tustvold commented on code in PR #6524:
URL: https://github.com/apache/arrow-rs/pull/6524#discussion_r1803829049
##########
parquet/src/arrow/arrow_writer/mod.rs:
##########
@@ -1614,6 +1614,58 @@ mod tests {
roundtrip(batch, Some(SMALL_SIZE / 2));
}
+ #[test]
+ fn arrow_writer_2_level_struct_mixed_null_2() {
+ // tests writing <struct<struct<primitive>>, where the primitive
columns are non-null.
+ let field_c = Field::new("c", DataType::Int32, false);
+ let field_d = Field::new("d", DataType::FixedSizeBinary(4), false);
+ let field_e = Field::new(
+ "e",
+ DataType::Dictionary(Box::new(DataType::Int32),
Box::new(DataType::Utf8)),
+ false,
+ );
+
+ let field_b = Field::new(
+ "b",
+ DataType::Struct(vec![field_c, field_d, field_e].into()),
+ false,
+ );
+ let type_a = DataType::Struct(vec![field_b.clone()].into());
+ let field_a = Field::new("a", type_a, true);
+ let schema = Schema::new(vec![field_a.clone()]);
+
+ // create data
+ let c = Int32Array::from_iter_values(0..6);
+ let d = FixedSizeBinaryArray::try_from_iter(
+ ["aaaa", "bbbb", "cccc", "dddd", "eeee", "ffff"].into_iter(),
+ )
+ .expect("four byte values");
+ let e = Int32DictionaryArray::from_iter(["one", "two", "three",
"four", "five", "one"]);
+ let b_data = ArrayDataBuilder::new(field_b.data_type().clone())
+ .len(6)
+ .add_child_data(c.into_data())
+ .add_child_data(d.into_data())
+ .add_child_data(e.into_data())
+ .build()
+ .unwrap();
+ let b = StructArray::from(b_data);
+ let a_data = ArrayDataBuilder::new(field_a.data_type().clone())
+ .len(6)
+ .null_bit_buffer(Some(Buffer::from([0b00100101])))
+ .add_child_data(b.into_data())
+ .build()
+ .unwrap();
+ let a = StructArray::from(a_data);
+
+ assert_eq!(a.null_count(), 3);
+ assert_eq!(a.column(0).null_count(), 0);
+
+ // build a record batch
+ let batch = RecordBatch::try_new(Arc::new(schema),
vec![Arc::new(a)]).unwrap();
+
+ roundtrip(batch, Some(SMALL_SIZE / 2));
+ }
+
Review Comment:
```suggestion
#[test]
fn test_empty_dict() {
let struct_fields = Fields::from(vec![Field::new(
"dict",
DataType::Dictionary(Box::new(DataType::Int32),
Box::new(DataType::Utf8)),
false,
)]);
let schema = Schema::new(vec![Field::new_struct(
"struct",
struct_fields.clone(),
true,
)]);
let dictionary = Arc::new(DictionaryArray::new(
Int32Array::new_null(5),
Arc::new(StringArray::new_null(0)),
));
let s = StructArray::new(
struct_fields,
vec![dictionary],
Some(NullBuffer::new_null(5)),
);
let batch = RecordBatch::try_new(Arc::new(schema),
vec![Arc::new(s)]).unwrap();
roundtrip(batch, None);
}
```
--
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]
