Repository navigation
Bug: bottom beam height is not correctly calculated #5
Description
Consider system
M:4/4
L:1/4
V:1
V:2 clef=bass
K:C
[V:1] z1/2z1/2z1/2[EAC]1/2[EF]1/2[EF]1/2z |
[V:2] z1/2z1/2z1/2[E,A,C]1/2[G,A,]1/2[G,A,]1/2z|
Result:
If I change
ABCUnity/Runtime/Scripts/NoteCreator.cs
Line 668 in 279e5ac
to
else if (beam.stemHeight != Beam.unspecifiedStemHeight)
{
// beam contains notes at different heights, used the calculated stem height
stemHeight = Mathf.Abs(stemHeight);
}
then it changes to
Not sure if the Mathf.Abs is needed there anymore. Also, simple notes do not seem to have this problem. I tested various systems after this change and it seems to work correctly now (although bottom stems are tiny bit longer).
I am not going to create PR because I have too many modifications.
Activity
matthewcpp commentedon Dec 9, 2024
Hey, thanks for submitting a fix for this issue! I did a little digging and it looks like the height issue you mentioned can be fixed by changing the line in question to:
Note: this will require you to make Beam.beamHeight public.
All the other test renders look good. That being said, I dont think the rendering in this case is 100% correct. The middle stem height should be extended to create a linear beam. I will look into that and see if I can come up with a fix, and get this merged into main
EpsilonD3lta commentedon Dec 9, 2024
Thanks for the reply. Bit of a side question, why don't beams use LineRenderer but custom procedural mesh? Was it problematic or any other particular reason? I want to change the colors of the beams (gradient), so instead of implementing custom vertex coloring I might as well rewrite it to LineRenderer, so I am asking if you stumbled on some problems.
Edit: nevermind, I realized that LineRenderer won't work so easily, because it has slanted ends.
EpsilonD3lta commentedon Dec 13, 2024
I found out that the line should probably look like this
stemHeight = Mathf.Abs(beam.stemHeight - stemPos.y);Similarly for the chords pointing upwards. This fixes almost all cases, however some still remain.
Consider system:
Result
The middle beam is slightly bent, and the right secondary beam is completely wrong.