Re: Punctuation character (inverted interrobang) proposed

From: Christopher Fynn (cfynn@gmx.net)
Date: Thu Oct 20 2005 - 04:00:50 CST

  • Next message: Michael Everson: "Re: Punctuation character (inverted interrobang) proposed"

    Denis Jacquerye wrote:

    > The inverted interrobang glyph could be a ligature of inverterted
    > U+00BF + U+200D + U+00A1 or the other way around, it would just need
    > to be defined in the font.

    Look-ups in a font could be defined for both ways - both pointing to the
    same glyph

    A problem I foresee is that some OpenType shaping engines make certain
    assumptions and internally classify characters or their glyphs. They
    then only apply particular lookups to particular categories of glyphs.
    In this kind of shaping engine glyphs for punctuation characters would
    normally be in a category which excluded them from most shaping lookups.

    In other words you might include a lookup in a font: U+00BF + U+200D +
    U+00A1 = inverted interrobang glyph but I suspect this might not work
    with all OpenType shaping engines (though this is something probably
    better discussed on th OpenType mailing list).

    > But I don't know if this is the proper use of ZWJ.

    I think this is proper use - after all you effectively want to represent
    something which is like glyphs for each of the two other characters
    *joined* sharing a common dot.

    If you don't want to use a combination of Unicode characters, then in a
    font the inverted interrobang glyph could classified as a discretionary
    alternative form for the regular interrobang and accessed using OpenType
    features.

    - Chris



    This archive was generated by hypermail 2.1.5 : Thu Oct 20 2005 - 04:01:49 CST