Subsetting Historic Ligatures

[bryankaplan]

Is it possible to specify a subset of hlig? If so, how?

[psb1558]

Are you using pyftsubset from fontTools? I ask because this is the only tool I've made much use of, and it allows fine control of your subsetting.

[bryankaplan]

pyftsubset looks quite useful, and it may well answer this question. I don't know yet.

But I may have chosen the wrong word when I said “subset”. Provided a non-subsetted junicode font, is it possible to indicate some (and only some) of the sub-features of hlig?

[psb1558]

Ah. I see. Sorry, but I don't know of any way to make hist work that way. The feature has just got to be on when you want it and off when you don't. I don't need it myself, but I can see that it could be a pain to use.

[psb1558]

I have been thinking, actually, of offering an alternative way of getting these ligatures: by placing U+200D zero width joiner between two characters. Then the three-character sequence could be resolved in liga Standard Ligatures instead of in hist. This might make it easier to get individual ligatures without having to turn a feature on and off, possibly disrupting kerning and contextual alternates.

[bryankaplan]

The ZWJ idea is smart, and I'm sure if you add it, some people will use it.

That being said, for covering an entire text it's generally easier to enable a feature. I just wish they could be enabled individually.

[psb1558]

Yeah, it would be nice. I've just reviewed the OpenType GSUB lookup types again, and I can't see a way to make it work. If anyone passing by here has an idea, I'd love to hear it.

[eclecticfluff]

If you can use LuaLaTeX for your document, selnolig readily covers this problem with the \nolig macro.

[psb1558]

I think there is nothing to do here.