MARC::File::XML 1.0

[Galen Charlton]

Hi,

I have uploaded version 1.0 of MARC::File::XML to CPAN. Changes from
version 0.93 are:

1.0 Sun Feb 10 21:32:32 PST 2013
- Use XML::LibXML rather than XML::SAX to parse MARCXML records.
This offers a significant speed improvement for record parsing;
in some tests, parsing time on a large file was reduced by 80%.
This also avoids problems caused by character encoding bugs in
XML::SAX::PurePerl.
- Add method to let client code give MARC::File::XML a shared
XML::LibXML parser object.

As you can see, this is a major change from the XML::SAX days, and I would
appreciate feedback, particularly if folks encounter porting problems.

Also, if there is anybody whose code depends on MARC::File::XML using SAX
-- presumably only if you were subclassing MARC::File::SAX -- please let me
know.

Regards,

Galen
--
Galen Charlton
gmch...@gmail.com

[Jonathan Druart]

Hi Gene,

The error means $tag is not defined.
You should check your code where you call the field method.
Could you provide a minimal bug test case to show us your problem?
Note that the most recent version is 2.0.6.

Regards,
Jonathan

2014-11-05 21:00 GMT+01:00 Gene Damon <gda...@vccs.edu>:
> I haven't commented on this list for a long time, but hit a problem today.
> I upgraded to the most recent? version of (MARC-Record-2.0.5) and using
> programs that haven't changed in a long time, but run everyday on an older
> version, I'm getting the following message.
>
> Use of uninitialized value in hash element at . . ./Record.pm line 202.
> The code at that spot is as following.
>
> for my $tag ( @specs ) {
> my $regex = $field_regex{ $tag };
>
> Before I start digging around, can someone tell me what might have changed
> in this version that would tell me why the message?
>
>
>> --
>> Gene Damon
>> Director of Library Automation &
>> Learning Resources
>> Virginia Community College System
>> (804) 819-4981 (voice)
>> (804) 819-4773 (fax)
>>
>> Address:
>> James Monroe Bldg.
>> 15th Floor
>> 101 N. 14th St.
>> Richmond, VA 23219

[Gene Damon]

Thanks that was the hint I needed to find the problem. Must be some
changes to the code that detects that problem as there was a bug in one
sequence that's been there for a very long time. Works fine now. And I
do have 2.0.6. Fat fingers I guess.