def predefined(self, queryUrl=None, **kwargs):
assert queryUrl is not None, 'queryUrl parameter required'
try:
bindings = {key: URIRef(kwargs[key]) for key in kwargs}
response = Graph(store=self._store).query(prepareQuery(
self._resolver.resolve(queryUrl)),
initBindings=bindings)
register_query(queryUrl, bindings, self._resolver)
if response.type == 'CONSTRUCT': #These cannot be JSON-serialized so we extract the data with a SELECT
g = Graph()
g += response
response = g.query(
prepareQuery("SELECT ?s ?p ?o WHERE {?s ?p ?o}"))
return response
finally:
self._store.flush_log(logging.DEBUG)
def _next(self):
xml_input = urllib2.urlopen(self._url, timeout=20)
page = XMLGraph(xml_input)
self._handle_resumption_token(page)
puredomain = URIRef('http://{}/'.format(self._location))
prepd = prepareQuery(self._query, initNs={'puredomain': puredomain})
return page.query(prepd)
def build(self, graph):
try:
result = graph.query(prepareQuery(self._query))
task = result.__iter__().next()
logging.debug(task['pureurl'])
return self._url_to_iterator_fun(task['pureurl'])
except StopIteration:
return None
def _next(self):
xml_input = urllib2.urlopen(self._url, timeout=20)
page = XMLGraph(xml_input)
self._handle_resumption_token(page)
puredomain = URIRef('http://{}/'.format(self._location))
prepd = prepareQuery(self._query,
initNs={'puredomain' : puredomain})
return page.query(prepd)
def dynamic(self, name=None, query=None):
assert name is not None, 'name parameter required'
assert query is not None, 'query parameter required'
self._store.log(name)
try:
response = Graph(store=self._store).query(prepareQuery(query))
return response
finally:
self._store.flush_log(logging.DEBUG)
def __init__(self, xml_input = None, url = None):
'''Specify either xml_input or url
@param xml_input An XML string
@param url A URL string referring to an XML document
'''
self._timestamp_triple = None
if xml_input is None:
#Load XML from URL into xml_input
assert url is not None, 'Need xml_input or url'
self._timestamp_triple = (URIRef(url), SEPAKE.wasDetailedAtTime, Literal(datetime.utcnow()))
xml_input = urllib2.urlopen(url, timeout = 30)
if url is None:
#Dummy URL for internal data
url = 'file:///'
#Transform XML to a an RDF graph, removing unnecessary parts in the process
self._xml_as_rdf = _slimmed_xml_as_rdf(xml_input)
#Parse the two queries used to create triples from self._xml_as_rdf
self._queries = [prepareQuery(_CONSTRUCT_PROJECT, initNs = {'rest_url' : URIRef(url)}),
prepareQuery(_CONSTRUCT_PEOPLE, initNs = {'rest_url' : URIRef(url)})]
def __init__(self, url):
self._url = url
self._original_url = url
logging.debug(url)
url_parsed = urlparse(url)
self._location = url_parsed.netloc
self._more = True
self._query = prepareQuery(_CONSTRUCT_PAPERS,
initNs={'puredomain' : URIRef('http://{}/'.format(self._location)),
'oai_url' : URIRef(url)})
def __init__(self, url):
self._url = url
self._original_url = url
logging.debug(url)
url_parsed = urlparse(url)
self._location = url_parsed.netloc
self._more = True
self._query = prepareQuery(_CONSTRUCT_PAPERS,
initNs={
'puredomain':
URIRef('http://{}/'.format(
self._location)),
'oai_url':
URIRef(url)
})
def __init__(self, graph, tasks, verbose=False):
self._graph = graph
self._tasks = tasks
self._queries = [prepareQuery(q) for q in _CONSTRUCTS]
self._verbose = verbose
def details_iterator_generator(graph):
tasks = [task for task in graph.query(prepareQuery(_TASKS))]
if len(tasks) > 0:
return PureRESTPublicationHarvester(graph, tasks)
else:
return None
def __init__(self, graph, tasks, verbose = False):
self._graph = graph
self._tasks = tasks
self._queries = [prepareQuery(q) for q in _CONSTRUCTS]
self._verbose = verbose
def details_iterator_generator(graph):
tasks = [task for task in graph.query(prepareQuery(_TASKS))]
if len(tasks) > 0:
return PureRESTPublicationHarvester(graph, tasks)
else:
return None