ColorVoices only colours noteheads

• Aug 12, 2023 - 11:16

Musescore 3.6.2

I thought colorvoices coloured stems & beams as well but I'm only getting the following:

Comments

• Aug 12, 2023 - 12:24

Look close: it does color the stems (and the flags), but indeed not the beams.
As per the code it should do beams too, need to investigate why it doesn't

Jojo-Schmitz

• Aug 12, 2023 - 12:30

Actually the code reads:

         if (element.beam) 
            // beams would need special treatment as they belong to more than
            // one chord, esp. if they belong to an even number of chords,
            // so for now leave (or make) them defaultColor
            toggleColor(element.beam, msSetScore.defaultColor)

So this is by current design, as it'd only work for beams connecting an odd number of chords. Better solution wanted.

rothers

• Aug 12, 2023 - 12:32

Thanks

yonah_ag

• Aug 12, 2023 - 15:05

The beam could take the colour of the note at the start of the beam, which is often the same as the note at the end of the beam. Where these are different the user could manually stop beaming or change the colour manually, (or it could be automatically to a dark grey, for example.)

Jojo-Schmitz

• Aug 12, 2023 - 16:06

Good idea (esp. as there are no cross-voice beams, as far as I can tell), but how to find out (in a plugin) whether a chord is starting a beam / whether a beam starts at a chord? Beam::Mode won't tell.

yonah_ag

• Aug 12, 2023 - 18:32

I'm not sure whether this helps:
https://musescore.github.io/MuseScore_PluginAPI_Docs/plugins/html/class…

For some plugin situations I have had to select a range to capture the elements that I'm interested in and build a segment position map to use in conjunction with the plugin's main processing.

Jojo-Schmitz

• Aug 12, 2023 - 18:43

No, that doesn't help.
Indeed building a map, collecting already worked on beam IDs and skipping those might be a way.
Quite some effort for a small effect

yonah_ag

• Aug 12, 2023 - 19:39

Inded — but satisfying if the have the time. 🙂

I'm not sure that you would have to detect "already worked on beams". If you can map the segment tick and voice of the beams, (in pass 1), then you should be able to check each beam-tick as you process the notes ticks and voices in pass 2.

I have taken this approach in tying articulations, like staccato, to tick position for processing a score's notes measure-by-measure as the plugin API does not expose articulations to this method of processing.

Jojo-Schmitz

• Aug 12, 2023 - 19:57

Yes, in a 2-pass approach. Building a list of beam IDs and skipping those already recorded would only need one pass

yonah_ag

• Aug 13, 2023 - 02:19

Neat: I might redo my code. Could the recorded beam IDs be deleted from the map rather than flagged as recorded or is this a waste of cpu?

Jojo-Schmitz

• Aug 13, 2023 - 08:40

Probably not even a map is needed

 prevID = -1
...
If (element.beam) {
    if (beam.element.id() != prevID) {
        prevID = beam.element.id()
        toggleColor(element.beam, colors[voice % 4])
    }
}