Red Hat Bugzilla – Bug 137534
Wrong char width for Japanese ideographs
Last modified: 2007-11-30 17:10:53 EST
Description of problem: Full-width qoute char's width is wrong in Emacs by default, So .emacs or other emacs settings needs the appropriate addition. Version-Release number of selected component (if applicable): How reproducible: Everytime Steps to Reproduce: 1. Add full width quote word to Canna dict. 2. Run emacs. 3. Input full width quote word into emacs through Canna. Actual results: Half width Expected results: Full Width Additional info: http://d.hatena.ne.jp/rna/20041027#p2 (Japanese)
Created attachment 105933 [details] Diff for lang-coding-systems-init.el This language setting file include all languages settings. Previous language settings were separated by each language. I wonder why you unified them while Red Hat splitted input method settings in /etc/X11/xinit/xinit.d/ for each langauge.
'ascii must be on top of the list, or you can't input '\'(backslash). And complete charset list would be better. (un-define-change-charset-order (list 'ascii 'japanese-jisx0208 'japanese-jisx0212 'katakana-jisx0201 'latin-iso8859-1 'latin-iso8859-2 'latin-iso8859-3 'latin-iso8859-4 'cyrillic-iso8859-5 'greek-iso8859-7 'hebrew-iso8859-8 'latin-iso8859-9 'latin-iso8859-14 'latin-iso8859-15 'ipa 'chinese-gb2312 'chinese-cns11643-1 'chinese-cns11643-2 'chinese-cns11643-3 'chinese-cns11643-4 'chinese-cns11643-5 'chinese-cns11643-6 'chinese-cns11643-7 'chinese-big5-1 'chinese-big5-2 'korean-ksc5601 'latin-jisx0201 'thai-tis620 'ethiopic 'indian-is13194 'chinese-sisheng 'lao 'vietnamese-viscii-lower 'vietnamese-viscii-upper 'mule-unicode-0100-24ff 'mule-unicode-2500-33ff 'mule-unicode-e000-ffff 'mule-ucs-unicode-multichar))
Perhaps this should indeed only be changed for ja, but does the following patch do the right thing for Japanese? And how does it affect affect input for other languages? diff -u Mule-UCS-current/lisp/un-define.el~ Mule-UCS-current/lisp/un-define.el --- Mule-UCS-current/lisp/un-define.el~ 2004-11-01 21:22:16.556475869 +0900 +++ Mule-UCS-current/lisp/un-define.el 2004-11-01 21:22:16.571473901 +0900 @@ -143,6 +143,8 @@ ipa japanese-jisx0208 japanese-jisx0212 + latin-jisx0201 + katakana-jisx0201 chinese-gb2312 chinese-cns11643-1 chinese-cns11643-2 @@ -154,8 +156,6 @@ chinese-big5-1 chinese-big5-2 korean-ksc5601 - latin-jisx0201 - katakana-jisx0201 thai-tis620 ethiopic indian-is13194
With regards to comment 1, at the time it seemed all the language files were so similar that unifying to one file made sense - more so because of the common code needed to setup MuleUCS for utf-8. But in retrospect perhaps it would be better to split up lang-coding-systems-init.el again: specially if different coding-system orders are going to be needed for different locale...
No, comment #3 patch doesn't make sense.
Why not? :)
I mean it doesn't work on my desktop. I don't know the reason.
According to the url in the original report this has been improved in the coming version of emacs. Since there doesn't seem to be obvious clear solution for 21.3 closing for now.