Loading ...
Sorry, an error occurred while loading the content.
 

Re: how to encode chill~ glyphs in unicode

Expand Messages
  • baijum81
    I think it is necesary to include chillu~ letters in unicode standared. (Consider input mechanism, internal representation of font and font display
    Message 1 of 16 , Aug 4, 2002
      I think it is necesary to include chillu~
      letters in unicode standared.
      (Consider input mechanism, internal representation
      of font and font display characteristics, will explain
      my ideas later)

      Whom should we contact at first? directly Ministry
      for Information Technlogy or any body here (kerala).

      Pavanaja, Can you help us for this (I think kannada
      has done some changes in unicode by your
      group effort).

      We have to make this fast, because without
      standardization Malayalam computing
      will not be possible in its complete range.

      Regards,
      Baiju M

      --- In malayalamlinux@y..., Cibu Johny <cibu@y...> wrote:
      > Dear Mahesh,
      >
      > Invariably, Unicode people will point to ISCII
      > document (I have a local copy at my varamozhi site).
      > (or they will be mute) ISCII infact describes how to
      > generate chill~ letters. Only problem with adopting
      > ISCII is that the original meaning of
      > Zero-Width-Joiner will be lost. Rules from ISCII are:
      >
      > conjuct of X and Y = X + virama + Y
      > chill~ of X = X + virama + zwj
      >
      > So if the language engine see X + virama + zwj + Y,
      > for all other indian languages, X will form the
      > conjunt with Y. But for Malayalam it will be: chill~
      > of X + full Y.
      >
      > If you look closely, Malayalam has two half forms for
      > a letter(at least for 'Ra') - 1. when it join from the
      > left(eg: chill~ Ra) 2. when it join from right(eg:
      > vakram). virama+ZWJ covers only the case 1.
      >
      > thanks,
      > Cibu
      >
      >
      > --- Mahesh T Pai <paivakil@v...> wrote:
      > > I think that the only solution ( and the proper one,
      > > I feel ) is to
      > > include the chillus in the Unicode Standard. I have
      > > been going through
      > > some old typewritten / printed matter with me, and
      > > find that even when
      > > the barest minimum number of characters are used,
      > > the chillus have been
      > > given their due place.
      > >
      > > Cibu had pointed out on this list (or was it the
      > > indlinux list?) that
      > > microsoft is erroneously treating chillu encoding as
      > > a rendering
      > > problem. This mistake has been cause of issues like
      > > this, and it is
      > > time to take this up, and get chillus included in
      > > the Unicode standard.
      > >
      > > On Thu, 25 Jul 2002 15:30:43, cibu wrote:
      > >
      > > >If this is correct, following sequences could be
      > > encoded as:
      > > ><chill~ n> <na> = 'na' + virama + zwj + 'na'
      > > ><conjunct nna> = 'na' + virama + 'na'
      > > ><na> attached to <half va> = 'na' + virama + 'va'
      > > >
      > > >There is a small confusion here because, the
      > > semantics of zwj is to
      > > >join with the next literal. Here it forces the
      > > preceding literal to
      > > >stand seperate forming the 'chill~'. Also
      > > virama+zwj sequence or just
      > > >the 'virama' is supposed to generate the half form
      > > of the literal
      > > >preceding it. But in the third example('nva'), it
      > > will generate the
      > > >half form of the literal after it.
      > > >
      > > >
      > >
      > >
      > >
      > >
      >
      ______________________________________________________________________
      __
      > > Want to sell your car? advertise on Yahoo Autos
      > > Classifieds. It's Free!!
      > > visit http://in.autos.yahoo.com
      > >
      >
      >
      > =====
      > ...............................................................
      > ....... Download Transliteration based Malayalam Editor .......
      > ........... from http://varamozhi.sourceforge.net .............
      > ...............................................................
      >
      > __________________________________________________
      > Do You Yahoo!?
      > Yahoo! Health - Feel better, live better
      > http://health.yahoo.com
    • baijum81
      ... Suppose, I have another font (eg:- Hindi,Hebru,Tai,Tamil etc.) with unicode encoding. And it uses any of Private location which is already taken by
      Message 2 of 16 , Aug 9, 2002
        --- In malayalamlinux@y..., "nvshaji" <nvshaji@n...> wrote:
        >
        > Hi Mahesh,
        >
        > Sorry for not responding to you before.
        >
        > Your observation is not right. (I too had wasted my time searching for
        > unicode representation for these glyphs.) The characters in ED15 to
        > ED55 are not to be part of unicode. They are only used as substitution
        > mechanism in Pango. Let me explain how it works.
        >
        > if you want to have "kka" you can represent it in unicode as
        > 0x0D15 0x0D4D 0x0D15
        > ka ^ ka (ka + chandrakkala + ka).
        > When Pango sees this combination in the text, then it looks up a
        > translation table, which substitutes the three chars with one single
        > glyph (kka - 0xED15). 0xExxx is reserved for private usage in Unicode.
        >

        Suppose, I have another font (eg:- Hindi,Hebru,Tai,Tamil etc.)
        with unicode encoding. And it uses any of Private location
        which is already taken by malayalam.ttf . Then, if I am opening a
        document which contains both languages. (I have both fonts and pango
        running) Then what will happen?
        Which character from the private area will show?
      Your message has been successfully submitted and would be delivered to recipients shortly.