Conversation
…narrowed commonTypeForBinaryComparison() picked whichever operand was approximate as the common type for a comparison against DECIMAL, narrowing the DECIMAL side to that operand's precision. A 32-bit REAL/FLOAT holds only ~7 significant digits, so a DECIMAL literal or column with more digits than that can silently collide with a different value after narrowing. For example, 59999943 and 59999945 both round to the same float. By widening the common type to DOUBLE whenever the exact-numeric operand is a DECIMAL, we don't lose precision. Updated ArrowAdapterTest's expected plan, which now casts a FLOAT column to DOUBLE when compared against a DECIMAL literal.
…ercion override Updated the design based on input from the JIRA so that users could opt in to thei behavior and avoid breaking changes for existing users. Extract the decision into a new protected AbstractTypeCoercion#approximateExactComparisonType(...) hook. A system that needs the precision-safe behavior can opt in with a small TypeCoercionFactory overriding this one method, the same pattern already used by TypeCoercionImpl and demonstrated by SqlToRelConverterTest#testNaturalJoinCastNoCoercion. Reworked the tests accordingly.
Contributor
|
This looks like a reasonable approach, I will review this |
mihaibudiu
approved these changes
Sep 29, 2026
|
mihaibudiu
added
the
LGTM-will-merge-soon
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
commonTypeForBinaryComparison() picked whichever operand was approximate as the common type for a comparison against DECIMAL, narrowing the DECIMAL side to that operand's precision.
A 32-bit REAL/FLOAT holds only ~7 significant digits, so a DECIMAL literal or column with more digits than that can silently collide with a different value after narrowing. For example, 59999943 and 59999945 both round to the same float.
By widening the common type to DOUBLE whenever the exact-numeric operand is a DECIMAL, we don't lose precision.
Updated ArrowAdapterTest's expected plan, which now casts a FLOAT column to DOUBLE when compared against a DECIMAL literal.
Jira Link
CALCITE-7827