[SPARK-49836][SQL][SS] Fix possibly broken query when window is provided to window/session_window fn #48309
Conversation
This file contains 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
…o window/session_window fn
|
https://dotty.epfl.ch/docs/reference/dropped-features/nonlocal-returns.html Going further, this is deprecated in newer Scala version (though it's like 3.2 which Spark wouldn't catch up for years) and will be removed. This clearly indicates that programmers have problems with this and they want to make this be much more explicit rather than just the same with local return. |
|
cc. @cloud-fan Would you mind take a look at this? Thanks! |
HeartSaVioR
commented
Oct 1, 2024
| @@ -87,85 +87,86 @@ object TimeWindowing extends Rule[LogicalPlan] { | |||
|
|
|||
| val window = windowExpressions.head | |||
|
|
|||
| // time window is provided as time column of window function, replace it with WindowTime | |||
There was a problem hiding this comment.
For reviewer: except this code comment, the only change is to remove return and replace it with a huge if-else statement.
HeartSaVioR
commented
Oct 1, 2024
| @@ -210,74 +211,74 @@ object SessionWindowing extends Rule[LogicalPlan] { | |||
| val session = sessionExpressions.head | |||
|
|
|||
| if (StructType.acceptsType(session.timeColumn.dataType)) { | |||
| return p transformExpressions { | |||
| p transformExpressions { | |||
There was a problem hiding this comment.
For reviewer: the only change is to remove return and replace it with a huge if-else statement.
|
cc. @viirya |
HeartSaVioR
changed the title
viirya
approved these changes
Oct 4, 2024
|
Thanks! Merging to master/3.5/3.4. |
HeartSaVioR
added a commit
that referenced
this pull request
Oct 4, 2024
…ded to window/session_window fn This PR fixes the correctness issue about losing operators during analysis - it happens when window is provided to window()/session_window() function. The rule `TimeWindowing` and `SessionWindowing` are responsible to resolve the time window functions. When the window function has `window` as parameter (time column) (in other words, building time window from time window), the rule wraps window with WindowTime function so that the rule ResolveWindowTime will further resolve this. (And TimeWindowing/SessionWindowing will resolve this again against the result of ResolveWindowTime.) The issue is that the rule uses "return" for the above, which intends to have "early return" as the other branch is too long compared to this branch. This unfortunately does not work as intended - the intention is just to go out of current local scope (mostly end of curly brace), but it seems to break the loop of execution in "outer" side. (I haven't debugged further but it's simply clear that it doesn't work as intended.) Quoting from Scala doc: > Nonlocal returns are implemented by throwing and catching scala.runtime.NonLocalReturnException-s. It's not super clear where NonLocalReturnException is caught in the call stack; it might exit the execution for much broader scope (context) than expected. And it's finally deprecated in Scala 3.2 and likely be removed in future. https://dotty.epfl.ch/docs/reference/dropped-features/nonlocal-returns.html Interestingly it does not break every query for chained time window aggregations. Spark already has several tests with DataFrame API and they haven't failed. The reproducer in community report is using SQL statement - where each aggregation is considered as subquery. This PR fixes the rule to NOT use early return and instead have a huge if else. Described in above. Yes, this fixes the possible query breakage. The impacted workloads may not be very huge as chained time window aggregations is an advanced usage, and it does not break every query for the usage. New UTs. No. Closes #48309 from HeartSaVioR/SPARK-49836. Lead-authored-by: Jungtaek Lim <[email protected]> Co-authored-by: Andrzej Zera <[email protected]> Signed-off-by: Jungtaek Lim <[email protected]> (cherry picked from commit d8c04cf) Signed-off-by: Jungtaek Lim <[email protected]>
HeartSaVioR
added a commit
that referenced
this pull request
Oct 5, 2024
…ded to window/session_window fn This PR fixes the correctness issue about losing operators during analysis - it happens when window is provided to window()/session_window() function. The rule `TimeWindowing` and `SessionWindowing` are responsible to resolve the time window functions. When the window function has `window` as parameter (time column) (in other words, building time window from time window), the rule wraps window with WindowTime function so that the rule ResolveWindowTime will further resolve this. (And TimeWindowing/SessionWindowing will resolve this again against the result of ResolveWindowTime.) The issue is that the rule uses "return" for the above, which intends to have "early return" as the other branch is too long compared to this branch. This unfortunately does not work as intended - the intention is just to go out of current local scope (mostly end of curly brace), but it seems to break the loop of execution in "outer" side. (I haven't debugged further but it's simply clear that it doesn't work as intended.) Quoting from Scala doc: > Nonlocal returns are implemented by throwing and catching scala.runtime.NonLocalReturnException-s. It's not super clear where NonLocalReturnException is caught in the call stack; it might exit the execution for much broader scope (context) than expected. And it's finally deprecated in Scala 3.2 and likely be removed in future. https://dotty.epfl.ch/docs/reference/dropped-features/nonlocal-returns.html Interestingly it does not break every query for chained time window aggregations. Spark already has several tests with DataFrame API and they haven't failed. The reproducer in community report is using SQL statement - where each aggregation is considered as subquery. This PR fixes the rule to NOT use early return and instead have a huge if else. Described in above. Yes, this fixes the possible query breakage. The impacted workloads may not be very huge as chained time window aggregations is an advanced usage, and it does not break every query for the usage. New UTs. No. Closes #48309 from HeartSaVioR/SPARK-49836. Lead-authored-by: Jungtaek Lim <[email protected]> Co-authored-by: Andrzej Zera <[email protected]> Signed-off-by: Jungtaek Lim <[email protected]> (cherry picked from commit d8c04cf) Signed-off-by: Jungtaek Lim <[email protected]>
himadripal
pushed a commit
to himadripal/spark
that referenced
this pull request
Oct 19, 2024
…ded to window/session_window fn ### What changes were proposed in this pull request? This PR fixes the correctness issue about losing operators during analysis - it happens when window is provided to window()/session_window() function. The rule `TimeWindowing` and `SessionWindowing` are responsible to resolve the time window functions. When the window function has `window` as parameter (time column) (in other words, building time window from time window), the rule wraps window with WindowTime function so that the rule ResolveWindowTime will further resolve this. (And TimeWindowing/SessionWindowing will resolve this again against the result of ResolveWindowTime.) The issue is that the rule uses "return" for the above, which intends to have "early return" as the other branch is too long compared to this branch. This unfortunately does not work as intended - the intention is just to go out of current local scope (mostly end of curly brace), but it seems to break the loop of execution in "outer" side. (I haven't debugged further but it's simply clear that it doesn't work as intended.) Quoting from Scala doc: > Nonlocal returns are implemented by throwing and catching scala.runtime.NonLocalReturnException-s. It's not super clear where NonLocalReturnException is caught in the call stack; it might exit the execution for much broader scope (context) than expected. And it's finally deprecated in Scala 3.2 and likely be removed in future. https://dotty.epfl.ch/docs/reference/dropped-features/nonlocal-returns.html Interestingly it does not break every query for chained time window aggregations. Spark already has several tests with DataFrame API and they haven't failed. The reproducer in community report is using SQL statement - where each aggregation is considered as subquery. This PR fixes the rule to NOT use early return and instead have a huge if else. ### Why are the changes needed? Described in above. ### Does this PR introduce _any_ user-facing change? Yes, this fixes the possible query breakage. The impacted workloads may not be very huge as chained time window aggregations is an advanced usage, and it does not break every query for the usage. ### How was this patch tested? New UTs. ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#48309 from HeartSaVioR/SPARK-49836. Lead-authored-by: Jungtaek Lim <[email protected]> Co-authored-by: Andrzej Zera <[email protected]> Signed-off-by: Jungtaek Lim <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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 fixes the correctness issue about losing operators during analysis - it happens when window is provided to window()/session_window() function.
The rule
TimeWindowingandSessionWindowingare responsible to resolve the time window functions. When the window function haswindowas parameter (time column) (in other words, building time window from time window), the rule wraps window with WindowTime function so that the rule ResolveWindowTime will further resolve this. (And TimeWindowing/SessionWindowing will resolve this again against the result of ResolveWindowTime.)The issue is that the rule uses "return" for the above, which intends to have "early return" as the other branch is too long compared to this branch. This unfortunately does not work as intended - the intention is just to go out of current local scope (mostly end of curly brace), but it seems to break the loop of execution in "outer" side.
(I haven't debugged further but it's simply clear that it doesn't work as intended.)
Quoting from Scala doc:
It's not super clear where NonLocalReturnException is caught in the call stack; it might exit the execution for much broader scope (context) than expected. And it's finally deprecated in Scala 3.2 and likely be removed in future.
https://dotty.epfl.ch/docs/reference/dropped-features/nonlocal-returns.html
Interestingly it does not break every query for chained time window aggregations. Spark already has several tests with DataFrame API and they haven't failed. The reproducer in community report is using SQL statement - where each aggregation is considered as subquery.
This PR fixes the rule to NOT use early return and instead have a huge if else.
Why are the changes needed?
Described in above.
Does this PR introduce any user-facing change?
Yes, this fixes the possible query breakage. The impacted workloads may not be very huge as chained time window aggregations is an advanced usage, and it does not break every query for the usage.
How was this patch tested?
New UTs.
Was this patch authored or co-authored using generative AI tooling?
No.