Skip to content

fix: interval cast expression can't work in range query, #5805 #5813

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

Merged
evenyag merged 3 commits into from
Apr 2, 2025
Merged

fix: interval cast expression can't work in range query, #5805 #5813

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
12c132b
fix: interval cast expression can't work in range query, #5805
killme2008 Apr 2, 2025
4df92e9
fix: nested cast
killme2008 Apr 2, 2025
fb18e1c
test: make vector test stable
killme2008 Apr 2, 2025
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
64 changes: 62 additions & 2 deletions src/query/src/range_select/plan_rewrite.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,8 +32,8 @@ use datafusion_expr::execution_props::ExecutionProps;
use datafusion_expr::expr::WildcardOptions;
use datafusion_expr::simplify::SimplifyContext;
use datafusion_expr::{
Aggregate, Analyze, Explain, Expr, ExprSchemable, Extension, LogicalPlan, LogicalPlanBuilder,
Projection,
Aggregate, Analyze, Cast, Explain, Expr, ExprSchemable, Extension, LogicalPlan,
LogicalPlanBuilder, Projection,
};
use datafusion_optimizer::simplify_expressions::ExprSimplifier;
use datatypes::prelude::ConcreteDataType;
Expand Down Expand Up @@ -548,19 +548,37 @@ fn have_range_in_exprs(exprs: &[Expr]) -> bool {
fn interval_only_in_expr(expr: &Expr) -> bool {
let mut all_interval = true;
let _ = expr.apply(|expr| {
// A cast expression for an interval.
if matches!(
expr,
Expr::Cast(Cast{
expr,
data_type: DataType::Interval(_)
}) if matches!(&**expr, Expr::Literal(ScalarValue::Utf8(_)))
) {
// Stop checking the sub `expr`,
// which is a `Utf8` type and has already been tested above.
return Ok(TreeNodeRecursion::Stop);
}

if !matches!(
expr,
Expr::Literal(ScalarValue::IntervalDayTime(_))
| Expr::Literal(ScalarValue::IntervalMonthDayNano(_))
| Expr::Literal(ScalarValue::IntervalYearMonth(_))
| Expr::BinaryExpr(_)
| Expr::Cast(Cast {
data_type: DataType::Interval(_),
..
})
) {
all_interval = false;
Ok(TreeNodeRecursion::Stop)
} else {
Ok(TreeNodeRecursion::Continue)
}
});

all_interval
}

Expand All @@ -569,6 +587,7 @@ mod test {

use std::error::Error;

use arrow::datatypes::IntervalUnit;
use catalog::memory::MemoryCatalogManager;
use catalog::RegisterTableRequest;
use common_catalog::consts::{DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME};
Expand Down Expand Up @@ -844,6 +863,15 @@ mod test {
parse_duration_expr(&args, 0).unwrap(),
parse_duration("1y4w").unwrap()
);
// test cast expression
let args = vec![Expr::Cast(Cast {
expr: Box::new(Expr::Literal(ScalarValue::Utf8(Some("15 minutes".into())))),
data_type: DataType::Interval(IntervalUnit::MonthDayNano),
})];
assert_eq!(
parse_duration_expr(&args, 0).unwrap(),
parse_duration("15m").unwrap()
);
// test index err
assert!(parse_duration_expr(&args, 10).is_err());
// test evaluate expr
Expand Down Expand Up @@ -958,5 +986,37 @@ mod test {
)))),
});
assert!(interval_only_in_expr(&expr));

let expr = Expr::BinaryExpr(BinaryExpr {
left: Box::new(Expr::Cast(Cast {
expr: Box::new(Expr::Literal(ScalarValue::Utf8(Some(
"15 minute".to_string(),
)))),
data_type: DataType::Interval(IntervalUnit::MonthDayNano),
})),
op: Operator::Minus,
right: Box::new(Expr::Literal(ScalarValue::IntervalDayTime(Some(
IntervalDayTime::new(10, 0).into(),
)))),
});
assert!(interval_only_in_expr(&expr));

let expr = Expr::Cast(Cast {
expr: Box::new(Expr::BinaryExpr(BinaryExpr {
left: Box::new(Expr::Cast(Cast {
expr: Box::new(Expr::Literal(ScalarValue::Utf8(Some(
"15 minute".to_string(),
)))),
data_type: DataType::Interval(IntervalUnit::MonthDayNano),
})),
op: Operator::Minus,
right: Box::new(Expr::Literal(ScalarValue::IntervalDayTime(Some(
IntervalDayTime::new(10, 0).into(),
)))),
})),
data_type: DataType::Interval(IntervalUnit::MonthDayNano),
});

assert!(interval_only_in_expr(&expr));
}
}
2 changes: 1 addition & 1 deletion tests/cases/standalone/common/function/vector/vector.result
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -342,7 +342,7 @@ FROM (
UNION ALL
SELECT '[8.0]' AS v
)
WHERE vec_kth_elem(v, 0) > 2.0;
WHERE vec_kth_elem(v, 0) > 2.0 ORDER BY first_elem;

+----------------------+------------+
| v | first_elem |
Expand Down
2 changes: 1 addition & 1 deletion tests/cases/standalone/common/function/vector/vector.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ FROM (
UNION ALL
SELECT '[8.0]' AS v
)
WHERE vec_kth_elem(v, 0) > 2.0;
WHERE vec_kth_elem(v, 0) > 2.0 ORDER BY first_elem;

SELECT vec_to_string(vec_subvector('[1.0,2.0,3.0,4.0,5.0]', 0, 3));

Expand Down
57 changes: 57 additions & 0 deletions tests/cases/standalone/common/range/interval.result
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,63 @@ SELECT ts, host, min(val) RANGE (INTERVAL '1' day) FROM host ALIGN (INTERVAL '1'
| 1971-01-02T00:00:00 | host2 | 6 |
+---------------------+-------+---------------------------------------------------------------------------------------------------------+

SELECT ts, host, min(val) RANGE ('1 day'::INTERVAL) FROM host ALIGN ('1 day'::INTERVAL) ORDER BY host, ts;

+---------------------+-------+-----------------------------------+
| ts | host | min(host.val) RANGE Utf8("1 day") |
+---------------------+-------+-----------------------------------+
| 1970-01-01T00:00:00 | host1 | 0 |
| 1971-01-02T00:00:00 | host1 | 2 |
| 1970-01-01T00:00:00 | host2 | 4 |
| 1971-01-02T00:00:00 | host2 | 6 |
+---------------------+-------+-----------------------------------+

SELECT ts, host, min(val) RANGE ('1 hour'::INTERVAL) FROM host ALIGN ('1 hour'::INTERVAL) ORDER BY host, ts;

+---------------------+-------+------------------------------------+
| ts | host | min(host.val) RANGE Utf8("1 hour") |
+---------------------+-------+------------------------------------+
| 1970-01-01T01:00:00 | host1 | 0 |
| 1970-01-01T02:00:00 | host1 | 1 |
| 1971-01-02T03:00:00 | host1 | 2 |
| 1971-01-02T04:00:00 | host1 | 3 |
| 1970-01-01T01:00:00 | host2 | 4 |
| 1970-01-01T02:00:00 | host2 | 5 |
| 1971-01-02T03:00:00 | host2 | 6 |
| 1971-01-02T04:00:00 | host2 | 7 |
+---------------------+-------+------------------------------------+

SELECT ts, host, min(val) RANGE ('30 minute'::INTERVAL + '30 minute'::INTERVAL) FROM host ALIGN ('30 minute'::INTERVAL + '30 minute'::INTERVAL) ORDER BY host, ts;

+---------------------+-------+-----------------------------------------------------------+
| ts | host | min(host.val) RANGE Utf8("30 minute") + Utf8("30 minute") |
+---------------------+-------+-----------------------------------------------------------+
| 1970-01-01T01:00:00 | host1 | 0 |
| 1970-01-01T02:00:00 | host1 | 1 |
| 1971-01-02T03:00:00 | host1 | 2 |
| 1971-01-02T04:00:00 | host1 | 3 |
| 1970-01-01T01:00:00 | host2 | 4 |
| 1970-01-01T02:00:00 | host2 | 5 |
| 1971-01-02T03:00:00 | host2 | 6 |
| 1971-01-02T04:00:00 | host2 | 7 |
+---------------------+-------+-----------------------------------------------------------+

--- Test nested cast, even though it is meaningless ----
SELECT ts, host, min(val) RANGE ((INTERVAL '1' hour)::INTERVAL) FROM host ALIGN ('1 hour'::INTERVAL::INTERVAL) ORDER BY host, ts;

+---------------------+-------+---------------------------------------------------------------------------------------------------------------------+
| ts | host | min(host.val) RANGE IntervalMonthDayNano("IntervalMonthDayNano { months: 0, days: 0, nanoseconds: 3600000000000 }") |
+---------------------+-------+---------------------------------------------------------------------------------------------------------------------+
| 1970-01-01T01:00:00 | host1 | 0 |
| 1970-01-01T02:00:00 | host1 | 1 |
| 1971-01-02T03:00:00 | host1 | 2 |
| 1971-01-02T04:00:00 | host1 | 3 |
| 1970-01-01T01:00:00 | host2 | 4 |
| 1970-01-01T02:00:00 | host2 | 5 |
| 1971-01-02T03:00:00 | host2 | 6 |
| 1971-01-02T04:00:00 | host2 | 7 |
+---------------------+-------+---------------------------------------------------------------------------------------------------------------------+

DROP TABLE host;

Affected Rows: 0
Expand Down
9 changes: 9 additions & 0 deletions tests/cases/standalone/common/range/interval.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,4 +18,13 @@ SELECT ts, host, min(val) RANGE (INTERVAL '1 year') FROM host ALIGN (INTERVAL '1

SELECT ts, host, min(val) RANGE (INTERVAL '1' day) FROM host ALIGN (INTERVAL '1' day) ORDER BY host, ts;

SELECT ts, host, min(val) RANGE ('1 day'::INTERVAL) FROM host ALIGN ('1 day'::INTERVAL) ORDER BY host, ts;

SELECT ts, host, min(val) RANGE ('1 hour'::INTERVAL) FROM host ALIGN ('1 hour'::INTERVAL) ORDER BY host, ts;

SELECT ts, host, min(val) RANGE ('30 minute'::INTERVAL + '30 minute'::INTERVAL) FROM host ALIGN ('30 minute'::INTERVAL + '30 minute'::INTERVAL) ORDER BY host, ts;

--- Test nested cast, even though it is meaningless ----
SELECT ts, host, min(val) RANGE ((INTERVAL '1' hour)::INTERVAL) FROM host ALIGN ('1 hour'::INTERVAL::INTERVAL) ORDER BY host, ts;

DROP TABLE host;