When starting with Glyphs.app about two years ago, I made the painful discovery that you can have 1 unicode per character only. But at least it was possible to assign the same Unicode to more than 1 character.
Now I actually have a use case for the latter, because I am working on a font that has instances for a document and a UI version. There are 4 encoded characters that have 2 versions. With the "Keep Glyphs" parameter one version for the document instances, the other for UI instances. But in the mean time the GlyphsApp UI bars me from entering the same Unicode twice.
So I tried to use the "Reencode Glyphs" parameter in the instances for the UI fonts
"Reencode Glyphs" = (
This tries to re-encode llvocalicbeng with U+09E1 and so on. It has no effect, these 4 unicodes are missing from the fonts.
In the mean time I discovered that I can bypass the UI and assign the unicodes twice with Python and the generated UI fonts have the desired cmap. This works for my current version 2.4 939, but I suppose there is no guarantee that it continues. Not to mention there is a side-effect in that when you have 2 glyphs with the same Unicode, it seems impossible to edit the first one. Whatever I try to view llvocalicbeng in the edit tab, I end up with llvocalicUIbeng ;-(.
So I guess the "Reencode Glyphs" is the best option moving forward. But I dearly want to know if works at all, or whether there are hidden assumptions that are required to be met, before this does something ??