Redundant code?

[bobh0303]

I'm curious whether both of these are necessary:

fontbakery/Lib/fontbakery/utils.py

Lines 777 to 781 in eab6d90

	def get_mark_class_glyphnames(ttFont):
	from fontbakery.constants import GlyphClass
	class_defs = ttFont["GDEF"].table.GlyphClassDef.classDefs.items()
	return {name for (name, value) in class_defs if value == GlyphClass.MARK}

fontbakery/Lib/fontbakery/utils.py

Lines 855 to 863 in eab6d90

	def mark_glyphs(ttFont):
	marks = []
	if "GDEF" in ttFont and ttFont["GDEF"].table.GlyphClassDef:
	class_def = ttFont["GDEF"].table.GlyphClassDef.classDefs
	glyphOrder = ttFont.getGlyphOrder()
	for name in glyphOrder:
	if name in class_def and class_def[name] == 3:
	marks.append(name)
	return marks

As far as I can tell, they both return the names of all the glyphs identified in GDEF as marks, the key differences being:

• the first returns a set; the second returns a list sorted by glyph order
• the first assumes presence of the GDEF table

Am I missing some important distinction?

Interestingly, one use of mark_glyphs() that I could find immediately turns the list into a set:

fontbakery/Lib/fontbakery/checks/tabular_kerning.py

Line 169 in eab6d90

set(ttFont.getGlyphOrder()) - set(IGNORE_GLYPHS) - set(mark_glyphs(ttFont))