Chinese (Taiwan) shows garbled TOC in chm

[lade]

Chinese (Taiwan) shows TOC in chm in wrong codepage. I'm using language code 'zh-tw'.

Problem seems to be the following: File map2hhpImpl.xsl adds '-' to the language code before analysing it (zh-tw-), but later on when ConvertLang.java checks languages.properties, the language code seems to be 'zh-tw' while languages.properties supposes it should be 'zh-tw-'.

By adding
zh-tw=0x0404 Chinese (TAIWAN, PROVINCE OF CHINA)
to languages.properties I got the Taiwanese TOC to show correctly. Without this change the .hhp file will assume that the encoding is 0x0804 Chinese (CHINA), which is incorrect for Taiwanese.

[jelovirt]

Related to #1271