def test_keep_existing_reaction_id(self):
message = reaction_pb2.Reaction()
message.reaction_id = 'foo'
message.provenance.record_created.time.value = '11 am'
updates.update_reaction(message)
self.assertEqual(message.reaction_id, 'foo')
self.assertLen(message.provenance.record_modified, 0)
def test_keep_existing_reaction_id(self):
message = reaction_pb2.Reaction()
message.reaction_id = 'ord-c0bbd41f095a44a78b6221135961d809'
message.provenance.record_created.time.value = '2020-01-01'
updates.update_reaction(message)
self.assertEqual(message.reaction_id,
'ord-c0bbd41f095a44a78b6221135961d809')
self.assertLen(message.provenance.record_modified, 0)
def test_with_no_updates(self):
message = reaction_pb2.Reaction()
message.provenance.record_created.time.value = '2020-05-08'
message.reaction_id = 'ord-c0bbd41f095a44a78b6221135961d809'
copied = reaction_pb2.Reaction()
copied.CopyFrom(message)
updates.update_reaction(copied)
self.assertEqual(copied, message)
def test_with_resolve_names(self):
reaction = reaction_pb2.Reaction()
component = reaction.inputs['ethylamine'].components.add()
component.identifiers.add(type='NAME', value='ethylamine')
updates.update_reaction(reaction)
self.assertLen(component.identifiers, 2)
self.assertEqual(component.identifiers[1].value, 'CCN')
self.assertEqual(component.identifiers[1].type,
reaction_pb2.CompoundIdentifier.IdentifierType.SMILES)
self.assertRegex(component.identifiers[1].details, 'NAME resolved')
def test_with_resolve_names(self):
reaction = reaction_pb2.Reaction()
component = reaction.inputs['ethylamine'].components.add()
component.identifiers.add(type='NAME', value='ethylamine')
updates.update_reaction(reaction)
self.assertLen(component.identifiers, 2)
self.assertEqual(
component.identifiers[1],
reaction_pb2.CompoundIdentifier(
type='SMILES', value='CCN',
details='NAME resolved by PubChem'))
def main(argv):
del argv # Only used by app.run().
inputs = sorted(_get_inputs())
if not inputs:
logging.info('nothing to do')
return # Nothing to do.
datasets = {}
for file_status in inputs:
datasets[file_status.filename] = message_helpers.load_message(
file_status.filename, dataset_pb2.Dataset)
if FLAGS.validate:
validations.validate_datasets(datasets, FLAGS.write_errors)
if not FLAGS.update:
logging.info('nothing else to do; use --update for more')
return # Nothing else to do.
for dataset in datasets.values():
for reaction in dataset.reactions:
updates.update_reaction(reaction)
# Offload large Data values.
data_filenames = data_storage.extract_data(dataset,
FLAGS.root,
min_size=FLAGS.min_size,
max_size=FLAGS.max_size)
if data_filenames:
args = ['git', 'add'] + data_filenames
logging.info('Running command: %s', ' '.join(args))
subprocess.run(args, check=True)
combined = _combine_datasets(datasets)
# Final validation to make sure we didn't break anything.
validations.validate_datasets({'_COMBINED': combined}, FLAGS.write_errors)
if FLAGS.output:
output_filename = FLAGS.output
else:
_, suffix = os.path.splitext(inputs[0].filename)
output_filename = os.path.join(
FLAGS.root,
message_helpers.id_filename(f'{combined.dataset_id}{suffix}'))
os.makedirs(os.path.dirname(output_filename), exist_ok=True)
if FLAGS.cleanup:
cleanup(inputs, output_filename)
logging.info('writing combined Dataset to %s', output_filename)
message_helpers.write_message(combined, output_filename)
def test_add_reaction_id(self):
message = reaction_pb2.Reaction()
updates.update_reaction(message)
self.assertNotEmpty(message.reaction_id)
self.assertLen(message.provenance.record_modified, 1)
def test_with_updates_simple(self):
message = reaction_pb2.Reaction()
updates.update_reaction(message)
self.assertNotEqual(message, reaction_pb2.Reaction())
self.assertLen(message.provenance.record_modified, 1)
