[SPARK-14019] [SQL] Remove noop SortOrder in Sort #11840
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| * Removes no-op SortOrder from Sort | ||
| */ | ||
| object PruneSorts extends Rule[LogicalPlan] { | ||
| def apply(plan: LogicalPlan): LogicalPlan = plan transform { |
Contributor
There was a problem hiding this comment.
can we just test whether the expressions are constant foldable?
Member
Author
There was a problem hiding this comment.
Yeah, we can do it. Thanks!
Contributor
|
Can you change the title to SPARK instead of Spark? |
Contributor
|
I will let @cloud-fan review and merge this. |
|
Test build #53605 has finished for PR 11840 at commit
|
|
Test build #53608 has finished for PR 11840 at commit
|
|
Test build #53610 has finished for PR 11840 at commit
|
gatorsmile
changed the title
Member
Author
|
Thank you! |
Contributor
|
LGTM, thanks, merging to master! |
roygao94
pushed a commit
to roygao94/spark
that referenced
this pull request
Mar 22, 2016
#### What changes were proposed in this pull request?
This PR is to add a new Optimizer rule for pruning Sort if its SortOrder is no-op. In the phase of **Optimizer**, if a specific `SortOrder` does not have any reference, it has no effect on the sorting results. If `Sort` is empty, remove the whole `Sort`.
For example, in the following SQL query
```SQL
SELECT * FROM t ORDER BY NULL + 5
```
Before the fix, the plan is like
```
== Analyzed Logical Plan ==
a: int, b: int
Sort [(cast(null as int) + 5) ASC], true
+- Project [a#92,b#93]
+- SubqueryAlias t
+- Project [_1#89 AS a#92,_2#90 AS b#93]
+- LocalRelation [_1#89,_2#90], [[1,2],[1,2]]
== Optimized Logical Plan ==
Sort [null ASC], true
+- LocalRelation [a#92,b#93], [[1,2],[1,2]]
== Physical Plan ==
WholeStageCodegen
: +- Sort [null ASC], true, 0
: +- INPUT
+- Exchange rangepartitioning(null ASC, 5), None
+- LocalTableScan [a#92,b#93], [[1,2],[1,2]]
```
After the fix, the plan is like
```
== Analyzed Logical Plan ==
a: int, b: int
Sort [(cast(null as int) + 5) ASC], true
+- Project [a#92,b#93]
+- SubqueryAlias t
+- Project [_1#89 AS a#92,_2#90 AS b#93]
+- LocalRelation [_1#89,_2#90], [[1,2],[1,2]]
== Optimized Logical Plan ==
LocalRelation [a#92,b#93], [[1,2],[1,2]]
== Physical Plan ==
LocalTableScan [a#92,b#93], [[1,2],[1,2]]
```
cc rxin cloud-fan marmbrus Thanks!
#### How was this patch tested?
Added a test suite for covering this rule
Author: gatorsmile <[email protected]>
Closes apache#11840 from gatorsmile/sortElimination.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
This PR is to add a new Optimizer rule for pruning Sort if its SortOrder is no-op. In the phase of Optimizer, if a specific
SortOrderdoes not have any reference, it has no effect on the sorting results. IfSortis empty, remove the wholeSort.For example, in the following SQL query
Before the fix, the plan is like
After the fix, the plan is like
cc @rxin @cloud-fan @marmbrus Thanks!
How was this patch tested?
Added a test suite for covering this rule