On Mon, 2002-02-25 at 12:55, Junichi Uekawa wrote: > Well, use utf-8, is probably the most simple answer > In fact, I have an impression that -utf8 libs will work > with any locale, but I might be wrong. It uses > the locale-charset conversion routines...
They will work with any locale so long as you stick to characters that are valid in that locale. In the "C" locale this is just the 7-bit ASCII set; none of the Latin1 accented characters are available. The part that causes trouble is where the libraries call mbrtowc() in order to work out how much screen real-estate a string needs. This function will give up and return an error code if it encounters a character that it thinks is invalid, which gives you the classic "strings truncated at the first accented character" syndrome. This is the reason that the UTF-8 disks will only fall back to English if framebuffer is unavailable. It would be possible to fix the problem by shipping an appropriate locale on the root disk, but there isn't enough space. p. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
