By the way - if you are wondering how necessary the special tweaks
in the lynx code for UTF-8 are, or to just test how lynx behaves
on UTF-8 text in an UTF-8 display without those tweaks - say for
testing an UTF-8 aware ncurses - :
View UTF-8 document with display character set == Transparent.
This foregoes any special handling for UTF-8 in the display engine.
This won't work if the characters are given as NCRs (&#nnn;) or
entities, since lynx will just show those literally with "Transparent".
But it works if the characters are in the source in actual UTF-8 form.
To convert the former[*] to the latter, the following procedure happens to
mostly[*] work"
1. Visit page with 'lynx -preparsed', display character set = UTF-8
(make sure document charset is properly recognized as what it is)
2. Switch to (preparsed) source mode with '\'.
3. 'P'rint, Save to a local file
4. The local file should now have raw UTF-8 characters.
[*] Actually, this works not just for texts that already have an utf-8
charset - any charset can be converted to raw UTF-8 this way; try
iso8859*.html files from lynx's test directory, compare all three columns.
[**] for characters in element contents, not in attribute values like
ALT text.
Klaus
; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to [EMAIL PROTECTED]
