On Tue Mar 3, 2026 at 5:16 PM CET, onf wrote:
> That is to say, fonts providing non-standard glyphs (i.e. ones that lack
> corresponding Unicode codepoint) need to continue naming their glyphs if
> consumers are expected to use them directly (rather than as a result of
> a GSUB rule).

Alternatively, there are some schemes which use codepoints in Unicode's
Private Use Area (see e.g. the Medieval Unicode Font Initiative).

This is what STIX fonts use for their non-standard math symbols, however
AFAIK they haven't coordinated this with other math fonts nor do their
fonts give understandable names to these symbols. As a result without
implementing explicit support for their mappings[1] (e.g. using a TeX
package) or using the exact same PUA codepoints in a document's source
code, these symbols are not accessible.

[1] https://github.com/stipub/stixfonts/tree/master/docs

> I believe neatroff currently doesn't support fonts without glyph names
> at all, but I have yet to see a single font that lack them, so I could
> not test this.

"At all" is probably an overstatement; I would expect glyphs mapped to
Unicode codepoints to work at the very least.

onf

  • ... onf
    • ... G. Branden Robinson
  • ... onf
    • ... Deri via discussion of the GNU roff typesetting system and related software
      • ... onf
        • ... G. Branden Robinson
          • ... Damian McGuckin
            • ... G. Branden Robinson
        • ... Deri via discussion of the GNU roff typesetting system and related software
          • ... onf
            • ... onf
            • ... Deri via discussion of the GNU roff typesetting system and related software
    • ... Damian McGuckin
  • ... onf
  • ... onf
  • ... onf
    • ... Deri via discussion of the GNU roff typesetting system and related software
      • ... onf
        • ... Deri via discussion of the GNU roff typesetting system and related software

Reply via email to