Tested 3-17-99 Win32 build. The page contains Kanji and hankaku katakana. After the double-byte characters are displayed, the page does not display anything. The part of hankaku katakana is the point the page does not display anything. I put the simple test case in http://babel/tests/browser/bft/jis/aiu-jis.html which contains only hankaku katakana. It does not display anything even though I click on Reload button.

It is ESC ( I 123 ESC ( B erik: what is ESC ( I ? I don't think hankaku is part of ISO-2022-JP. We should think carefully about the error handling of our converter. Should accept ESC ( I ? ( in my opinion, we may not want to but 4.x did. We may be foreced to do it for backward compatability) or we should at least make sure it convert correctly after the ESC ( B. Erik, please help to give us opinion.

The iso-2022-jp escape sequences are as follows: ESC(B ASCII ESC(J JIS X 0201-1976 Roman ESC$@ JIS X 0208-1978 ESC$B JIS X 0208-1983 The following escape sequences are not part of iso-2022-jp: ESC(I JIS X 0201-1976 Kana ESC$(D JIS X 0212-1990 However, our old code base accepted these escape sequences. Also, in networking it is better to be liberal in what you accept, and conservative in what you send out. So we should accept these escape sequences at least for receiving, and we may even need to use one or both of them for sending to remain compatible with our old code base. Naoki may have further input on the Hankaku Kana issue for mail/news.

Moved to M6 to give Cata time to assist Nisheeth on external DTD support in the XML parser.

push to M10

Mark this back to M7 since I have already have fix in local dist. Need to check in monday morning

check in and Mark it fix. Please verify ASAP

I changed the URL. I verified this in 6-14-09 build.