[PD] nanoKontrol2 dropping MIDI messages (OSX)

[Ivica Ico Bukvic]

Cool! Thanks!

> -----Original Message-----
> From: Miller Puckette [mailto:msp at ucsd.edu]
> Sent: Sunday, June 24, 2012 1:36 PM
> To: Ivica Ico Bukvic
> Cc: chris clepper; pd-list
> Subject: Re: [PD] nanoKontrol2 dropping MIDI messages (OSX)
> 
> Here's the relevant diff - should apply OK to 0.42 I think.
> 
> cheers
> Miller
> 
> ------------------
> 
> diff --git a/src/s_midi_pm.c b/src/s_midi_pm.c index fe123ff..856f121
100644
> --- a/src/s_midi_pm.c
> +++ b/src/s_midi_pm.c
> @@ -241,44 +241,51 @@ void nd_sysex_inword(int midiindev, int status, int
> data1, int data2, int data3)
> 
>  void sys_poll_midi(void)
>  {
> -    int i, nmess;
> +    int i, nmess, throttle = 100;
>      PmEvent buffer;
>      for (i = 0; i < mac_nmidiindev; i++)
>      {
> -        int nmess = Pm_Read(mac_midiindevlist[i], &buffer, 1);
> -        if (nmess > 0)
> +        while (1)
>          {
> -            int status = Pm_MessageStatus(buffer.message);
> -            int data1  = Pm_MessageData1(buffer.message);
> -            int data2  = Pm_MessageData2(buffer.message);
> -            int data3 = ((buffer.message >> 24) & 0xFF);
> -            int msgtype = (status >> 4) - 8;
> -            switch (msgtype)
> +            if (!throttle--)
> +                goto overload;
> +            nmess = Pm_Read(mac_midiindevlist[i], &buffer, 1);
> +            if (nmess > 0)
>              {
> -            case 0:
> -            case 1:
> -            case 2:
> -            case 3:
> -            case 6:
> -                sys_midibytein(i, status);
> -                sys_midibytein(i, data1);
> -                sys_midibytein(i, data2);
> -                break;
> -            case 4:
> -            case 5:
> -                sys_midibytein(i, status);
> -                sys_midibytein(i, data1);
> -                break;
> -            case 7:
> -                nd_sysex_mode=1;
> -                nd_sysex_inword(i, status, data1, data2, data3);
> -                break;
> -            default:
> -                if (nd_sysex_mode)
> +                int status = Pm_MessageStatus(buffer.message);
> +                int data1  = Pm_MessageData1(buffer.message);
> +                int data2  = Pm_MessageData2(buffer.message);
> +                int data3 = ((buffer.message >> 24) & 0xFF);
> +                int msgtype = (status >> 4) - 8;
> +                switch (msgtype)
> +                {
> +                case 0:
> +                case 1:
> +                case 2:
> +                case 3:
> +                case 6:
> +                    sys_midibytein(i, status);
> +                    sys_midibytein(i, data1);
> +                    sys_midibytein(i, data2);
> +                    break;
> +                case 4:
> +                case 5:
> +                    sys_midibytein(i, status);
> +                    sys_midibytein(i, data1);
> +                    break;
> +                case 7:
> +                    nd_sysex_mode=1;
>                      nd_sysex_inword(i, status, data1, data2, data3);
> +                    break;
> +                default:
> +                    if (nd_sysex_mode)
> +                        nd_sysex_inword(i, status, data1, data2, data3);
> +                }
>              }
> +            else break;
>          }
>      }
> +    overload: ;
>  }
> 
>  void midi_getdevs(char *indevlist, int *nindevs,
> ------------------
> 
> On Sun, Jun 24, 2012 at 10:23:52AM -0400, Ivica Ico Bukvic wrote:
> > Does this affect the 0.42 branch as well? If so, would you mind sharing
the
> relevant patch or svn commit?
> >
> > Many thanks!
> >
> > Ivica Ico Bukvic, D.M.A
> > Composition, Music Technology
> > Director, DISIS Interactive Sound & Intermedia Studio Director, L2Ork
> > Linux Laptop Orchestra Head, ICAT IMPACT Studio Virginia Tech
> > Department of Music Blacksburg, VA 24061-0240
> > (540) 231-6139
> > (540) 231-5034 (fax)
> > disis.music.vt.edu
> > l2ork.music.vt.edu
> > ico.bukvic.net