update GlyphData from Glyphs 3.2

[schriftgestalt]

Putting that up to see what the test are doing ;).

[khaledhosny]

A few tests need to be updated and most of the regression tests failures look like improvements, except may be this one?

 ! tests/data/gf/DotGothic16/DotGothic16-Regular.ufo/lib.plist - /tmp/tmpzlyzjnmk/DotGothic16-Regular.ufo/lib.plist--- 
+++ 
@@ -11190,9 +11190,9 @@
       <key>micro.rotat</key>
       <string>uni00B5.rotat</string>
       <key>middledot-kata.half</key>
-      <string>uniFF65</string>
+      <string>middledotkata.half</string>
       <key>middledot-kata.half.rotat</key>
-      <string>uniFF65.rotat</string>
+      <string>middledotkata.half.rotat</string>
       <key>miriSquare</key>
       <string>uni3349</string>
       <key>miriSquare.vert</key>

[schriftgestalt]

There was an altName in that is not in my data.
there is one more thing that probably needs updating code in more places. The case information moved from subcategory="Upper/Lowercase" to case="upper/lower". Not sure if that is used anywhere.

[khaledhosny]

There was an altName in that is not in my data. there is one more thing that probably needs updating code in more places. The case information moved from subcategory="Upper/Lowercase" to case="upper/lower". Not sure if that is used anywhere.

I don’t think glyphsLib uses this. It only cares about GDEF, so Mark/[Nonspacing,Spacing Combining], and */Ligature:

glyphsLib/Lib/glyphsLib/builder/features.py

Lines 202 to 219 in 3554be4

	def _build_public_opentype_categories(ufo: Font) -> dict[str, str]:
	"""Returns a dictionary mapping glyph names to GDEF categories.
	Does not handle ligature carets. A GDEF table with both categories and
	ligature carets is generated by ufo2ft's GdefFeatureWriter at compile time,
	using the categories dict and glyph data.
	Determining the categories requires anchor propagation or user care to work
	as expected, as Glyphs.app also looks at anchors for classification:
	* Base: any glyph that has an attaching anchor (such as "top"; "_top" does
	not count) and is neither classified as Ligature nor Mark using the
	definitions below;
	* Ligature: if subCategory is "Ligature" and the glyph has at least one
	attaching anchor;
	* Mark: if category is "Mark" and subCategory is either "Nonspacing" or
	"Spacing Combining";
	* Compound: never assigned by Glyphs.app.

[schriftgestalt]

It does seem to look at the altName. That way it can connect the glyph name "middledot-kata.half" to the info entry "dot-kata.half" and get its production name.

It could have used the fallback to search with the unicode, but that doesn't seem to work.

[markhdavid]

Can anyone summarize where this stands, and if anything is holding this PR up?

[anthrotype]

Was this copied from the files in https://github.com/schriftgestalt/GlyphsInfo? Usually we write the commit hash in the commit message when we update these files to know where they came from. Can you do that?
Also there seem to be merge conflicts.

[markhdavid]

Was this copied from the files in https://github.com/schriftgestalt/GlyphsInfo? Usually we write the commit hash in the commit message when we update these files to know where they came from. Can you do that? Also there seem to be merge conflicts.

@schriftgestalt ping...

[schriftgestalt]

The regression tests need to be updated. There are a lot fails with the production name of the brevecomb-cy. It was brevecombcy and now is uni0306.cy. Not sure how to do that in CI.

So it needs to run tests/tools/generate_regression_test_files.py on with the latest GlyphData.

[anthrotype]

The regression tests will fix themselves automatically after you merge the PR

[markhdavid]

@schriftgestalt are you able to take any action to move this forward? E.g., along the lines of @anthrotype's comment #959 (comment) ?

[schriftgestalt]

I can’t do anything else on this. As I understand cosimos comment is that the test will work after it is merged?

[anthrotype]

As I understand cosimos comment is that the test will work after it is merged?

i think so, let's see

[anthrotype]

@schriftgestalt why this change? What does it have to do with updating the GlyphData database?

[anthrotype]

I see you commented on this already

The case information moved from subcategory="Upper/Lowercase" to case="upper/lower". Not sure if that is used anywhere.

If it were me, i'd just have left it as it was. Let's hope nothing else breaks

[schriftgestalt]

But it was failing some tests.

[anthrotype]

the regression tests passed