Github user TomaszGaweda commented on a diff in the pull request:
https://github.com/apache/spark/pull/21875#discussion_r205227335
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/jdbc/JDBCRelation.scala
---
@@ -172,7 +172,11 @@ private[sql] case class JDBCRelation(
// Check if JDBCRDD.compileFilter can accept input filters
override def unhandledFilters(filters: Array[Filter]): Array[Filter] = {
- filters.filter(JDBCRDD.compileFilter(_,
JdbcDialects.get(jdbcOptions.url)).isEmpty)
+ if (jdbcOptions.pushDownPredicate) {
--- End diff --
Are you sure, that this is the only place? JDBCRDD.scanTable defines
filters as all filters that may be pushed down. Probably we should use `filters
-- unhandledFilters` in JdbcRelation.buildScan
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
