[FEAT] Support null equal safe join in SQL #3166
Conversation
advancedxy
changed the title
advancedxy
changed the title
CodSpeed Performance ReportMerging #3166 will not alter performanceComparing Summary
|
|
@advancedxy, #3161 is merged now, so feel free to ping me once this one has all conflicts resolved & ready for review! |
universalmind303
requested review from
universalmind303
and removed request for
universalmind303
advancedxy
force-pushed
the
null_equal_safe_sql_support
branch
from
06471a2
to
cb10930
Compare
advancedxy
force-pushed
the
null_equal_safe_sql_support
branch
from
cb10930
to
0278ad1
Compare
|
@universalmind303 I think it's ready for review once the CI passes. |
| @@ -41,6 +41,8 @@ impl OptimizerRule for EliminateCrossJoin { | |||
| LogicalPlan::Join(Join { | |||
| join_type: JoinType::Inner, | |||
| join_strategy: None, | |||
| // TODO: consider support eliminate cross join with null_equals_nulls | |||
| null_equals_nulls: None, | |||
There was a problem hiding this comment.
Seems that I didn't cover this in my previous PR. I found elimate_cross_join is not compatible with join with null equal safe yet when adding a test in the python side(a.k.a tests/sql/test_joins.py).
I simply disable this for now as it seems it might require a decent amount of work to support that.
| right_on: right_keys, | ||
| null_equals_nulls: None, |
There was a problem hiding this comment.
this is just style change, to align it with the struct definition.
advancedxy
force-pushed
the
null_equal_safe_sql_support
branch
from
0278ad1
to
f1e96f3
Compare
|
Thanks for reviewing and merging. It would be great if you could take a look at #3134 as well. |
One follow-up of #3161: add
on a <=> bsupport for SQL's Join operation.