dongjoon-hyun commented on a change in pull request #27728: [SPARK-17636][SQL]
Nested Column Predicate Pushdown for Parquet
URL: https://github.com/apache/spark/pull/27728#discussion_r387213594
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilterSuite.scala
##########
@@ -128,37 +134,47 @@ abstract class ParquetFilterSuite extends QueryTest with
ParquetTest with Shared
val ts3 = data(2)
val ts4 = data(3)
- withParquetDataFrame(data.map(i => Tuple1(i))) { implicit df =>
- checkFilterPredicate('_1.isNull, classOf[Eq[_]], Seq.empty[Row])
- checkFilterPredicate('_1.isNotNull, classOf[NotEq[_]], data.map(i =>
Row.apply(i)))
-
- checkFilterPredicate('_1 === ts1, classOf[Eq[_]], ts1)
- checkFilterPredicate('_1 <=> ts1, classOf[Eq[_]], ts1)
- checkFilterPredicate('_1 =!= ts1, classOf[NotEq[_]],
- Seq(ts2, ts3, ts4).map(i => Row.apply(i)))
-
- checkFilterPredicate('_1 < ts2, classOf[Lt[_]], ts1)
- checkFilterPredicate('_1 > ts1, classOf[Gt[_]], Seq(ts2, ts3, ts4).map(i
=> Row.apply(i)))
- checkFilterPredicate('_1 <= ts1, classOf[LtEq[_]], ts1)
- checkFilterPredicate('_1 >= ts4, classOf[GtEq[_]], ts4)
-
- checkFilterPredicate(Literal(ts1) === '_1, classOf[Eq[_]], ts1)
- checkFilterPredicate(Literal(ts1) <=> '_1, classOf[Eq[_]], ts1)
- checkFilterPredicate(Literal(ts2) > '_1, classOf[Lt[_]], ts1)
- checkFilterPredicate(Literal(ts3) < '_1, classOf[Gt[_]], ts4)
- checkFilterPredicate(Literal(ts1) >= '_1, classOf[LtEq[_]], ts1)
- checkFilterPredicate(Literal(ts4) <= '_1, classOf[GtEq[_]], ts4)
-
- checkFilterPredicate(!('_1 < ts4), classOf[GtEq[_]], ts4)
- checkFilterPredicate('_1 < ts2 || '_1 > ts3, classOf[Operators.Or],
Seq(Row(ts1), Row(ts4)))
- }
- }
-
- private def testDecimalPushDown(data: DataFrame)(f: DataFrame => Unit): Unit
= {
- withTempPath { file =>
- data.write.parquet(file.getCanonicalPath)
- readParquetFile(file.toString)(f)
- }
+ Seq(
+ (
+ spark.sqlContext.createDataFrame(data.map(x => ColA(Some(x)))),
Review comment:
Shall we use `spark` instead of `spark.sqlContext`?
`SQLContext.createDataFrame` is an alias for `SparkSession.createDataFrame`.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
