RE: Zhuyin font similar to Hangeul

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Mon May 11 2009 - 15:30:31 CDT

  • Next message: John H. Jenkins: "Re: Zhuyin font similar to Hangeul"

    If you have resources showing how Zhuin syllables can be infered from the inline encoding, and then how to stack them properly so that the resulting display will not be confusive and will have an unambiguous reading, nothing forbids you to create a mapping from Zhuin initial, medial and final letters to a linear index that can be combined like in Hangul to create the composed syllables squares.

    But can you cite a reference showing that shuch stacking is used, and how it can be rendered ? Are there layout constraints defined ? How are the Zhyuin letters modified or simplified when they are stacked in the same square ? Do these letters overlap in the composition square (like in the Han ideographs) ? Now, if you have a well defined set of syllables that can be stacked this way, using full orthogonal composition without exceptions (like in Hangul), nothing fordids you, then, to define SUBST rules in your font that will remap the syllables into a single glyph id (possibly a PUA only within the font itself, for the single purpose of this composition), and so you'll be able to create a font behaving like Hangul.

    The difference with Hangul, is that Hangul does not needs that fonts define the 11000 or so "SUBST rules", as Hangul syllables just need to be mapped from the canonical precomposed code points. Note however that Hangul fonts may contain more precomposed syllables than just the predefined Hangul syllables (those that are used in modern Korean): there are also other syllables that cannot be represented in Unicode using a single code point for the precomposed syllable, notably for squares that are using more than just a single modern initial consonnant, a single medial modern vowel and a single modern final consonnant: in Unicode, these syllables are still representable, even if they are not precomposed, and support for those additional syllables still need SUBST rules, just like you'll need to define for your Zhuin project.

    Note however that Zhuin letters do not differentiate initial and final letters, so you first need a rule to define how the syllables are broken, as I don't think that your composed squares should stack more than one syllable (and optionally its respective tone mark). Tone marks are good indicators of where the syllable breaks can occur, but they are certainly not enough.

    Forsuch project, you need more knowledge about the Zhuin script. The first thing to find will then be samples of printed books or manuscripts showing these squares and if there's some rule governing their layout.

    > -----Message d'origine-----
    > De : unicode-bounce@unicode.org
    > [mailto:unicode-bounce@unicode.org] De la part de Gerrit
    > Envoyé : lundi 11 mai 2009 21:21
    > À : unicode@unicode.org
    > Objet : Zhuyin font similar to Hangeul
    >
    >
    > Hello,
    >
    > I’m wondering if it is possible to make a font for zhuyin
    > which behaves similar as Hangeul Jamo.
    >
    > Currently, if I type ㄓㄨˋ ㄧㄣ, it will be displayed in a line,
    > with the tone mark after it. This not only does not look very
    > nice, but also is too long.
    >
    > I thus wonder, if I can create a font which stacks the
    > characters on top of each other and then puts the tone mark
    > at the right. I guess, the method would be similar, or even
    > identical, to Hangeul. But, Zhuyin characters are not
    > designed by Unicode to behave like that (at least as far as I
    > know). I wonder, if I can create a font with this behaviour
    > nonetheless.
    >
    > I think, this would be possible by creating many ligatures,
    > but this is not convenient and does not allow to create
    > customized syllables, so the Hangeul method would be better.
    >
    > Does anyone know something about this?
    >
    > Thanks,
    > Gerrit
    >
    >
    > Aucun virus détecté dans ce courrier reçu.
    > Analyse effectuée par AVG - www.avg.fr
    > Version: 8.5.325 / Base de données virale: 270.12.20/2100 -
    > Date: 05/11/09 05:52:00
    >



    This archive was generated by hypermail 2.1.5 : Mon May 11 2009 - 15:34:06 CDT