[Koha-devel] a charmed letter about characters....

[Ignacio Javier]

Hi all, I'm stuck on a **madness bubble**, in Koha 3 i18n.

I've trying to import ISO records that use the upper part of 8 bit
ASCII(latin1), and I've found that drives to an important warning:

[Thu Jan 31 18:28:46 2008] import.pl: no mapping found for [0xCD] at
position 23 in / Manuel Murguía; MURGUÍA, Manuel; g0=ASCII_DEFAULT
g1=EXTENDED_LATIN at /usr/local/share/perl/5.8.8/MARC/Charset.pm line 210,
<fh000011201706011.mrc> line 1.

...that result in not recognition of latin characters when imported

Seems that g0 and g1 are reckon OK, the input encoding of the MARC record is
in 8-bit ASCII extended, but the imported record is not imported
correctly!!!

In Koha.pm is the butter:

sub FixEncoding {
  my $marc=shift;
  my $record = MARC::Record->new_from_usmarc($marc); #HERE<<<********

We are using MARC21.

The rare part is the MARC::Charset issue, [0xCD] is latin1 OK, where is the
problem?

USMARC has not mapping for extended latin 1 chars?

What perl module::method should I use for decoding MARC?

Thanks in advance!

Ignacio Javier Gómez Rodríguez
Analista - Programador
Tfno: 902905590 - Fax:  981571425
address@hidden
www.coremain.com