QuakeWang commented on code in PR #411:
URL: https://github.com/apache/paimon-rust/pull/411#discussion_r3478772088
##########
crates/paimon/src/spec/types.rs:
##########
@@ -241,6 +250,196 @@ impl ArrayType {
}
}
+/// VectorType for paimon.
+///
+/// Data type of a fixed-size dense vector `VECTOR<element, length>`. Elements
are densely
+/// stored. The element type must be one of BOOLEAN, TINYINT, SMALLINT, INT,
BIGINT, FLOAT,
+/// DOUBLE, and `length` must be between 1 and `i32::MAX` (both inclusive).
+///
+/// Impl Reference:
<https://github.com/apache/paimon/blob/master/paimon-api/src/main/java/org/apache/paimon/types/VectorType.java>.
+#[serde_as]
+#[derive(Debug, Clone, PartialEq, Eq, Hash, Serialize)]
+pub struct VectorType {
+ #[serde(rename = "type")]
+ #[serde_as(as =
"FromInto<serde_utils::NullableType<serde_utils::VECTOR>>")]
+ nullable: bool,
+ #[serde(rename = "element")]
+ element_type: Box<DataType>,
+ length: u32,
+}
+
+impl VectorType {
+ /// Maximum vector length, matching Java `Integer.MAX_VALUE`.
+ pub const MAX_LENGTH: u32 = i32::MAX as u32;
+ /// Minimum vector length.
+ pub const MIN_LENGTH: u32 = 1;
+
+ pub fn new(length: u32, element_type: DataType) -> Result<Self> {
+ Self::try_new(true, length, element_type)
+ }
+
+ pub fn with_nullable(nullable: bool, length: u32, element_type: DataType)
-> Result<Self> {
+ Self::try_new(nullable, length, element_type)
+ }
+
+ pub fn try_new(nullable: bool, length: u32, element_type: DataType) ->
Result<Self> {
+ if !Self::is_valid_element_type(&element_type) {
+ return Err(Error::DataTypeInvalid {
+ message: format!("Invalid element type for vector:
{element_type:?}"),
+ });
+ }
+ if !(Self::MIN_LENGTH..=Self::MAX_LENGTH).contains(&length) {
+ return Err(Error::DataTypeInvalid {
+ message: format!(
+ "Vector length must be between {} and {} (both inclusive),
got {length}.",
+ Self::MIN_LENGTH,
+ Self::MAX_LENGTH
+ ),
+ });
+ }
+ Ok(Self {
+ nullable,
+ element_type: Box::new(element_type),
+ length,
+ })
+ }
+
+ fn is_valid_element_type(dt: &DataType) -> bool {
+ matches!(
+ dt,
+ DataType::Boolean(_)
+ | DataType::TinyInt(_)
+ | DataType::SmallInt(_)
+ | DataType::Int(_)
+ | DataType::BigInt(_)
+ | DataType::Float(_)
+ | DataType::Double(_)
+ )
+ }
+
+ pub fn family(&self) -> DataTypeFamily {
+ DataTypeFamily::CONSTRUCTED | DataTypeFamily::COLLECTION
+ }
+
+ pub fn element_type(&self) -> &DataType {
+ &self.element_type
+ }
+
+ pub fn length(&self) -> u32 {
+ self.length
+ }
+
+ /// SQL name of a valid vector element type.
+ fn element_sql_name(&self) -> &'static str {
+ match self.element_type.as_ref() {
+ DataType::Boolean(_) => "BOOLEAN",
+ DataType::TinyInt(_) => "TINYINT",
+ DataType::SmallInt(_) => "SMALLINT",
+ DataType::Int(_) => "INT",
+ DataType::BigInt(_) => "BIGINT",
+ DataType::Float(_) => "FLOAT",
+ DataType::Double(_) => "DOUBLE",
+ other => {
+ unreachable!("vector element type validated at construction:
{other:?}")
+ }
+ }
+ }
+}
+
+impl Display for VectorType {
+ fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+ write!(f, "VECTOR<{}, {}>", self.element_sql_name(), self.length)?;
Review Comment:
`Display` drops the element type nullability. For example, a `VECTOR<FLOAT
NOT NULL, 4>`-like type would render as `VECTOR<FLOAT, 4>`, while serde
preserves the element nullability. Please either render `element_type` with its
SQL string or reject/normalize non-null vector element types.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
