def get_context_data(self, **kwargs):
context = super(AggregatorExportView, self).get_context_data(**kwargs)
context.update({
'sd_prefix': settings.TRIPLE_DATABASE['PREFIXES']['sdv1'],
'sparql_endpoint': get_virtuoso_endpoint(),
'mastergraph_host': settings.TRIPLE_DATABASE_MASTER['HOST'],
'mastergraph_port':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['rexpro_port'],
'mastergraph_graphname':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['graph'],
'resource_namespace':
settings.TRIPLE_DATABASE_MASTER['PREFIXES']['sdres'],
})
return context
def get_context_data(self, **kwargs):
context = super(AggregatorExportView, self).get_context_data(**kwargs)
context.update({
'sd_prefix':
settings.TRIPLE_DATABASE['PREFIXES']['sdv1'],
'sparql_endpoint':
get_virtuoso_endpoint(),
'mastergraph_host':
settings.TRIPLE_DATABASE_MASTER['HOST'],
'mastergraph_port':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['rexpro_port'],
'mastergraph_graphname':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['graph'],
'resource_namespace':
settings.TRIPLE_DATABASE_MASTER['PREFIXES']['sdres'],
})
return context
def process_aggregator(aggregator, force=False):
""" execute the aggregator workflow: run silk on every archive item
associated to the aggregator.
"""
from tempfile import mkdtemp
from webui.cnmain.utils import get_virtuoso_endpoint
logger_name = process_aggregator.request.id
loggy = get_redis_logger(logger_name)
local_manager.cleanup()
local.logger = loggy
tmpdir = mkdtemp()
scheduler = Scheduler.objects.create(
content_type=ContentType.objects.get_for_model(aggregator),
object_id=aggregator.pk,
status=Scheduler.RUNNING,
logger_name=logger_name,
)
try:
loggy.info("Processing aggregator %s", unicode(aggregator))
loggy.debug("Working dir: %s", tmpdir)
context = {
'aggregator':
aggregator,
'sd_prefix':
settings.TRIPLE_DATABASE['PREFIXES']['sdv1'],
'sparql_endpoint':
get_virtuoso_endpoint(),
'mastergraph_host':
settings.TRIPLE_DATABASE_MASTER['HOST'],
'mastergraph_port':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['rexpro_port'],
'mastergraph_graphname':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['graph'],
'resource_namespace':
settings.TRIPLE_DATABASE_MASTER['PREFIXES']['sdres'],
}
loggy.info("Connecting to virtuoso")
aggregator_archiveitems = aggregator.aggregatorarchiveitem_set\
.all().order_by('first_workflow_success')
if not force:
res = []
for aggregator_archiveitem in aggregator_archiveitems:
if aggregator_archiveitem.needs_update():
res.append(aggregator_archiveitem)
else:
loggy.info('Skipped archiveitem %s',
unicode(aggregator_archiveitem.archiveitem))
aggregator_archiveitems = res
_aggregator_process_archiveitems(aggregator_archiveitems, scheduler,
tmpdir, context)
loggy.info('Workflow completed')
except Exception, e:
loggy.exception('Generic exception in the workflow')
scheduler.status = Scheduler.FAIL
scheduler.error = e.message or str(e)
# send the exception to sentry
raise
def process_aggregator(aggregator, force=False):
""" execute the aggregator workflow: run silk on every archive item
associated to the aggregator.
"""
from tempfile import mkdtemp
from webui.cnmain.utils import get_virtuoso_endpoint
logger_name = process_aggregator.request.id
loggy = get_redis_logger(logger_name)
local_manager.cleanup()
local.logger = loggy
tmpdir = mkdtemp()
scheduler = Scheduler.objects.create(
content_type=ContentType.objects.get_for_model(aggregator),
object_id=aggregator.pk,
status=Scheduler.RUNNING,
logger_name=logger_name,
)
try:
loggy.info("Processing aggregator %s", unicode(aggregator))
loggy.debug("Working dir: %s", tmpdir)
context = {
'aggregator': aggregator,
'sd_prefix': settings.TRIPLE_DATABASE['PREFIXES']['sdv1'],
'sparql_endpoint': get_virtuoso_endpoint(),
'mastergraph_host': settings.TRIPLE_DATABASE_MASTER['HOST'],
'mastergraph_port':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['rexpro_port'],
'mastergraph_graphname':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['graph'],
'resource_namespace':
settings.TRIPLE_DATABASE_MASTER['PREFIXES']['sdres'],
}
loggy.info("Connecting to virtuoso")
aggregator_archiveitems = aggregator.aggregatorarchiveitem_set\
.all().order_by('first_workflow_success')
if not force:
res = []
for aggregator_archiveitem in aggregator_archiveitems:
if aggregator_archiveitem.needs_update():
res.append(aggregator_archiveitem)
else:
loggy.info('Skipped archiveitem %s',
unicode(aggregator_archiveitem.archiveitem))
aggregator_archiveitems = res
_aggregator_process_archiveitems(
aggregator_archiveitems, scheduler, tmpdir, context
)
loggy.info('Workflow completed')
except Exception, e:
loggy.exception('Generic exception in the workflow')
scheduler.status = Scheduler.FAIL
scheduler.error = e.message or str(e)
# send the exception to sentry
raise
def test_silk_project_file_is_valid(self):
import xml.etree.ElementTree as ET
self.client_login('admin')
item1 = ArchiveItemFactory()
item2 = ArchiveItemFactory()
for item in (item1, item2):
AggregatorArchiveItem.objects.create(
aggregator=self.aggregator,
archiveitem=item
)
response = self.client.get(self.export_url)
tree = ET.fromstring(response.content)
self.assertIn(
(settings.TRIPLE_DATABASE['PREFIXES']['sdv1'], 'sd'),
[(x.get('namespace'), x.get('id'))
for x in tree.findall('.//Prefix')]
)
# check datasources
datasources = tree.findall('.//DataSource')
self.assertEqual(len(datasources), 3)
self.assertEqual(datasources[0].get('id'), 'master-graph')
mastergraph = datasources[0]
datasources = datasources[1:]
# check datasources endpoints
self.assertEqual(
mastergraph.find('Param[@name="host"]').get('value'),
settings.TRIPLE_DATABASE_MASTER['HOST']
)
self.assertEqual(
[get_virtuoso_endpoint()] * 2,
[x.find('Param[@name="endpointURI"]').get("value")
for x in datasources]
)
# check datasources graph names
self.assertEqual(
mastergraph.find('Param[@name="graph"]').get('value'),
settings.TRIPLE_DATABASE_MASTER["KWARGS"]["graph"]
)
self.assertEqual(
[item1.datagraph_mapped_name, item2.datagraph_mapped_name],
[x.find('Param[@name="graph"]').get("value")
for x in datasources]
)
# check tasks
datasource_ids = [x.get('id') for x in datasources]
tasks = tree.findall('.//LinkingTask')
self.assertEqual(len(tasks), 2)
self.assertEqual(
datasource_ids,
[x.find('.//Interlink').get('id') for x in tasks]
)
# check task parameters
for datasource_id, task in zip(datasource_ids, tasks):
self.assertEqual(
task.find('.//SourceDataset').get('dataSource'),
datasource_id
)
self.assertEqual(
task.find('.//TargetDataset').get('dataSource'),
'master-graph'
)
self.assertEqual(
task.find('.//SourceDataset').find('RestrictTo').text.strip(),
'?a rdf:type <{}> .'.format(self.aggregator.entity_type)
)
self.assertEqual(
task.find('.//TargetDataset').find('RestrictTo').text.strip(),
'b -> {}'.format(self.aggregator.vertex_selector)
)
self.assertIsNone(task.find('.//LinkageRule').text)
self.assertIsNone(task.find('.//Filter').text)
self.assertIsNone(task.find('.//Outputs').text)
self.assertIsNone(task.find('.//PositiveEntities').text)
self.assertIsNone(task.find('.//NegativeEntities').text)
self.assertIsNone(
task.find('.//Alignment/')
.find('{}Alignment'.format('{http://knowledgeweb.'
'semanticweb.org'
'/heterogeneity/alignment#}')
).text
)
def test_silk_project_file_is_valid(self):
import xml.etree.ElementTree as ET
self.client_login('admin')
item1 = ArchiveItemFactory()
item2 = ArchiveItemFactory()
for item in (item1, item2):
AggregatorArchiveItem.objects.create(aggregator=self.aggregator,
archiveitem=item)
response = self.client.get(self.export_url)
tree = ET.fromstring(response.content)
self.assertIn((settings.TRIPLE_DATABASE['PREFIXES']['sdv1'], 'sd'),
[(x.get('namespace'), x.get('id'))
for x in tree.findall('.//Prefix')])
# check datasources
datasources = tree.findall('.//DataSource')
self.assertEqual(len(datasources), 3)
self.assertEqual(datasources[0].get('id'), 'master-graph')
mastergraph = datasources[0]
datasources = datasources[1:]
# check datasources endpoints
self.assertEqual(
mastergraph.find('Param[@name="host"]').get('value'),
settings.TRIPLE_DATABASE_MASTER['HOST'])
self.assertEqual([get_virtuoso_endpoint()] * 2, [
x.find('Param[@name="endpointURI"]').get("value")
for x in datasources
])
# check datasources graph names
self.assertEqual(
mastergraph.find('Param[@name="graph"]').get('value'),
settings.TRIPLE_DATABASE_MASTER["KWARGS"]["graph"])
self.assertEqual(
[item1.datagraph_mapped_name, item2.datagraph_mapped_name],
[x.find('Param[@name="graph"]').get("value") for x in datasources])
# check tasks
datasource_ids = [x.get('id') for x in datasources]
tasks = tree.findall('.//LinkingTask')
self.assertEqual(len(tasks), 2)
self.assertEqual(datasource_ids,
[x.find('.//Interlink').get('id') for x in tasks])
# check task parameters
for datasource_id, task in zip(datasource_ids, tasks):
self.assertEqual(
task.find('.//SourceDataset').get('dataSource'), datasource_id)
self.assertEqual(
task.find('.//TargetDataset').get('dataSource'),
'master-graph')
self.assertEqual(
task.find('.//SourceDataset').find('RestrictTo').text.strip(),
'?a rdf:type <{}> .'.format(self.aggregator.entity_type))
self.assertEqual(
task.find('.//TargetDataset').find('RestrictTo').text.strip(),
'b -> {}'.format(self.aggregator.vertex_selector))
self.assertIsNone(task.find('.//LinkageRule').text)
self.assertIsNone(task.find('.//Filter').text)
self.assertIsNone(task.find('.//Outputs').text)
self.assertIsNone(task.find('.//PositiveEntities').text)
self.assertIsNone(task.find('.//NegativeEntities').text)
self.assertIsNone(
task.find('.//Alignment/').find('{}Alignment'.format(
'{http://knowledgeweb.'
'semanticweb.org'
'/heterogeneity/alignment#}')).text)
def test_can_silk_rules_file_is_valid(self):
import xml.etree.ElementTree as ET
from django.template.loader import render_to_string
archive_item = ArchiveItemFactory()
AggregatorArchiveItem.objects.create(
aggregator=self.aggregator,
archiveitem=archive_item
)
self.aggregator.silk_rule = \
'<LinkageRule><smart data="now" /></LinkageRule>'
self.aggregator.save()
output_filename = 'a_really_cool_filename.thm'
context = {
'aggregator': self.aggregator,
'sd_prefix': settings.TRIPLE_DATABASE['PREFIXES']['sdv1'],
'sparql_endpoint': get_virtuoso_endpoint(),
'archive_item': archive_item,
'output_filename': output_filename,
'mastergraph_host': settings.TRIPLE_DATABASE_MASTER['HOST'],
'mastergraph_port':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['rexpro_port'],
'mastergraph_graphname':
settings.TRIPLE_DATABASE_MASTER['KWARGS']['graph'],
'resource_namespace':
settings.TRIPLE_DATABASE_MASTER['PREFIXES']['sdres'],
}
tree = ET.fromstring(render_to_string(
'controller/aggregator/silk_rules.xml', context
))
self.assertIn(
(settings.TRIPLE_DATABASE['PREFIXES']['sdv1'], 'sd'),
[(x.get('namespace'), x.get('id'))
for x in tree.findall('.//Prefix')]
)
# check datasources
datasources_dom = tree.findall('.//DataSource')
self.assertEqual(len(datasources_dom), 2)
self.assertEqual(datasources_dom[0].get('id'), 'master-graph')
mastergraph, datasource = datasources_dom
# check datasource endpoints
self.assertEqual(
get_virtuoso_endpoint(),
datasource.find('Param[@name="endpointURI"]').get("value"),
)
# check datasources graph names
self.assertEqual(
mastergraph.find('Param[@name="graph"]').get('value'),
settings.TRIPLE_DATABASE_MASTER["KWARGS"]["graph"]
)
self.assertEqual(
archive_item.datagraph_mapped_name,
datasource.find('Param[@name="graph"]').get("value")
)
# check tasks
datasource_id = datasource.get('id')
rules = tree.findall('.//Interlink')
self.assertEqual(len(rules), 1)
self.assertEqual(datasource_id, rules[0].get('id'))
# check rules parameters
rule = rules[0]
self.assertEqual(
rule.find('.//SourceDataset').get('dataSource'),
datasource_id
)
self.assertEqual(
rule.find('.//TargetDataset').get('dataSource'),
'master-graph'
)
self.assertEqual(
ET.tostring(rule.find('.//LinkageRule')).strip(),
self.aggregator.silk_rule
)
self.assertEqual(
rule.find('.//SourceDataset').find('RestrictTo').text.strip(),
'?a rdf:type <{}> .'.format(self.aggregator.entity_type)
)
self.assertEqual(
rule.find('.//TargetDataset').find('RestrictTo').text.strip(),
'b -> {}'.format(self.aggregator.vertex_selector)
)
self.assertIsNone(rule.find('.//Filter').text)
output = rule.find('.//Outputs').find('Output')
self.assertEqual(output.get('type'), 'file')
self.assertEqual(output.findall('Param')[0].get('name'), 'file')
self.assertEqual(
output.findall('Param')[0].get('value'), output_filename)
self.assertEqual(output.findall('Param')[1].get('name'), 'format')
self.assertEqual(output.findall('Param')[1].get('value'), 'ntriples')
