def test_incremental_assembler_constructor():
ia = IncrementalAssembler([s1, s2])
assert ia.prepared_stmts == [s1, s2]
assert ia.stmts_by_hash == {s1h: s1, s2h: s2}
assert ia.evs_by_stmt_hash == {s1h: [ev1], s2h: [ev2]}, ia.evs_by_stmt_hash
assert ia.refinement_edges == {(s1h, s2h)}
assert set(ia.refinements_graph.nodes()) == {s1h, s2h}
assert set(ia.get_all_supporting_evidence(s1h)) == {ev1, ev2}
assert set(ia.get_all_supporting_evidence(s2h)) == {ev2}
def test_incremental_assembler_add_statement_duplicate():
ev3 = Evidence('eidos', text='3')
s3 = Influence(e1, e2, ev3)
s3h = s3.get_hash(matches_fun=location_matches_compositional)
ia = IncrementalAssembler([s1, s2])
delta = ia.add_statements([s3])
assert not delta.new_stmts, delta.new_stmts
assert delta.new_evidences == {s3h: [ev3]}, delta.new_evidences
assert not delta.new_refinements, delta.new_refinements
# TODO: test beliefs
assert set(ia.get_all_supporting_evidence(s1h)) == {ev1, ev2, ev3}
assert set(ia.get_all_supporting_evidence(s2h)) == {ev2}
def load_project(self, project_id, record_keys=None):
# 1. Select records associated with project
if record_keys is None:
record_keys = self.db.get_records_for_project(project_id)
# 2. Select statements from prepared stmts table
prepared_stmts = []
for record_key in record_keys:
prepared_stmts += self.db.get_statements_for_record(record_key)
# 3. Select curations for project
curations = self.get_project_curations(project_id)
# 4. Initiate an assembler
assembler = IncrementalAssembler(prepared_stmts, curations=curations)
self.assemblers[project_id] = assembler
def test_post_processing_new_stmts():
stmts = copy.deepcopy([s1, s2])
ia = IncrementalAssembler([stmts[0]])
delta = ia.add_statements([stmts[1]])
assert len(delta.new_stmts) == 1
stmt = list(delta.new_stmts.values())[0]
assert stmt.subj.concept.name == 'crop'
# Check that we added annotations
assert 'agents' in stmt.evidence[0].annotations
assert stmt.evidence[0].annotations['agents'] == {
'raw_text': ['some_text2', 'some_text2']
}, stmt.evidence[0].annotations['agents']
def test_incremental_assembler_add_statement_new_refinement():
ev4 = Evidence('eidos', text='4')
s4 = Influence(e2, e4, ev4)
s4h = s4.get_hash(matches_fun=location_matches_compositional)
ia = IncrementalAssembler([s1, s2])
delta = ia.add_statements([s4])
assert delta.new_stmts, {s4h: s4}
assert delta.new_evidences == {s4h: [ev4]}, delta.new_evidences
assert delta.new_refinements == {(s1h, s4h), (s2h, s4h)}, \
delta.new_refinements
# TODO: test beliefs
assert set(ia.get_all_supporting_evidence(s1h)) == {ev1, ev2, ev4}
assert set(ia.get_all_supporting_evidence(s2h)) == {ev2, ev4}
assert set(ia.get_all_supporting_evidence(s4h)) == {ev4}
def test_post_processing_all_stmts():
stmts = copy.deepcopy([s1, s2])
ia = IncrementalAssembler(stmts)
stmts_out = ia.get_statements()
# Check that we normalized concept names
assert stmts_out[0].subj.concept.name == 'agriculture'
# Check that we added flattened groundings
flat_grounding = [{'grounding': 'wm/concept/agriculture',
'name': 'agriculture', 'score': 1.0}]
assert stmts_out[0].subj.concept.db_refs['WM_FLAT'] == \
flat_grounding, flat_grounding
# Check that we added annotations
assert 'agents' in stmts_out[0].evidence[0].annotations
assert stmts_out[0].evidence[0].annotations['agents'] == {
'raw_text': ['some_text1', 'some_text2']
}, stmts_out[0].evidence[0].annotations['agents']
def assemble(self):
"""Run assembly on the prepared statements.
This function loads all the prepared statements associated with the
corpus and then runs assembly on them.
"""
all_stmts = []
for record in self.dart_records:
stmts = self.sc.db.get_statements_for_document(
document_id=record['document_id'],
reader=record['reader'],
reader_version=record['reader_version'])
all_stmts += stmts
ia = IncrementalAssembler(all_stmts)
self.assembled_stmts = ia.get_statements()
self.metadata['num_statements'] = len(self.assembled_stmts)
def assemble(self):
"""Run assembly on the prepared statements.
This function loads all the prepared statements associated with the
corpus and then runs assembly on them.
"""
all_stmts = []
logger.info('Loading statements from DB for %d records' %
len(self.dart_records))
for record in tqdm.tqdm(self.dart_records):
stmts = self.sc.db.get_statements_for_record(record['storage_key'])
all_stmts += stmts
logger.info('Instantiating incremental assembler with %d statements' %
len(all_stmts))
ia = IncrementalAssembler(all_stmts)
logger.info('Getting assembled statements')
self.assembled_stmts = ia.get_statements()
logger.info('Got %d assembled statements' % len(self.assembled_stmts))
self.metadata['num_statements'] = len(self.assembled_stmts)