def delete_datasets(self):
"""delete the datasets from the database and the triplestore"""
sparql = SparqlQueryLauncher(self.app, self.session)
for dataset in self.datasets:
# Delete from triplestore
if dataset.graph_name:
Utils.redo_if_failure(self.log, 3, 1, sparql.drop_dataset,
dataset.graph_name)
# Delete from db
dataset.delete_from_db()
def delete_datasets(self):
"""delete the datasets from the database and the triplestore"""
sparql = SparqlQueryLauncher(self.app, self.session)
tse = TriplestoreExplorer(self.app, self.session)
for dataset in self.datasets:
# Delete from triplestore
if dataset.graph_name:
Utils.redo_if_failure(self.log, 3, 1, sparql.drop_dataset,
dataset.graph_name)
# Delete from db
dataset.delete_from_db()
# Uncache abstraction
tse.uncache_abstraction(public=dataset.public)
def delete_user_rdf(self, username):
"""Delete a user rdf graphs
Delete in DB, TS and filesystem
Parameters
----------
username : string
Username to delete
"""
tse = TriplestoreExplorer(self.app, self.session)
query_launcher = SparqlQueryLauncher(self.app, self.session)
graphs = tse.get_graph_of_user(username)
for graph in graphs:
Utils.redo_if_failure(self.log, 3, 1, query_launcher.drop_dataset, graph)
def integrate(self, public=False):
"""Integrate the file into the triplestore
Parameters
----------
public : bool, optional
Integrate in private or public graph
"""
sparql = SparqlQueryLauncher(self.app, self.session)
tse = TriplestoreExplorer(self.app, self.session)
self.public = public
method = self.settings.get('triplestore', 'upload_method')
# Load file into a RDF graph
self.graph_chunk.parse(self.path, format=self.type_dict[self.type])
# get metadata
self.set_metadata()
# Remove metadata from data
self.delete_metadata_location()
# insert metadata
sparql.insert_data(self.graph_metadata, self.file_graph, metadata=True)
if method == "load":
# write rdf into a tmpfile and load it
temp_file_name = 'tmp_{}_{}.{}'.format(Utils.get_random_string(5),
self.name,
self.rdf_extention)
# Try to load data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, self.load_graph,
self.graph_chunk, temp_file_name)
else:
# Insert
# Try to insert data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, sparql.insert_data,
self.graph_chunk, self.file_graph)
# Remove chached abstraction
tse.uncache_abstraction(public=self.public)
self.set_triples_number()
def integrate(self, dataset_id=None):
"""Integrate the file into the triplestore"""
sparql = SparqlQueryLauncher(self.app, self.session)
# insert metadata
sparql.insert_data(self.get_metadata(), self.file_graph, metadata=True)
content_generator = self.generate_rdf_content()
# Insert content
chunk_number = 0
for _ in content_generator:
if self.graph_chunk.ntriple >= self.max_chunk_size:
if self.graph_chunk.percent and dataset_id:
self.update_percent_in_db(self.graph_chunk.percent,
dataset_id)
if self.method == 'load':
# write rdf into a tmpfile and load it
temp_file_name = 'tmp_{}_{}_chunk_{}.{}'.format(
Utils.get_random_string(5), self.name, chunk_number,
self.rdf_extention)
# Try to load data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, self.load_graph,
self.graph_chunk, temp_file_name)
else:
# Insert
# Try to insert data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, sparql.insert_data,
self.graph_chunk, self.file_graph)
chunk_number += 1
self.graph_chunk = RdfGraph(self.app, self.session)
# Load the last chunk
if self.graph_chunk.percent and dataset_id:
self.update_percent_in_db(100, dataset_id)
if self.method == 'load':
temp_file_name = 'tmp_{}_{}_chunk_{}.{}'.format(
Utils.get_random_string(5), self.name, chunk_number,
self.rdf_extention)
# Try to load data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, self.load_graph,
self.graph_chunk, temp_file_name)
else:
# Insert
# Try to insert data. if failure, wait 5 sec and retry 5 time
Utils.redo_if_failure(self.log, 5, 5, sparql.insert_data,
self.graph_chunk, self.file_graph)
# Content is inserted, now insert abstraction and domain_knowledge
self.set_rdf_abstraction_domain_knowledge()
if self.method == 'load':
temp_file_name = 'tmp_{}_{}_abstraction_domain_knowledge.{}'.format(
Utils.get_random_string(5), self.name, self.rdf_extention)
self.load_graph(self.graph_abstraction_dk, temp_file_name)
else:
# Insert
sparql.insert_data(self.graph_abstraction_dk, self.file_graph)
self.set_triples_number()
