Add pitch range CC handling

[liushuyu]

Added one basic and widely used RPN CC handling: Pitch Sensitivity (aka Pitch Range)
CC code sequence:

101, 0  ; Indicate a RPN-MSB entry
100, 0  ; Indicate a RPN-MSB entry
; The combination above (0, 0) indicates to change Pitch Sensitivity (aka Pitch Range)
6, <value in semitones / 127>
38, 0 ; End

This will fix #1967

---

This change is 

[senseab]

I would say two flags might be better. The prevCC = 201 looks strange.

[liushuyu]

Because I need to prepare for other CC message other than (0,0)

[senseab]

I see that. 在 2017年01月04日 12:49, liushuyu 写道:

…

Because I need to prepare for other CC message other than (0,0) — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3206 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACEtHowpdNyFf--D4tH4-dRD-NSMZB5Eks5rOyTHgaJpZM4LaINQ>.

[softrabbit]

I'm not entirely sure about the code. What if there's a CC 101, X ; CC 100,0, won't that be interpreted the same for any X? I believe the order of these CCs isn't dictated by the MIDI standard.

Here's how I'd implement the RPN logic: https://github.com/softrabbit/lmms/blob/midi_import/plugins/MidiImport/MidiImport.cpp#L520. I think it's the more future-proof way, in case other RPNs are needed at some point.

[senseab]

Okay, CC101 means RPN MSB and CC100 means RPN LSB. RPN MSB seems to be always 0 or 127(Null). While RPN LSB = 0 and MSB = 0, it means Pitch Bend Sensitivity.

CC101 RPN MSB	CC100 RPN LSB	Meaning
0	0	Pitch Range
0	1	Fine Tune
0	2	Coarse Tune
127	127	Null

I was read your code. Pitch bend should be multiplied by pitchRangeModel, or you will hear incorrect tones after you import MIDI file. here is a MIDI file you can test https://cfp.vim-cn.com/cbcNg

Base64 decoder is needed. And it will be correct while you hear a octave.

For more info about MIDI controller, see http://www.indiana.edu/~emusic/cntrlnumb.html

[softrabbit]

I was read your code. Pitch bend should be multiplied by pitchRangeModel, or you will hear incorrect tones after you import MIDI file. here is a MIDI file you can test https://cfp.vim-cn.com/cbcNg

I never said it was complete 😄. Looks like I was going to post-process the pitch ranges in case they happened in a different track in the file, but that might be a pretty rare situation. Just simply multiplying while reading the track will probably work fine for something like 99% of all files.

I guess #1967 can be closed when this is merged.

[senseab]

Aha, it's done. 在 2017年01月04日 17:21, Raine M. Ekman 写道:

…

I was read your code. Pitch bend should be multiplied by pitchRangeModel, or you will hear incorrect tones after you import MIDI file. here is a MIDI file you can test https://cfp.vim-cn.com/cbcNg I never said it was complete 😄. Looks like I was going to post-process the pitch ranges in case they happened in a different track in the file, but that might be a pretty rare situation. Just simply multiplying while reading the track will probably work fine for something like 99% of all files. I guess #1967 <#1967> can be closed when this is merged. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3206 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACEtHv6FGm0NT20UuF79oKDsYZMk-ESgks5rO2SRgaJpZM4LaINQ>.

[senseab]

Now looks like my style.....
And now i reailze that, bool is bad for this

[liushuyu]

Well, some more annoying things occurred and proved this solution is not applicable. Close this out for future reference.

[softrabbit]

Well, some more annoying things occurred and proved this solution is not applicable.

Please, document them for future reference.

[liushuyu]

Okay. So write this to wiki or just leave here as comments? "Raine M. Ekman" <[email protected]> wrote: Well, some more annoying things occurred and proved this solution is not applicable. Please, document them for future reference. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#3206 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGg1IYW4sjWP_cfEOstK59yu0LJ5G-A-ks5rPWgugaJpZM4LaINQ> .

[softrabbit]

Comment here, this is the most likely place for people to look for info.

[liushuyu]

Comment here, this is the most likely place for people to look for info.

Okay, I will do