Github user jiangxb1987 commented on a diff in the pull request:
https://github.com/apache/spark/pull/16168#discussion_r91234171
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/command/views.scala ---
@@ -207,31 +205,56 @@ case class CreateViewCommand(
}
/**
- * Returns a [[CatalogTable]] that can be used to save in the catalog.
This comment canonicalize
- * SQL based on the analyzed plan, and also creates the proper schema
for the view.
+ * Returns a [[CatalogTable]] that can be used to save in the catalog.
This stores the following
+ * properties for a view:
+ * 1. The `viewText` which is used to generate a logical plan when we
resolve a view;
+ * 2. The `currentDatabase` which sets the current database on Analyze
stage;
+ * 3. The `schema` which ensure we generate the correct output.
*/
private def prepareTable(sparkSession: SparkSession, aliasedPlan:
LogicalPlan): CatalogTable = {
- val viewSQL: String = new SQLBuilder(aliasedPlan).toSQL
+ val currentDatabase =
sparkSession.sessionState.catalog.getCurrentDatabase
- // Validate the view SQL - make sure we can parse it and analyze it.
- // If we cannot analyze the generated query, there is probably a bug
in SQL generation.
- try {
- sparkSession.sql(viewSQL).queryExecution.assertAnalyzed()
- } catch {
- case NonFatal(e) =>
- throw new RuntimeException(s"Failed to analyze the canonicalized
SQL: $viewSQL", e)
- }
+ if (originalText.isDefined) {
--- End diff --
I should add comments over this code. The `originalText` is non-empty only
if the command is generated from SQL.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
