create individual MARCXML files for archivists toolkit from a MARC21 file

marc2xml-at.py

"""creates individual MARCXML files for archivists toolkit from a MARC21 file"""

import pymarc
import os
import sys

header = u"""<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/xsl" href="MARC21slim2HTML.xsl" ?>
<collection xmlns="http://www.loc.gov/MARC21/slim"
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xsi:schemaLocation="http://www.loc.gov/MARC21/slim
    http://www.loc.gov/standards/marcxml/schema/MARC21slim.xsd">"""

reader = pymarc.MARCReader(file(sys.argv[1]))
if len(sys.argv) >= 4:
    tag = sys.argv[3]
else:
    tag = '090'

count = 0
for record in reader:
    count += 1
    if record[tag] is not None:
        recnum = record[tag].format_field().replace('\\',' ').strip()
    else:
        recnum = ''
    filename = os.path.join(sys.argv[2], '%08d-%s'.strip() + '.xml') % (count, recnum)
    fh = open(filename, 'w')
    out = "%s%s</collection>" % (header, pymarc.record_to_xml(record))
    fh.write(out.encode("utf-8"))
    fh.close
print '%s total records' % count