[Bug]: Beam SQL Extension raised an error when the input row contained iterable fields

[yyfhust]

What happened?

Upon utilizing the Beam SQL within our pipeline, we encountered a noteworthy exception. It appears that when the input row encompasses fields of iterable types, the process fails irrespective of whether the iterable fields are included in the SQL filter condition or not. This issue stems from the fact that the Beam SQL extension attempts to construct an output row schema based on the input schema, and unfortunately, it currently lacks support for iterable types.

Consider the following example:

Given an inputRow in the schema:

field1: String
field2: Integer
field3: Array<String>
field4: ITERABLE

And ANY Beam SQL condition such as :
field2 > 1
or Even
1 = 1

The pipeline will invariably fail, yielding the following error: Exception in thread "main" java.lang.UnsupportedOperationException: Unable to get ITERABLE at org.apache.beam.sdk.extensions.sql.impl.rel.BeamCalcRel$InputGetterImpl.getBeamField(BeamCalcRel.java:603).

Issue Priority

Priority: 2 (default / most bugs should be filed as P2)

Issue Components

• Component: Python SDK
• Component: Java SDK
• Component: Go SDK
• Component: Typescript SDK
• Component: IO connector
• Component: Beam YAML
• Component: Beam examples
• Component: Beam playground
• Component: Beam katas
• Component: Website
• Component: Spark Runner
• Component: Flink Runner
• Component: Samza Runner
• Component: Twister2 Runner
• Component: Hazelcast Jet Runner
• Component: Google Cloud Dataflow Runner