Re: A bug report: Pango cannnot handle some CJK character sets

[Mark Leisher <mleisher crl nmsu edu>]

    Ienup> Hi Owen, I believe there are a couple of other issues regarding
    Ienup> this mapping mechanism. The first thing is the tables are hard
    Ienup> coded in a source file which make people to change and re-compile
    Ienup> whenever there is a need for a new mapping. There are several other
    Ienup> ways to externalize the tables, including especially charset[],
    Ienup> char_mask_map[], and char_masks[] tables.

This is something I brought up with various people at various times.  We have
a large number of mappings our customers use that have emerged from many years
of use and are not in any standard.  I hate to compile them all into iconv to
make a special distribution version.

Our current mapping program is really, really slow, but it loads mapping
tables at runtime which has kept it low-maintenance for a number of years
now.

But at the same time I feel that in an effort to promote one common encoding,
users should be responsible for dealing with conversion issues.  That will
make them *want* to convert all their text to UTF-8 or UTF-16.

My current opinion is that encoding conversion support should not be made
available as part of all these different projects, but should be left to one
package, iconv perhaps, to do all conversion chores on all platforms.

Gotta go!
-----------------------------------------------------------------------------
Mark Leisher
Computing Research Lab            Cinema, radio, television, magazines are a
New Mexico State University       school of inattention: people look without
Box 30001, Dept. 3CRL             seeing, listen without hearing.
Las Cruces, NM  88003                            -- Robert Bresson