Github user jiangxb1987 commented on a diff in the pull request:
https://github.com/apache/spark/pull/20397#discussion_r164347650
--- Diff:
sql/core/src/main/java/org/apache/spark/sql/sources/v2/reader/SupportsScanColumnarBatch.java
---
@@ -30,21 +30,21 @@
@InterfaceStability.Evolving
public interface SupportsScanColumnarBatch extends DataSourceV2Reader {
@Override
- default List<ReadTask<Row>> createReadTasks() {
+ default List<DataReaderFactory<Row>> createDataReaderFactories() {
--- End diff --
I would also mention this:
```
data reader creation must be done at executor side
```
so it makes it more clear why we shall have a list of `DataReaderFactory`s,
but it's only my personal opinions, it's totally fine if we don't include the
implementation details in the comment.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
