RE: Arabic letters separated by markup

From: Jony Rosenne (rosennej@qsm.co.il)
Date: Fri Jun 10 2005 - 14:40:15 CDT

  • Next message: Mete Kural: "Re: Arabic letters separated by markup"

    I made up test pages with Hebrew and Latin, where the diacritics are
    separated by markup indicating a different color.

    On my computer (Windows XP), MSIE ignores the color for all diacritics, and
    FireFox respects the black color for Hebrew and drops the Latin diacritics.

    http://www.qsm.co.il/Hebrew/HebrewTest/ColorHtml.htm (HTML, uses the user
    default font)

    http://www.qsm.co.il/Hebrew/HebrewTest/ColorCss.htm (CSS, uses Arial
    (Hebrew) and Arial fonts)

    Jony

    > -----Original Message-----
    > From: unicode-bounce@unicode.org
    > [mailto:unicode-bounce@unicode.org] On Behalf Of John Hudson
    > Sent: Friday, June 10, 2005 6:57 PM
    > To: unicode@unicode.org
    > Subject: Re: Arabic letters separated by markup
    >
    >
    > Jony Rosenne wrote:
    >
    > > See http://www.w3.org/TR/html401/struct/global.html#h-7.5.3
    > >
    > > It is clear that block level elements interrupt text flow
    > and that inline
    > > elements should not.
    >
    > Some stylistic inline elements, e.g. changing font size or
    > style, will unavoidably create
    > run boundaries for rendering. In many scripts, these will not
    > cause any problems. For
    > instance, one can make one letter in the middle of a
    > Latin-script work bold or italic
    > without doing anything to the rendering that would strike the
    > reader as incorrect shaping.
    > But if one tries the same trick in the middle of an Arabic
    > word, one will break the
    > shaping, since font level glyph layout can't be processed
    > across run boundaries.
    >
    > 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 : Fri Jun 10 2005 - 13:41:54 CDT