[SPARK-32018][FOLLOWUP][Doc] Add migration guide for decimal value overflow in sum aggregation #29458
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
|
Test build #127528 has finished for PR 29458 at commit
|
cloud-fan
reviewed
Aug 18, 2020
docs/sql-migration-guide.md
Outdated
| @@ -36,6 +36,10 @@ license: | | |||
|
|
|||
| - In Spark 3.1, NULL elements of structures, arrays and maps are converted to "null" in casting them to strings. In Spark 3.0 or earlier, NULL elements are converted to empty strings. To restore the behavior before Spark 3.1, you can set `spark.sql.legacy.castComplexTypesToString.enabled` to `true`. | |||
|
|
|||
| - In Spark 3.1, when `spark.sql.ansi.enabled` is false, sum aggregation of decimal type column always returns `null` on decimal value overflow. In Spark 3.0 or earlier, when `spark.sql.ansi.enabled` is false and decimal value overflow happens in sum aggregation of decimal type column: | |||
| - If it is hash aggregation with `group by` clause, a runtime exception is thrown. | |||
There was a problem hiding this comment.
not many users know the physical nodes. How about
In Spark 3.1, Spark always returns null if the sum of decimal overflows under non-ANSI
mode (`spark.sql.ansi.enabled` is false). In Spark 3.0 or earlier, the sum of decimal may
fail at runtime under non-ANSI mode (when the query has GROUP BY and is planned as hash aggregate)
cloud-fan
reviewed
Aug 18, 2020
docs/sql-migration-guide.md
Outdated
| @@ -36,6 +36,10 @@ license: | | |||
|
|
|||
| - In Spark 3.1, NULL elements of structures, arrays and maps are converted to "null" in casting them to strings. In Spark 3.0 or earlier, NULL elements are converted to empty strings. To restore the behavior before Spark 3.1, you can set `spark.sql.legacy.castComplexTypesToString.enabled` to `true`. | |||
|
|
|||
| - In Spark 3.1, when `spark.sql.ansi.enabled` is false, sum aggregation of decimal type column always returns `null` on decimal value overflow. In Spark 3.0 or earlier, when `spark.sql.ansi.enabled` is false and decimal value overflow happens in sum aggregation of decimal type column: | |||
| - If it is hash aggregation with `group by` clause, a runtime exception is thrown. | |||
There was a problem hiding this comment.
not many users know the physical nodes. How about
In Spark 3.1, Spark always returns null if the sum of decimal overflows under non-ANSI
mode (`spark.sql.ansi.enabled` is false). In Spark 3.0 or earlier, the sum of decimal may
fail at runtime under non-ANSI mode (when the query has GROUP BY and is planned as hash aggregate)
There was a problem hiding this comment.
The name "non-ANSI mode" is a bit wired.
Also, we have to mention that Spark 3.0 or earlier returns null under certain conditions.
There was a problem hiding this comment.
We can use "default mode".
I don't see a difference between "may fail at runtime" or "may return null". They are mutually exclusive.
There was a problem hiding this comment.
Thanks, I have updated the doc and screenshot
|
Test build #127559 has finished for PR 29458 at commit
|
maropu
reviewed
Aug 18, 2020
| @@ -36,6 +36,8 @@ license: | | |||
|
|
|||
| - In Spark 3.1, NULL elements of structures, arrays and maps are converted to "null" in casting them to strings. In Spark 3.0 or earlier, NULL elements are converted to empty strings. To restore the behavior before Spark 3.1, you can set `spark.sql.legacy.castComplexTypesToString.enabled` to `true`. | |||
|
|
|||
| - In Spark 3.1, when `spark.sql.ansi.enabled` is false, Spark always returns null if the sum of decimal type column overflows. In Spark 3.0 or earlier, when `spark.sql.ansi.enabled` is false, the sum of decimal type column may return null or incorrect result, or even fails at runtime (depending on the actual query plan execution). | |||
|
|
|||
There was a problem hiding this comment.
nit: we need to describe spark.sql.ansi.enabled is false two times? I think its okay just to describe it like this;
In Spark 3.0 or earlier, the sum of...
or
In Spark 3.0 or earlier, in the case, the sum of...
maropu
approved these changes
Aug 18, 2020
|
Merging to master |
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?
Add migration guide for decimal value overflow behavior in sum aggregation, introduced in #29026
Why are the changes needed?
Add migration guide for the behavior changes from 3.0 to 3.1.
See also: #29450 (comment)
Does this PR introduce any user-facing change?
No
How was this patch tested?
Build docs and preview:
