& XSLT Thanks Grant....that fixed it.

The database originally started with a DOS application and still encodes 
'£' symbols weirdly, so my other (non-Perl) Windows application was 
extracting the data into XML and adding the correct '£' symbol, but I 
was not specifying the encoding as you pointed out.

If I specify "WINDOWS-1252" or "ISO-8859-1" my XML::Twig code can parse 
it without a problem. I'll probably use the former because I suspect 
there are Euro symbols in the data somewhere.

Much appreciated
-- 
Neil Hughes


On 27/6/09 22:39, Grant McLean wrote:
>  Hi Neil
>  
>  You need to determine what encoding has been used by the database export
>  process.
>  
>  If you're working with a Windows system then the most likely guess is
>  that the data is encoded with CP1252 or 'Win-Latin-1'.  In which case
>  the first line of the XML file should specify an encoding like this:
>  
>    <?xml version="1.0" encoding="WINDOWS-1252" ?>
>  
>  If the data was encoded with UTF-8 then the XML parser module would have
>  recognised it automatically, so you can eliminate that option.
>  
>  You can also safely assume that the data is not in ISO-8859-1 (Latin-1),
>  because that encoding pre-dates the definition of the Euro symbol.
>  
>  Encodings in XML (and Perl) are a largish subject with many subtleties,
>  you can read more here:
>  
>    http://perl-xml.sourceforge.net/faq/#encodings
>  
>  Cheers
>  Grant
>  
>  On Sat, 2009-06-27 at 22:24 +0100, Neil Hughes wrote:
> > I've hit a problem in XML::Twig trying to handle data exported from a 
> > legacy database, but I suspect this is an issue I need to get some 
> > advice on regardless of the parser...
> >
> > The data contains '£' symbols which I'm struggling to format in XML for 
> > processing later on. The following code might help explain:
> >
> > ------------ BEGIN --------------
> >
> > use strict;
> > use warnings;
> >
> > use XML::Twig;
> >
> > my $t= XML::Twig->new();
> >
> > # this is OK
> > #my $input = '<?xml 
> > version="1.0"?><root><item>one</item><item>two</item><item>three</item></root>'; 
> >
> >
> > # this is invalid
> > #my $input = '<?xml version="1.0"?><root><item>one 
> > £</item><item>two</item><item>three</item></root>';
> >
> > # this is OK
> > #my $input = '<?xml 
> > version="1.0"?><root><item><![CDATA[one]]></item><item><![CDATA[two]]></item><item><![CDA
TA[three]]> </item></root>'; 
> >
> >
> > # this is invalid
> > my $input = '<?xml version="1.0"?><root><item><![CDATA[one 
> > £]]></item><item><![CDATA[two]]></item><item><![CDATA[three]]></item></root>'; 
> >
> >
> > $t->parse($input);
> > $t->print;
> >
> > ------------ END --------------
> >
> > Whether I wrap my text data in CDATA or not, as soon as I include a 
> > pound sterling symbol I get the following error:
> >
> > not well-formed (invalid token) at line 1, column 46, byte 46 at 
> > /usr/local/ActivePerl-5.8/lib/XML/Parser.pm line 187
> >   at /Users/nkh/Documents/Dev/Perl/xml_twig/pound_test1.pl line 14
> >
> > Byte 46 seems to align with the '£', so I'm wondering what I need to do 
> > to get this character not to break the parser.
> >
>  
>  _______________________________________________
>  Perl-XML mailing list
>  Perl-XML@[...].com
>  To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs

_______________________________________________
Perl-XML mailing list
Perl-XML@[...].com
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs