Non-SMuFL-to-SMuFL glyph mapping projects

[rpatters1]

This is a question that is not directly about MNX but impinges on the ability to export MNX files from older (pre-SMuFL) programs such as Finale.

Is anyone aware of an open source project that provides mappings from older music symbol fonts to SMuFL glyphs or glyph names? It seems like such an obvious need that I am hesitant to start such a project without doing some due diligence to see if it anything like it is already being done.

Ideally it would

• have mappings per font from the font's glyph codes to equivalent SMuFL glyph codes or glyph names. (I'm thinking semantic equivalents: "this glyph is a G clef" kind of thing.)
• be expandable at runtime by means of some commonplace format such as json.
• be C++ compatible.
• lightweight (no dependencies other than perhaps a json parser, preferably through an interface so that any json parser could be used by implementing the interface).

[samuelbradshaw]

I had a similar need a few weeks ago, and created a mapping file based on Finale and SMuFL source data. I'd be very interested in an open source project to expand the mappings and make them more useful! Here's what I have so far: https://github.com/samuelbradshaw/music-font-mapper

[rpatters1]

This is a tremendous start. (I just discovered those Finale json files this morning myself.) What would be a really helpful next step is to expand the .py script to generate per-font and/or combined C headers as well as the combined json. I can work on this if you are amenable to it.

FWIW: the SMuFL json is not that necessary as an add-on for my purposes. The codebases I am currently working on (my own MNX exporter and the MuseScore codebase) already have SMuFL under control. I imagine any other modern notation software does as well. The goldmine here is the legacy font mappings. I think it would be useful to have them separated (as I am proposing for the C headers.)

Also, it will need to have a CMake file so that it can be easily incorporated into other projects. Again, something I can work on if you are amenable.

[samuelbradshaw]

Sounds great!

[rpatters1]

I don't know if Samuel is subscribed to issues on the repo he created, but I've opened a couple of them.

[mscuthbert]

I don't know if there's a way to move a discussion, but I think this should be on the SMuFL repo -- I totally endorse hosting a set of files there in a standard format that maps SMuFL codes to non-SMuFL fonts (and vice-versa)!

[rpatters1]

Ask and it shall be done.