Looking for help with Contextual Alternates

2»

Comments

  • CalicoStonewolf
    CalicoStonewolf Posts: 21
    edited March 2022
    Thomas Phinney said:
    None of your OpenType layout processing will work across writing systems, unfortunately. Which is to say, Theta simply cannot interact, in terms of OpenType code, with IPA glyphs. (As I think you have discovered, Richard’s suggestion will not resolve the problem.)

    Your viable options include:
    - get rid of the Theta
    - encode all your characters in a private use area instead of using pre-existing codepoints

    But if I did private codepoints, then it still wouldn't work as intended, would it? Would it be viable to have a substitution of Theta as an input to point to a unmarked glyph, and have it work that way?
    Or, what about ligature substitution? Force it that in any situation where it is, depending on the backtrack or lookahead, replace it and the vowel with a predefined ligature?
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    Or, what about ligature substitution? Force it that in any situation where it is, depending on the backtrack or lookahead, replace it and the vowel with a predefined ligature?

    As others have already said, Greek and Latin glyph runs are evaluated separately, so you will never get a theta to join in a ligature with a Latin glyph. They never get to «see» each other.

  • CalicoStonewolf
    CalicoStonewolf Posts: 21
    Damn. That's just extremely disappointing at this point. Feel like I've really accomplished something, only to be shorthanded by a literal language barrier.
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    It should still be easy enough to use, just Search/Replace the copy beforehand. (Honestly, typing in IPA is going to be the main difficulty!)
  • CalicoStonewolf
    CalicoStonewolf Posts: 21
    Christian Thalmann said:
    It should still be easy enough to use, just Search/Replace the copy beforehand. (Honestly, typing in IPA is going to be the main difficulty!)
    Not if I'm running it through a lookup/translator first.
  • John Savard
    John Savard Posts: 1,135
    An issue I would see - obviously, this is intended for implementing a conlang - is that when a consonant is between two vowels, there will be no indication of which syllable it belongs to. Although for some reason this isn't an issue in this case, as the vowel which precedes the consonant is to take priority, I'd think that is the normal reason this sort of thing doesn't work.
    And the answer, of course, would be to use Korean jamo instead of IPA characters or Latin letters in this more usual case.
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    I think the goal is less to accurately map the syllable structure but rather to «stow» each vowel inside a nearby consonant for the purpose of neatness...

    In any case, if you use an automated lookup/translator, you might as well implement the theta/þ replacement there...
  • CalicoStonewolf
    CalicoStonewolf Posts: 21
    Christian Thalmann said:
    I think the goal is less to accurately map the syllable structure but rather to «stow» each vowel inside a nearby consonant for the purpose of neatness...

    In any case, if you use an automated lookup/translator, you might as well implement the theta/þ replacement there...
    which, while I don't entirely like it, this is what I chose to do. I've still kept the theta symbol mapped for my own sake, but I just have to remember for it to work correctly, to swap them out. It's very annoying, being the only thing holding it back from working as intended, but it's not my choice. I think it would have to take somehow convincing whoever is in charge of changes to Unicode mappings or something to have a Latin theta, but that's never gonna happen. :P

Categories