Spark: Fix issue when partitioning by UUID #8250
Merged
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
18e15e5 to
60f39d1
Compare
Member
|
Which issue is this fixing? |
1355e85 to
3fe816b
Compare
nastra
commented
Aug 8, 2023
| required(116, "dec_38_10", Types.DecimalType.of(38, 10)) // maximum precision | ||
| ); | ||
| required(116, "dec_38_10", Types.DecimalType.of(38, 10)), // maximum precision | ||
| optional(117, "uuid", Types.UUIDType.get())); |
Contributor
Author
There was a problem hiding this comment.
this is being used by Spark tests across all versions, so we need to apply the fix for all Spark versions in a single PR unfortunately
hililiwei
reviewed
Aug 9, 2023
spark/v3.4/spark/src/main/java/org/apache/iceberg/spark/source/InternalRowWrapper.java
Show resolved
Hide resolved
Contributor
Author
|
@aokolnychyi could you take a look at this please? |
singhpk234
reviewed
Oct 14, 2023
|
|
||
| @SuppressWarnings("unchecked") | ||
| InternalRowWrapper(StructType rowType) { | ||
| InternalRowWrapper(StructType rowType, Types.StructType icebergStruct) { |
Contributor
There was a problem hiding this comment.
should we have a Pre-condition check that icebergStruct.fields().length is equal to types.lenght ?
spark/v3.3/spark/src/main/java/org/apache/iceberg/spark/source/InternalRowWrapper.java
Show resolved
Hide resolved
161538c to
ed424a9
Compare
singhpk234
approved these changes
Oct 17, 2023
Contributor
singhpk234
left a comment
There was a problem hiding this comment.
LGTM, Thanks @nastra !
amogh-jahagirdar
approved these changes
May 16, 2024
Contributor
amogh-jahagirdar
left a comment
There was a problem hiding this comment.
Sorry for the delayed review, overall I think the fix is good, just a naming nit.
spark/v3.3/spark/src/main/java/org/apache/iceberg/spark/source/InternalRowWrapper.java
Outdated
Show resolved
Hide resolved
Contributor
Author
|
thanks for the reviews @hililiwei, @singhpk234, @amogh-jahagirdar |
sasankpagolu
pushed a commit
to sasankpagolu/iceberg
that referenced
this pull request
Oct 27, 2024
zachdisc
pushed a commit
to zachdisc/iceberg
that referenced
this pull request
Dec 23, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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.
fixes #8247
The isse (as can be seen in #8247) is that
InternalRowWrapperuses the underlying Java class of a particular type (UUID.classin this case) to do a cast. Since Spark doesn't support UUIDs, they are mapped as strings and the cast fails.To solve the issue, we're passing the underlying Iceberg schema as struct to
InternalRowWrapper, so that we can convert to the right type before doing the cast. A similar approach is being done for Flink inRowDataWrapper.Unfortunately this required to fix this issue across all Spark versions, because
RecordWrapperTestis being used by all Spark versions. The fix itself is insideInternalRowWrapperand all other changes are due to passing the underlying Schema as struct toInternalRowWrapper.