Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 17 additions & 3 deletions sdks/python/apache_beam/pipeline.py
Original file line number Diff line number Diff line change
Expand Up @@ -995,12 +995,26 @@ def contains_external_transforms(pipeline):
pipeline.visit(visitor)
return visitor._contains_external_transforms

def visit_transform(self, transform_node):
# type: (AppliedPTransform) -> None
def _perform_exernal_transform_test(self, transform):
if not transform:
return
from apache_beam.transforms import ExternalTransform
if isinstance(transform_node.transform, ExternalTransform):
if isinstance(transform, ExternalTransform):
self._contains_external_transforms = True

def visit_transform(self, transform_node):
# type: (AppliedPTransform) -> None
self._perform_exernal_transform_test(transform_node.transform)

def enter_composite_transform(self, transform_node):
# type: (AppliedPTransform) -> None
# Python SDK object graph may represent an external transform that is a leaf
# of the pipeline graph as a composite without sub-transforms.
# Note that this visitor is just used to identify pipelines with external
# transforms. A Runner API pipeline proto generated from the Pipeline object
# will include external sub-transform.
self._perform_exernal_transform_test(transform_node.transform)


class AppliedPTransform(object):
"""For internal use only; no backwards-compatibility guarantees.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -56,6 +56,9 @@
TEST_PARTITION_URN = "beam:transforms:xlang:test:partition"
TEST_PYTHON_BS4_URN = "beam:transforms:xlang:test:python_bs4"

# A transform that does not produce an output.
TEST_NO_OUTPUT_URN = "beam:transforms:xlang:test:nooutput"


@ptransform.PTransform.register_urn('beam:transforms:xlang:count', None)
class CountPerElementTransform(ptransform.PTransform):
Expand Down Expand Up @@ -296,6 +299,23 @@ def from_runner_api_parameter(unused_ptransform, level, unused_context):
return FibTransform(int(level.decode('ascii')))


@ptransform.PTransform.register_urn(TEST_NO_OUTPUT_URN, None)
class NoOutputTransform(ptransform.PTransform):
def expand(self, pcoll):
def log_val(val):
logging.debug('Got value: %r', val)

# Logging without returning anything
_ = (pcoll | 'TestLabel' >> beam.ParDo(log_val))

def to_runner_api_parameter(self, unused_context):
return TEST_NO_OUTPUT_URN, None

@staticmethod
def from_runner_api_parameter(unused_ptransform, payload, unused_context):
return NoOutputTransform(parse_string_payload(payload)['data'])


def parse_string_payload(input_byte):
payload = ExternalConfigurationPayload()
payload.ParseFromString(input_byte)
Expand Down
27 changes: 27 additions & 0 deletions sdks/python/apache_beam/transforms/external_test.py
Original file line number Diff line number Diff line change
Expand Up @@ -259,6 +259,33 @@ def test_unique_name(self):
self.assertEqual(
len(set(pcolls)), len(pcolls), msg='PCollection names are not unique.')

def test_external_transform_finder_non_leaf(self):
pipeline = beam.Pipeline()
_ = (
pipeline
| beam.Create(['a', 'b'])
| beam.ExternalTransform(
'beam:transforms:xlang:test:prefix',
ImplicitSchemaPayloadBuilder({'data': u'0'}),
expansion_service.ExpansionServiceServicer())
| beam.Map(lambda x: x))
pipeline.run().wait_until_finish()

self.assertTrue(pipeline.contains_external_transforms)

def test_external_transform_finder_leaf(self):
pipeline = beam.Pipeline()
_ = (
pipeline
| beam.Create(['a', 'b'])
| beam.ExternalTransform(
'beam:transforms:xlang:test:nooutput',
ImplicitSchemaPayloadBuilder({'data': u'0'}),
expansion_service.ExpansionServiceServicer()))
pipeline.run().wait_until_finish()

self.assertTrue(pipeline.contains_external_transforms)


if __name__ == '__main__':
logging.getLogger().setLevel(logging.INFO)
Expand Down