[GitHub] [hudi] alexeykudinkin commented on a change in pull request #4789: [HUDI-1296] Support Metadata Table in Spark Datasource

[GitBox <gi...@apache.org>]

alexeykudinkin commented on a change in pull request #4789:
URL: https://github.com/apache/hudi/pull/4789#discussion_r805114615



##########
File path: hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieBaseRelation.scala
##########
@@ -41,15 +58,91 @@ abstract class HoodieBaseRelation(
 
   protected val sparkSession: SparkSession = sqlContext.sparkSession
 
-  protected val tableAvroSchema: Schema = {
+  protected lazy val tableAvroSchema: Schema = {
     val schemaUtil = new TableSchemaResolver(metaClient)
-    Try (schemaUtil.getTableAvroSchema).getOrElse(SchemaConverters.toAvroType(userSchema.get))
+    Try(schemaUtil.getTableAvroSchema).getOrElse(
+      // If there is no commit in the table, we can't get the schema
+      // t/h [[TableSchemaResolver]], fallback to provided the [[userSchema]] instead.
+      userSchema match {
+        case Some(s) => SchemaConverters.toAvroType(s)
+        case _ => throw new IllegalArgumentException("User-provided schema is required in case the table is empty")
+      }
+    )
   }
 
   protected val tableStructSchema: StructType = AvroConversionUtils.convertAvroSchemaToStructType(tableAvroSchema)
 
   protected val partitionColumns: Array[String] = metaClient.getTableConfig.getPartitionFields.orElse(Array.empty)
 
-  override def schema: StructType = userSchema.getOrElse(tableStructSchema)
+  override def schema: StructType = tableStructSchema
+}
+
+object HoodieBaseRelation {
+
+  def isMetadataTable(metaClient: HoodieTableMetaClient) =
+    HoodieTableMetadata.isMetadataTable(metaClient.getBasePath)
 
+  /**
+   * Returns file-reader routine accepting [[PartitionedFile]] and returning an [[Iterator]]
+   * over [[InternalRow]]
+   */
+  def createBaseFileReader(spark: SparkSession,
+                           tableSchemas: HoodieTableSchemas,
+                           filters: Array[Filter],
+                           options: Map[String, String],
+                           hadoopConf: Configuration): PartitionedFile => Iterator[InternalRow] = {
+    val hfileReader = createHFileReader(

Review comment:
       We can not b/c it have to be executed by the Driver




-- 
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: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org