Insert glyph without unicode def

[Andreasox]

Description of the bug

I assume it is not possible to reference a glyph without a unicode name in insert_text and insert_textbox for export to PDF.

How to reproduce the bug

At least Gotham Narrow fonts have defined a separate range of digits 0-9 without unicode names.
Downloaded from https://freefontsfamily.com/gotham-font-family/
Glyphs 0183 - 018d, they are named in the font as zero.tab, one.tab, two.tab etc. (checked for Gotham Narrow Medium)

PyMuPDF version

1.23.8

Operating system

MacOS

Python version

3.9

[JorjMcKie]

I'm not sure I understand the issue.
Both methods Page.insert_text/Page.insert_textbox simply use the Unicode number of a character to lookup the desired glyph and then output the glyph's hex value.
There is no other check whatsoever.

[Andreasox]

I interpret your answer as my assumption is correct, pymupdf (Python/HTML) can only access glyphs with a valid Unicode codepoint.
To make such glyphs usable, we must update the fonts with valid Unicode codepoints and ensure that the fonts are embedded in the PDF.

[JorjMcKie]

To make such glyphs usable, we must update the fonts with valid Unicode codepoints and ensure that the fonts are embedded in the PDF.

Correct - with the additional comment that the embedding itself is happening automatically, once at least one Unicode of the font has been written.
Initially, a font is embedded in its entirety. You have the option to execute doc.subset_fonts() right before saving. This will scan the PDF for fonts that benefit from a smaller size, when unused glyphs are stripped off of it. Beneficial for large fonts (e.g. Asian scripts).