Hi, On 8/22/05, Colin Watson <[EMAIL PROTECTED]> wrote: > On Sat, Aug 20, 2005 at 05:04:55PM +0800, Huang, Zhangrong wrote: > > Currently, Chinese users can't read manpages-zh well, because of man-db > > and groff lacking of some CJK encodings. > > > > This patch will add more CJK encodings for groff, with this patch, a > > patched man-db will work fine in CJK locale. > > Looks fair enough to me. > > Why did you exclude zh_*.UTF-8 and ko_*.UTF-8, though? Does the UTF-8 > support added for Japanese in groff 1.18.1-5 not work for Chinese or > Korean? I've attached the diff, which looks reasonably language-agnostic > to me. It's include. As you know, UTF-8 support works fine for Chinese and Korean too, but groff doesn't have any stuff like tmac/euc-jp.tmac for EUC-CN(GBK/GB2312) or EUC-KR, that means if we use encoding EUC-CN(GBK/GB2312) or EUC-KR like EUC-JP, groff can't handle it well. So, we fall back to ascii8 and iso8859, and it works ok for Chinese and Korean, otherwise, we get wrong output in locales like zh_CN.GBK or ko_KR.EUC-KR.
> > I'm also slightly surprised that no files such as tmac/euc-jp.tmac are > needed for EUC-CN or EUC-KR. That's why the dirty patch only include zh_*.UTF-8 and ko_*.UTF-8 :-), add I think such files should be added by the upstream. > > Thanks, > > -- > Colin Watson [EMAIL PROTECTED] >
