[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev bad behavior: changing assumed document charset

From: Leonid Pauzner
Subject: Re: lynx-dev bad behavior: changing assumed document charset
Date: Tue, 29 Feb 2000 14:54:56 +0300 (MSK)

27-Feb-2000 20:55 Ivan Zakharyaschev wrote:
>       Hello!

> I use Lynx Version 2.8.3dev.14 and I don't like the following behavior of
> it when reading Russian documents:

> I have these parameters pre-set:
> display character set: koi8-r
> assumed document charset: koi8-r

To read russian texts try
  assumed document charset: windows-1251
  preferred charset: windows-1251, *;q=0.5

this will made all windows-1251 texts readable for lynx
and (few) koi8-r texts will be readable by pressing `@' (raw mode for
your display character set). Such a behaviour is preferrable
since a majority of WWW visitors use Microsoft platform,
so texts in koi8-r usually set charset= attribute properly
and automatically displayed on reasonable modern browsers
without a problem.

> Then I browse to a page written in another (windows-1251) charset, and, of
> course, the browser shows some rave. Then go to the options screen, change
> the assumed document charset to the appropriate one and accept the
> changes. After this the document is reloaded and I can read it (it is
> displayed correctly). So far everything seems to be all right.

> But here comes the interesting thing: when I open another page with the
> same charset (win-1251), for instance, following a link from the original
> page (that was also in win-1251), I get again the document displayed in a
> wrong way, still the options remain unchanged (one can see it going to the
> options screen - the document charset is still the second one - win-1251).
> Just accepting the options another time doesn't help anymore. In order to
> tell the browser that the encoding of the document is win-1251, I have to,
> first, to change it to some other value, to accept this change, and then,
> to revert to win-1251 and accept it another time. Now - after the third
> loading, the document looks pretty good.
Could you write precise examples with URLs, I have not seen the problem
you described above. Perhaps you deal with some cached data that was not
flashed properly (probably a remote proxy or lynx HText cache) - please
give real life examples.

> Saving the options to disk doesn't make any difference.
yes, `assumed document charset' doesn't save to the disk from options
menu intentionally (read lynx.cfg for details).

> I wonder whether this behavior is normal, and if not - whether this bug
> has been fixed. Maybe my comment will be of some use for you.

> Regards
>       Ivan Z.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]