Re: [Jsynthlib-devel] Converter Problems

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

Hi Hiroo,

a quick test shows that this hack doesn't work, alas.

Hiroo Hayashi wrote:
> Torsten,
> 
> Does this solve your problem?
> 
> Index: Converter.java
> ===================================================================
> RCS file: /cvsroot/jsynthlib/JSynthLib/core/Converter.java,v
> retrieving revision 1.14
> diff -u -r1.14 Converter.java
> --- Converter.java	6 Sep 2004 21:13:29 -0000	1.14
> +++ Converter.java	10 Sep 2004 04:27:14 -0000
> @@ -31,8 +31,10 @@
>              String patchString = patarray[i].getPatchHeader();
>              for (int jdrv = 0; jdrv < dev.driverCount(); jdrv++) {
>                  IPatchDriver drv = (IPatchDriver) dev.getDriver(jdrv);
> -                if (drv.supportsPatch(patchString, patarray[i].getByteArray()))
> +                if (drv.supportsPatch(patchString, patarray[i].getByteArray())) {
>                      patarray[i].setDriver(drv);
> +                    break;
> +                }
>              }
>          }
>          return patarray;
> 

Hiroo Hayashi wrote:
 > tt_ml_2_g7> But in the core.Driver.createPatch(SysexMessage[]) method I found only the lines
 > tt_ml_2_g7>          // if Conveter for the patch exist, convert the patch.
 > tt_ml_2_g7>          IPatch[] patarray = createPatch(patchSysex);
 > And Patch.valueOf() calls it.

Alas Patch.valueof() isn't called anytime.

I've made a small fix with calling
IPatch[] patarray = Patch.valueOf(patchSysex);
instead of
IPatch[] patarray = createPatch(patchSysex);

And now the converter is concerned properly.

I think that the old call missed the proper driver information to work correctly.
But Patch.valueOf() determine the right driver and call the proper createPatch() method.

If you agree, I will commit this fix.


Thanks and bye
Torsten