Skip to content

Use shared statistics merge for union stats #21430

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kumarUjjawal wants to merge 3 commits into
base: main
Choose a base branch
from
+344 −369
Open

Use shared statistics merge for union stats #21430

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8f43abe
Use shared statistics merge for union stats
kumarUjjawal Apr 7, 2026
5214b45
test: restore shared NDV coverage and reset testing submodule
kumarUjjawal Apr 7, 2026
d85309d
test: add full shared NDV edge-case table
kumarUjjawal Apr 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
212 changes: 212 additions & 0 deletions datafusion/common/src/stats.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1461,6 +1461,39 @@ mod tests {
}
}

fn make_single_i64_ndv_stats(
distinct_count: Precision<usize>,
min_value: Option<i64>,
max_value: Option<i64>,
) -> Statistics {
let to_precision = |value| Precision::Exact(ScalarValue::Int64(Some(value)));

Statistics::default()
.with_num_rows(Precision::Exact(10))
.add_column_statistics(
ColumnStatistics::new_unknown()
.with_distinct_count(distinct_count)
.with_min_value(
min_value.map(to_precision).unwrap_or(Precision::Absent),
)
.with_max_value(
max_value.map(to_precision).unwrap_or(Precision::Absent),
),
)
}

fn merge_single_i64_ndv_distinct_count(
left: Statistics,
right: Statistics,
) -> Precision<usize> {
let schema = Schema::new(vec![Field::new("a", DataType::Int64, true)]);

Statistics::try_merge_iter([&left, &right], &schema)
.unwrap()
.column_statistics[0]
.distinct_count
}

#[test]
fn test_try_merge() {
// Create a schema with two columns
Expand Down Expand Up @@ -1906,6 +1939,185 @@ mod tests {
);
}

#[test]
fn test_try_merge_ndv_original_union_edge_cases() {
struct NdvTestCase {
name: &'static str,
left_ndv: Precision<usize>,
left_min: Option<i64>,
left_max: Option<i64>,
right_ndv: Precision<usize>,
right_min: Option<i64>,
right_max: Option<i64>,
expected: Precision<usize>,
}

let cases = vec![
NdvTestCase {
name: "disjoint ranges",
left_ndv: Precision::Exact(5),
left_min: Some(0),
left_max: Some(10),
right_ndv: Precision::Exact(3),
right_min: Some(20),
right_max: Some(30),
expected: Precision::Inexact(8),
},
NdvTestCase {
name: "identical ranges",
left_ndv: Precision::Exact(10),
left_min: Some(0),
left_max: Some(100),
right_ndv: Precision::Exact(8),
right_min: Some(0),
right_max: Some(100),
expected: Precision::Inexact(10),
},
NdvTestCase {
name: "partial overlap",
left_ndv: Precision::Exact(100),
left_min: Some(0),
left_max: Some(100),
right_ndv: Precision::Exact(50),
right_min: Some(50),
right_max: Some(150),
expected: Precision::Inexact(125),
},
NdvTestCase {
name: "right contained in left",
left_ndv: Precision::Exact(100),
left_min: Some(0),
left_max: Some(100),
right_ndv: Precision::Exact(50),
right_min: Some(25),
right_max: Some(75),
expected: Precision::Inexact(100),
},
NdvTestCase {
name: "same constant value",
left_ndv: Precision::Exact(1),
left_min: Some(5),
left_max: Some(5),
right_ndv: Precision::Exact(1),
right_min: Some(5),
right_max: Some(5),
expected: Precision::Inexact(1),
},
NdvTestCase {
name: "different constant values",
left_ndv: Precision::Exact(1),
left_min: Some(5),
left_max: Some(5),
right_ndv: Precision::Exact(1),
right_min: Some(10),
right_max: Some(10),
expected: Precision::Inexact(2),
},
NdvTestCase {
name: "left constant within right range",
left_ndv: Precision::Exact(1),
left_min: Some(5),
left_max: Some(5),
right_ndv: Precision::Exact(10),
right_min: Some(0),
right_max: Some(10),
expected: Precision::Inexact(10),
},
NdvTestCase {
name: "left constant outside right range",
left_ndv: Precision::Exact(1),
left_min: Some(20),
left_max: Some(20),
right_ndv: Precision::Exact(10),
right_min: Some(0),
right_max: Some(10),
expected: Precision::Inexact(11),
},
NdvTestCase {
name: "right constant within left range",
left_ndv: Precision::Exact(10),
left_min: Some(0),
left_max: Some(10),
right_ndv: Precision::Exact(1),
right_min: Some(5),
right_max: Some(5),
expected: Precision::Inexact(10),
},
NdvTestCase {
name: "right constant outside left range",
left_ndv: Precision::Exact(10),
left_min: Some(0),
left_max: Some(10),
right_ndv: Precision::Exact(1),
right_min: Some(20),
right_max: Some(20),
expected: Precision::Inexact(11),
},
NdvTestCase {
name: "missing bounds exact plus exact",
left_ndv: Precision::Exact(10),
left_min: None,
left_max: None,
right_ndv: Precision::Exact(5),
right_min: None,
right_max: None,
// Shared merge falls back to max without bounds.
expected: Precision::Inexact(10),
},
NdvTestCase {
name: "missing bounds exact plus inexact",
left_ndv: Precision::Exact(10),
left_min: None,
left_max: None,
right_ndv: Precision::Inexact(5),
right_min: None,
right_max: None,
// Shared merge falls back to max without bounds.
expected: Precision::Inexact(10),
},
NdvTestCase {
name: "missing bounds inexact plus inexact",
left_ndv: Precision::Inexact(7),
left_min: None,
left_max: None,
right_ndv: Precision::Inexact(3),
right_min: None,
right_max: None,
// Shared merge falls back to max without bounds.
expected: Precision::Inexact(7),
},
NdvTestCase {
name: "exact plus absent",
left_ndv: Precision::Exact(10),
left_min: None,
left_max: None,
right_ndv: Precision::Absent,
right_min: None,
right_max: None,
expected: Precision::Absent,
},
NdvTestCase {
name: "inexact plus absent",
left_ndv: Precision::Inexact(4),
left_min: None,
left_max: None,
right_ndv: Precision::Absent,
right_min: None,
right_max: None,
expected: Precision::Absent,
},
];

for case in cases {
let actual = merge_single_i64_ndv_distinct_count(
make_single_i64_ndv_stats(case.left_ndv, case.left_min, case.left_max),
make_single_i64_ndv_stats(case.right_ndv, case.right_min, case.right_max),
);

assert_eq!(actual, case.expected, "case {} failed", case.name);
}
}

#[test]
fn test_with_fetch_basic_preservation() {
// Test that column statistics and byte size are preserved (as inexact) when applying fetch
Expand Down
Loading
Loading