[COMMIT] Switch away from ISO 2022 for the ISO-8859 character sets.

[Stephen J. Turnbull]

VETO

Aidan Kehoe writes:
 > 
 > APPROVE COMMIT
 > 
 > NOTE: This patch has been committed.
 > 
 > It would perhaps have been a little nicer to use the Unicode mapping tables
 > under etc/unicode-consortium for this, but it wouldn’t read any better.

This needs to be done eventually, but will probably happen en passant
with the move to Unicode internal coding.

 > GNU provide iso-latin-[1-16]-with-esc. Might be worth considering, with an
 > autoload infrastructure.

This must be done *immediately*, as otherwise there is no way to read
files that have inadvertantly been saved with ISO-2022 extensions.
There's no reason in 21.5 that you can't simply rename the builtins to
-with-esc and duplicate or alias the ones defined in
latin-euro-standards.  Then the autoload feature (which will almost
surely be hard to implement fully because of the large number of
places in the C initialization code where coding systems are used) can
be added later.

If you find that too revolting, revert this change until you've
provided some way to read "XEmacs ISO-8859" files that contain
ISO-2022 designation sequences.