[
https://issues.apache.org/jira/browse/SPARK-30353?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ulysses you updated SPARK-30353:
--------------------------------
Description:
Now Spark can propagate constraint during sql optimization when
`spark.sql.constraintPropagation.enabled` is true, then `where c = 1` will
convert to `where c = 1 and c is not null`. We also can use constraint in
`SimplifyBinaryComparison`.
`SimplifyBinaryComparison` will simplify expression which is not nullable and
semanticEquals. And we also can simplify if one expression is infered
`IsNotNull`.
For e.g.
```
create table test (c1 string);
explain extended select c1 from test where c1 = c1 limit 10;
-- before
GlobalLimit 10
+- LocalLimit 10
+- Filter (isnotnull(c1#20) AND (c1#20 = c1#20))
+- Relation[c1#20]
-- after
GlobalLimit 10
+- LocalLimit 10
+- Filter (isnotnull(c1#20)
+- Relation[c1#20]
explain extended select c1 from test where c1 > c1 limit 10;
-- before
GlobalLimit 10
+- LocalLimit 10
+- Filter (isnotnull(c1#20) && (c1#20 > c1#20))
+- Relation[c1#20]
-- after
LocalRelation <empty>, [c1#20]
```
was:
Now Spark can propagate constraint during sql optimization when
`spark.sql.constraintPropagation.enabled` is true, then `where c = 1` will
convert to `where c = 1 and c is not null`. We also can use constraint in
`SimplifyBinaryComparison`.
`SimplifyBinaryComparison` will simplify expression which is not nullable and
semanticEquals. And we also can simplify if one expression is infered
`IsNotNull`.
For e.g.
```
create table test (c1 string);
explain extended select c1 from test where c1 = c1 limit 10;
-- before
GlobalLimit 10
+- LocalLimit 10
+- Filter (isnotnull(c1#20) AND (c1#20 = c1#20))
+- Relation[c1#20]
-- after
GlobalLimit 10
+- LocalLimit 10
+- Relation[c1#20]
explain extended select c1 from test where c1 > c1 limit 10;
-- before
GlobalLimit 10
+- LocalLimit 10
+- Filter (isnotnull(c1#20) && (c1#20 > c1#20))
+- Relation[c1#20]
-- after
LocalRelation <empty>, [c1#20]
```
> Use constraints in SimplifyBinaryComparison optimization
> --------------------------------------------------------
>
> Key: SPARK-30353
> URL: https://issues.apache.org/jira/browse/SPARK-30353
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 3.0.0
> Reporter: ulysses you
> Priority: Minor
>
> Now Spark can propagate constraint during sql optimization when
> `spark.sql.constraintPropagation.enabled` is true, then `where c = 1` will
> convert to `where c = 1 and c is not null`. We also can use constraint in
> `SimplifyBinaryComparison`.
> `SimplifyBinaryComparison` will simplify expression which is not nullable and
> semanticEquals. And we also can simplify if one expression is infered
> `IsNotNull`.
> For e.g.
> ```
> create table test (c1 string);
> explain extended select c1 from test where c1 = c1 limit 10;
> -- before
> GlobalLimit 10
> +- LocalLimit 10
> +- Filter (isnotnull(c1#20) AND (c1#20 = c1#20))
> +- Relation[c1#20]
> -- after
> GlobalLimit 10
> +- LocalLimit 10
> +- Filter (isnotnull(c1#20)
> +- Relation[c1#20]
> explain extended select c1 from test where c1 > c1 limit 10;
> -- before
> GlobalLimit 10
> +- LocalLimit 10
> +- Filter (isnotnull(c1#20) && (c1#20 > c1#20))
> +- Relation[c1#20]
> -- after
> LocalRelation <empty>, [c1#20]
> ```
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
