How do copy metric/kerning from uppercase letters to lowercase in Fontlab?

Posted by willryan42 in

11 Aug 2012 — 4:17pm

I'm designing a display face in Fontlab, and I'm only creating uppercase letters. I planned on finishing the uppercase letters (metric and kerning included) and then just copy them to the lowercase letter spots so the same A shows up whether you type a or A.

However, it doesn't seem like metric and kerning data gets included when I copy and paste letters. I don't know much about Fontlab, but I would've thought I could somehow copy this data. Any ideas?

HVB
11 Aug 2012 — 4:44pm

I've got the same question; and to add to it, once you've done that, you still have to consider the upper/lower combinations. It would be nice if
AV could be copied to all combinations of those glyphs: AV, av, aV, and Av.

hrant
11 Aug 2012 — 5:02pm

Save the metrics out as a text file, manually* duplicate/modify the kerning part, then re-import.

* Or recruit a word processor's find/replace function.

hhp

HVB
11 Aug 2012 — 5:07pm

Too simple. Thank you :)

John Hudson
11 Aug 2012 — 5:34pm

Use the 'Paste Special' function from the Edit menu.

hrant
11 Aug 2012 — 5:39pm

I feel like an old dog who needs to learn new tricks.

hhp

David Vereschagin
11 Aug 2012 — 5:42pm

Why not create an OpenType feature to substitute the uppercase characters when their lowercase counterparts are typed?

David

Theunis de Jong
12 Aug 2012 — 4:28am

Or assign the same glyph to different Unicode values.

willryan42
12 Aug 2012 — 11:01am

@HVB: I didn't even think about aV, av, and Av... that introduces a whole new problem...

@John Hudson: Thanks, this worked! Using paste special and selecting left and right kerning option works beautifully.

@Quadrat: I've never messed with OT features, and regardless I think this font is ultimately going to be a TTF. For some reason when I generate the font it looks very wrong when I export as an OpenType font, but fine when I choose TrueType. The characters are very complex so I'm sure I could just be doing something incorrect when generating the font, but for now it looks like I'll have to stick with TTF.

@Theunis: Didn't even know you could do that! Seems a little complicated for me, but I'll look into it.

eliason
12 Aug 2012 — 11:54am

Or assign the same glyph to different Unicode values

That's what I did with the Ambicase fonts.

gargoyle
12 Aug 2012 — 12:45pm

For all-caps fonts, I find double-encoding the glyphs to be more convenient during development, even if I end up expanding them at the end. Since the "lowercase" is literally the same as the uppercase, it doesn't matter which one you edit/kern/etc., and there's no need to worry about case when navigating or typing out test strings. Then if I'm feeling especially standards-compliant, before generating I can run a macro that creates properly-encoded lowercase copies and adds them to appropriate kerning classes.

Frode Bo Helland
12 Aug 2012 — 2:33pm