srielau opened a new pull request, #58057: URL: https://github.com/apache/spark/pull/58057
### What changes were proposed in this pull request? **Depends on #58033** ([SPARK-58794](https://issues.apache.org/jira/browse/SPARK-58794) foundation). Please review/merge that first; rebase this onto `master` afterward. Delta vs foundation is a single commit: https://github.com/srielau/spark/compare/SPARK-58794...SPARK-58802 Expands HybridAnalyzer dual-run coverage under `spark.sql.charVarchar.standardSemantics.enabled` so fixed-point and single-pass analyzers are proven to agree on (D19): - CAST / try_cast type introduction (R3) - LCT: COALESCE / CASE / NULL / CHAR+VARCHAR / CHAR+STRING / IN / UNION / INTERSECT (R2) - R1 transforming exprs: upper/lower/`||`/concat/substr/trim/regexp_replace/mask/split, including R1 after LCT - Nested ARRAY/STRUCT of CHAR - Collated CAST - Bare column references from a CHAR/VARCHAR table The single-pass resolver has no Char/Varchar-specific logic; it inherits `Expression.dataType` and shared TypeCoercion. Dual-run with sample rate 1.0 is the proof of parity. ### Why are the changes needed? Parent design decision D19: single-pass must match fixed-point on LCT/CAST/R1 before calling the standardSemantics foundation complete. #58033 only had a small smoke test. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? ``` ./build/sbt "sql/testOnly org.apache.spark.sql.BasicCharVarcharTestSuite -- -z SPARK-58802" ``` Result: 1 test succeeded (dual-run matrix under standardSemantics). -- 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]
