> I want to use Russian KOI8-encoded Type 1 fonts with Groff. > Provided that I am using UTF-8-encoded Groff sourcres, does it mean > that I should create my own font_map file in order to specify the > translation from Groff internal characters, used in the UTF-8 mode, > into PostScript character names?
No. groff accesses Type 1 fonts always by glyph names. > More generally, does the map_file required to produce correct native > Groff font files from .afm metrics depend on whether Groff is to be > used with -KUTF-8 option or not? I think it does, because the > internal character names of the Cyrillic letters will be different. > Am I right? No. Internally, all input characters, regardless of the input encoding, are converted to entities from the Groff Glyph List (GGL; you can find the details in groff.texinfo). afmtodit does the same on the font side. Consequently, they fit within groff. > Should I use Cyrillic character names as specified here: > > > http://www.adobe.com/content/dam/Adobe/en/devnet/font/pdfs/5013.Cyrillic_Font_Spec.pdf Today you should use the Adobe Glyph List (AGL). I suspect that the values for Cyrillic glyphs are still identical to the #5013 report, though. > What does the encoding of a PostScript font mean, if the characters > can be accessed by name? It allows you to use \N'...' with a glyph index number to access glyphs without a name; this can happen if afmtodit can't automatically deduce a GGL compliant name, for example, if you have glyph variants like `foo' and `foo.small' in the same file. Werner
