Re: Mapping tables available

From: Mark Leisher (
Date: Wed Aug 14 1996 - 19:10:36 EDT

[Sorry, I forgot to forward my reply to this list.]

    Asmus> I'm a bit confused. Are these mappings really Unicode 1.1 specific,
    Asmus> or did you just not consider the information on Unicode 2.0.

I was unclear about this. The mapping tables are Unicode 1.1 specific for
Hangul but the others should work for Unicode 2.0. I haven't had time to
verify them against the Unicode 2.0 Character Database, so I thought I'd
better stick with saying "Unicode 1.1."

    Asmus> Also, what exactly is the value added of these tables compared to
    Asmus> the ones on the Unicode site?

Three valued added features:

1. All the tables are three tab-separated columns. The first two columns
   contain either a space separated list of hex values or a sequence of ASCII
   characters. The third column has the Unicode character name prefixed by
   "# ".

   I don't know what the definition of "Format A" is with regard to the
   Unicode Consortium supplied tables, but some have three columns and some
   have four. Not much of a problem to code around, I'll grant you, but a
   trifle irregular.

2. Our representation allows one-to-one, one-to-many, many-to-one, and
   many-to-many mappings. This representation regularizes what we feel is
   needed for our current and future encoding/transliteration conversion

3. Mapping tables are provided that are not provided by the Unicode

    Asmus> Is it just a transformation in the layout?


    Asmus> Did you change the mapping values for certain characters? If so,
    Asmus> is there any information that you would like to feedback to the
    Asmus> Consortium.

The vendor-specific codepoints in the Private Use Area were changed to U+FFFD
until I can solicit more current info on vendor-specific allocations from the
Private Use Area. Our version of the Mac Arabic table has been complicated a
bit by removing the explicit directionality codes. This wasn't strictly
necessary, but at the time we needed the table, we did not support explicit
bidi, only implicit.

Other than that, no character mappings were changed.

Unless the variable number of columns in "Format A" tables is an issue, there
is no real feedback yet. Any problems with the mapping information we come
across will certainly be reported.
Mark Leisher "A designer knows he has achieved perfection
Computing Research Lab not when there is nothing left to add, but
New Mexico State University when there is nothing left to take away."
Box 30001, Dept. 3CRL -- Antoine de Saint-Exup'ery
Las Cruces, NM 88003

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:31 EDT