wombatu-kun commented on PR #16346:
URL: https://github.com/apache/iceberg/pull/16346#issuecomment-4461002055
Good call, thanks @pvary — done in 6d2f77785.
Rather than a Flink-only test I added a `StructWithDecimals` generator
(`decimal(9,2)` / `(15,3)` / `(38,10)`, covering the Parquet INT32 / INT64 /
FIXED_LEN_BYTE_ARRAY encodings) to `DataGenerators.ALL`, so decimals now flow
through the whole `BaseFormatModelTests` matrix for Flink **and** Spark across
Avro/Parquet/ORC.
One gap I hit while doing this: the generator-parameterized cases were only
`engineWrite→genericRead` and `genericWrite→engineRead`, so the
engine-writer→engine-reader roundtrip the original TODO asked for
("write-read-validate decimal via FlinkParquetWrite/Reader") wasn't actually
covered by any single case. I added a `testDataWriterEngineWriteEngineRead`
case to `BaseFormatModelTests` to close that symmetry — for all
types/engines/formats, not just decimals.
The duplicated `TestFlinkParquetReader` decimal test is removed from
v1.20/v2.0/v2.1; the `FlinkParquetReaders` TODO stays resolved by the broader
shared coverage. Verified green: `TestFlinkFormatModel` (Flink 1.20/2.0/2.1)
and `TestSparkFormatModel` (Spark 4.1).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
