Re: Arabic & Arabic Extended Languages have 4 or 2 Glyps ...

Date: Thu Dec 23 1999 - 12:36:37 EST

       Dear Mr. Liwal:

       Yes, your glyphs need glyph IDs, but they don't need to
       directly associated with Unicode values; i.e. you don't need to
       have an entry in the cmap pointing to every glyph.

       Your decision to use the Compatibility Area for those things
       that have been defined is what I would do.

       There have not been any changes to the Arabic Presentation
       Forms-A and -B blocks between versions 2.0 and 3.0. Those
       codepoints that were previously undefined are still undefied.

       Please note, though, that you should *not* assign anything to
       undefined characters. Doing so will make your fonts
       non-conformant. In other words, if you have glyphs that don't
       appear in any of the defined positions in the Unicode 2 charts,
       then your only options are to assign them to PUA values (U+E000
       - U+F8FF), or to not assign them to any Unicode value but have
       the glyphs accessible *only* by their glyph IDs within your
       OpenType tables. I recommend the latter.



       From: <> AT Internet on 12/23/99 10:06 AM

       Received on: 12/23/99

       To: Peter Constable/IntlAdmin/WCT, <> AT
       Subject: Re: Arabic & Arabic Extended Languages have 4 or 2
             Glyps ...

       Dear Mr. Peter;

       Thanks for the advise, back in 1998 I have developed my inhouse
       tools that can Encode OpenType (TrueType Open) tables to
       TrueType Fonts Even in OpenType you need Glyp IDs.

       What I want that may fonts should be 100% Unicode Compatable,
       therefore I post question to Unicode.

       What I decided now that I will utilize Compatablity Area as
       much as possible.

       What about the unassinged Code Points Compatabilty Area, I have
       Unicode Ver.20, are they assigned to any charachter in Ver 3.x?


       ----- Original Message -----
       From: <>
> In fact, the best option is probably not exactly
> Map the glyph for the isolate from the cmap using the
> value in the Arabic block; if you use glyphs that
       correspond to
> characters in the compatibility areas (>= U+FB00),
       then also
> map these glyphs from the cmap using the Unicode
       values in that
> range. For any other glyphs, however, do not map them
> from the cmap; i.e. don't assign them to *any* Unicode
> PUA or otherwise.

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:57 EDT