prints out PubMedCentra IDs and their associated license URL (if one is found in the article XML)

pmc-licenses.py

#!/usr/bin/env python

"""
Script to go through all the OAI-PMH records in the PubMedCentral database and
print out a tab delimited list of record identifiers and a license url (if one
is included).

You'll need lxml installed to run this.
"""


from lxml import etree

# namespaces we'll use
ns = {'oai': 'http://www.openarchives.org/OAI/2.0/',
      'nlm': 'http://dtd.nlm.nih.gov/2.0/xsd/archivearticle',
      'xlink': 'http://www.w3.org/1999/xlink'}

# initial url, for the first set of records
url = 'http://www.pubmedcentral.nih.gov/oai/oai.cgi?verb=ListRecords&metadataPrefix=pmc&set=pmc-open'


while True:
    doc = etree.parse(url)

    # iterate through each record, printing out the id and license (if there)
    for record in doc.xpath('oai:ListRecords/oai:record', namespaces=ns):
        id = record.xpath('string(oai:header/oai:identifier)', namespaces=ns)
        license = record.xpath('string(oai:ListRecords/oai:record/oai:metadata/nlm:article/nlm:front/nlm:article-meta/nlm:permissions/nlm:license/@xlink:href)', namespaces=ns)
        print "%s\t%s" % (id, license)

    # use resumption token to construct the url for the next bunch of records
    t = doc.xpath('string(oai:ListRecords/oai:resumptionToken)', namespaces=ns)
    if not t:
        break
    url = "http://www.pubmedcentral.nih.gov/oai/oai.cgi?verb=ListRecords&resumptionToken=%s" % t