From: Ryuichi Arafune <[EMAIL PROTECTED]> Subject: WHEN do you upload to grep and sed with multi-byte extension? > If you left them not supported multi-byte, Japanese users will not be > happy to use debian. There is not enough time to release potato. If > you cannot upload grep and sed with multibyte extension until the > frozen status (November 1st?), I want to upload them as another > package (grep-ja or sed-ja).
I've maintained grep-ja/sed-ja package in Debian JP Project. (I'm now waiting to become new maintainer, so Ryuichi Arafune intended to upload/merge grep-ja/sed-ja to Debian Project instead of me.) I'm currently working grep-2.3 multi-byte patch because original grep-ja package based on grep-2.0. This multi-byte patch has ability to deal with EUC/SJIS (muti-byte characters). If I complete working, I plan to send it. BTW, sed-ja patch has more ability handling EUC/SJIS, and UTF-8. It's nice that sed is UTF-8 ready. If Mr.Wichart merge between grep/sed and their-ja patch, it's useful for Japanese, and besides for non-Japanese people. Properly speaking, these `i18n/m17n' related issue should fix/apply in upstream level. Many essential tools (currently even libc!) don't have much ability to handle multi-byte characters. I, however, understand that it 's difficult problem to achieve `m17n'; for example, generally regular expressions don't consider multi-byte characters. But, it is to be regretted that these current m17n circumstance, status, situation have continued. In addition, I think that a software's `i18n/m17n' does not mean to deal with *only* Unicode. Regards, -- GOTO Masanori Department of Computational Intelligence and Systems Science, Tokyo Institue of Technology.
