Challenge: Generate code for Joplin - Gladiolus Rag

[rvilarl]

The good news: no crash reading the XML file.
The challenge: quite a few things to fix.

[rvilarl]

Our target :)

[jaredjj3]

I like this approach. It gives us concrete bugs to work on. The cross-staff notations are particularly interesting.

[jaredjj3]

@rvilarl, would you attach the MusicXML file somewhere in this issue please?

[rvilarl]

The score is this one https://musescore.com/user/58480/scores/1749906

[rvilarl]

@jaredjj3 the score produces this fault at https://vexml.dev/

Gladiolus_Rag__Scott_Joplin__1907.zip

17:12:14.620 (1296px) Error: [RuntimeError] BadArguments: Too few notes for beam. at new sf (https://vexml.dev/assets/index-e7Ei05G8.js:60:246688) at Wo.render (https://vexml.dev/assets/index-e7Ei05G8.js:60:1213988) at https://vexml.dev/assets/index-e7Ei05G8.js:60:1224036 at Array.map () at Jh.render (https://vexml.dev/assets/index-e7Ei05G8.js:60:1224027) at Zs.render (https://vexml.dev/assets/index-e7Ei05G8.js:61:5105) at mr.render (https://vexml.dev/assets/index-e7Ei05G8.js:63:174) at https://vexml.dev/assets/index-e7Ei05G8.js:63:7625 at Dc (https://vexml.dev/assets/index-e7Ei05G8.js:40:24283) at Ss (https://vexml.dev/assets/index-e7Ei05G8.js:40:42406)

[jaredjj3]

Good catch, I think we can fix this by adding a check here and making the BeamRendering.vexflow.beam nullable.

[rvilarl]

The cross staff notations are no longer there in measure 5

[rvilarl]

@jaredjj3 also starting on measure 10 the second staff is missing.

[jaredjj3]

There must be a bug with beaming. This will be a good score to thoroughly test the system.

[jaredjj3]

I'm going to take another stab at this before going back to lilypond specs. Here's what I plan to work on:

• Ensure that the top of a rendering is present.
• Prevent ties from rendering across systems.
• Investigate increasing StaveTie height, part distance, or both.
• Render repeats.