def _query_and_parameters(self):
""" A tuple of the Cypher query and parameters used to select
the nodes that match the criteria for this selection.
:return: Cypher query string
"""
clauses = [
"MATCH (_%s)" % "".join(":%s" % cypher_escape(label)
for label in self._labels)
]
parameters = {}
if self._conditions:
conditions = []
for condition in self._conditions:
if isinstance(condition, tuple):
condition, param = condition
parameters.update(param)
conditions.append(condition)
clauses.append("WHERE %s" % " AND ".join(conditions))
clauses.append("RETURN _")
if self._order_by:
clauses.append("ORDER BY %s" % (", ".join(self._order_by)))
if self._skip:
clauses.append("SKIP %d" % self._skip)
if self._limit is not None:
clauses.append("LIMIT %d" % self._limit)
return " ".join(clauses), parameters
def __db_create__(self, tx):
from py2neo.database.cypher import cypher_escape
nodes = list(self.nodes())
reads = []
writes = []
parameters = {}
returns = {}
for i, node in enumerate(nodes):
node_id = "a%d" % i
param_id = "x%d" % i
remote_node = remote(node)
if remote_node:
reads.append("MATCH (%s) WHERE id(%s)={%s}" %
(node_id, node_id, param_id))
parameters[param_id] = remote_node._id
else:
label_string = "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))
writes.append("CREATE (%s%s {%s})" %
(node_id, label_string, param_id))
parameters[param_id] = dict(node)
node._set_remote_pending(tx)
returns[node_id] = node
for i, relationship in enumerate(self.relationships()):
if not remote(relationship):
rel_id = "r%d" % i
start_node_id = "a%d" % nodes.index(relationship.start_node())
end_node_id = "a%d" % nodes.index(relationship.end_node())
type_string = cypher_escape(relationship.type())
param_id = "y%d" % i
writes.append("CREATE UNIQUE (%s)-[%s:%s]->(%s) SET %s={%s}" %
(start_node_id, rel_id, type_string, end_node_id,
rel_id, param_id))
parameters[param_id] = dict(relationship)
returns[rel_id] = relationship
relationship._set_remote_pending(tx)
statement = "\n".join(reads + writes +
["RETURN %s LIMIT 1" % ", ".join(returns)])
tx.entities.append(returns)
tx.run(statement, parameters)
def _property_equality_conditions(properties, offset=1):
for i, (key, value) in enumerate(properties.items(), start=offset):
if key == "__id__":
condition = "id(_)"
else:
condition = "_.%s" % cypher_escape(key)
if isinstance(value, (tuple, set, frozenset)):
condition += " IN {%d}" % i
parameters = {"%d" % i: list(value)}
else:
condition += " = {%d}" % i
parameters = {"%d" % i: value}
yield condition, parameters
def __db_create__(self, tx):
from py2neo.database.cypher import cypher_escape
nodes = list(self.nodes())
reads = []
writes = []
parameters = {}
returns = {}
for i, node in enumerate(nodes):
node_id = "a%d" % i
param_id = "x%d" % i
remote_node = remote(node)
if remote_node:
reads.append("MATCH (%s) WHERE id(%s)={%s}" % (node_id, node_id, param_id))
parameters[param_id] = remote_node._id
else:
label_string = "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))
writes.append("CREATE (%s%s {%s})" % (node_id, label_string, param_id))
parameters[param_id] = dict(node)
node._set_remote_pending(tx)
returns[node_id] = node
for i, relationship in enumerate(self.relationships()):
if not remote(relationship):
rel_id = "r%d" % i
start_node_id = "a%d" % nodes.index(relationship.start_node())
end_node_id = "a%d" % nodes.index(relationship.end_node())
type_string = cypher_escape(relationship.type())
param_id = "y%d" % i
writes.append("CREATE UNIQUE (%s)-[%s:%s]->(%s) SET %s={%s}" %
(start_node_id, rel_id, type_string, end_node_id, rel_id, param_id))
parameters[param_id] = dict(relationship)
returns[rel_id] = relationship
relationship._set_remote_pending(tx)
statement = "\n".join(reads + writes + ["RETURN %s LIMIT 1" % ", ".join(returns)])
tx.entities.append(returns)
tx.run(statement, parameters)
def __db_merge__(self, tx, primary_label=None, primary_key=None):
from py2neo.database.cypher import cypher_escape, cypher_repr
nodes = list(self.nodes())
match_clauses = []
merge_clauses = []
parameters = {}
returns = {}
for i, node in enumerate(nodes):
node_id = "a%d" % i
param_id = "x%d" % i
remote_node = remote(node)
if remote_node:
match_clauses.append("MATCH (%s) WHERE id(%s)={%s}" %
(node_id, node_id, param_id))
parameters[param_id] = remote_node._id
else:
merge_label = getattr(node, "__primarylabel__",
None) or primary_label
if merge_label is None:
label_string = "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))
elif node.labels():
label_string = ":" + cypher_escape(merge_label)
else:
label_string = ""
merge_keys = getattr(node, "__primarykey__",
None) or primary_key
if merge_keys is None:
merge_keys = ()
elif is_collection(merge_keys):
merge_keys = tuple(merge_keys)
else:
merge_keys = (merge_keys, )
if merge_keys:
property_map_string = cypher_repr({
k: v
for k, v in dict(node).items() if k in merge_keys
})
else:
property_map_string = cypher_repr(dict(node))
merge_clauses.append(
"MERGE (%s%s %s)" %
(node_id, label_string, property_map_string))
if node.labels():
merge_clauses.append(
"SET %s%s" %
(node_id, "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))))
if merge_keys:
merge_clauses.append("SET %s={%s}" % (node_id, param_id))
parameters[param_id] = dict(node)
node._set_remote_pending(tx)
returns[node_id] = node
clauses = match_clauses + merge_clauses
for i, relationship in enumerate(self.relationships()):
if not remote(relationship):
rel_id = "r%d" % i
start_node_id = "a%d" % nodes.index(relationship.start_node())
end_node_id = "a%d" % nodes.index(relationship.end_node())
type_string = cypher_escape(relationship.type())
param_id = "y%d" % i
clauses.append("MERGE (%s)-[%s:%s]->(%s) SET %s={%s}" %
(start_node_id, rel_id, type_string,
end_node_id, rel_id, param_id))
parameters[param_id] = dict(relationship)
returns[rel_id] = relationship
relationship._set_remote_pending(tx)
statement = "\n".join(clauses +
["RETURN %s LIMIT 1" % ", ".join(returns)])
tx.entities.append(returns)
tx.run(statement, parameters)
def process_results(gearman_worker, gearman_job):
disease = gearman_job.data['disease']
node_label = gearman_job.data['node_label']
retstart = gearman_job.data['retstart']
retmax = gearman_job.data['retmax']
results = search_field("\"{}\"".format(disease), retstart, retmax)
id_list = results['IdList']
papers = fetch_details(id_list)
countries = dict()
driver = GraphDatabase.driver(neo4j_url,
auth=basic_auth(neo4j_user, neo4j_password))
session = driver.session()
for i, paper in enumerate(papers):
if 'MedlineCitation' in paper.keys():
if 'Article' in paper['MedlineCitation'].keys():
if 'AuthorList' in paper['MedlineCitation']['Article'].keys():
author_list = (
paper['MedlineCitation']['Article']['AuthorList'])
authors = Set()
for index in range(0, min(len(author_list), 12)):
a = author_list[index]
name = ''
if 'LastName' in a.keys() and 'ForeName' in a.keys():
name = "{}_{}".format(
a['ForeName'].encode('utf-8').replace(
' ', '_'),
a['LastName'].encode('utf-8').replace(' ', ''))
authors.add(name)
if (name not in countries.keys()
or countries[name] == "Not found"):
if 'AffiliationInfo' in a.keys():
if len(a['AffiliationInfo']) > 0:
if ('Affiliation'
in a['AffiliationInfo'][0].keys()):
s_country = (find_between_r(
(a['AffiliationInfo'][0]
['Affiliation']), ',', '.'))
if 'China' in s_country:
s_country = 'China'
if 'United States' in s_country:
s_country = 'USA'
countries[name] = s_country
else:
countries[name] = "Not found"
authors = list(authors)
# create network
for i in range(len(authors)):
u = authors[i]
for j in range(i + 1, len(authors)):
v = authors[j]
session.run("MERGE (p1:`{node}` {{name:\"{}\"}}) "
"MERGE (p2:`{node}` {{name:\"{}\"}}) "
"CREATE (p1)-[:KNOWS]->(p2)".format(
cypher_escape(u).encode('utf-8'),
cypher_escape(v).encode('utf-8'),
node=node_label))
session.close()
return countries
def __db_merge__(self, tx, primary_label=None, primary_key=None):
from py2neo.database.cypher import cypher_escape, cypher_repr
nodes = list(self.nodes())
match_clauses = []
merge_clauses = []
parameters = {}
returns = {}
for i, node in enumerate(nodes):
node_id = "a%d" % i
param_id = "x%d" % i
remote_node = remote(node)
if remote_node:
match_clauses.append("MATCH (%s) WHERE id(%s)={%s}" % (node_id, node_id, param_id))
parameters[param_id] = remote_node._id
else:
merge_label = getattr(node, "__primarylabel__", None) or primary_label
if merge_label is None:
label_string = "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))
elif node.labels():
label_string = ":" + cypher_escape(merge_label)
else:
label_string = ""
merge_keys = getattr(node, "__primarykey__", None) or primary_key
if merge_keys is None:
merge_keys = ()
elif is_collection(merge_keys):
merge_keys = tuple(merge_keys)
else:
merge_keys = (merge_keys,)
if merge_keys:
property_map_string = cypher_repr({k: v for k, v in dict(node).items()
if k in merge_keys})
else:
property_map_string = cypher_repr(dict(node))
merge_clauses.append("MERGE (%s%s %s)" % (node_id, label_string, property_map_string))
if node.labels():
merge_clauses.append("SET %s%s" % (
node_id, "".join(":" + cypher_escape(label)
for label in sorted(node.labels()))))
if merge_keys:
merge_clauses.append("SET %s={%s}" % (node_id, param_id))
parameters[param_id] = dict(node)
node._set_remote_pending(tx)
returns[node_id] = node
clauses = match_clauses + merge_clauses
for i, relationship in enumerate(self.relationships()):
if not remote(relationship):
rel_id = "r%d" % i
start_node_id = "a%d" % nodes.index(relationship.start_node())
end_node_id = "a%d" % nodes.index(relationship.end_node())
type_string = cypher_escape(relationship.type())
param_id = "y%d" % i
clauses.append("MERGE (%s)-[%s:%s]->(%s) SET %s={%s}" %
(start_node_id, rel_id, type_string, end_node_id, rel_id, param_id))
parameters[param_id] = dict(relationship)
returns[rel_id] = relationship
relationship._set_remote_pending(tx)
statement = "\n".join(clauses + ["RETURN %s LIMIT 1" % ", ".join(returns)])
tx.entities.append(returns)
tx.run(statement, parameters)