Customizing MARC::Errorchecks

[Shelley Doljack]

Hi,

I'm playing around with using MARC::Errorchecks for reviewing ebook records we get from vendors. I want to make some modifications to the module, but I find that if I do so in a similar manner described in the tutorial for customizing MARC::Lint, by making a subclass of the module, it doesn't work. Is this not possible with Errorchecks?

Thanks,
Shelley

----
Shelley Doljack
E-Resources Metadata Librarian
Metadata and Library Systems
Stanford University Libraries
sdol...@stanford.edu
650-725-0167

[Bryan Baldus]

On Tuesday, July 17, 2012 4:34 PM, Shelley Doljack [sdol...@stanford.edu] wrote:
>I'm playing around with using MARC::Errorchecks for reviewing ebook records we get from vendors. I want to make some modifications to the module, but I find that if I do so in a similar manner described in the tutorial for customizing MARC::Lint, by making a subclass of the module, it doesn't work. Is this not possible with Errorchecks?

Indeed, MARC::Errorchecks was not written in the object-oriented style that MARC::Lint uses. Skimming through the code just now (I've not worked with it as regularly as I might like to be able to keep it fresh in my memory), I believe it is essentially a collection of subs with a wrapper sub to call each check--check_all_subs() calls each of the checking subroutines and returns the arrayref of any errors found. When I wrote it I was still early in learning Perl (and while I've gotten better since then, lack of recent practice working with it hasn't necessarily improved my knowledge of the language), so I'm sure it's not the most optimized code possible. check_all_subs() and the POD comments could serve as an index to each of the checks, with the SYNOPSIS showing examples of how to call the individual checks.

That said, if you have ideas for additions or changes, or other questions, I welcome hearing about them, either to add to the base module or to help with creating a related module of your own. I do know that I need to get working on the changes required for RDA records, but haven't really even started looking into the challenges those will pose (though that will likely result in a new module or more devoted just to RDA, and will also likely require changes/subclasses to MARC::Lint and MARC::Lintadditions).

Also of note, I have a newer version I've just uploaded to CPAN [1] with the following changes (in addition to those listed below, I plan on removing MARC::Lint::CodeData from the Errorchecks distribution and then requiring MARC::Lint, which includes CodeData (to hopefully resolve issues with installing both module packages at the same time due to this file):

Version 1.16: Updated May 16-Nov. 14, 2011. Released 7-17-2012.
-Removed MARC::Lint::CodeData and require MARC::Lint
-Turned off check_fieldlength($record) in check_all_subs()
-Turned off checking of floating hyphens in 520 fields in findfloatinghyphens($record)
-Updated validate008 subs (and 006) related to 008/24-27 (Books and Continuing Resources) for MARC Update no. 10, Oct. 2009 and Update no. 11, 2010; no. 12, Oct. 2010; and no. 13, Sept. 2011.
-Updated %ldrbytes with leader/18 'c' and redefinition of 'i' per MARC Update no. 12, Oct. 2010.

Version 1.15: Updated June 24-August 16, 2009. Released , 2009.

-Updated checks related to 300 to better account for electronic resources.
-Revised wording in validate008($field008, $mattype, $biblvl) language code (008/35-37) for ' '/zxx.
-Updated validate008 subs (and 006) related to 008/24-27 (Books and Continuing Resources) for MARC Update no. 9, Oct. 2008.
-Updated validate008 sub (and 006) for Books byte 33, Literary form, invalidating code 'c' and referring it to 008/24-27 value 'c' .
-Updated video007vs300vs538($record) to allow Blu-ray in 538 and 's' in 07/04.

[1] While the CPAN indexer works on that: <http://www.cpan.org/authors/id/E/EI/EIJABB/MARC-Errorchecks-1.16.tar.gz>
, I've also posted the file to my website: <http://home.comcast.net/~eijabb/bryanmodules/MARC-Errorchecks-1.16.tar.gz>, with text versions of each file visible in:
<http://home.comcast.net/~eijabb/bryanmodules/MARC-Errorchecks-1.16>

#####################

Finally, I meant to mention it on this list earlier, but I've posted a new version of MARC::Lint, 1.45, to CPAN [2], with the current development version (as of now, same as CPAN's version), in SourceForge's Git repository [3]. Updates to that module include:
- Updated Lint::DATA section with Update No. 10 (Oct. 2009) through Update No. 14 (Apr. 2012)
- Updated _check_article with the exceptions: 'A & ', 'L is '

#####################

[2] <http://search.cpan.org/~eijabb/MARC-Lint-1.45/>
[3] <http://marcpm.git.sourceforge.net/git/gitweb.cgi?p=marcpm/marcpm;a=summary>

I hope this helps,

Bryan Baldus
Cataloger
Quality Books Inc.
1-800-323-4241x402
bryan....@quality-books.com
eij...@cpan.org
http://home.comcast.net/~eijabb/

[Shelley Doljack]

Hi Bryan,

Thanks for the information. I also downloaded Lintadditions. Right now, I'm trying to figure out what I want to do. I'm new to object-oriented programming in Perl, so it's a learning experience for me. What I think I need to do is create my own module for checks specific to ebook records, like checking that all records have an 856 field with 2nd indicator 0 or an electronic resource 007, among other things. I need to think about it more and play around with the different modules more.

Thanks,
Shelley

[Dobrowolsky, Amy]

Hi Shelley,

Just as another option: you don't necessarily need to write a new module to do this. I have a similar situation (verifying the presence of an 856 field(s) as well as other fields). I just use the MARC::File::USMARC module to read through each record in the .mrc file and look for the specific fields/indicators that I want.

/amy




Amy A. Dobrowolsky, Metadata Librarian, KnowledgeWorks Provider Data Acquisitions & Integration
Serials Solutions | 501 N. 34th Street, Ste. 300 | Seattle, WA 98103 USA | 1.866.737.4257 extension 1038
Amy.Dob...@serialssolutions.com
www.serialssolutions.com