dirty fix unicode char width

why

Unicode defined some character width as ambigious. Software determine the width of characters accroding the context. However sometimes it is hard or impossible to know the context. Thus I dirty fix some console softwares, let them recognize all characters in big5 encoding range as double width. Now you will be happy with utf8 terminal and big5 fonts :)

fix zh_TW.UTF-8/LC_CTYPE
- for wcwidth(3) and wcswidth(3)
- Before FreeBSD-5.3 or if you don't have utf8locale
  1. cd /usr/ports/misc/utf8locale/; make extract; cd work/utf8locale-*/mklocale
  2. big5width.pl big5uni.txt la_LN.UTF-8.src > zh_TW.UTF-8.src
  3. rm /usr/share/locale/zh_TW.UTF-8/LC_CTYPE
  4. mklocale -o /usr/share/locale/zh_TW.UTF-8/LC_CTYPE zh_TW.UTF-8.src
- FreeBSD-5
  1. cd /usr/src/share/mklocale; big5width.pl big5uni.txt UTF-8.src > zh_TW.UTF-8.src
  2. rm /usr/share/locale/zh_TW.UTF-8/LC_CTYPE
  3. mklocale -o /usr/share/locale/zh_TW.UTF-8/LC_CTYPE zh_TW.UTF-8.src
- FreeBSD-6, after 20050227
  - the procedure is same as FreeBSD-5
  - /usr/share/locale/zh_TW.UTF-8/LC_CTYPE
screen patch (against 4.00.02)
- dirty fix width of characters which range in big5 encoding. You will need this if your screen is in utf8 encoding and use big5 font.
- see also mhsin's patch, which make cjkwidth as screen option, and using Markus Kuhn's free wcwidth() implementation.

why

Reference