def z3950_query(target=None,
keyword=None,
qualifier='(1,1016)',
query_type='CCL'):
if target is not None:
host = target['host']
database = target['database']
port = target['port']
syntax = target['syntax']
user = target['user']
password = target['password']
try:
if user is not None \
and password is not None:
conn = zoom.Connection(host,
port,
user=user,
password=password)
else:
conn = zoom.Connection(host, port)
conn.databaseName = database
conn.preferredRecordSyntax = syntax
query_str = qualifier + '=' + keyword
query = zoom.Query(query_type, query_str)
res = conn.search(query)
return True, res
except zoom.ConnectionError:
raise
else:
raise ValueError('Z3950 target not provided.')
def run ():
conn = zoom.Connection ('amicus.nlc-bnc.ca', 210)
conn.databaseName = 'NL'
q = zoom.Query ('CCL', 'ti="1066"')
ss = conn.scan (q)
for s in ss[0:10]:
print s
def RunQuery():
"Run a Z39.50 query & save MARC results to files"
#open connection
conn = zoom.Connection ('z3950.loc.gov', 7090)
conn.databaseName = 'VOYAGER'
conn.preferredRecordSyntax = 'USMARC'
#setup query
query = zoom.Query('CCL', 'ti="1066 and all that"')
#run query
res = conn.search(query)
#for each record in the resultset, save as file
ifilecount = 0
for r in res:
sSaveAs = os.path.join(tempfile.gettempdir(),
"PyZ3950 search resultset %d.bin" % ifilecount)
print("Saving as file:", sSaveAs)
fx = open(sSaveAs, "wb")
fx.write(r.data)
fx.close()
ifilecount += 1
#parse each record as we save
ParseRecord(sSaveAs)
#close connection
conn.close()
def run(self):
if not parse_only:
self.conn = zoom.Connection('localhost', 2100)
self.conn.preferredRecordSyntax = 'SUTRS'
while 1:
self.count += 1
# if not (self.count % 100):
# print("Thread", currentThread (), "Count", self.count)
query_str = random.choice(self.queries)
try:
q = zoom.Query('CCL', query_str)
if not parse_only:
r = self.conn.search(q)
for rec in r:
self.consume(rec)
except zoom.Bib1Err as err:
pass
except zoom.QuerySyntaxError as e:
print("e", e, "q", query_str)
if self.count > 500:
if not parse_only:
self.conn.close()
# should randomly do clean vs. not clean exit
self.terminate_queue.put(self, 1)
break
def read_mc(sys_no):
"""
Loads marc data from aleph.unibas.ch for one single system number.
:param sys_no: System number to which the marc entry is to be loaded.
:return: marc binary for said system number.
"""
# print("reading: "+sys_no)
try:
conn = zoom.Connection('aleph.unibas.ch', 9909)
conn.databaseName = 'dsv05'
conn.preferredRecordSyntax = 'USMARC'
query = zoom.Query('PQF', '@attr 1=1032 ' + sys_no)
res = conn.search(query)
data = bytes(res[0].data)
except zoom.ConnectionError:
print("\n!!! Error: could not connect to aleph !!!\n")
return
__write_to_cache(data, sys_no)
reader = MARCReader(bytes(data), force_utf8=True, to_unicode=True)
tmp = next(reader)
# print("loaded data from aleph.")
return tmp
def gather_stage(self, harvest_job):
log = logging.getLogger(__name__ + '.WAF.gather')
log.debug('z3950Harvester gather_stage for job: %r', harvest_job)
self.harvest_job = harvest_job
# Get source URL
source_url = harvest_job.source.url
self._set_source_config(harvest_job.source.config)
# get current objects out of db
query = model.Session.query(HarvestObject.guid, HarvestObject.package_id).filter(HarvestObject.current==True).\
filter(HarvestObject.harvest_source_id==harvest_job.source.id)
guid_to_package_id = dict((res[0], res[1]) for res in query)
current_guids = set(guid_to_package_id.keys())
current_guids_in_harvest = set()
# Get contents
try:
conn = zoom.Connection(source_url,
int(self.source_config.get('port', 210)))
conn.databaseName = self.source_config.get('database', '')
conn.preferredRecordSyntax = 'XML'
conn.elementSetName = 'T'
query = zoom.Query('CCL', 'metadata')
res = conn.search(query)
ids = []
for num, result in enumerate(res):
hash = hashlib.md5(result.data).hexdigest()
if hash in current_guids:
current_guids_in_harvest.add(hash)
else:
obj = HarvestObject(
job=harvest_job,
guid=hash,
extras=[
HOExtra(key='status', value='new'),
HOExtra(key='original_document',
value=result.data.decode('latin-1')),
HOExtra(key='original_format', value='fgdc')
])
obj.save()
ids.append(obj.id)
for guid in (current_guids - current_guids_in_harvest):
obj = HarvestObject(
job=harvest_job,
guid=guid,
package_id=guid_to_package_id[guid],
extras=[HOExtra(key='status', value='delete')])
obj.save()
ids.append(obj.id)
return ids
except Exception, e:
self._save_gather_error('Unable to get content for URL: %s: %r' % \
(source_url, e),harvest_job)
return None
def run ():
conn = zoom.Connection ('amicus.nlc-bnc.ca', 210)
conn.databaseName = 'NL'
q = zoom.Query ('CCL', 'ti=A')
conn.numberOfEntries = 80
ss = conn.scan (q)
for i in range (len (ss)):
print ss.get_term (i), ss.get_fields (i)
def establishZ3950Connection(database_address, port, username, database_name):
try:
conn = zoom.Connection(database_address, port, user=username)
conn.databaseName = database_name
conn.preferredRecordSyntax = 'USMARC'
return conn
except zoom.ConnectionError:
print "GRACEFUL CLOSE ERROR -- ESTABLISHING CONNECTION"
waitSixSeconds(datetime.datetime.now().time())
return establishZ3950Connection(database_address, port, username,
database_name)
def _connect(self):
"""Return a connection to the configured database.
:returns: A connection.
"""
conn = zoom.Connection(self.host, self.port, user=self.user,
password=self.password)
conn.databaseName = self.db
conn.preferredRecordSyntax = self.syntax
conn.elementSetName = self.elem_set_name
return conn
def connect(self):
""" Connects to z3950 server.
Called by __init__() """
conn = zoom.Connection(
self.HOST,
int(self.PORT),
databaseName=self.DB_NAME,
preferredRecordSyntax=
u'OPAC', # Getting records in "opac" format. (Others were not more helpful.)
charset='utf-8')
log.debug('connection made.')
return conn
def __init__(self, host, port, db):
self.conn = zoom.Connection(host, port)
self.conn.databaseName = db
self.conn.elementSetName = 'F'
# self.conn.preferredRecordSyntax = 'XML'
# currently (2002-7-17) the dbiref.kub.nl seems broken for XML
# in two ways:
# 1) Much more data is returned for elementSetName 'B' than 'F'.
# 2) even for 'B', the term name for the top-level record isn't
# returned, but the term id is. The term name isn't marked as
# optional in the spec.
self.conn.preferredRecordSyntax = 'GRS-1'
def _make_connection(self):
"""
Returns a connection to the Z39.50 server
"""
# Create connection to database
connection = zoom.Connection(
self._host,
self._port,
charset=self._charset,
)
connection.databaseName = self._database
connection.preferredRecordSyntax = self._syntax
return connection
def get_marc(target_name, cclquery, result_offset):
target = targets[target_name]
m = re_identifier.match(target['identifier'])
(host, port, db) = m.groups()
port = int(port) if port else 210
conn = zoom.Connection (host, port)
if db:
conn.databaseName = db
conn.preferredRecordSyntax = 'USMARC'
query = zoom.Query ('PQF', cclquery)
res = conn.search (query)
offset = 0
for r in res:
return r.data
offset += 1
if offset == result_offset:
return r.data
def search(host, port, database, query, start=1, limit=10):
conn = zoom.Connection(host, port)
conn.databaseName = database
conn.preferredRecordSyntax = 'XML'
query = zoom.Query ('CCL', str(query))
res = conn.search (query)
collector = []
#if we were dealing with marc8 results, would probably need this
#m = zmarc.MARC8_to_Unicode ()
# how many to present? At most 10 for now.
to_show = min(len(res)-(start - 1), limit)
if limit:
to_show = min(to_show, limit)
#this seems to an efficient way of snagging the records
#would be good to cache the result set for iterative display
for r in range(start - 1,(start-1) + to_show):
#would need to translate marc8 records, evergreen doesn't need this
#collector.append(m.translate(r.data))
collector.append(str(res.__getitem__(r)).replace('\n',''))
conn.close ()
raw = "" . join(collector)
raw_records = []
err = None
pat = re.compile('<record .*?</record>', re.M)
raw_records = pat.findall(raw)
parsed = []
for rec in raw_records:
# TODO: fix this ascii/replace, once our z3950/marc encoding
# issues are sorted out.
rec = unicode(rec, 'ascii', 'replace')
# replace multiple 'unknown' characters with a single one.
rec = re.sub(u'\ufffd+', u'\ufffd', rec)
assert isinstance(rec, unicode) # this must be true.
parsed.append(ET.fromstring(rec.encode('utf-8')))
return parsed, len(res)
def run(self):
if not parse_only:
self.conn = zoom.Connection('localhost', 2100)
self.conn.preferredRecordSyntax = 'SUTRS'
while 1:
self.count += 1
# if not (self.count % 100):
# print "Thread", currentThread (), "Count", self.count
query_str = random.choice(self.queries)
try:
q = zoom.Query('CCL', query_str)
if not parse_only:
r = self.conn.search(q)
for rec in r:
self.consume(rec)
except zoom.Bib1Err, err:
pass
except zoom.QuerySyntaxError, e:
print "e", e, "q", query_str
def open_stream(self, stream):
server = stream.replace('z3950', 'https')
(transport, user, passwd, host, port, dirname, filename, args,
anchor) = self._parse_url(server)
conn = zoom.Connection(host, port)
conn.databaseName = dirname
q = args['query']
qo = zoom.Query('CQL', q)
if 'preferredRecordSyntax' in args:
conn.preferredRecordSyntax = args['preferredRecordSyntax']
else:
conn.preferredRecordSyntax = 'USMARC'
if 'elementSetName' in args:
conn.elementSetName = args['elementSetName']
else:
conn.elementSetName = 'F'
rs = conn.search(qo)
self.total = len(rs)
return rs
queryString = ''
if (isbn != None):
# isbn = replace (replace (isbn, "-", ""), " ", "");
queryString = 'isbn="' + isbn + '"'
# elif (field == "LCCN"):
# field = "lccn"
else:
queryString = 'ti="' + title + '"'
if (authors != ""):
queryString = queryString + ' and au="' + authors + '"'
conn = zoom.Connection('opac.sbn.it', 3950)
conn.databaseName = 'nopac'
conn.preferredRecordSyntax = 'USMARC'
query = zoom.Query('CCL', str(queryString))
query
doc = Document()
root = doc.createElement("importedData")
doc.appendChild(root)
collection = doc.createElement("List")
collection.setAttribute("name", "Importato dall\'OPAC SBN")
root.appendChild(collection)
res = conn.search(query)
def make_conn ():
conn = zoom.Connection ('z3950.loc.gov', 7090)
conn.databaseName = 'VOYAGER'
conn.preferredRecordSyntax = 'USMARC'
return conn
#!/usr/bin/env python
from PyZ3950 import zoom
conn = zoom.Connection('z3950.loc.gov', 7090)
conn.databaseName = 'VOYAGER'
conn.preferredRecordSyntax = 'USMARC'
query = zoom.Query('CCL', 'isbn=0253333490')
res = conn.search(query)
print res[0]
conn.close()
self.attributes[attr] = 1
if hasattr(ti, 'dbCombinations'):
dbcomb = ti.dbCombinations
for db1 in dbcomb:
for db2 in db1:
self.databases[db2] = 1
if hasattr(ti, 'recordSyntaxes'):
self.record_syntaxes = ti.recordSyntaxes
def disp_res(self, res):
for r in res:
print r
if __name__ == '__main__':
# XXX what record syntax does catalogue.bized.ac.uk:2105 want?
#conn = zoom.Connection ('z3950.copac.ac.uk', 210)
# conn = zoom.Connection ('www.cnshb.ru', 210)
# conn = zoom.Connection ('blpcz.bl.uk', 21021)
# conn = zoom.Connection('sherlock.berkeley.edu', 2100)
conn = zoom.Connection('gondolin.hist.liv.ac.uk', 210)
conn.databaseName = 'IR-Explain-1'
conn.preferredRecordSyntax = 'EXPLAIN'
# conn._cli.test = 1
e = Explainer(conn)
e.databases['l5r'] = 1
e.databases['scifi'] = 1
e.run()
conn.close()
#!/usr/bin/env python
from PyZ3950 import zoom
#conn = zoom.Connection ('z3950.loc.gov', 7090)
#conn.databaseName = 'VOYAGER'
conn = zoom.Connection('z3950.bibsys.no', 2100)
conn.databaseName = 'BIBSYS'
conn.preferredRecordSyntax = 'USMARC'
query1 = zoom.Query('CCL', 'au=Gould, Stephen Jay')
res1 = conn.search(query1)
query2 = zoom.Query('CCL', 'au=Pynchon, Thomas')
res2 = conn.search(query2)
for i in range(0, max(len(res1), len(res2))):
if i < len(res1):
print "1:", res1[i]
if i < len(res2):
print "2:", res2[i]
conn.close()
#!/usr/bin/env python
from __future__ import print_function, absolute_import
from PyZ3950 import zoom
conn = zoom.Connection('ilsz3950.nlm.nih.gov', 7090)
conn.databaseName = 'voyager'
conn.preferredRecordSyntax = 'MARC8'
query1 = zoom.Query('CCL', 'au=Gould, Stephen Jay')
res1 = conn.search(query1)
query2 = zoom.Query('CCL', 'au=Pynchon, Thomas')
res2 = conn.search(query2)
for i in range(0, max(len(res1), len(res2))):
if i < len(res1):
print("1:", res1[i])
if i < len(res2):
print("2:", res2[i])
conn.close()
def _getconn(self, ip, port, name, syntax):
conn = zoom.Connection(ip, port)
conn.databaseName = name
conn.preferredRecordSyntax = syntax
return conn
from PyZ3950 import zoom
conn = zoom.Connection('www.lib.csu.ru', 210)
conn.databaseName = 'arefd+knigi+liter+period'
conn.preferredRecordSyntax = 'USMARC'
query = zoom.Query('CCL', 'ti=Journal or au=Turgenev')
res = conn.search(query)
for r in res:
print "db:", repr(r.databaseName), r
def _get_offers_z3950(id, library):
offers = []
# determine which server to talk to
if library == 'Georgetown':
conf = settings.Z3950_SERVERS['GT']
elif library == 'George Mason':
id = id.strip('m')
conf = settings.Z3950_SERVERS['GM']
else:
raise Exception("unrecognized library %s" % library)
# search for the id, and get the first record
z = zoom.Connection(conf['IP'], conf['PORT'])
z.databaseName = conf['DB']
z.preferredRecordSyntax = conf['SYNTAX']
q = zoom.Query('PQF', '@attr 1=12 %s' % id.encode('utf-8'))
results = z.search(q)
if len(results) == 0:
return []
rec = results[0]
# normalize holdings information as schema.org offers
if hasattr(rec, 'data') and not hasattr(rec.data, 'holdingsData'):
return []
for holdings_data in rec.data.holdingsData:
h = holdings_data[1]
o = {'@type': 'Offer', 'seller': library}
if hasattr(h, 'callNumber'):
o['sku'] = h.callNumber.rstrip('\x00').strip()
if hasattr(h, 'localLocation'):
o['availabilityAtOrFrom'] = h.localLocation.rstrip('\x00')
if hasattr(h, 'publicNote') and library == 'Georgetown':
note = h.publicNote.rstrip('\x00')
if note == 'AVAILABLE':
o['availability'] = 'http://schema.org/InStock'
o['description'] = 'Available'
elif note in ('SPC USE ONLY', 'LIB USE ONLY'):
o['availability'] = 'http://schema.org/InStoreOnly'
o['description'] = 'Available'
else:
# set availabilityStarts from "DUE 09-15-14"
m = re.match('DUE (\d\d)-(\d\d)-(\d\d)', note)
if m:
m, d, y = [int(i) for i in m.groups()]
o['availabilityStarts'] = "20%02i-%02i-%02i" % (y, m, d)
o['availability'] = 'http://schema.org/OutOfStock'
o['description'] = 'Checked Out'
elif hasattr(h, 'circulationData'):
cd = h.circulationData[0]
if cd.availableNow is True:
o['availability'] = 'http://schema.org/InStock'
o['description'] = 'Available'
else:
if hasattr(cd, 'availabilityDate') and cd.availablityDate:
m = re.match("^(\d{4}-\d{2}-\d{2}).+", cd.availablityDate)
if m:
o['availabilityStarts'] = m.group(1)
o['availability'] = 'http://schema.org/OutOfStock'
o['description'] = 'Checked Out'
else:
logging.warn("unknown availability: bibid=%s library=%s h=%s", id,
library, h)
# some locations have a weird period before the name
o['availabilityAtOrFrom'] = o.get('availabilityAtOrFrom',
'').lstrip('.')
offers.append(o)
return offers
# coding: utf-8
"""
Simple script to search a Z39.50 target using Python
and PyZ3950.
"""
from PyZ3950 import zoom
ISBNs = ['978-1-905017-60-7', '2-86377-125-6']
conn = zoom.Connection('z3950.bnf.fr', 2211)
conn.databaseName = 'TOUT'
conn.preferredRecordSyntax = 'UNIMARC'
for isbn in ISBNs:
query = zoom.Query('CQL', 'find @attr 1=7 ' + isbn)
print(query)
res = conn.search(query)
for r in res:
print(str(r))
conn.close()
