Re: Matching opening and closing characters: How?

From: Joachim Durchholz (jo@durchholz.org)
Date: Sun Aug 16 2009 - 14:48:22 CDT

  • Next message: Karl Pentzlin: "Re: Support of U+00AD SOFT HYPHEN in current software"

    Am Samstag, den 08.08.2009, 23:18 +0200 schrieb Hans Aberg:
    > > In other words, different pairs of parentheses
    > > would have different, programmer-defined semantics, which means the
    > > programmer names the parenthese pair anyway and I don't need to rely
    > > on
    > > Unicode to classify parentheses.
    >
    > The grammar of the set matching pairs is not context-free.

    I'm not sure what exactly you meant here, so I can respond only in very
    general terms:

    * For most applications, type information is sufficient to disambiguate.

    * A library designer will have to think quite hard to select a useful
    set of bracket symbols. One thing that counts against reusing the
    original symbols is that these are usually not readily available on the
    keyboard.
    So I guess that while having mathematical bracket notations available
    would be super-cool, it probably won't happen.

    Regards,
    Jo



    This archive was generated by hypermail 2.1.5 : Sun Aug 16 2009 - 14:52:19 CDT