Note: You can specify the default character set by both command line argument and X resources.
Note:
When you set Mosaic*simpleInterface
resource
(see here)
to True
, localization facilities cannot be used.
As of today, to best of my knowledge,
strict interpretation of this request header field is not determined yet,
but use of a language code, which is an
ISO 639 language code
with an optional
ISO 3166 country code
to specify a national variant, is encourage now.
For example, en_UK
means that the content of the message is in British English,
while en
means that the language is English in one of its forms
(for technical details, see
here).
A few sample for this language preference feature is provided. Try these URL
http://palomine.nttam.com/mldocs/hello.html http://palomine.nttam.com/mldocs/radio-j.html http://palomine.nttam.com/Mule/FAQ.txtwith the following language code. If there is no document in the language you specified, English version is returned.
en,ja,ru,el,fr,de,es,it,in,sw,ms,sv,ko,iw,zh, zh_CN,zh_TW,zh_HK(BTW, do you know what languages do these codes mean completely? The answer is here :-)
Furthermore, you can specify a list of preferred languages like this:
xx; yy; de; zzIn this case, we have no documents in language such like
xx
, yy
and zz
;
so you get German. Note: You can specify the default accept languages by both command line argument and X resources.
Visual
or Implicit
.
A sample for this bi-directionality feature is provided.
Try these two URL.
http://www.ntt.com/Mosaic-l10n/hebrew-visual.html http://www.ntt.com/Mosaic-l10n/hebrew-implicit.htmlNote: Currently, implementation of
Implicit
directionality is still
immature. Explicit
directionality is not supported yet.
For technical details, see RFC-1556
. Note: You can specify the default bi-directionality by both command line argument and X resources.
ISO-2022-JP (RFC-1468)
or
ISO-2022-KR (RFC-1557)
are automatically displayed by
Japanese/Korean fonts without font setting via Font Menu. However, documents in ISO 8859-X, EUC-C/J/K, KOI and Big5 cannot be judged what character set is used without external information. This enhancement supports the following ISO 2022 initial designation sequences. (Note: We will always use G0 and G1 for GL and GR, respectively.)
"<ESC> - A" designate right-hand part of ISO 8859-1 into G1 "<ESC> - B" designate right-hand part of ISO 8859-2 into G1 "<ESC> - C" designate right-hand part of ISO 8859-3 into G1 "<ESC> - D" designate right-hand part of ISO 8859-4 into G1 "<ESC> - L" designate right-hand part of ISO 8859-5 into G1 "<ESC> - F" designate right-hand part of ISO 8859-7 into G1 "<ESC> - H" designate right-hand part of ISO 8859-8 into G1 "<ESC> - M" designate right-hand part of ISO 8859-9 into G1 "<ESC> $ ) A" designate GB 2312-1980 into G1 "<ESC> $ ) B" designate JIS X 0208-1983 into G1 "<ESC> $ ) C" designate KSC 5601-1987 into G1 "<ESC> ( B" designate 7-bit ASCII graphics into G0 "<ESC> $ B" designate JIS X 0208-1983 into G0If a document includes the one of these escape sequences, the document is displayed by appropriate fonts without font setting via Font Menu. These examples
are encoded in this way. Please check it.
Note: We also have a little note on Japanese encoding methods and a related problem. It would be helpful in considering handling multi-byte characters in WWW. Please check out here.
Note:
L10N-enhanced Mosaic remembers a specified character set of each
document in the Window History (not Hotlist),
and if you revisit the document, the current character set is
automatically changed.
If you don't like this, set Mosaic*keepDocumentCharset
resource to False
(see here).