Where (code) does MuseScore issue "Resets" to MIDI instruments?
Somewhere, MuseScore issues what it thinks is a "reset" to Midi-connected instruments, on VPO organs resetting all the stops. There are clearly (several) "Reset" actions in Joker Musician's MIDI Action list in his instruments.xml, but I'd like to see where in the code it is looking for this, what exactly it is looking for and where, and in what circumstances it does it. I am skilled in C++.
Thanks!
Comments
Is there a way to search the MS subset of the GitHub source repository?
In reply to Is there a way to search the… by [DELETED] 1831606
I think there you can only search for filenames
My guess would libmscore/rendermidi.cpp
OK, found the 2500-line module; all I have to do now is figure it out. It seems to know about specific nonstandard rendering engines (e.g., Aeolus)....
In reply to OK, found the 2500-line… by [DELETED] 1831606
From what I remember, Aeolus is a non-maintained experiment from MuseScore 2 and thus not supported. I believe you can safely skip all constructs relating to it when reading through the code.
In reply to From what I remember, Aeolus… by jeetee
The original author of Aeolus, Fons Adriaensen, back then violenlty disagreed it being included in MuseScore, regardless that he released it as open source (GPLv3 actually meanwhile, since June 2013). See https://de.wikipedia.org/wiki/Aeolus_(Software)
In reply to The original author of… by Jojo-Schmitz
"@Jester Musician" has "<synti>Aeolus" in his XML instrument definition, so I wonder if it's relying on something in that code; I haven't tried removing it and seeing the effect, but I suppose that's on my agenda for the day now.
In reply to "@Jester Musician" has "… by [DELETED] 1831606
I don't think that code was ever active in any released version though, just in development builds
e176e25, 6bfd9f1
In reply to I don't think that code was… by Jojo-Schmitz
OK. Which code consumes XML instrument definitions?
I have figured this out, das letzte Mysterium -- It is the "program" (i.e., which MIDI patch) ability being abused. MS apparently (reasonably) sends out the value of "program" before (understand!) and after (not so much) a "play". JM has the St Anne "general cancel" button bound not "program change 0" (easily undone -- then you have to manage clearing registrations beforehand manually yourself). But at least this is understood now).