siying commented on PR #60032: URL: https://github.com/apache/spark/pull/60032#issuecomment-6595903252
Pushed commit `77185e030d4` with explicit complete input/output parity checks. The exact pairs are the inline batch reference and streaming sink output within each of these SQL files: - `available-now-generators.sql`: generator query - `available-now-having.sql`: HAVING query - `available-now-lateral-subquery.sql`: lateral subquery - `available-now-natural-join.sql`: natural join - `available-now-pipe-operators.sql`: pipe operators - `available-now-select-except.sql`: SELECT EXCEPT - `available-now-sql-functions.sql`: scalar-function and table-function flows - `available-now-static-subqueries.sql`: scalar, IN, and EXISTS subquery flows - `available-now-unnest.sql`: UNNEST - `available-now-unpivot.sql`: UNPIVOT - `available-now-using-join.sql`: stream-static and stream-stream USING joins For every pair, the batch and streaming paths read the same physical fixture tables, so the complete data input is identical by construction. Complete outputs are compared in both directions with `EXCEPT ALL`, preserving duplicate sensitivity. All 15 generated `complete_output_identical` results are `true`. To be precise, these are exact batch-vs-streaming pairs inside each listed file. They are not claims that an entire streaming SQL file is byte-for-byte identical to a separate legacy batch SQL file. Golden generation and normal verification both passed all 42 matching execution/analyzer tests. -- 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]
