def search(query):
"""
Search against Worldcat Search API
"""
matches = []
wskey = os.environ['WSKEY']
s = SRURequest(wskey=wskey, query='srw.ti = "' + query + '"', maximumRecords='3')
o = s.get_response()
for r in pymarc_extract(o.data):
title = ''
control_num = 0
for f in r.get_fields():
if ( f.tag == '245' ):
title = f.format_field()
app.logger.info( title )
if ( f.tag == '001' ):
control_num = f.format_field()
app.logger.info(control_num)
matches.append({ "id": json.dumps(control_num), "name": json.dumps(title), "score": 100, "match": False, "type": [ {"id": "/entry/title", "name": "Entry Title"} ] })
return matches
def get_related(oclcnum):
bib = BibRequest(wskey=WSKEY, rec_num=oclcnum).get_response()
record = pymarc_extract(bib.data)
subjects = []
if len(record) == 0:
return 'record %s not found' % oclcnum
else:
# if record[0].author() is not None:
# subjects.append(record[0].author())
for s in record[0].subjects():
if s.indicator2 == '0':
subjects.append('"%s"' % s.format_field())
elif s.indicator2 == '7' and s.tag == '655':
subjects.append('"%s"' % s['a'])
# if len(subjects) > 5:
# subjects = subjects[0:5]
# print 'more than 5 subjects - truncating'
#print subjects
# 'not srw.no = oclcnum'
if len(subjects) > 0:
sru = SRURequest(wskey=WSKEY)
sru.args['query'] = '(srw.kw = %s) not srw.no = "%s"' % \
(" or srw.kw = ".join(subjects), oclcnum)
# sru.args['servicelevel'] = 'full'
sru.args['sortKeys'] = 'LibraryCount,,0 relevance'
#print sru.args['query']
results = pymarc_extract(sru.get_response().data)
results = [r['001'].format_field() for r in results]
out = []
for r in results:
c = CitationRequest(wskey=WSKEY, rec_num=r)
out.append(c.get_response().data)
return "".join(out)
else:
return None
def get_related(oclcnum):
bib = BibRequest(wskey=WSKEY, rec_num=oclcnum).get_response()
record = pymarc_extract(bib.data)
subjects = []
if len(record) == 0:
return 'record %s not found' % oclcnum
else:
# if record[0].author() is not None:
# subjects.append(record[0].author())
for s in record[0].subjects():
if s.indicator2 == '0':
subjects.append('"%s"' % s.format_field())
elif s.indicator2 == '7' and s.tag == '655':
subjects.append('"%s"' % s['a'])
# if len(subjects) > 5:
# subjects = subjects[0:5]
# print 'more than 5 subjects - truncating'
#print subjects
# 'not srw.no = oclcnum'
if len(subjects) > 0:
sru = SRURequest(wskey=WSKEY)
sru.args['query'] = '(srw.kw = %s) not srw.no = "%s"' % \
(" or srw.kw = ".join(subjects), oclcnum)
# sru.args['servicelevel'] = 'full'
sru.args['sortKeys'] = 'LibraryCount,,0 relevance'
#print sru.args['query']
results = pymarc_extract(sru.get_response().data)
results = [r['001'].format_field() for r in results]
out = []
for r in results:
c = CitationRequest(wskey=WSKEY, rec_num=r)
out.append(c.get_response().data)
return "".join(out)
else:
return None
def _sru_request(self, query, startNum = 1):
# wait a random amount of time
self._wait()
req = SRURequest(wskey = self.apikey)
req.args['query'] = query
req.args['maximumRecords'] = 50
if startNum > 1:
req.args['startRecord'] = startNum
try:
return req.get_response().data
except (urllib2.HTTPError, urllib2.URLError), errstr:
self._stopped = True
self.log("Error: %s" % errstr)
return None
def get(self):
query = self.request.get('query', 'srw.ti = "IISG"')
jsonp_callback = self.request.get('jsonp_callback')
maximumRecords = self.request.get('maximumRecords', 30)
wskey='You Need to stick your WS_KEY from OCLC in here'
s = SRURequest(wskey=wskey, query=query, maximumRecords=maximumRecords)
o = s.get_response()
self.response.headers['Content-Type'] = 'application/json'
buf = []
for r in pymarc_extract(o.data):
rr = {}
for f in r.get_fields():
rr[f.tag] = f.format_field()
buf.append(rr)
if jsonp_callback: self.response.out.write('%s(' % jsonp_callback)
self.response.out.write(simplejson.dumps(buf))
if jsonp_callback: self.response.out.write(')')
def generate_srurequest(self, query):
'''
Generates the SRURequest used to query OCLC API.
Most of the values are static, except for query.
'''
bib_rec = SRURequest(wskey=WSKEY,
sortKeys=sortkeys,
recordSchema=recordschema,
recordPacking=recordpacking,
servicelevel=servicelevel,
query=query,
startRecord=1,
maximumRecords=MAX_RECORDS,
frbrGrouping=frbrgrouping)
return bib_rec
libs='ALL'
lcabsNotHeld.append(makeItemList(results,libs,item[1]))
else: #Honnold holdings-ish
libs = 'HDC-ish'
lcabsHeld.append(makeItemList(results,libs,item[1]))
return lcabsHeld, lcabsNotHeld
if __name__ == "__main__":
config = getYAMLConfig(cfgFileName) #read in config values
WSKEY = config['Auth']['WSKEY_SEARCH']
LIBS = config['Config']['LIBS']
SVCLVL = config['Config']['SVCLVL']
SRUELEM = config['Config']['SRUELEM']
MARCCODES = config['Config']['MARCCODES']
lCodes=[]
sru = SRURequest(wskey=WSKEY) # Set up the worldcat SRU request object
sru.args['servicelevel'] = SVCLVL
sru.args['FRBRgrouping'] = FRBRgrouping
sru.url = 'http://www.worldcat.org/webservices/catalog/search/worldcat/sru'
fileIn = sys.argv[1]
fileOut = sys.argv[2]
csvHdr = ['Library','ISBN']+MARCCODES
with open(fileIn, 'r') as cabCodes , open(fileOut, 'w') as bibsOut:
csvOut = csv.writer(bibsOut, quoting=csv.QUOTE_NONNUMERIC)# set up a csv file
csvOut.writerow(csvHdr)
bList = codesList(cabCodes) # get the codes from the file
lCodes = bList.listed() #a de-duped list of the codes in the file
matches, nonmatches = search(lCodes) # look them all up by ISBN, return worldcat bib data
for row in matches: # write the direct matches (on ISBN) to a csv file
csvOut.writerow(row)
titlematches, nontitlematches = search245(nonmatches) #look for title/author matches
