| Subject: | bug / "not playing nice issue" with utf8 - MARC::Record v2.0.0 |
| Date: | Fri, 14 Dec 2007 12:40:40 +1100 |
| To: | bug-MARC-Record [...] rt.cpan.org |
| From: | Steven McPhillips <smcphillips [...] nla.gov.au> |
Hi,
I've got an issue with some marc records containing bad utf8 data. At
the moment, the MARC::Record module seems to barf at creation from
the MARC::Batch->next() call, reporting something like:
utf8 "\xE2" does not map to Unicode at /usr/local/perl/shared/perl/
5.8.5-09/lib/5.8.5/sun4-solaris/Encode.pm line 16
The problem I have here is that I can't catch this too easily, apart
from performing an eval{} around the MARC::Batch->next() call.
So I realise this isn't really a bug, but seeing as you have a
strict_off() subroutine for MARC::Batch, I was wondering if perhaps
you could fail a bit more gracefully in the event of decoding issues.
Thoughts?
my system details:
perl -v : 5.8.5
uname -a : SunOS meroo 5.9 Generic_118558-30 sun4u sparc SUNW,Sun-
Fire-880
----
Steven McPhillips <smcphillips@nla.gov.au>
IT Business Systems
National Library of Australia