This is an automated email from the ASF dual-hosted git repository.
kevinjqliu pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/iceberg-rust.git
The following commit(s) were added to refs/heads/main by this push:
new 33d97ba2 Add UUID support for the Avro schema (#1706)
33d97ba2 is described below
commit 33d97ba2906782ccd11178ae62a469d2049a4e8f
Author: Fokko Driesprong <fo...@apache.org>
AuthorDate: Wed Sep 24 04:00:43 2025 +0200
Add UUID support for the Avro schema (#1706)
## Which issue does this PR close?
`Fixed[16]` encoded UUIDs are now supported in Avro-rs:
https://github.com/apache/avro-rs/pull/255
This will automatically support String and Fixed[16] logical types if
the `logicalType` annotation is set correctly 👍
## What changes are included in this PR?
<!--
Provide a summary of the modifications in this PR. List the main changes
such as new features, bug fixes, refactoring, or any other updates.
-->
## Are these changes tested?
<!--
Specify what test covers (unit test, integration test, etc.).
If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?
-->
---
crates/iceberg/src/avro/schema.rs | 65 ++++++++++++++++-----------------------
1 file changed, 27 insertions(+), 38 deletions(-)
diff --git a/crates/iceberg/src/avro/schema.rs
b/crates/iceberg/src/avro/schema.rs
index b08a6730..fdbc6809 100644
--- a/crates/iceberg/src/avro/schema.rs
+++ b/crates/iceberg/src/avro/schema.rs
@@ -36,10 +36,7 @@ const ELEMENT_ID: &str = "element-id";
const FIELD_ID_PROP: &str = "field-id";
const KEY_ID: &str = "key-id";
const VALUE_ID: &str = "value-id";
-const UUID_BYTES: usize = 16;
-const UUID_LOGICAL_TYPE: &str = "uuid";
const MAP_LOGICAL_TYPE: &str = "map";
-// # TODO: https://github.com/apache/iceberg-rust/issues/86
// This const may better to maintain in avro-rs.
const LOGICAL_TYPE: &str = "logicalType";
@@ -237,8 +234,8 @@ impl SchemaVisitor for SchemaToAvroSchema {
PrimitiveType::TimestampNs => AvroSchema::TimestampNanos,
PrimitiveType::TimestamptzNs => AvroSchema::TimestampNanos,
PrimitiveType::String => AvroSchema::String,
- PrimitiveType::Uuid => avro_fixed_schema(UUID_BYTES,
Some(UUID_LOGICAL_TYPE))?,
- PrimitiveType::Fixed(len) => avro_fixed_schema((*len) as usize,
None)?,
+ PrimitiveType::Uuid => AvroSchema::Uuid,
+ PrimitiveType::Fixed(len) => avro_fixed_schema((*len) as usize)?,
PrimitiveType::Binary => AvroSchema::Bytes,
PrimitiveType::Decimal { precision, scale } => {
avro_decimal_schema(*precision as usize, *scale as usize)?
@@ -274,21 +271,13 @@ fn avro_record_schema(name: &str, fields:
Vec<AvroRecordField>) -> Result<AvroSc
}))
}
-pub(crate) fn avro_fixed_schema(len: usize, logical_type: Option<&str>) ->
Result<AvroSchema> {
- let attributes = if let Some(logical_type) = logical_type {
- BTreeMap::from([(
- LOGICAL_TYPE.to_string(),
- Value::String(logical_type.to_string()),
- )])
- } else {
- Default::default()
- };
+pub(crate) fn avro_fixed_schema(len: usize) -> Result<AvroSchema> {
Ok(AvroSchema::Fixed(FixedSchema {
name: Name::new(format!("fixed_{len}").as_str())?,
aliases: None,
doc: None,
size: len,
- attributes,
+ attributes: Default::default(),
default: None,
}))
}
@@ -533,30 +522,9 @@ impl AvroSchemaVisitor for AvroSchemaToSchema {
AvroSchema::Long => Type::Primitive(PrimitiveType::Long),
AvroSchema::Float => Type::Primitive(PrimitiveType::Float),
AvroSchema::Double => Type::Primitive(PrimitiveType::Double),
+ AvroSchema::Uuid => Type::Primitive(PrimitiveType::Uuid),
AvroSchema::String | AvroSchema::Enum(_) =>
Type::Primitive(PrimitiveType::String),
- AvroSchema::Fixed(fixed) => {
- if let Some(logical_type) = fixed.attributes.get(LOGICAL_TYPE)
{
- let logical_type = logical_type.as_str().ok_or_else(|| {
- Error::new(
- ErrorKind::DataInvalid,
- "logicalType in attributes of avro schema is not a
string type",
- )
- })?;
- match logical_type {
- UUID_LOGICAL_TYPE =>
Type::Primitive(PrimitiveType::Uuid),
- ty => {
- return Err(Error::new(
- ErrorKind::FeatureUnsupported,
- format!(
- "Logical type {ty} is not support in
iceberg primitive type.",
- ),
- ));
- }
- }
- } else {
- Type::Primitive(PrimitiveType::Fixed(fixed.size as u64))
- }
- }
+ AvroSchema::Fixed(fixed) =>
Type::Primitive(PrimitiveType::Fixed(fixed.size as u64)),
AvroSchema::Bytes => Type::Primitive(PrimitiveType::Binary),
AvroSchema::Null => return Ok(None),
_ => {
@@ -1223,4 +1191,25 @@ mod tests {
converter.primitive(&AvroSchema::Date).unwrap().unwrap()
);
}
+
+ #[test]
+ fn test_uuid_type() {
+ let avro_schema = {
+ AvroSchema::parse_str(
+ r#"
+ {"name": "test", "type": "fixed", "size": 16, "logicalType":
"uuid"}
+ "#,
+ )
+ .unwrap()
+ };
+
+ let mut converter = AvroSchemaToSchema;
+
+ let iceberg_type = Type::from(PrimitiveType::Uuid);
+
+ assert_eq!(
+ iceberg_type,
+ converter.primitive(&avro_schema).unwrap().unwrap()
+ );
+ }
}
