seojangho commented on a change in pull request #114: [NEMO-203] Beam SQL
Aggregation fails to match RowCoder in Combine transform
URL: https://github.com/apache/incubator-nemo/pull/114#discussion_r214564039
##########
File path:
compiler/frontend/beam/src/main/java/edu/snu/nemo/compiler/frontend/beam/PipelineTranslator.java
##########
@@ -216,11 +221,23 @@ private static void topologicalTranslator(final
TranslationContext ctx,
private static void combineTranslator(final TranslationContext ctx,
final CompositeTransformVertex
transformVertex,
final PTransform<?, ?> transform) {
+ // No optimization for BeamSQL that handles Beam 'Row's.
+ final boolean handlesBeamRow = Stream
+ .concat(transformVertex.getNode().getInputs().values().stream(),
+ transformVertex.getNode().getOutputs().values().stream())
+ .map(pValue -> (KvCoder) getCoder(pValue, ctx.getPipeline())) // Input
and output of combine should be KV
Review comment:
Suggestion: Use ctx.pipeline instead of ctx.getPipeline
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
