Re: Display of Isolated Nonspacing Marks (was Re: Questions on ZWNBS...)

From: Jim Allan (
Date: Fri Aug 08 2003 - 12:54:51 EDT

  • Next message: "Newbie Question - what are all those duplicated characters FOR?"

    Philip Verdy posted:

    > Could ZWS+combining diacritic may be the best solution for
    > isolated diacritics in text?


    << * Such characters may be large enough to effect the placement of
    their base character relative to preceding and succeeding base
    characters. For example, a circumflex applied to an "i" may effect
    spacing (""), as might the character U+20DD COMBINING ENCLOSED CIRCLE. >>

    Unless Unicode 4.0 as changed this the words "may" and "might" here
    would indicate that ZWSP is not *necessarily* the best solution.

    There is no specification about what an application *must* do to be
    conforming in this circumstance, merely indication that an application
    that does expand spacing for the sake of appearance is not
    non-confirming. It is *probably* implied that this is the right way to go.

    But I would guess that it would also be conforming for an application to
    not expand spacing at all on ZWSP so that coding of _o_ + ZWSP +
    COMBINING CIRCUMFLEX + _o_ would place the circumflex centered over _oo_
    with its center point between the two letters.

    Either result would be useful for different purposes.

    It certainly makes sense that in the case of space characters that have
    a defined width that this width is innate to the definition of the
    character and in such a case should take precidence over the width of
    the normally non-spacing combining character.

    I would welcome clear instructions by Unicode on this point where either
    result would be useful in order than applications may be expected to
    produce results that are consistent with each other. :-)

    I would think it would be consistant with Unicode for an application to
    shrink the width of normal space followed by a diacritic such as a
    single overdot as exact formatting behavior is not defined in such cases.

    Jim Allan

    This archive was generated by hypermail 2.1.5 : Fri Aug 08 2003 - 13:24:55 EDT