, but I couldn't
think of a better name.
--------------------------------------------------------------------

I agreed with Rib silently.

org.jsynthlib is for JSynthLib project which may release other programs.
org.jsynthlib.jsynthlib is for an application called 'JSynthLib'
released by JSynthLib project.
-- 
Hiroo Hayashi

Re: [Jsynthlib-devel] ComboBoxWidget: Suggestion for a change of the setValue method

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

Hi,

Joachim> I have the following suggestion for a change
Joachim> of the ComboBoxWidget.setValue behavior to avoid errors:
Joachim> 
Joachim> - The value has to be 0 or more

No.  The value can be minus value.

Joachim> - The index of the corresponding JComboBox is set to 0
Joachim>   if the value - the minimum value is less than 0

The current code;
    public void setValue(int v) {
	super.setValue(v);
	cb.setSelectedIndex(v - getValueMin());
    }

can handle such case.  It's programmers responsibility to set a value
between min value and max value which he specified on the constructor. 
All sysex widgets based on this rule.

Well,
    public void setValue(int v) {
	super.setValue(v);
	cb.setSelectedIndex(getValue() - getValueMin());
    }

this may be better.  efficiency vs. fail proof.  I'm not sure which is
better.

BTW personally I prefer letting SysexWidget.setValue throw
IllegalArgumentException than hiding a bug as it does now.  But it may
break many drivers.

I think it is very good idea to fix an error caused by a user, but it is
bad idea to fix (hide) an error caused by a programmer.  It makes more
difficult for the programmer to debug his code.
-- 
Hiroo Hayashi

[Jsynthlib-devel] Once again: Package structure

[Joachim B. <jba...@pi...>]

Hello,

some time ago in august we discussed about a new
package structure for JSynthLib.

I think it's hard to keep track of the suggested
changes in a mailing list.

The last message on this I think is from 26. August 2004 by denis =
queffeulou:

"Rib Rdb wrote:

> 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 this thread, here's what it looks like we've suggested so
> far (all under org.jsynthlib):
>=20
> .midi - MIDI support classes
> .widgets - Sysex widgets
> .desktop - JSL* classes
> .editorbuilder - GUI editorbuilder
> .utils - MacUtils, ErrorMsg
> .jsynthlib - current core

why not .core ?

> .jsynthlib.synthdrivers - all (xml and java) drivers

why not .synthdrivers ?

> .jsynthlib.xml - XML driver and support files

why not .xml ?

Just to say I don't understand the meaning of a jsynthlib.jsynthlib =
package (I=20
would expect to understand just by reading the package name)."

Actually we have (all under org.jsynthlib):

.jsynthlib.synthdrivers
.jsynthlib.xml
.utils
.plugins

So nobody read Dennis's mail or what?

I agree with Dennis that an org.jsynthlib.jsynthlib package
is superfluous and I've never seen such a doubled structure
somewhere else.

Best regards,
Joachim Backhaus

[Jsynthlib-devel] ComboBoxWidget: Suggestion for a change of the setValue method

[Joachim B. <jba...@pi...>]

Hello,

I have the following suggestion for a change
of the ComboBoxWidget.setValue behavior to avoid errors:

- The value has to be 0 or more
- The index of the corresponding JComboBox is set to 0
  if the value - the minimum value is less than 0

Here is the code:

/**
* Set the value for the ComboBoxWidget.
* The index for the corresponding JComboBox is
* the value subtracted by the minimum value.
*
* @param v  The value, has to be 0 or more
*/
public void setValue(int v) {
    // Only set value if it is 0 or more
    if (v >=3D 0) {
        super.setValue(v);

        if ( (v - getValueMin() ) >=3D 0 ) {
            cb.setSelectedIndex(v - getValueMin());
        }
        else {
            cb.setSelectedIndex(0);
        }
    }
}

Actual code:

public void setValue(int v) {
    super.setValue(v);
    cb.setSelectedIndex(v - getValueMin());
}

What do do you think of it?

Best regards,
Joachim Backhaus

Save command bug with bank editor (Re: [Jsynthlib-devel] Bug report: Actions.saveFrame())

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

Hi,

Well, I've checked the code and found an interesting fact.  I made this
bug which caused the class cast exception soon after we released 0.19.=20
But before that 'Save' and "Save As..." command did nothing for Bank
Editor.

By saving library window, we can save a bank even now.  But the menu
'Save' and "Save As..." must be disabled when Bank Editor is selected.
I'll make a fix.

Thanks.

Thorsten> I'm not sure if this is related to one of the last changes in sav=
eFrame()
Thorsten> or not --- it seems that it isn't possible to save a bank anymore
Thorsten>=20
Thorsten> The problem can be reproduced in the following way:
Thorsten>    o ensure that the Kawai K4 driver is available
Thorsten>    o create a new library
Thorsten>    o create a new patch, select Kawai K4/K4R Bank
Thorsten>    o edit the new bank
Thorsten>    o click on File->Save
Thorsten>=20
Thorsten> Following error messages will appear:
Thorsten> -----------------------------------------------------------------=
-
Thorsten> Exception in thread "AWT-EventQueue-0" java.lang.ClassCastExcepti=
on: core.BankEditorFrame
Thorsten>         at core.Actions.saveFrame(Actions.java:544)
=2E..
--=20
Hiroo Hayashi