wombatu-kun commented on code in PR #12772:
URL: https://github.com/apache/hudi/pull/12772#discussion_r2099986746
##########
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/CreateHoodieTableAsSelectCommand.scala:
##########
@@ -100,7 +101,8 @@ case class CreateHoodieTableAsSelectCommand(
HiveSyncConfigHolder.HIVE_TABLE_PROPERTIES.key ->
ConfigUtils.configToString(updatedTable.properties.asJava),
HoodieWriteConfig.COMBINE_BEFORE_INSERT.key -> "false",
DataSourceWriteOptions.SQL_INSERT_MODE.key ->
InsertMode.NON_STRICT.value(),
- DataSourceWriteOptions.SQL_ENABLE_BULK_INSERT.key -> "true"
+ DataSourceWriteOptions.SQL_ENABLE_BULK_INSERT.key -> "true",
+ "path" -> tablePath
Review Comment:
this method is called here:
```
case class HoodieFileIndex(spark: SparkSession,
metaClient: HoodieTableMetaClient,
schemaSpec: Option[StructType],
options: Map[String, String],
@transient fileStatusCache: FileStatusCache =
NoopCache,
includeLogFiles: Boolean = false,
shouldEmbedFileSlices: Boolean = false)
extends SparkHoodieTableFileIndex(
spark = spark,
metaClient = metaClient,
schemaSpec = schemaSpec,
configProperties = getConfigProperties(spark, options,
metaClient.getTableConfig),
queryPaths = HoodieFileIndex.getQueryPaths(options),
...) with FileIndex {
```
Is `glob.paths`-feature actual? If it's not, may be we can fix the test
without adding `path` option by changing to `queryPaths =
Seq(metaClient.getBasePath)`. Seems, i've seen some Jira about removing
`glob.paths` support, but i'm not sure.
--
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]
