Taking WOF (who's on first) concordance, and running against Wikidata SPARQL endpoint for new identifiers. (christina just playing around)

WOF2Wikidata.py

import csv
import rdflib
from SPARQLWrapper import SPARQLWrapper, JSON
import requests


def sparqlWD(label):
    sparql = SPARQLWrapper("https://query.wikidata.org/sparql")
    sparql.setQuery("""
        PREFIX wikibase: <http://wikiba.se/ontology#>
        PREFIX wd: <http://www.wikidata.org/entity/>
        PREFIX wdt: <http://www.wikidata.org/prop/direct/>
        PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
        SELECT ?wdURI ?loc ?fast ?psh ?dmoz ?mesh ?gnd WHERE {
          ?wdURI wdt:P1566 '""" + label + """' .
          OPTIONAL { ?wdURI wdt:P244 ?loc . }
          OPTIONAL { ?wdURI wdt:P2163 ?fast . }
          OPTIONAL { ?wdURI wdt:P1051 ?psh . }
          OPTIONAL { ?wdURI wdt:P998 ?dmoz . }
          OPTIONAL { ?wdURI wdt:P486 ?mesh . }
          OPTIONAL { ?wdURI wdt:P227 ?gnd . }
        }
    """)
    sparql.setReturnFormat(JSON)
    res = sparql.query().convert()
    if len(res['results']['bindings']) > 0:
        return(res)


def main():
    WOFconc = requests.get('https://raw.githubusercontent.com/whosonfirst/whosonfirst-data/master/meta/wof-concordances-latest.csv')
    with open('WOFconcordance.csv', 'wb') as WOFcsv:
        WOFcsv.write(WOFconc.text.encode('utf-8'))
    WOFreader = csv.DictReader(open('WOFconcordance.csv'))
    fn = WOFreader.fieldnames
    fn.extend(('fst:id', 'dmoz:id', 'gnd:id', 'mesh:id', 'psh:id'))
    WOFwriter = csv.DictWriter(open('WOFconcordanceUpdated.csv', 'wb'),
                               fieldnames=fn)
    WOFwriter.writeheader()
    start = added = 0
    for row in WOFreader:
        start += 1
        row['fst:id'] = row['psh:id'] = row['dmoz:id'] = ''
        row['mesh:id'] = row['gnd:id'] = ''
        resp = sparqlWD(row["gn:id"])
        if resp:
            if 'fast' in resp['results']['bindings'][0] and not row['fst:id']:
                row['fst:id'] = resp['results']['bindings'][0]['fast']['value']
                added += 1
            if 'dmoz' in resp['results']['bindings'][0] and not row['dmoz:id']:
                row['dmoz:id'] = resp['results']['bindings'][0]['dmoz']['value'].encode('utf-8')
                added += 1
            if 'gnd' in resp['results']['bindings'][0] and not row['gnd:id']:
                row['gnd:id'] = resp['results']['bindings'][0]['gnd']['value']
                added += 1
            if 'loc' in resp['results']['bindings'][0] and not row['loc:id']:
                row['loc:id'] = resp['results']['bindings'][0]['loc']['value']
                added += 1
            if 'mesh' in resp['results']['bindings'][0] and not row['mesh:id']:
                row['mesh:id'] = resp['results']['bindings'][0]['mesh']['value']
                added += 1
            if 'psh' in resp['results']['bindings'][0] and not row['psh:id']:
                row['psh:id'] = resp['results']['bindings'][0]['psh']['value']
                added += 1
            if 'wdURI' in resp['results']['bindings'][0] and not row['wd:id']:
                row['wd:id'] = resp['results']['bindings'][0]['wdURI']['value']
                added += 1
        WOFwriter.writerow(row)
    print("Number of records: " + str(start) + " and added identifiers: " +
          str(added))


if __name__ == "__main__":
    main()