>> Hmm. I don't think this glyph looks inconsistent...
>
> Here's what I meant. In the following image, we see two normal flats
> superimposed (in green) on top of the doubleflat glyph (in
> black). The stems are matched up to show what I'm referring to as
> being inconsistent: [...] It would seem better to me to at least
> have the the left counter look more like the right one.
Indeed, the code aims for having two almost identically sized
`holes'. As other have already remarked, we need real-world samples
to better judge this.
>> > 2. scripts.pralldown, scripts.prallup: The wiggle's thick/thin parts
>> > are opposite of the other similar glyphs, which seems odd to me.
>
> Well, whatever the reason, in fetatrills.mf, pralldown is the
> created by scaling upprall by -X, which certainly does save a bunch
> of extra code, but looks funny (and, again, inconsistent to the rest
> of the trill-like glyphs) to me. Looks like its a similar story
> with prallup being a -X of downprall. At least the code could be
> changed to be pralldown = rotate(downprall, 180deg) and prallup =
> rotate(upprall, 180deg). That would keep the thick/thin lines
> pointing in the correct direction for each. What say ye?
The same as with the double-flat: we need evidence from real-world
scores to decide this.
Do you have time to check this?
Werner
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
