Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
4 changes: 4 additions & 0 deletions arrow-cast/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -61,6 +61,10 @@ insta = { workspace = true }
num-bigint = { version = "0.5", default-features = false, features = ["std"] }
rand = "0.10"

[[test]]
name = "cast"
path = "./tests/cast/mod.rs"

[[bench]]
name = "parse_timestamp"
harness = false
Expand Down
223 changes: 0 additions & 223 deletions arrow-cast/src/cast/mod.rs
Original file line number Diff line number Diff line change
Expand Up @@ -10274,229 +10274,6 @@ mod tests {
))
}

#[test]
fn test_cast_map_dont_allow_change_of_order() {
let string_builder = StringBuilder::new();
let value_builder = StringBuilder::new();
let mut builder = MapBuilder::new(None, string_builder, value_builder);

builder.keys().append_value("0");
builder.values().append_value("test_val_1");
builder.append(true).unwrap();
builder.keys().append_value("1");
builder.values().append_value("test_val_2");
builder.append(true).unwrap();

// map builder returns unsorted map by default
let array = builder.finish();

let new_ordered = true;
let new_type = DataType::Map(
Arc::new(Field::new(
Field::MAP_ENTRIES_FIELD_DEFAULT_NAME,
DataType::Struct(
vec![
Field::new(Field::MAP_KEY_FIELD_DEFAULT_NAME, DataType::Utf8, false),
Field::new(Field::MAP_VALUE_FIELD_DEFAULT_NAME, DataType::Utf8, false),
]
.into(),
),
false,
)),
new_ordered,
);

let new_array_result = cast(&array, &new_type.clone());
assert!(!can_cast_types(array.data_type(), &new_type));
let Err(ArrowError::CastError(t)) = new_array_result else {
panic!();
};
assert_eq!(
t,
r#"Casting from Map("entries": non-null Struct("key": non-null Utf8, "value": Utf8), unsorted) to Map("entries": non-null Struct("key": non-null Utf8, "value": non-null Utf8), sorted) not supported"#
);
}

#[test]
fn test_cast_map_dont_allow_when_container_cant_cast() {
let string_builder = StringBuilder::new();
let value_builder = IntervalDayTimeArray::builder(2);
let mut builder = MapBuilder::new(None, string_builder, value_builder);

builder.keys().append_value("0");
builder.values().append_value(IntervalDayTime::new(1, 1));
builder.append(true).unwrap();
builder.keys().append_value("1");
builder.values().append_value(IntervalDayTime::new(2, 2));
builder.append(true).unwrap();

// map builder returns unsorted map by default
let array = builder.finish();

let new_ordered = true;
let new_type = DataType::Map(
Arc::new(Field::new(
Field::MAP_ENTRIES_FIELD_DEFAULT_NAME,
DataType::Struct(
vec![
Field::new(Field::MAP_KEY_FIELD_DEFAULT_NAME, DataType::Utf8, false),
Field::new(
Field::MAP_VALUE_FIELD_DEFAULT_NAME,
DataType::Duration(TimeUnit::Second),
false,
),
]
.into(),
),
false,
)),
new_ordered,
);

let new_array_result = cast(&array, &new_type.clone());
assert!(!can_cast_types(array.data_type(), &new_type));
let Err(ArrowError::CastError(t)) = new_array_result else {
panic!();
};
assert_eq!(
t,
r#"Casting from Map("entries": non-null Struct("key": non-null Utf8, "value": Interval(DayTime)), unsorted) to Map("entries": non-null Struct("key": non-null Utf8, "value": non-null Duration(s)), sorted) not supported"#
);
}

#[test]
fn test_cast_map_field_names() {
let string_builder = StringBuilder::new();
let value_builder = StringBuilder::new();
let mut builder = MapBuilder::new(
Some(MapFieldNames {
// Explicitly writing the name so it will be apparent from what names to what names are we converting to
entry: Field::MAP_ENTRIES_FIELD_DEFAULT_NAME.to_string(),
key: Field::MAP_KEY_FIELD_DEFAULT_NAME.to_string(),
value: Field::MAP_VALUE_FIELD_DEFAULT_NAME.to_string(),
}),
string_builder,
value_builder,
);

builder.keys().append_value("0");
builder.values().append_value("test_val_1");
builder.append(true).unwrap();
builder.keys().append_value("1");
builder.values().append_value("test_val_2");
builder.append(true).unwrap();
builder.append(false).unwrap();

let array = builder.finish();

let new_type = DataType::Map(
Arc::new(Field::new(
"entries_new",
DataType::Struct(
vec![
Field::new("key_new", DataType::Utf8, false),
Field::new("value_values", DataType::Utf8, false),
]
.into(),
),
false,
)),
false,
);

assert_ne!(new_type, array.data_type().clone());

let new_array = cast(&array, &new_type.clone()).unwrap();
assert_eq!(new_type, new_array.data_type().clone());
let map_array = new_array.as_map();

assert_ne!(new_type, array.data_type().clone());
assert_eq!(new_type, map_array.data_type().clone());

let key_string = map_array
.keys()
.as_any()
.downcast_ref::<StringArray>()
.unwrap()
.into_iter()
.flatten()
.collect::<Vec<_>>();
assert_eq!(&key_string, &vec!["0", "1"]);

let values_string_array = cast(map_array.values(), &DataType::Utf8).unwrap();
let values_string = values_string_array
.as_any()
.downcast_ref::<StringArray>()
.unwrap()
.into_iter()
.flatten()
.collect::<Vec<_>>();
assert_eq!(&values_string, &vec!["test_val_1", "test_val_2"]);

assert_eq!(
map_array.nulls(),
Some(&NullBuffer::from(vec![true, true, false]))
);
}

#[test]
fn test_cast_map_contained_values() {
let string_builder = StringBuilder::new();
let value_builder = Int8Builder::new();
let mut builder = MapBuilder::new(None, string_builder, value_builder);

builder.keys().append_value("0");
builder.values().append_value(44);
builder.append(true).unwrap();
builder.keys().append_value("1");
builder.values().append_value(22);
builder.append(true).unwrap();

let array = builder.finish();

let new_type = DataType::Map(
Arc::new(Field::new(
Field::MAP_ENTRIES_FIELD_DEFAULT_NAME,
DataType::Struct(
vec![
Field::new(Field::MAP_KEY_FIELD_DEFAULT_NAME, DataType::Utf8, false),
Field::new(Field::MAP_VALUE_FIELD_DEFAULT_NAME, DataType::Utf8, false),
]
.into(),
),
false,
)),
false,
);

let new_array = cast(&array, &new_type.clone()).unwrap();
assert_eq!(new_type, new_array.data_type().clone());
let map_array = new_array.as_map();

assert_ne!(new_type, array.data_type().clone());
assert_eq!(new_type, map_array.data_type().clone());

let key_string = map_array
.keys()
.as_any()
.downcast_ref::<StringArray>()
.unwrap()
.into_iter()
.flatten()
.collect::<Vec<_>>();
assert_eq!(&key_string, &vec!["0", "1"]);

let values_string_array = cast(map_array.values(), &DataType::Utf8).unwrap();
let values_string = values_string_array
.as_any()
.downcast_ref::<StringArray>()
.unwrap()
.into_iter()
.flatten()
.collect::<Vec<_>>();
assert_eq!(&values_string, &vec!["44", "22"]);
}

#[test]
fn test_utf8_cast_offsets() {
// test if offset of the array is taken into account during cast
Expand Down
Loading