icu4c proposal: remove _generic_ ISO-2022 converter

I propose to remove the _generic_ ISO-2022 converter from ICU.

I believe that this converter is not used, and not very useful.

Conversion to Unicode: It does not handle the full ISO-2022/ECMA-35 framework. For example, it 
switches immediately after escape sequences announcing G1/G2/G3 character sets, instead of 
registering them and waiting for SO/SS2/SS3 shifts. It also does not handle a switch-with-no-return 
to UTF-16BE or similar.

Conversion from Unicode: The ICU ISO-2022 converter always emits an escape sequence for shifting to 
UTF-8, followed by UTF-8 for the rest of the text.


The ISO-2022-JP, -CN, -KR variants and their subvariants are different: They provide a selection of 
subsets of supported charsets and escape sequences. This proposal does not cover any of these 
specific variants.


Please reply if you have comments on
- the removal of the generic ISO-2022 converter in ICU 2.8
- the removal of the generic ISO-2022 converter in ICU 3.0

Sincerely,
markus

_______________________________________________
[email protected] - icu4c-support mailing list
To Un/Subscribe:
http://oss.software.ibm.com/developerworks/oss/mailman/listinfo/icu4c-support