To permit the use of Kanji characters in atoms, variable names,
predicate names, messages and comments you need to set the environment
variable QUINTUS_KANJI_FLAG
to true
before starting Prolog.
When the Kanji flag is set, an 8-bit character codes (that is, a
character code between 128 and 255) is assumed to be part of a
multi-byte sequence representing a Kanji character. Such characters
are treated, for the purposes of Prolog's syntax, as if they were
lower-case alphabetic characters. The Kanji flag does not affect
character I/O, or the conversion between atoms and lists of character
codes using name/2
or atom_chars/2
. In the case of I/O, two (or more)
character codes will need to be read or written for every Kanji
character. Similarly, an atom made up of Kanji characters will be
transformed by name/2
or atom_chars/2
into a list of twice (or more
than twice) as many 8-bit character codes.
codeset byte(s) representation 0: ASCII 0xxxxxxx 1: JIS-X0208 1xxxxxxx 1xxxxxxx 2: JIS-X0201 10001110 1xxxxxxx 3: Gaiji 10001111 1xxxxxxx 1xxxxxxx
Codeset 1 is JIS code with the hit bits set; Gaiji is a user-defined
character set.