def populate_db(self):
for d in self.data:
d['Property']['IdUnique'] = d['IdUnique']
if d['Kind'] == 'node':
if d['DeDuplication'] == None and self.g.nodes.match(
IdUnique=d['IdUnique']).count() == 0:
create_nodes(self.g.auto(), [d['Property']],
labels={*d['Label']})
else:
merge_nodes(self.g.auto(), [d['Property']],
(tuple(d['Label']), 'IdUnique'),
labels={*d['Label']})
else:
da = ((d['FromIdMaster']), d['Property'], (d['ToIdMaster']))
if d['DeDuplication'] == None and self.g.relationships.match(
IdUnique=d['IdUnique']).count() == 0:
create_relationships(self.g.auto(), [da],
d['Type'],
start_node_key=(d['FromLabel'],
'IdMaster'),
end_node_key=(d['ToLabel'],
'IdMaster'))
else:
merge_relationships(self.g.auto(), [da],
(d['Type'], 'IdUnique'),
start_node_key=(d['FromLabel'],
'IdMaster'),
end_node_key=(d['ToLabel'],
'IdMaster'))
def test_merge_nodes_from_property_dicts(graph):
graph.delete_all()
with graph.begin() as tx:
merge_nodes(tx,
DATA_DICTS, ("Person", "name"),
labels={"Person", "Employee"})
matched = graph.nodes.match("Person")
assert matched.count() == 3
assert all(node.labels == {"Person", "Employee"} for node in matched)
def test_merge_nodes_from_property_lists(graph):
graph.delete_all()
tx = graph.begin()
merge_nodes(tx,
DATA_LISTS, ("Person", "name"),
labels={"Person", "Employee"},
keys=HEADERS)
graph.commit(tx)
matched = graph.nodes.match("Person")
assert matched.count() == 3
assert all(node.labels == {"Person", "Employee"} for node in matched)
def test_merge_nodes_from_property_lists(graph):
graph.delete_all()
tx = graph.begin()
tx.update("CREATE (a:Person {name:'Alice', age:99})")
merge_nodes(tx,
DATA_LISTS, ("Person", "name"),
labels={"Person", "Employee"},
keys=HEADERS)
graph.commit(tx)
matched = graph.nodes.match("Person")
assert matched.count() == 3
assert all(node.labels == {"Person", "Employee"} for node in matched)
assert graph.nodes.match("Person", name="Alice").first()["age"] == 66
def merge_nodes(self, g, data, merge_key=None, keys=None):
"""
:param tx: Transaction in which to carry out this operation
:param data: node data supplied as a list of lists (if keys are provided) or a list of dictionaries (if keys is None
:param merge_key: tuple of (label, key1, key2…) on which to merge
:param labels: additional labels to apply to the merged nodes
:param keys: optional set of keys for the supplied data (if supplied as value lists)
"""
merge_key = merge_key or self.merge_key
keys = keys or list(self.fields.keys())
if len(data) > self.LIMIT:
for small_data in self.split_data(data):
merge_nodes(g.auto(), small_data, merge_key, keys=keys)
else:
merge_nodes(g.auto(), data, merge_key, keys=keys)
def merge(self, graph, merge_properties=None, batch_size=None, preserve=None, append_props=None):
"""
Merge nodes from NodeSet on merge properties.
:param merge_properties: The merge properties.
"""
# overwrite if preserve is passed
if preserve:
self.preserve = preserve
# overwrite if array_props is passed
if append_props:
self.append_props = append_props
log.debug('Merge NodeSet on {}'.format(merge_properties))
if not batch_size:
batch_size = self.batch_size
if not merge_properties:
merge_properties = self.merge_keys
log.debug('Batch Size: {}'.format(batch_size))
# use py2neo base functions if no properties are preserved
if not self.preserve and not self.append_props:
for batch in chunks(self.node_properties(), size=batch_size):
merge_nodes(graph, batch, (tuple(self.labels), *merge_properties))
elif self.preserve and not self.append_props:
q = nodes_merge_unwind_preserve(self.labels, self.merge_keys, property_parameter='props')
for batch in chunks(self.node_properties(), size=batch_size):
graph.run(q, props=list(batch), preserve=self.preserve)
elif not self.preserve and self.append_props:
q = nodes_merge_unwind_array_props(self.labels, self.merge_keys, self.append_props,
property_parameter='props')
for batch in chunks(self.node_properties(), size=batch_size):
graph.run(q, props=list(batch), append_props=self.append_props)
elif self.preserve and self.append_props:
q = nodes_merge_unwind_preserve_array_props(self.labels, self.merge_keys, self.append_props, self.preserve,
property_parameter='props')
print(q)
for batch in chunks(self.node_properties(), size=batch_size):
graph.run(q, props=list(batch), append_props=self.append_props, preserve=self.preserve)
def merge_nodes_from_data_frame(tx, df, merge_key, labels=None, preserve=None):
""" Merge nodes from a DataFrame.
This function wraps the :func:`py2neo.bulk.merge_nodes` function,
allowing a `DataFrame <https://pandas.pydata.org/pandas-docs/stable/dsintro.html#series>`__
object to be passed in place of the regular `data` argument.
:param tx:
:param df:
:param merge_key:
:param labels:
:param preserve:
:return:
"""
merge_nodes(tx,
df.itertuples(index=False, name=None),
merge_key,
labels=labels,
keys=df.keys(),
preserve=preserve)