houqp commented on a change in pull request #605:
URL: https://github.com/apache/arrow-datafusion/pull/605#discussion_r663445283
##########
File path: datafusion/src/logical_plan/builder.rs
##########
@@ -440,65 +424,19 @@ impl LogicalPlanBuilder {
pub fn build_join_schema(
left: &DFSchema,
right: &DFSchema,
- on: &[(Column, Column)],
join_type: &JoinType,
- join_constraint: &JoinConstraint,
) -> Result<DFSchema> {
let fields: Vec<DFField> = match join_type {
- JoinType::Inner | JoinType::Left | JoinType::Full => {
- let duplicate_keys = match join_constraint {
- JoinConstraint::On => on
- .iter()
- .filter(|(l, r)| l == r)
- .map(|on| on.1.clone())
- .collect::<HashSet<_>>(),
- // using join requires unique join columns in the output
schema, so we mark all
- // right join keys as duplicate
- JoinConstraint::Using => {
- on.iter().map(|on| on.1.clone()).collect::<HashSet<_>>()
- }
- };
-
+ JoinType::Inner | JoinType::Left | JoinType::Full | JoinType::Right =>
{
Review comment:
Join schemas are now consistent across all join types and constraints.
We implement the join column "merge" semantic externally in plan builder and
optimizer.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
