Skip to content
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

make array_union/array_except/array_intersect handle empty/null arrays rightly #8269

Merged
merged 5 commits into from
Nov 21, 2023
Merged

make array_union/array_except/array_intersect handle empty/null arrays rightly #8269

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
32a8e7c
make array_union handle empty/null arrays rightly
Veeupup Nov 19, 2023
9dbaf5f
make array_except handle empty/null arrays rightly
Veeupup Nov 19, 2023
c08d6cb
make array_intersect handle empty/null arrays rightly
Veeupup Nov 19, 2023
be32301
fix sql_array_literal
Veeupup Nov 20, 2023
e154ead
fix comments
Veeupup Nov 21, 2023
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
27 changes: 24 additions & 3 deletions datafusion/expr/src/built_in_function.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -599,12 +599,33 @@ impl BuiltinScalarFunction {
BuiltinScalarFunction::ArrayReplaceAll => Ok(input_expr_types[0].clone()),
BuiltinScalarFunction::ArraySlice => Ok(input_expr_types[0].clone()),
BuiltinScalarFunction::ArrayToString => Ok(Utf8),
BuiltinScalarFunction::ArrayIntersect => Ok(input_expr_types[0].clone()),
BuiltinScalarFunction::ArrayUnion => Ok(input_expr_types[0].clone()),
BuiltinScalarFunction::ArrayIntersect => {
Veeupup marked this conversation as resolved.
Show resolved Hide resolved
match (input_expr_types[0].clone(), input_expr_types[1].clone()) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes!

(DataType::Null, DataType::Null) => Ok(DataType::List(Arc::new(
Field::new("item", DataType::Null, true),
))),
(dt, _) => Ok(dt),
}
}
BuiltinScalarFunction::ArrayUnion => {
match (input_expr_types[0].clone(), input_expr_types[1].clone()) {
(DataType::Null, DataType::Null) => Ok(DataType::List(Arc::new(
Field::new("item", DataType::Null, true),
))),
(dt, _) => Ok(dt),
}
}
BuiltinScalarFunction::Range => {
Ok(List(Arc::new(Field::new("item", Int64, true))))
}
BuiltinScalarFunction::ArrayExcept => Ok(input_expr_types[0].clone()),
BuiltinScalarFunction::ArrayExcept => {
match (input_expr_types[0].clone(), input_expr_types[1].clone()) {
(DataType::Null, DataType::Null) => Ok(DataType::List(Arc::new(
Field::new("item", DataType::Null, true),
))),
(dt, _) => Ok(dt),
}
}
BuiltinScalarFunction::Cardinality => Ok(UInt64),
BuiltinScalarFunction::MakeArray => match input_expr_types.len() {
0 => Ok(List(Arc::new(Field::new("item", Null, true)))),
Expand Down
138 changes: 96 additions & 42 deletions datafusion/physical-expr/src/array_expressions.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -580,6 +580,21 @@ pub fn array_except(args: &[ArrayRef]) -> Result<ArrayRef> {
let array2 = &args[1];

match (array1.data_type(), array2.data_type()) {
(DataType::Null, DataType::Null) => {
// NullArray(1): means null, NullArray(0): means []
// except([], []) = [], except([], null) = [], except(null, []) = null, except(null, null) = null
let nulls = match (array1.len(), array2.len()) {
(1, _) => Some(NullBuffer::new_null(1)),
_ => None,
};
let arr = Arc::new(ListArray::try_new(
Arc::new(Field::new("item", DataType::Null, true)),
OffsetBuffer::new(vec![0; 2].into()),
Arc::new(NullArray::new(0)),
nulls,
)?) as ArrayRef;
Ok(arr)
Veeupup marked this conversation as resolved.
Show resolved Hide resolved
}
(DataType::Null, _) | (_, DataType::Null) => Ok(array1.to_owned()),
(DataType::List(field), DataType::List(_)) => {
check_datatypes("array_except", &[array1, array2])?;
Expand Down Expand Up @@ -1510,6 +1525,21 @@ pub fn array_union(args: &[ArrayRef]) -> Result<ArrayRef> {
let array1 = &args[0];
let array2 = &args[1];
match (array1.data_type(), array2.data_type()) {
(DataType::Null, DataType::Null) => {
// NullArray(1): means null, NullArray(0): means []
// union([], []) = [], union([], null) = [], union(null, []) = [], union(null, null) = null
let nulls = match (array1.len(), array2.len()) {
(1, 1) => Some(NullBuffer::new_null(1)),
_ => None,
};
let arr = Arc::new(ListArray::try_new(
Arc::new(Field::new("item", DataType::Null, true)),
OffsetBuffer::new(vec![0; 2].into()),
Arc::new(NullArray::new(0)),
nulls,
)?) as ArrayRef;
Ok(arr)
}
(DataType::Null, _) => Ok(array2.clone()),
(_, DataType::Null) => Ok(array1.clone()),
(DataType::List(field_ref), DataType::List(_)) => {
Expand Down Expand Up @@ -1998,55 +2028,79 @@ pub fn string_to_array<T: OffsetSizeTrait>(args: &[ArrayRef]) -> Result<ArrayRef
pub fn array_intersect(args: &[ArrayRef]) -> Result<ArrayRef> {
assert_eq!(args.len(), 2);

let first_array = as_list_array(&args[0])?;
let second_array = as_list_array(&args[1])?;
let first_array = &args[0];
let second_array = &args[1];

if first_array.value_type() != second_array.value_type() {
return internal_err!("array_intersect is not implemented for '{first_array:?}' and '{second_array:?}'");
}
let dt = first_array.value_type();
match (first_array.data_type(), second_array.data_type()) {
(DataType::Null, DataType::Null) => {
// NullArray(1): means null, NullArray(0): means []
// intersect([], []) = [], intersect([], null) = [], intersect(null, []) = [], intersect(null, null) = null
let nulls = match (first_array.len(), second_array.len()) {
(1, 1) => Some(NullBuffer::new_null(1)),
_ => None,
};
let arr = Arc::new(ListArray::try_new(
Arc::new(Field::new("item", DataType::Null, true)),
OffsetBuffer::new(vec![0; 2].into()),
Arc::new(NullArray::new(0)),
nulls,
)?) as ArrayRef;
Ok(arr)
}
_ => {
let first_array = as_list_array(&first_array)?;
let second_array = as_list_array(&second_array)?;

let mut offsets = vec![0];
let mut new_arrays = vec![];

let converter = RowConverter::new(vec![SortField::new(dt.clone())])?;
for (first_arr, second_arr) in first_array.iter().zip(second_array.iter()) {
if let (Some(first_arr), Some(second_arr)) = (first_arr, second_arr) {
let l_values = converter.convert_columns(&[first_arr])?;
let r_values = converter.convert_columns(&[second_arr])?;

let values_set: HashSet<_> = l_values.iter().collect();
let mut rows = Vec::with_capacity(r_values.num_rows());
for r_val in r_values.iter().sorted().dedup() {
if values_set.contains(&r_val) {
rows.push(r_val);
}
if first_array.value_type() != second_array.value_type() {
return internal_err!("array_intersect is not implemented for '{first_array:?}' and '{second_array:?}'");
}

let last_offset: i32 = match offsets.last().copied() {
Some(offset) => offset,
None => return internal_err!("offsets should not be empty"),
};
offsets.push(last_offset + rows.len() as i32);
let arrays = converter.convert_rows(rows)?;
let array = match arrays.get(0) {
Some(array) => array.clone(),
None => {
return internal_err!(
"array_intersect: failed to get array from rows"
)
let dt = first_array.value_type();

let mut offsets = vec![0];
let mut new_arrays = vec![];

let converter = RowConverter::new(vec![SortField::new(dt.clone())])?;
for (first_arr, second_arr) in first_array.iter().zip(second_array.iter()) {
if let (Some(first_arr), Some(second_arr)) = (first_arr, second_arr) {
let l_values = converter.convert_columns(&[first_arr])?;
let r_values = converter.convert_columns(&[second_arr])?;

let values_set: HashSet<_> = l_values.iter().collect();
let mut rows = Vec::with_capacity(r_values.num_rows());
for r_val in r_values.iter().sorted().dedup() {
if values_set.contains(&r_val) {
rows.push(r_val);
}
}

let last_offset: i32 = match offsets.last().copied() {
Some(offset) => offset,
None => return internal_err!("offsets should not be empty"),
};
offsets.push(last_offset + rows.len() as i32);
let arrays = converter.convert_rows(rows)?;
let array = match arrays.get(0) {
Some(array) => array.clone(),
None => {
return internal_err!(
"array_intersect: failed to get array from rows"
)
}
};
new_arrays.push(array);
}
};
new_arrays.push(array);
}

let field = Arc::new(Field::new("item", dt, true));
let offsets = OffsetBuffer::new(offsets.into());
let new_arrays_ref =
new_arrays.iter().map(|v| v.as_ref()).collect::<Vec<_>>();
let values = compute::concat(&new_arrays_ref)?;
let arr = Arc::new(ListArray::try_new(field, offsets, values, None)?);
Ok(arr)
}
}

let field = Arc::new(Field::new("item", dt, true));
let offsets = OffsetBuffer::new(offsets.into());
let new_arrays_ref = new_arrays.iter().map(|v| v.as_ref()).collect::<Vec<_>>();
let values = compute::concat(&new_arrays_ref)?;
let arr = Arc::new(ListArray::try_new(field, offsets, values, None)?);
Ok(arr)
}

#[cfg(test)]
Expand Down
44 changes: 42 additions & 2 deletions datafusion/sqllogictest/test_files/array.slt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2014,7 +2014,7 @@ drop table arrays_with_repeating_elements_for_union;
query ?
select array_union([], []);
----
NULL
[]

# array_union scalar function #7
query ?
Expand All @@ -2032,7 +2032,7 @@ select array_union([null], [null]);
query ?
select array_union(null, []);
----
NULL
[]

# array_union scalar function #10
query ?
Expand Down Expand Up @@ -2687,6 +2687,26 @@ SELECT array_intersect(make_array(1,2,3), make_array(2,3,4)),
----
[2, 3] [] [aa, cc] [true] [2.2, 3.3] [[2, 2], [3, 3]]

query ?
select array_intersect([], []);
----
[]

query ?
select array_intersect([], null);
----
[]

query ?
select array_intersect(null, []);
----
[]

query ?
select array_intersect(null, null);
----
NULL

query ??????
SELECT list_intersect(make_array(1,2,3), make_array(2,3,4)),
list_intersect(make_array(1,3,5), make_array(2,4,6)),
Expand Down Expand Up @@ -2842,6 +2862,26 @@ NULL
statement ok
drop table array_except_table_bool;

query ?
select array_except([], null);
----
[]

query ?
select array_except([], []);
----
[]

query ?
select array_except(null, []);
----
NULL

query ?
select array_except(null, null)
----
NULL

### Array operators tests


Expand Down