Re: Arabic letters separated by markup

From: John Hudson (tiro@tiro.com)
Date: Sat Jun 11 2005 - 14:05:14 CDT

  • Next message: John Hudson: "Re: Arabic letters separated by markup"

    Kent Karlsson wrote:

    >>>(always) be blocked over markup tags.
    >>
    >>But if one letter is at 20pt and another is at 12pt, how can
    >>the form a ligature?

    > I wrote: BLOCKED.

    I think I misinterpreted that term: I took it to refer to a 'block' of text that included
    text on either side of the markup tags. My mistake.

    >>as you change the font, you
    >>are dealing with completely separate runs of glyphs that will
    >>be independently shaped.

    > Are you saying that you cannot process ZWJ?

    ZWJ is a character, and it plays a role in glyph processing either directly or indirectly.
    ZWJ is part of the text, while something like html markup is not part of the text.

    > That way the shaping behaviour, both for cursive joining and for
    > ligature
    > formation, is the same independently of whether it is just a colour
    > change,
    > a size "change" (regardless of if the the new size is the same as
    > before), or
    > style "change" (regardless of if the the new style is the same as
    > before), etc.

    I'm not sure that I follow all of this. As I wrote before, I can see some possibilities
    for handling basic Arabic shaping, i.e. substitution of positional forms, regardless of
    intervening markup, and colour is a relatively trivial and separate issue since it is in
    no way font-specific. But I don't see at all how something like this

            |lam|<em>|alif|</em>

    Can be correctly rendered, with an obligatory ligature, without either ignoring the markup
    or applying it to both letters.

    John Hudson

    -- 
    Tiro Typeworks        www.tiro.com
    Vancouver, BC        tiro@tiro.com
    Currently reading:
    Truth and tolerance, by Benedict XVI, Cardinal Ratzinger as was
    War (revised edition), by Gwynne Dyer
    


    This archive was generated by hypermail 2.1.5 : Sat Jun 11 2005 - 14:06:59 CDT