gene-bordegaray commented on code in PR #23680:
URL: https://github.com/apache/datafusion/pull/23680#discussion_r3632181118
##########
datafusion/physical-expr/src/partitioning.rs:
##########
@@ -768,320 +811,114 @@ mod tests {
}
#[test]
- fn test_partitioning_satisfy_by_subset() -> Result<()> {
+ fn test_partitioning_key_satisfaction() -> Result<()> {
let fixture = PartitioningTestFixture::int64(&["a", "b", "c"])?;
+ let unknown: Arc<dyn PhysicalExpr> =
Arc::new(UnKnownColumn::new("dropped"));
let test_cases = vec![
(
- "KeyPartitioned([a, b]) satisfied by Hash([a])",
- fixture.hash_partitioning([0], 4),
+ "exact: KeyPartitioned([a, b]) satisfied by Hash([a, b])",
+ fixture.hash_partitioning([0, 1], 4),
fixture.key_distribution([0, 1]),
- PartitioningSatisfaction::Subset,
- PartitioningSatisfaction::NotSatisfied,
+ PartitioningSatisfaction::Exact,
+ PartitioningSatisfaction::Exact,
),
(
- "KeyPartitioned([a, b, c]) satisfied by Hash([a])",
+ "subset: KeyPartitioned([a, b]) satisfied by Hash([a])",
fixture.hash_partitioning([0], 4),
- fixture.key_distribution([0, 1, 2]),
- PartitioningSatisfaction::Subset,
- PartitioningSatisfaction::NotSatisfied,
- ),
- (
- "KeyPartitioned([a, b, c]) satisfied by Hash([a, b])",
- fixture.hash_partitioning([0, 1], 4),
- fixture.key_distribution([0, 1, 2]),
+ fixture.key_distribution([0, 1]),
PartitioningSatisfaction::Subset,
PartitioningSatisfaction::NotSatisfied,
),
(
- "KeyPartitioned([a, b, c]) satisfied by Hash([b])",
+ "subset: KeyPartitioned([a, b, c]) satisfied by Hash([b])",
fixture.hash_partitioning([1], 4),
fixture.key_distribution([0, 1, 2]),
PartitioningSatisfaction::Subset,
PartitioningSatisfaction::NotSatisfied,
),
(
- "KeyPartitioned([a, b, c]) satisfied by Hash([b, a])",
+ "subset reordered: KeyPartitioned([a, b, c]) satisfied by
Hash([b, a])",
fixture.hash_partitioning([1, 0], 4),
fixture.key_distribution([0, 1, 2]),
PartitioningSatisfaction::Subset,
PartitioningSatisfaction::NotSatisfied,
),
- ];
-
- for (desc, partition, required, expected_with_subset,
expected_without_subset) in
- test_cases
- {
- let result = partition.satisfaction(&required,
&fixture.eq_properties, true);
- assert_eq!(
- result, expected_with_subset,
- "Failed for {desc} with subset enabled"
- );
-
- let result = partition.satisfaction(&required,
&fixture.eq_properties, false);
- assert_eq!(
- result, expected_without_subset,
- "Failed for {desc} with subset disabled"
- );
- }
-
- Ok(())
- }
-
- #[test]
- fn test_partitioning_current_superset() -> Result<()> {
- let fixture = PartitioningTestFixture::int64(&["a", "b", "c"])?;
-
- let test_cases = vec![
(
- "KeyPartitioned([a]) satisfied by Hash([a, b])",
+ "superset: KeyPartitioned([a]) not satisfied by Hash([a, b])",
fixture.hash_partitioning([0, 1], 4),
fixture.key_distribution([0]),
PartitioningSatisfaction::NotSatisfied,
PartitioningSatisfaction::NotSatisfied,
),
(
- "KeyPartitioned([a]) satisfied by Hash([a, b, c])",
+ "superset: KeyPartitioned([a, b]) not satisfied by Hash([a, b,
c])",
fixture.hash_partitioning([0, 1, 2], 4),
- fixture.key_distribution([0]),
+ fixture.key_distribution([0, 1]),
PartitioningSatisfaction::NotSatisfied,
PartitioningSatisfaction::NotSatisfied,
),
(
- "KeyPartitioned([a, b]) satisfied by Hash([a, b, c])",
- fixture.hash_partitioning([0, 1, 2], 4),
+ "partial overlap: KeyPartitioned([a, b]) not satisfied by
Hash([a, c])",
+ fixture.hash_partitioning([0, 2], 4),
fixture.key_distribution([0, 1]),
PartitioningSatisfaction::NotSatisfied,
PartitioningSatisfaction::NotSatisfied,
),
- ];
-
- for (desc, partition, required, expected_with_subset,
expected_without_subset) in
- test_cases
- {
- let result = partition.satisfaction(&required,
&fixture.eq_properties, true);
- assert_eq!(
- result, expected_with_subset,
- "Failed for {desc} with subset enabled"
- );
-
- let result = partition.satisfaction(&required,
&fixture.eq_properties, false);
- assert_eq!(
- result, expected_without_subset,
- "Failed for {desc} with subset disabled"
- );
- }
-
- Ok(())
- }
-
- #[test]
- fn test_partitioning_partial_overlap() -> Result<()> {
Review Comment:
renamed and went over tests to ensure we didn't lose coverage.
I added back: ` Hash([unknown]) vs KeyPartitioned([unknown,a])` to check an
unknown cant match itself in subset
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
