Github user gatorsmile commented on a diff in the pull request:
https://github.com/apache/spark/pull/13496#discussion_r66725456
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
---
@@ -452,6 +452,17 @@ class Analyzer(
def apply(plan: LogicalPlan): LogicalPlan = plan resolveOperators {
case i @ InsertIntoTable(u: UnresolvedRelation, parts, child, _, _)
if child.resolved =>
+ // A partitioned relation's schema can be different from the input
logicalPlan, since
+ // partition columns are all moved after data columns. We Project
to adjust the ordering.
+ val input = if (parts.nonEmpty) {
+ val (inputPartCols, inputDataCols) = child.output.partition {
attr =>
+ parts.contains(attr.name)
+ }
+ Project(inputDataCols ++ inputPartCols, child)
+ } else {
+ child
+ }
--- End diff --
Let me show an example,
```SQL
INSERT OVERWRITE TABLE $table
partition (p1='a',p2)
SELECT 'blarr3' as p2, 'blarr1'
```
Without this change, the output is like
```
+------+---+------+
| c1| p1| p2|
+------+---+------+
|blarr3| a|blarr1|
+------+---+------+
```
After this change, the output becomes
```
+------+---+------+
| c1| p1| p2|
+------+---+------+
|blarr1| a|blarr3|
+------+---+------+
```
Based on the [Hive
specification](https://cwiki.apache.org/confluence/display/Hive/DynamicPartitions),
>>In INSERT ... SELECT ... queries, the dynamic partition columns must be
specified last among the columns in the SELECT statement and in the same order
in which they appear in the PARTITION() clause.
Now, the new behavior does not follow the Hive spec. I think we should not
do this.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
