Closed Finii closed 2 months ago
Ah, another downside of the way the ligs are formulated here (substitude two glyphs to one glyph with increased width) is
Changing the ligature formulation would probably solve both (but I feel the pain in reformulating the ligs).
Split off the notdef glyph width into #159
Fixed in ca8015b27bdc6d8e1201b1ad722f4ae3c38970d0
Probably you want the font to be detected as monospaced.
Apart from the missing Panose values
125
the font is not really monospaced, although it could be. The width of the font is generally 1240.
The most obvious errors in being monospaced are:
.notdef
-> width 965uni2123
-> width 7440 (!!!)Apart from the wrong width (you could in principle just make the advance width standard and have a very big right side bearing), the glyph itself is questionable (see #51).
All the other non-conforming widths are from the ligatures.
Here a complete list of all glyphs with a differing width (patched the
fontforge
font generation code):The ligatures substitute multiple 1240 wide glyphs with one extrawide liga glyph. That could also be formulated differently, so that the ligature glyph itself has the standard width but a wide right side bearing, here a comparison for example with Cascadia:
They substitude
(
and*
byparenleft-asterix.liga
andLIG
, where bothparenleft-asterix.liga
andLIG
are of standard width,LIG
is just empty. You have a different approach. I'm not sure what is better; most fonts I have examined have the 'all glyphs have the same width' approach. At least with that approach it getsisFixedPitch
set.