Follow-up Comment #6, bug #44235 (group groff): Yes, exactly the same. There is currently no way to ask groff to use a real glyph from one of the fonts listed in the DESC file before using a fallback (synthetic) glyph.
The fonts listed in DESC have the "special" keyword in them but are not
treated the same as fonts introduced with .special. I think they should be
treated the same.
The glyph search order should be:-
.char
Current font
.fchar for current font
.fspecials for current font
.fschar for current font
.specials first then DESC fonts
.schar then becomes "the glyph of last resort".
No longer search any previously loaded fonts by font number, which is where
DESC fonts currently get searched.
The only change required is that on startup DESC fonts are treated as though
they were .special(ed). On encountering .special groff would build a new
specials list and add the DESC fonts to the end of the list.
The only change in behaviour would be schar fallbacks won't block searching
the DESC fonts, and the odds'n'sods roulette game groff plays by searching all
loaded special fonts in numerical order as a last resort can be dropped (since
the ordering of specials is now fully specified).
I agree the sentiment of comment #2, which can be achieved in one of two ways:
either a new .?char (I suggested .zchar) which is used after the current
numerical font search fails, or promote DESC fonts as though they were fully
paid up members of .special in which case they are no longer masked out by
.schar, the final numerical font search can be dropped.
However this will not help with your wish to "mask out" a specific font/glyph
combination the way .char does with the current font.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?44235>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
signature.asc
Description: PGP signature
