yihua commented on code in PR #12584:
URL: https://github.com/apache/hudi/pull/12584#discussion_r1913967820
##########
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/MergeIntoHoodieTableCommand.scala:
##########
@@ -176,9 +176,9 @@ case class MergeIntoHoodieTableCommand(mergeInto:
MergeIntoTable) extends Hoodie
expressionSet.remove((attr, expr))
(attr, expr)
}
- if (resolving.isEmpty && rk._1.equals("primaryKey")
+ if (resolving.isEmpty && rk._1.equals("recordKey")
Review Comment:
Same here.
##########
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/MergeIntoHoodieTableCommand.scala:
##########
@@ -137,19 +137,19 @@ case class MergeIntoHoodieTableCommand(mergeInto:
MergeIntoTable) extends Hoodie
if (primaryKeyFields.isPresent) {
//pkless tables can have more complex conditions
if (!conditions.forall(p => p.isInstanceOf[EqualTo])) {
- throw new AnalysisException(s"Currently only equality predicates are
supported in MERGE INTO statement on primary key table" +
+ throw new AnalysisException(s"Currently only equality predicates are
supported in MERGE INTO statement on record key table" +
s"(provided ${mergeInto.mergeCondition.sql}")
}
}
val resolver = sparkSession.sessionState.analyzer.resolver
val partitionPathFields = hoodieCatalogTable.tableConfig.getPartitionFields
- //ensure all primary key fields are part of the merge condition
+ //ensure all record key fields are part of the merge condition
//allow partition path to be part of the merge condition but not required
val targetAttr2ConditionExpressions = doCasting(conditions,
primaryKeyFields.isPresent)
val expressionSet = scala.collection.mutable.Set[(Attribute,
Expression)](targetAttr2ConditionExpressions:_*)
var partitionAndKeyFields: Seq[(String,String)] = Seq.empty
if (primaryKeyFields.isPresent) {
- partitionAndKeyFields = partitionAndKeyFields ++
primaryKeyFields.get().map(pk => ("primaryKey", pk)).toSeq
+ partitionAndKeyFields = partitionAndKeyFields ++
primaryKeyFields.get().map(pk => ("recordKey", pk)).toSeq
Review Comment:
This has to be kept the same to avoid confusion on the reserved word for SQL.
##########
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/MergeIntoHoodieTableCommand.scala:
##########
@@ -489,6 +479,24 @@ case class MergeIntoHoodieTableCommand(mergeInto:
MergeIntoTable) extends Hoodie
}
}
+ private def isPartialUpdateActionForMOR(parameters: Map[String, String]) = {
+ (targetTableType == MOR_TABLE_TYPE_OPT_VAL
+ && UPSERT_OPERATION_OPT_VAL == getOperationType(parameters)
+ && parameters.getOrElse(
+ ENABLE_MERGE_INTO_PARTIAL_UPDATES.key,
+ ENABLE_MERGE_INTO_PARTIAL_UPDATES.defaultValue.toString).toBoolean
+ && updatingActions.nonEmpty)
+ }
+
+ private def getOperationType(parameters: Map[String, String]) = {
+ val operation = if
(StringUtils.isNullOrEmpty(parameters.getOrElse(PRECOMBINE_FIELD.key, "")) &&
updatingActions.isEmpty) {
+ INSERT_OPERATION_OPT_VAL
+ } else {
+ UPSERT_OPERATION_OPT_VAL
+ }
+ operation
Review Comment:
```suggestion
if (StringUtils.isNullOrEmpty(parameters.getOrElse(PRECOMBINE_FIELD.key,
"")) && updatingActions.isEmpty) {
INSERT_OPERATION_OPT_VAL
} else {
UPSERT_OPERATION_OPT_VAL
}
```
##########
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/MergeIntoHoodieTableCommand.scala:
##########
@@ -31,10 +31,11 @@ import org.apache.hudi.exception.{HoodieException,
HoodieNotSupportedException}
import org.apache.hudi.hive.HiveSyncConfigHolder
import org.apache.hudi.sync.common.HoodieSyncConfig
import org.apache.hudi.util.JFunction.scalaFunction1Noop
-
import org.apache.avro.Schema
+import org.apache.hudi.common.table.HoodieTableConfig
Review Comment:
nit: need to group hudi imports together. Check your scala import layout in
IntelliJ
<img width="680" alt="Screenshot 2025-01-13 at 16 36 39"
src="https://github.com/user-attachments/assets/291272a9-de64-4919-8235-128f694398de";
/>
--
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]
