Repository navigation
[HIVE-30228]: Vectorized LEAD/LAG returns wrong default value when the default literal's type differs from the window column's type - #6933
Open
Indhumathi27 wants to merge 1 commit into
Conversation
Indhumathi27
force-pushed
the
HIVE-30228
branch
from
November 3, 2026 15:40
f9fd1bb to
c3215ea
Compare
…ral's type differs from the window column's type
Indhumathi27
force-pushed
the
HIVE-30228
branch
from
November 4, 2026 06:41
c3215ea to
aa16cba
Compare
|
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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?
Vectorized LEAD/LAG now correctly coerces the default value argument to the window column's type.
Why are the changes needed?
The default value in lead/lag wasn't being coerced to the column's type before vectorization — only the non-vectorized execution path handled that conversion at runtime. So a query like lag(mydate, 1, '1999-01-01') on a DATE column would read the string default incorrectly and silently return the wrong value (epoch 1970-01-01) instead of erroring or returning the correct default, causing vectorized and non-vectorized execution to disagree on the same query.
Does this PR introduce any user-facing change?
No
How was this patch tested?
Testcase added