Transform CSV data from DOAJ into MARC suitable for ingestion into Koha

doaj2koha.sh

#!/bin/bash

# doaj2koha.sh
# 
# Get the DOAJ (http://www.doaj.org/) data in CSV format here:
# http://www.doaj.org/doaj?func=csv
# Save the data as doaj.csv
#
# If you are only interested in some of the journals in DOAJ you
# can probably use the grep command to extract only the lines that
# contain words that are relevant to you. Make sure you keep the 
# first line with the column definiteions, though! 
# 
# Get csvtomarc.pl and csvutils.pm from here: 
# http://git.catalyst.net.nz/gw?p=koha.git;a=tree;f=import/csv;h=82facb6056faf0d70cab1264c121b67ef4c00c1f;hb=refs/heads/import_branch
# 
# Put all of these files in the same directory as this script. 
# 
# Adjust these values:
KOHACONF=/home/magnus/sites/kohanor32-dev/etc/koha-conf.xml
KOHALIBS=/home/magnus/scripts/kohanor32/
# 
# Run as ./doaj2koha.sh and you will get the MARC records in a file
# called out.mrc
# 
# FIXME
# There is a problem with encoding which results in some garbled chars

perl ./csvtomarc.pl \
  --kohaconf $KOHACONF \
  --kohalibs $KOHALIBS \
  -m 'Title=title' \
  -m 'ISSN=biblioitems.issn?' \
  -m 'EISSN=biblioitems.issn?' \
  -m 'Title Alternative=marc:245_b?' \
  -m 'Country=marc:260_a?' \
  -m 'Publisher=marc:260_b?' \
  -m 'Further Information=notes?' \
  -m 'func:prefix:Language(s): :Language=notes?' \
  -m 'Subjects=marc:653_a?' \
  -m 'Keyword=marc:653_a?' \
  -m 'Identifier=biblioitems.url!' \
  -o out.mrc \
  -i doaj.csv \
  --format usmarc 

# TODO
# Columns that still need to be mapped:
# Start Year
# End Year
# 
# This did not work:
# -m 'func:combine:append::Start Year:append: - :append::End Year=notes?' \