LYNX-DEV lynx and charsets

[Pavel Kasparek]

Hello everybody !

I have just joined this mailing list, because I have some troubles
with Lynx. 

We try to use Lynx for internet access for blind people. They have
usually only some DOS terminal program and modem and call to our Linux
machine, on which they run Lynx. They use some screen-reader, which
converts screen output to voice, so text output is much simpler than
graphics environment and Lynx is great for this.

The problem is with many versions of Czech language coding. The ususal
ones are ISO latin 2, PC latin 2, Windows 1250 and more others.
If I set Lynx charset to ISO Latin 2 and html page is in ISO Latin2,
everything is fine. But when I use other coding and I use some charset
in Lynx and even if I use Raw 8 bit mode, there are still some chars
that Lynx converts or doesn't display. I tried all charset in Lynx.

At last I addaed new charset in source code, I call it "no convesion",
and set it not to convert anything, and it works - it displays any char
that html page contains.

The second problem was when I tried to browse some pages that have the
charset=... in Content-type header. If the charset is ISO-8859-1 there
is no problem. If it is ISO-8859-2 and selected charset is ISO Latin 2
it's OK too. But if the chaset is for example Windows1250, Lynx doesn't
display the page and offers download. I had to rewrite souurce so that
if the selected charset is my new "no conversion" it doesn't care for
the charset and displays any page.

Is this the right way or can I do this some simpler way ?

-- 
Pavel Kasparek
email: address@hidden
Brailcom o.p.s., Vysehradska 3, Praha 2, Czech Republic, Europe
tel. 42-2-299151, www: http://www.brailcom.cz/
;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;