Mixed character encoding in lpstat output

[frifl]

This issue was referenced here from apple/CUPS : apple/cups#5896

I run CUPS 2.3.1 on Ubuntu 20.04 and I get mixed character encoding in the output from lpstat. The system locale is set to zh_CN.GB18030. The message is UTF-8 and the timestamp is GB18030.

$ lpstat -a
printer 鑷粠 2021年02月04日 星期四 07时51分54秒 寮�濮嬫帴鍙楄姹�

The timestamp is correct but there are replacement characters in the message part which indicate bad encoding. I checked that the message part was indeed UTF-8 by manually decoding the bytes.

This is my locale setting:

$ locale
LANG=zh_CN.GB18030
LANGUAGE=
LC_CTYPE="zh_CN.GB18030"
LC_NUMERIC="zh_CN.GB18030"
LC_TIME="zh_CN.GB18030"
LC_COLLATE="zh_CN.GB18030"
LC_MONETARY="zh_CN.GB18030"
LC_MESSAGES="zh_CN.GB18030"
LC_PAPER="zh_CN.GB18030"
LC_NAME="zh_CN.GB18030"
LC_ADDRESS="zh_CN.GB18030"
LC_TELEPHONE="zh_CN.GB18030"
LC_MEASUREMENT="zh_CN.GB18030"
LC_IDENTIFICATION="zh_CN.GB18030"
LC_ALL=

This is the locales I have on my system:

$ locale -a
C
C.UTF-8
POSIX
zh_CN.gb18030

[michaelrsweet]

The issue is that libcups doesn't recognize the character set name "GB18030". Investigating...

[michaelrsweet]

GB18030 replaces GB2312 aka Windows code page 936. So we need to add it to the cups_encoding_t enum and the name table, then the right things should happen...

[michaelrsweet]

[master e0c9fdc] Add GB18030 charset support for localization (Issue #159)