apilloud commented on a change in pull request #14729:
URL: https://github.com/apache/beam/pull/14729#discussion_r698828276
##########
File path:
sdks/java/extensions/sql/src/test/java/org/apache/beam/sdk/extensions/sql/impl/rule/BeamAggregateProjectMergeRuleTest.java
##########
@@ -59,9 +60,19 @@ public void buildUp() {
Table projectTable = getTable("TEST_PROJECT", PushDownOptions.PROJECT);
Table filterTable = getTable("TEST_FILTER", PushDownOptions.FILTER);
Table noneTable = getTable("TEST_NONE", PushDownOptions.NONE);
+
tableProvider.createTable(projectTable);
tableProvider.createTable(filterTable);
tableProvider.createTable(noneTable);
+
+ // Rules are cost based, need rows to optimize!
+ tableProvider.addRows(
+ "TEST_PROJECT", Row.withSchema(BASIC_SCHEMA).addValues(1, 2, "3",
4).build());
Review comment:
No, the row count just needs to be non-zero. Cost based optimization
uses the row count as the basis for optimization, if the row count is zero then
project and filter pushdown will never be applied. It is a multiply by zero
issue.
--
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]
