flaneur2020 commented on code in PR #871:
URL: https://github.com/apache/iceberg-rust/pull/871#discussion_r1901836558
##########
crates/iceberg/src/metadata_scan.rs:
##########
@@ -134,44 +137,122 @@ pub struct ManifestsTable<'a> {
}
impl<'a> ManifestsTable<'a> {
- fn partition_summary_fields(&self) -> Vec<Field> {
- vec![
- Field::new("contains_null", DataType::Boolean, false),
- Field::new("contains_nan", DataType::Boolean, true),
- Field::new("lower_bound", DataType::Utf8, true),
- Field::new("upper_bound", DataType::Utf8, true),
- ]
- }
-
- /// Returns the schema of the manifests table.
- pub fn schema(&self) -> Schema {
- Schema::new(vec![
- Field::new("content", DataType::Int8, false),
- Field::new("path", DataType::Utf8, false),
- Field::new("length", DataType::Int64, false),
- Field::new("partition_spec_id", DataType::Int32, false),
- Field::new("added_snapshot_id", DataType::Int64, false),
- Field::new("added_data_files_count", DataType::Int32, false),
- Field::new("existing_data_files_count", DataType::Int32, false),
- Field::new("deleted_data_files_count", DataType::Int32, false),
- Field::new("added_delete_files_count", DataType::Int32, false),
- Field::new("existing_delete_files_count", DataType::Int32, false),
- Field::new("deleted_delete_files_count", DataType::Int32, false),
- Field::new(
+ /// Returns the iceberg schema of the manifests table.
+ pub fn schema(&self) -> crate::spec::Schema {
+ let fields = vec![
+ NestedField::new(14, "content",
Type::Primitive(PrimitiveType::Int), true),
+ NestedField::new(1, "path",
Type::Primitive(PrimitiveType::String), true),
+ NestedField::new(2, "length",
Type::Primitive(PrimitiveType::Long), true),
+ NestedField::new(
+ 3,
+ "partition_spec_id",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 4,
+ "added_snapshot_id",
+ Type::Primitive(PrimitiveType::Long),
+ true,
+ ),
+ NestedField::new(
+ 5,
+ "added_data_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 6,
+ "existing_data_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 7,
+ "deleted_data_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 15,
+ "added_delete_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 16,
+ "existing_delete_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 17,
+ "deleted_delete_files_count",
+ Type::Primitive(PrimitiveType::Int),
+ true,
+ ),
+ NestedField::new(
+ 8,
"partition_summaries",
- DataType::List(Arc::new(Field::new_struct(
- "item",
- self.partition_summary_fields(),
- false,
- ))),
- false,
+ Type::List(ListType {
+ element_field: Arc::new(NestedField::new(
+ 9,
+ "item",
+ Type::Struct(StructType::new(vec![
+ Arc::new(NestedField::new(
+ 10,
+ "contains_null",
+ Type::Primitive(PrimitiveType::Boolean),
+ true,
+ )),
+ Arc::new(NestedField::new(
+ 11,
+ "contains_nan",
+ Type::Primitive(PrimitiveType::Boolean),
+ false,
+ )),
+ Arc::new(NestedField::new(
+ 12,
+ "lower_bound",
+ Type::Primitive(PrimitiveType::String),
+ false,
+ )),
+ Arc::new(NestedField::new(
+ 13,
+ "upper_bound",
+ Type::Primitive(PrimitiveType::String),
+ false,
+ )),
+ ])),
+ true,
+ )),
+ }),
+ true,
),
- ])
+ ];
+
+ crate::spec::Schema::builder()
+ .with_fields(fields.into_iter().map(|f| f.into()))
+ .build()
+ .unwrap()
}
/// Scans the manifests table.
pub async fn scan(&self) -> Result<RecordBatch> {
- let mut content = PrimitiveBuilder::<Int8Type>::new();
+ let schema = schema_to_arrow_schema(&self.schema())?;
+ let partition_summary_fields = if let DataType::List(list_type) =
+ schema.field_with_name("partition_summaries")?.data_type()
+ {
+ if let DataType::Struct(fields) = list_type.data_type() {
+ fields.to_vec()
+ } else {
+ unreachable!()
+ }
+ } else {
+ unreachable!()
+ };
Review Comment:
very ugly here 😲
--
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: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
