implement the except (distinct | all ) for table model #16760
Merged
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Outdated suggestions cannot be applied.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.
This PR implements the EXCEPT DISTINCT and EXCEPT ALL operators. This is achieved by introducing two new optimizer rules that translate the ExceptNode into a more fundamental execution plan:
(1)ImplementExceptDistinctAsUnion (EXCEPT DISTINCT)
This rule translates EXCEPT DISTINCT into a Union + Aggregation + Filter plan.
It utilizes SetOperationNodeTranslator to append marker columns, then uses Aggregation (GROUP BY) to count the occurrences of each row in each sub-relation.
Finally, a FilterNode selects rows that appear in the first relation (count1 >= 1) but not in any of the subsequent relations (count2...n == 0).
(2)ImplementExceptAllAsUnion (EXCEPT ALL)
This rule translates EXCEPT ALL into a Union + Window + Filter plan.
It uses a Window function to compute the exact occurrence count for each row from each sub-relation (count1, count2...) along with a ROW_NUMBER().
A FilterNode then applies the expression row_number <= greatest(greatest(count1 - count2, 0) - count3...) to precisely implement the bag semantics subtraction required by EXCEPT ALL.
The SetOperationNodeTranslator class contains the core, shared logic used by both translation rules.