I am having problems with a number of glyphs that are flagged as having “Bad Path Orientation” by MekkaBlue’s Path Problem Finder. Problem is when I run “Correct Path direction” on each master, their interpolation compatibility is broken. Then if I restore it by means of “Shape Order”, they will be again flagged as having “Bad Path Orientation”.
Also, if I later run MekkaBlue’s Path Problem Finder on the generated fonts extra glyphs, not previously flagged in the sources are listed as having “Bad Path Orientation”.
Any idea of what might be happening?
(I’ve sent the files to Glyphs Support)
The correct path direction algorithm is not perfect. It can help in a lot cases but fails here and there.
And the “Bad Path Orientation” script uses that same method and thous leads to the wrong flagging of that glyphs.
How then can I be 100% certain there are no glyphs with incorrect path direction? Shall I use Fontlab?
Just look at them. Use the script to find potential candidates.
I’ll use your examples for my tests my implementation and maybe I can improve it.
The bad path orientation option lists the glyphs in which the built-in correction algorithm would have changed the direction. To some extent, this will always have to be a guess.
As with all the results of the Path Problem Finder, they just list glyphs and let you decide. I would expect some false positives.