dmd/flatten: Support more than MARC 533$a

crkn-rcdr / cihm-metadatabus

Documentation and Docker build environment for key portions of the metadata bus.

BSD 2-Clause "Simplified" License

0 stars 1 forks source link

dmd/flatten: Support more than MARC 533$a #68

Closed RussellMcOrmond closed 1 year ago

RussellMcOrmond commented 1 year ago

This is: 533 - Reproduction Note https://www.loc.gov/marc/bibliographic/bd533.html

https://github.com/crkn-rcdr/cihm-metadatabus/blob/afb41cc2c71be131ed276547f11f2c85d9d12a45/CIHM-Meta/lib/CIHM/Meta/dmd/flatten.pm#L264-L269

RussellMcOrmond commented 1 year ago

Example @nataliemacdonald gave for https://www.canadiana.ca/view/oocihm.8_00559 , indicating that the no_source field should read "Electronic resource: Ottawa : Canadiana.ca, 2017."

  <datafield tag="533" ind1=" " ind2=" ">
    <subfield code="a">Electronic resource:</subfield>
    <subfield code="b">Ottawa :</subfield>
    <subfield code="c">Canadiana.org,</subfield>
    <subfield code="d">2017.</subfield>
  </datafield>

RussellMcOrmond commented 1 year ago

Questions for @nataliemacdonald ,

Do we need to scan existing MARC records to look for examples where $b, $c, or $d are used to ensure that we aren't exposing anything that wasn't recorded correctly, or can be assume all MARC records currently stored would meet the new criteria?
Are there other 533 subfields that should be supported, and can you offer suggestions of how they should be displayed as well as MARC sample files?

RussellMcOrmond commented 1 year ago

Full record as XML (Use MarcEdit to convert back to a MARC binary file for testing with uploader).

oocihm.8_00559-MARC.xml.txt

RussellMcOrmond commented 1 year ago

I'm authoring a tool to scan all existing MARC records for there being more than a subfield 'a', and then output the concatenated record as a file for someone in Heritage Services to scan.

It isn't only "new" records that have to work with any change to the MAP, but all existing records.

https://metacpan.org/pod/MARC::Field#as_string(-[$subfields]-[,-$delimiter]-)

By default all subfields will be used in the XML order with a space between them. We may want to do something different, and only include a subset of subfields in the record.

Possibly only 'abcd' or 'abcdef'?

RussellMcOrmond commented 1 year ago

The tool skips any that only have an "a" subfield, and thus would be the same before and after any MAP change.

russell@eclipse:/media/crkn-nas-wip$ wc -l  _Metadata_Synchronised/analysis/MARC533lists/*
       2 _Metadata_Synchronised/analysis/MARC533lists/access-ams-MARC533.csv
       6 _Metadata_Synchronised/analysis/MARC533lists/access-omcn-MARC533.csv
   89035 _Metadata_Synchronised/analysis/MARC533lists/access-oocihm-MARC533.csv
      95 _Metadata_Synchronised/analysis/MARC533lists/access-oocihm.N-MARC533.csv
      12 _Metadata_Synchronised/analysis/MARC533lists/access-ooe-MARC533.csv
       5 _Metadata_Synchronised/analysis/MARC533lists/access-oop-MARC533.csv
       2 _Metadata_Synchronised/analysis/MARC533lists/access-qmma-MARC533.csv
   89157 total
russell@eclipse:/media/crkn-nas-wip$