viirya commented on a change in pull request #33525:
URL: https://github.com/apache/spark/pull/33525#discussion_r676909418
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/jsonExpressions.scala
##########
@@ -561,7 +561,17 @@ case class JsonToStructs(
override def checkInputDataTypes(): TypeCheckResult = nullableSchema match {
case _: StructType | _: ArrayType | _: MapType =>
- super.checkInputDataTypes()
+ val checkMapKeyType = nullableSchema.existsRecursively( dataType =>
dataType match {
+ case MapType(keyType, _, _) if (keyType != StringType) => true
Review comment:
nit: `if (keyType != StringType)` => `if keyType != StringType`
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/jsonExpressions.scala
##########
@@ -561,7 +561,17 @@ case class JsonToStructs(
override def checkInputDataTypes(): TypeCheckResult = nullableSchema match {
case _: StructType | _: ArrayType | _: MapType =>
- super.checkInputDataTypes()
+ val checkMapKeyType = nullableSchema.existsRecursively( dataType =>
dataType match {
Review comment:
nit: `checkMapKeyType` => `invalidMapType`?
##########
File path:
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/JsonExpressionsSuite.scala
##########
@@ -853,4 +854,32 @@ class JsonExpressionsSuite extends SparkFunSuite with
ExpressionEvalHelper with
checkEvaluation(JsonObjectKeys(Literal(input)), expected)
}
}
+
+ def checkMapTypeKeyError(schema: DataType, jsonData: String): Unit = {
+ assert( JsonToStructs(schema, Map.empty,
Literal(jsonData)).checkInputDataTypes() match {
+ case TypeCheckFailure(_) => true
+ case _ => false
+ })
+ }
+
+ test("SPARK-35320: from_json should fail with a key type different of
StringType") {
+ checkMapTypeKeyError(MapType(IntegerType, StringType), """{"1": "test"}""")
+ }
+
+ test("SPARK-35320: from_json should fail with a key type different of
StringType " +
+ "anidated in a StructType") {
+ checkMapTypeKeyError(StructType(Seq(StructField("test",
MapType(IntegerType, StringType)))),
+ """"test": {"1": "test"}""")
+ }
+
+ test("SPARK-35320: from_json should fail with a key type different of
StringType " +
+ "anidated in a ArrayType") {
+ checkMapTypeKeyError(ArrayType(MapType(IntegerType, StringType)),
"""[{"1": "test"}]""")
+ }
+
+ test("SPARK-35320: from_json should fail with a key type different of
StringType " +
+ "anidated in a MapType") {
+ checkMapTypeKeyError(MapType(StringType, MapType(IntegerType, StringType)),
+ """{"key": {"1" : "test"}}""")
+ }
Review comment:
Looks like you can have just one test and run through the combination of
map types and json string.
##########
File path:
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/JsonExpressionsSuite.scala
##########
@@ -853,4 +854,32 @@ class JsonExpressionsSuite extends SparkFunSuite with
ExpressionEvalHelper with
checkEvaluation(JsonObjectKeys(Literal(input)), expected)
}
}
+
+ def checkMapTypeKeyError(schema: DataType, jsonData: String): Unit = {
+ assert( JsonToStructs(schema, Map.empty,
Literal(jsonData)).checkInputDataTypes() match {
Review comment:
`assert( JsonToStructs(...` => `assert(JsonToStructs(...`
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/jsonExpressions.scala
##########
@@ -561,7 +561,17 @@ case class JsonToStructs(
override def checkInputDataTypes(): TypeCheckResult = nullableSchema match {
case _: StructType | _: ArrayType | _: MapType =>
- super.checkInputDataTypes()
+ val checkMapKeyType = nullableSchema.existsRecursively( dataType =>
dataType match {
Review comment:
nit: `nullableSchema.existsRecursively( dataType` =>
`nullableSchema.existsRecursively(dataType`
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/jsonExpressions.scala
##########
@@ -561,7 +561,17 @@ case class JsonToStructs(
override def checkInputDataTypes(): TypeCheckResult = nullableSchema match {
case _: StructType | _: ArrayType | _: MapType =>
- super.checkInputDataTypes()
+ val checkMapKeyType = nullableSchema.existsRecursively( dataType =>
dataType match {
+ case MapType(keyType, _, _) if (keyType != StringType) => true
+ case _ => false
+ })
+ if(checkMapKeyType) {
Review comment:
style: `if(checkMapKeyType)` => `if (checkMapKeyType)`
--
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]
