Re: lynx-dev Display of SGML Greek Math entities -- solution

[Jungshik Shin]

Steve White wrote:

> I just looked at your page again, and see the changes you made
>         http://www.irm.nara.kindai.ac.jp/lynxdev/greek.html
>
> What I see is,
>
>         Raw EUC-JP code is displayed as Greek letters
>         SGML character entities are transliterated into Latin

  I got the same result under other Japanese terminal emulators
(gnome-terminal launched under ja_JP.eucJP locale, luit launched
under ja_JP.eucJP locale that is run inside UTF-8 xterm, kterm,
rxvt ).  That's exactly what I expected because lynx does NOT have
any ability to map Unicode/ISO 10646 characters to characters in CJK
(multibyte) legacy encodings as I wrote in my message kindly forwarded
by Thomas (http://www.flora.org/lynx-dev/html/month082002/msg00038.html).
Lynx CAN map Unicode/ISO 10646 characters to single byte legacy encodings
(ISO-8859-x, windows-12xx, KOI8-R/U), but it CANNOT do that for multibyte
legacy encodings. That's why I suggested Lynx makes use of iconv(3)
for conversions between various encodings.


> This is
>         kterm -km euc
>         lynx -display_charset=euc-jp
>
> It's still a mystery.  It's clear it is something lynx is failing to do,
> but why does it fail for me, but not for you?

  It's really strange that Henry's screenshot showed Greek characters
represented both in EUC-JP encoding and in NCRs as Greek characters
(i.e. without transliteration). At first, I thought the conversion
between Unicode and CJK legacy encodings might have been added to the
newest snapshot of Lynx. I downloaded and compiled it, but the result
was the same. Henry, could you tell me what you did we don't know about
to make that happen?

  Jungshik


; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden