Build simple glyphs (static glyf, 2 of 3) #131
Conversation
rsheeter
force-pushed
the
simple_glyphs
branch
3 times, most recently
from
4b10dcd to
b50438a
Compare
rsheeter
changed the title
cmyr
left a comment
cmyr
left a comment
There was a problem hiding this comment.
I feel my brain has largely departed me for the day, but nothing here jumps out as egregious 👍
|
@anthrotype although now merged I would still love your feedback on this |
| .collect(); | ||
| new_glyphs.sort(); | ||
| trace!( | ||
| "Added {} additional glyphs: {new_glyphs:?}", |
There was a problem hiding this comment.
why just "added {} additional glyphs"? It is not conceivable that glyphs are also removed when current != new glyph_order?
| fn add_glyph_be_job(workload: &mut Workload, fe_root: &FeContext, glyph_name: GlyphName) { | ||
| let glyph_ir = fe_root.get_glyph_ir(&glyph_name); | ||
|
|
||
| // To build a glyph we need it's components, plus static metadata |
There was a problem hiding this comment.
typo
+ we need it's components
- we need its components| let successes = self.read_completions(&recv, RecvType::Blocking)?; | ||
| successes | ||
| .into_iter() | ||
| .for_each(|s| self.handle_success(&fe_root, s)); |
There was a problem hiding this comment.
noob question: what's the advantage of using iterator+for_each closure over an explicit for loop here?
Is it just taste? I should probably get used to it ..
There was a problem hiding this comment.
I actually agree with you, I think this is confusing, and maybe doing too much.
| .collect(); | ||
|
|
||
| // At long last, actually convert something to quadratic | ||
| // TODO what should we pass for accuracy |
There was a problem hiding this comment.
python cu2qu defaults to a relative max_error of 1/1000th of the font's UPEM; e.g. for a usual font with head.unitsPerEm=1000, the absolute max_error is 1 font unit.
rsheeter
added this to the
Build a simple, static, font that renders in a browser milestone
3 participants
Split out from #120 to try to make review more tractable.
Compile simple static glyphs. cu2qu in a manner that should work for variations.
Blocked on googlefonts/fontations#258, but I don't expect that to appreciably change the logic so I believe this is reviewable. Edit: fixed.