On Thu, Nov 27, 2008 at 04:49:17PM +0100, Mario Salzer wrote:
> I've reverted back to Joe 3.5-1.1, because with 3.5-2 there was a
> regression regarding charset/encoding detection.
>
> Having still a few non-UTF8 text files over, I'm using the joerc
> option -guess_non_utf8. I didn't notice till now, but this feature
> has been working exceptionally well.
> With 3.5-2 it stopped.
>
> When opening a file in ISO-8859-1, joe 3.5-2 now assumes it was UTF-8
> from environment vars or the current locale. Its previous sanity check
> does no longer work. (And as far as I tested, the actual char/encoding
> of a file cannot be set via commandline arguments or LC_* vars any
> longer either.)
If you set the locale to ISO-8859-1, either implicitly or explicitly
(depending on your locale) it should open the file with that charset.
Can you paste the output of locale(1) in the shell where there's a problem?
Regarding the guess option, that is probably another bug, but we'll deal
with that with the pending joe 3.7, which also modifies this logic but in
a more predictable way.
--
2. That which causes joy or happiness.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
