On Friday, April 5, 2002, at 08:40 , Nick Ing-Simmons wrote:
It is _really_ sad, Tk only realy _needs_ one encoding which it expects
it to be called ucs-2be or iso10464-1
We don't support either name.
Instead we claim UCS-2 without specifying and endian :-(
As a matter of fact, I don't like that way either. My first thought
was iso10464-1. But speaking of those two, they are not exactly equal
to UTF-16BE.
I really think Encode/10646_1.pm should have a definitive name which has
its endian.
UTF-16 does support the surrogate pair. Does our version support
surrogates correctly? If so the canonical should be UTF-16BE, not UCS-2
or iso10464-1.
Tk would also like 'jis0208' and 'jis0201' but those are not found
by default either.
I cannot even find the table for jis0208 any more :-(
I have dropped it while gb2312 and ksc5601 does have raw version. It is
inconsistent. Okay. I will restore raw jis0208 and jis0201 if jhi
agrees (we don't need raw jis0212, do we?)
Dan the Encode Maintainer
