[SPARK-31904][SQL] Fix case sensitive problem of char and varchar partition columns
### What changes were proposed in this pull request?
```sql
CREATE TABLE t1(a STRING, B VARCHAR(10), C CHAR(10)) STORED AS parquet;
CREATE TABLE t2 USING parquet PARTITIONED BY (b, c) AS SELECT * FROM t1;
SELECT * FROM t2 WHERE b = 'A';
```
Above SQL throws MetaException
> Caused by: java.lang.reflect.InvocationTargetException
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.spark.sql.hive.client.Shim_v0_13.getPartitionsByFilter(HiveShim.scala:810)
... 114 more
Caused by: MetaException(message:Filtering is supported only on partition keys of type string, or integral types)
at org.apache.hadoop.hive.metastore.parser.ExpressionTree$FilterBuilder.setError(ExpressionTree.java:184)
at org.apache.hadoop.hive.metastore.parser.ExpressionTree$LeafNode.getJdoFilterPushdownParam(ExpressionTree.java:439)
at org.apache.hadoop.hive.metastore.parser.ExpressionTree$LeafNode.generateJDOFilterOverPartitions(ExpressionTree.java:356)
at org.apache.hadoop.hive.metastore.parser.ExpressionTree$LeafNode.generateJDOFilter(ExpressionTree.java:278)
at org.apache.hadoop.hive.metastore.parser.ExpressionTree.generateJDOFilterFragment(ExpressionTree.java:583)
at org.apache.hadoop.hive.metastore.ObjectStore.makeQueryFilterString(ObjectStore.java:3315)
at org.apache.hadoop.hive.metastore.ObjectStore.getPartitionsViaOrmFilter(ObjectStore.java:2768)
at org.apache.hadoop.hive.metastore.ObjectStore.access$500(ObjectStore.java:182)
at org.apache.hadoop.hive.metastore.ObjectStore$7.getJdoResult(ObjectStore.java:3248)
at org.apache.hadoop.hive.metastore.ObjectStore$7.getJdoResult(ObjectStore.java:3232)
at org.apache.hadoop.hive.metastore.ObjectStore$GetHelper.run(ObjectStore.java:2974)
at org.apache.hadoop.hive.metastore.ObjectStore.getPartitionsByFilterInternal(ObjectStore.java:3250)
at org.apache.hadoop.hive.metastore.ObjectStore.getPartitionsByFilter(ObjectStore.java:2906)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.hadoop.hive.metastore.RawStoreProxy.invoke(RawStoreProxy.java:101)
at com.sun.proxy.$Proxy25.getPartitionsByFilter(Unknown Source)
at org.apache.hadoop.hive.metastore.HiveMetaStore$HMSHandler.get_partitions_by_filter(HiveMetaStore.java:5093)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.hadoop.hive.metastore.RetryingHMSHandler.invokeInternal(RetryingHMSHandler.java:148)
at org.apache.hadoop.hive.metastore.RetryingHMSHandler.invoke(RetryingHMSHandler.java:107)
at com.sun.proxy.$Proxy26.get_partitions_by_filter(Unknown Source)
at org.apache.hadoop.hive.metastore.HiveMetaStoreClient.listPartitionsByFilter(HiveMetaStoreClient.java:1232)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.hadoop.hive.metastore.RetryingMetaStoreClient.invoke(RetryingMetaStoreClient.java:173)
at com.sun.proxy.$Proxy27.listPartitionsByFilter(Unknown Source)
at org.apache.hadoop.hive.ql.metadata.Hive.getPartitionsByFilter(Hive.java:2679)
... 119 more
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Add a unit test.
Closes #28724 from LantaoJin/SPARK-31904.
Authored-by: LantaoJin <jinlantao@gmail.com>
Signed-off-by: Takeshi Yamamuro <yamamuro@apache.org>
(cherry picked from commit 5079831106ba22f81a26b4a8104a253422fa1b6a)
Signed-off-by: Takeshi Yamamuro <yamamuro@apache.org>
diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/client/HiveShim.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/client/HiveShim.scala
index 2b80660..8df43b7 100644
--- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/client/HiveShim.scala
+++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/client/HiveShim.scala
@@ -708,7 +708,8 @@
.map(col => col.getName).toSet
def unapply(attr: Attribute): Option[String] = {
- if (varcharKeys.contains(attr.name)) {
+ val resolver = SQLConf.get.resolver
+ if (varcharKeys.exists(c => resolver(c, attr.name))) {
None
} else if (attr.dataType.isInstanceOf[IntegralType] || attr.dataType == StringType) {
Some(attr.name)
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala
index e8548fd..e8cf4ad 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala
@@ -2720,4 +2720,14 @@
checkAnswer(sql("SHOW PARTITIONS ta_part"), Row("ts=10") :: Nil)
}
}
+
+ test("SPARK-31904: Fix case sensitive problem of char and varchar partition columns") {
+ withTable("t1", "t2") {
+ sql("CREATE TABLE t1(a STRING, B VARCHAR(10), C CHAR(10)) STORED AS parquet")
+ sql("CREATE TABLE t2 USING parquet PARTITIONED BY (b, c) AS SELECT * FROM t1")
+ // make sure there is no exception
+ assert(sql("SELECT * FROM t2 WHERE b = 'A'").collect().isEmpty)
+ assert(sql("SELECT * FROM t2 WHERE c = 'A'").collect().isEmpty)
+ }
+ }
}
