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

feat(functions): add map_insert function #15567

Merged
merged 11 commits into from
Nov 5, 2024
Merged

feat(functions): add map_insert function #15567

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
b5537e0
feat(functions): add map_insert function
hanxuanliang May 18, 2024
d527a7b
feat(functions): check ci
hanxuanliang May 18, 2024
fa0c45d
feat(functions): domain fix
hanxuanliang May 19, 2024
ea0c1d4
feat(functions): add new params format in map_insert
hanxuanliang May 19, 2024
35c6a6e
feat(functions): cargo fmt
hanxuanliang May 19, 2024
b287126
feat(functions): fix NULL mapType in insert
hanxuanliang May 25, 2024
9133ba0
merge
b41sh Nov 5, 2024
0a4711b
fix
b41sh Nov 5, 2024
03e4a7c
fix tests
b41sh Nov 5, 2024
019c6d5
Merge branch 'main' into feat/15295_map_insert
b41sh Nov 5, 2024
b531703
fix typos
b41sh Nov 5, 2024
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
88 changes: 88 additions & 0 deletions src/query/functions/src/scalars/map.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,8 @@ use databend_common_expression::types::SimpleDomain;
use databend_common_expression::types::ValueType;
use databend_common_expression::vectorize_1_arg;
use databend_common_expression::vectorize_with_builder_2_arg;
use databend_common_expression::vectorize_with_builder_3_arg;
use databend_common_expression::vectorize_with_builder_4_arg;
use databend_common_expression::ColumnBuilder;
use databend_common_expression::Function;
use databend_common_expression::FunctionDomain;
Expand Down Expand Up @@ -348,6 +350,92 @@ pub fn register(registry: &mut FunctionRegistry) {
},
);

registry.register_3_arg_core::<NullableType<MapType<GenericType<0>, GenericType<1>>>, GenericType<0>, GenericType<1>, MapType<GenericType<0>, GenericType<1>>, _, _>(
"map_insert",
|_, map_domain, key_domain, value_domain| {
let domain = map_domain
.value
.as_ref()
.map(|box inner_domain| {
inner_domain
.as_ref()
.map(|(inner_key_domain, inner_value_domain)| (inner_key_domain.merge(key_domain), inner_value_domain.merge(value_domain)))
.unwrap_or((key_domain.clone(), value_domain.clone()))
});
FunctionDomain::Domain(domain)
},
vectorize_with_builder_3_arg::<NullableType<MapType<GenericType<0>, GenericType<1>>>, GenericType<0>, GenericType<1>, MapType<GenericType<0>, GenericType<1>>>(
|map, key, val, output, ctx| {
let key_type = &ctx.generics[0];
if !check_valid_map_key_type(key_type) {
ctx.set_error(output.len(), format!("map keys can not be {}", key_type));
output.commit_row();
return;
}
if let Some(map) = map {
for (k, _) in map.iter() {
if k == key {
ctx.set_error(output.len(), format!("map key `{}` duplicate", key));
output.commit_row();
return;
}
}
for (k, v) in map.iter() {
output.put_item((k, v));
}
}
output.put_item((key, val));
output.commit_row();
})
);

registry.register_4_arg_core::<NullableType<MapType<GenericType<0>, GenericType<1>>>, GenericType<0>, GenericType<1>, BooleanType, MapType<GenericType<0>, GenericType<1>>, _, _>(
"map_insert",
|_, map_domain, key_domain, value_domain, _| {
let domain = map_domain
.value
.as_ref()
.map(|box inner_domain| {
inner_domain
.as_ref()
.map(|(inner_key_domain, inner_value_domain)| (inner_key_domain.merge(key_domain), inner_value_domain.merge(value_domain)))
.unwrap_or((key_domain.clone(), value_domain.clone()))
});
FunctionDomain::Domain(domain)
},
vectorize_with_builder_4_arg::<NullableType<MapType<GenericType<0>, GenericType<1>>>, GenericType<0>, GenericType<1>, BooleanType, MapType<GenericType<0>, GenericType<1>>>(
|map, key, val, update_flag, output, ctx| {
let key_type = &ctx.generics[0];
if !check_valid_map_key_type(key_type) {
ctx.set_error(output.len(), format!("map keys can not be {}", key_type));
output.commit_row();
return;
}
let mut is_duplicate = false;
if let Some(map) = map {
for (k, _) in map.iter() {
if k == key && !update_flag {
ctx.set_error(output.len(), format!("map key `{}` duplicate", key));
output.commit_row();
return;
}
}
for (k, v) in map.iter() {
if k == key {
is_duplicate = true;
output.put_item((k, val.clone()));
} else {
output.put_item((k, v));
}
}
}
if !is_duplicate {
output.put_item((key, val));
}
output.commit_row();
})
);

registry.register_function_factory("map_pick", |_, args_type: &[DataType]| {
let return_type = check_map_arg_types(args_type)?;
Some(Arc::new(Function {
Expand Down
49 changes: 49 additions & 0 deletions src/query/functions/tests/it/scalars/map.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,7 @@ fn test_map() {
test_map_delete(file);
test_map_contains_key(file);
test_map_pick(file);
test_map_insert(file)
}

fn test_map_cat(file: &mut impl Write) {
Expand Down Expand Up @@ -419,3 +420,51 @@ fn test_map_pick(file: &mut impl Write) {
&columns,
);
}

fn test_map_insert(file: &mut impl Write) {
run_ast(file, "map_insert({}, 'k1', 'v1')", &[]);
run_ast(file, "map_insert({'k1': 'v1'}, 'k2', 'v2')", &[]);
run_ast(
file,
"map_insert({'k1': 'v1', 'k2': 'v2'}, 'k1', 'v10', false)",
&[],
);
run_ast(
file,
"map_insert({'k1': 'v1', 'k2': 'v2'}, 'k1', 'v10', true)",
&[],
);

let columns = [
("a_col", StringType::from_data(vec!["a", "b", "c"])),
("b_col", StringType::from_data(vec!["d", "e", "f"])),
("c_col", StringType::from_data(vec!["x", "y", "z"])),
(
"d_col",
StringType::from_data_with_validity(vec!["v1", "v2", "v3"], vec![true, true, true]),
),
(
"e_col",
StringType::from_data_with_validity(vec!["v4", "v5", ""], vec![true, true, false]),
),
(
"f_col",
StringType::from_data_with_validity(vec!["v6", "", "v7"], vec![true, false, true]),
),
];
run_ast(
file,
"map_insert(map([a_col, b_col, c_col], [d_col, e_col, f_col]), 'k1', 'v10')",
&columns,
);
run_ast(
file,
"map_insert(map([a_col, b_col, c_col], [d_col, e_col, f_col]), 'a', 'v10', true)",
&columns,
);
run_ast(
file,
"map_insert(map([a_col, b_col, c_col], [d_col, e_col, f_col]), 'a', 'v10', false)",
&columns,
);
}
2 changes: 2 additions & 0 deletions src/query/functions/tests/it/scalars/testdata/function_list.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2546,6 +2546,8 @@ Functions overloads:
1 map_contains_key(Map(T0, T1), T0) :: Boolean
2 map_contains_key(Map(T0, T1) NULL, T0 NULL) :: Boolean NULL
0 map_delete FACTORY
0 map_insert(Map(T0, T1) NULL, T0, T1) :: Map(T0, T1)
1 map_insert(Map(T0, T1) NULL, T0, T1, Boolean) :: Map(T0, T1)
0 map_keys(Map(Nothing)) :: Array(Nothing)
1 map_keys(Map(T0, T1)) :: Array(T0)
2 map_keys(Map(T0, T1) NULL) :: Array(T0) NULL
Expand Down
Loading
Loading