Re: Odd RE error at start up - what's the cause?

On Wed, Dec 05, 2007 at 06:59:04PM +0100, Urs Janßen wrote:
> On Wed, Dec 05, 2007 at 04:23:17PM +0000, Chris G wrote:
> > My hosting provider has is migrating my shell account from BSD to
> > Linux.
> 
> and from iso-8859-x to UTF-8
> 
Hadn't noticed that, but you're right of course!

> > I have most things transferred and working but there's an
> > annoying message when tin starts up:-
> > 
> >     Reading config file...
> >     tin 1.8.3 release 20070201 ("Scotasay") [UNIX] (c) Copyright 1991-2006 Iain Lea.
> >     Error in regex: invalid UTF-8 string at pos. 21 '.\(([Ww]a[rs]|[Bb]y[l³]o):.*\)\
>                                                        0123456789012345678901
>                                                                  1         2
> > So where is the expression at "pos. 21 "?  .... and why am I getting
> 
> it's the codepoint 0xb3 '³' (from iso-8859-2) which is no valid UTF-8 char,
> possible fixes are: convert the tinrc into UTF-8 via
> 
> mv ${TIN_HOMEDIR:-"$HOME"}/.tin/tinrc /tmp/tinrc.bak &&
> iconv -f iso-8859-1 -t utf-8 < /tmp/tinrc.bak > ${TIN_HOMEDIR:-"$HOME"}/.tin/tinrc
> 
> or fix the regexp and ommit the polish "bylo" part:
> strip_was_regex=.\((?:[Ww]a[rs]):.*\)\s*$
> 
Thank you, I now have to decide which is the easiest/tidiest way to
fix it.  :-)

-- 
Chris Green