Sidebearings bug?

Hello, I already ran into this issue last week and again today, with no clue whatsoever what it might be caused by. In my glyph I, the sidebearings are 73 (yes, metrics key or manual makes no difference). The right side of my I is completely straight (as is shown on the screenshot). Still, Glyphs decides to show the sidebearing at the measurement line height to be 72.5. What causes this? I wouldn’t usually mind, but it messes with Kern On, so it results in different kerning for I and H. Any pointers appreciated, thanks a lot!

Can you send me that file?

I’m afraid I can’t, but Rainer has access to it :wink: I “fixed” this issue by deleting the paths and redrawing them. I’ll send you a file next time I experience the issue.

My first guess was that the vertical line was not completely vertical, but there was a difference of 1u between the top and the bottom node.

What do you mean? The top and bottom node were a unit apart? I don’t understand, why was there 0 difference between the x coordinates of the two nodes in the edit view? Even after I moved them back and forth, the problem still persisted.

I do not know, it was a guess. Happens all the time with grid rounding. Can you recreate the issue?

I’ll let you know which file it is on the repo.

OK, some widths had retained the extrapolated value. And since the difference was less than 0.5u, the deviation was tolerated. Paste this in Macro window and press Run:

for g in Font.glyphs:
	print(g.name)
	for l in g.layers:
		l.width=round(l.width)
		l.updateMetrics()