diff options
author | Yijie Shen <henry.yijieshen@gmail.com> | 2015-06-17 23:46:57 -0700 |
---|---|---|
committer | Reynold Xin <rxin@databricks.com> | 2015-06-17 23:46:57 -0700 |
commit | e86fbdb1e6f1538f65ef78d90bbc41604f6bd580 (patch) | |
tree | 3fa7d726cf2eea3f9f97e94617e50f898204c4fb | |
parent | fee3438a32136a8edbca71efb566965587a88826 (diff) | |
download | spark-e86fbdb1e6f1538f65ef78d90bbc41604f6bd580.tar.gz spark-e86fbdb1e6f1538f65ef78d90bbc41604f6bd580.tar.bz2 spark-e86fbdb1e6f1538f65ef78d90bbc41604f6bd580.zip |
[SPARK-8283][SQL] Resolve udf_struct test failure in HiveCompatibilitySuite
This PR aimed to resolve udf_struct test failure in HiveCompatibilitySuite.
Currently, this is done by loosening CreateStruct's children type from NamedExpression to Expression and automatically generating StructField name for non-NamedExpression children.
The naming convention for unnamed children follows the udf's counterpart in Hive:
`col1, col2, col3, ...`
Author: Yijie Shen <henry.yijieshen@gmail.com>
Closes #6828 from yijieshen/SPARK-8283 and squashes the following commits:
6052b73 [Yijie Shen] Doc fix
677e0b7 [Yijie Shen] Resolve udf_struct test failure by automatically generate structField name for non-NamedExpression children
2 files changed, 10 insertions, 5 deletions
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/complexTypes.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/complexTypes.scala index 1aaf9b309e..72fdcebb4c 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/complexTypes.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/complexTypes.scala @@ -53,7 +53,7 @@ case class CreateArray(children: Seq[Expression]) extends Expression { * Returns a Row containing the evaluation of all children expressions. * TODO: [[CreateStruct]] does not support codegen. */ -case class CreateStruct(children: Seq[NamedExpression]) extends Expression { +case class CreateStruct(children: Seq[Expression]) extends Expression { override def foldable: Boolean = children.forall(_.foldable) @@ -62,9 +62,14 @@ case class CreateStruct(children: Seq[NamedExpression]) extends Expression { override lazy val dataType: StructType = { assert(resolved, s"CreateStruct contains unresolvable children: ${children.filterNot(_.resolved)}.") - val fields = children.map { child => - StructField(child.name, child.dataType, child.nullable, child.metadata) - } + val fields = children.zipWithIndex.map { case (child, idx) => + child match { + case ne: NamedExpression => + StructField(ne.name, ne.dataType, ne.nullable, ne.metadata) + case _ => + StructField(s"col${idx + 1}", child.dataType, child.nullable, Metadata.empty) + } + } StructType(fields) } diff --git a/sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala b/sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala index 432de2564d..f88e62763c 100644 --- a/sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala +++ b/sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala @@ -933,7 +933,7 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter { "udf_stddev_pop", "udf_stddev_samp", "udf_string", - // "udf_struct", TODO: FIX THIS and enable it. + "udf_struct", "udf_substring", "udf_subtract", "udf_sum", |