L2/04-017 Source: Markus Scherer Date: January 12, 2004 Subject: Proposed update to UTR #22 Please add as a document and agenda item. I propose to update UTR #22 Character Mapping Markup Language so that mappings between multiple Unicode code points and multiple codepage character sequences are defined. As far as I can tell, the DTD already allows such m:n mappings, but some of the text assumed mappings between single Unicode code points and single codepage character sequences. m:n mappings are used in mapping tables such as between Unicode and charsets with the JIS X 0213 repertoire. JIS X 0213 contains precomposed characters for Hiragana/Katakana Ka/Ki/Ku/Ke/Ko + semi-voiced mark. IBM has Unicode conversion tables for some EBCDIC CCSIDs for JIS X 0213. http://www.mindspring.com/~markus.scherer/unicode/tr22-4-20040107.html I uploaded the file via binary FTP. Pointing a browser to http://www.mindspring.com/~markus.scherer/unicode/ , right-clicking on tr22-4-20040107.html and "save as" or similar should download it properly. Best regards, markus