AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-570447062
> LGTM, waiting for the updated benchmark result. Great job @AngersZh !
Updated all, thanks
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-570445224
**With current pr**
```
scala> (1 to 1).toDF("id").createOrReplaceTempView("s1")
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-570132270
@cloud-fan
Could you help to trigger retest?
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-570127420
> We don't need to add `Project(Nil, ...)` because we convert to `WHERE
(SELECT 1 FROM (subquery)
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-570112465
> @AngersZh any updates?
If use `ScalaSubquery` we can't add `Project(Nil,..)` because of
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-558476143
> For top-level limit, we use a special physical plan `CollectLimitExec`,
which doesn't have a
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-558463297
> @AngersZh after some more thoughts, I think we can still reuse
non-correlated scalar subquery.
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-558387275
> @AngersZh, I think there's probably some thing wrong with your latest
master merge...
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-558386503
> hmmm why are there so many conflicts?
Last merge master branch bring unexpected code,..
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-557636283
cc @cloud-fan
Simply look at the calculation process, the calculation of non-correlated
exists
AngersZh commented on issue #26437: [SPARK-29800][SQL] Plan non-correlated
Exists 's subquery in PlanSubqueries
URL: https://github.com/apache/spark/pull/26437#issuecomment-553415168
> @AngersZh I think it's pretty close now. Can we have some benchmark to
show the benefit of this
11 matches
Mail list logo