rok commented on code in PR #16738:
URL: https://github.com/apache/datafusion/pull/16738#discussion_r2294031511


##########
datafusion/datasource-parquet/src/file_format.rs:
##########
@@ -1723,28 +1708,47 @@ async fn output_single_parquet_file_parallelized(
     let (serialize_tx, serialize_rx) =
         mpsc::channel::<SpawnedTask<RBStreamSerializeResult>>(max_rowgroups);
 
+    let parquet_schema = ArrowSchemaConverter::new()
+        .with_coerce_types(parquet_props.coerce_types())
+        .convert(&output_schema)?;
+    let merged_buff = SharedBuffer::new(INITIAL_BUFFER_BYTES);
+    let parquet_writer = SerializedFileWriter::new(
+        merged_buff.clone(),
+        parquet_schema.root_schema_ptr(),
+        parquet_props.clone().into(),
+    )?;
+    let arrow_row_group_writer_factory = ArrowRowGroupWriterFactory::new(
+        &parquet_writer,
+        parquet_schema,
+        Arc::clone(&output_schema),
+        parquet_props.clone().into(),
+    );
+
     let arc_props = Arc::new(parquet_props.clone());
     let launch_serialization_task = spawn_parquet_parallel_serialization_task(
+        Arc::new(arrow_row_group_writer_factory),
         data,
         serialize_tx,
         Arc::clone(&output_schema),
         Arc::clone(&arc_props),
         parallel_options,
         Arc::clone(&pool),
     );
+
+    launch_serialization_task
+        .join_unwind()
+        .await
+        .map_err(|e| DataFusionError::ExecutionJoin(Box::new(e)))??;
+
     let file_metadata = concatenate_parallel_row_groups(
+        parquet_writer,
+        merged_buff,
         serialize_rx,
-        Arc::clone(&output_schema),
-        Arc::clone(&arc_props),
         object_store_writer,
         pool,
     )
     .await?;
 
-    launch_serialization_task
-        .join_unwind()
-        .await
-        .map_err(|e| DataFusionError::ExecutionJoin(Box::new(e)))??;
     Ok(file_metadata)
 }
 

Review Comment:
   I feel these were mostly addressed or the code was significantly refactored 
so they might not apply anymore.



-- 
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: github-unsubscr...@datafusion.apache.org

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


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org

Reply via email to