[arrow-rs] branch master updated: Fix DataTypeLayout for LargeList (#3503)

[tustvold]

This is an automated email from the ASF dual-hosted git repository.

tustvold pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git


The following commit(s) were added to refs/heads/master by this push:
     new 55c87c114 Fix DataTypeLayout for LargeList (#3503)
55c87c114 is described below

commit 55c87c114443739ed73ebc28d0ba53bf875ecd9a
Author: Liang-Chi Hsieh <vii...@gmail.com>
AuthorDate: Wed Jan 11 13:41:07 2023 -0800

    Fix DataTypeLayout for LargeList (#3503)
    
    * Fix DataTypeLayout for LargeList
    
    * Add datalayout test
---
 arrow-array/src/types.rs | 44 ++++++++++++++++++++++++++++++++++++++++++++
 arrow-data/src/data.rs   |  2 +-
 2 files changed, 45 insertions(+), 1 deletion(-)

diff --git a/arrow-array/src/types.rs b/arrow-array/src/types.rs
index e7d92d2d0..7c41a469e 100644
--- a/arrow-array/src/types.rs
+++ b/arrow-array/src/types.rs
@@ -767,6 +767,8 @@ pub type LargeBinaryType = GenericBinaryType<i64>;
 #[cfg(test)]
 mod tests {
     use super::*;
+    use arrow_data::{layout, BufferSpec};
+    use std::mem::size_of;
 
     #[test]
     fn month_day_nano_should_roundtrip() {
@@ -803,4 +805,46 @@ mod tests {
         let value = IntervalYearMonthType::make_value(-1, -2);
         assert_eq!(IntervalYearMonthType::to_months(value), -14);
     }
+
+    fn test_layout<T: ArrowPrimitiveType>() {
+        let layout = layout(&T::DATA_TYPE);
+
+        assert_eq!(layout.buffers.len(), 1);
+
+        let spec = &layout.buffers[0];
+        assert_eq!(
+            spec,
+            &BufferSpec::FixedWidth {
+                byte_width: size_of::<T::Native>()
+            }
+        );
+    }
+
+    #[test]
+    fn test_layouts() {
+        test_layout::<Int8Type>();
+        test_layout::<Int16Type>();
+        test_layout::<Int32Type>();
+        test_layout::<Int64Type>();
+        test_layout::<UInt8Type>();
+        test_layout::<UInt16Type>();
+        test_layout::<UInt32Type>();
+        test_layout::<UInt64Type>();
+        test_layout::<Float16Type>();
+        test_layout::<Float32Type>();
+        test_layout::<Float64Type>();
+        test_layout::<TimestampSecondType>();
+        test_layout::<Date32Type>();
+        test_layout::<Date64Type>();
+        test_layout::<Time32SecondType>();
+        test_layout::<Time32MillisecondType>();
+        test_layout::<Time64MicrosecondType>();
+        test_layout::<Time64NanosecondType>();
+        test_layout::<IntervalMonthDayNanoType>();
+        test_layout::<IntervalDayTimeType>();
+        test_layout::<IntervalYearMonthType>();
+        test_layout::<DurationNanosecondType>();
+        test_layout::<DurationMicrosecondType>();
+        test_layout::<DurationMillisecondType>();
+    }
 }
diff --git a/arrow-data/src/data.rs b/arrow-data/src/data.rs
index 31dad5e82..14dbe9387 100644
--- a/arrow-data/src/data.rs
+++ b/arrow-data/src/data.rs
@@ -1470,7 +1470,7 @@ pub fn layout(data_type: &DataType) -> DataTypeLayout {
         DataType::LargeUtf8 => DataTypeLayout::new_binary(size_of::<i64>()),
         DataType::List(_) => DataTypeLayout::new_fixed_width(size_of::<i32>()),
         DataType::FixedSizeList(_, _) => DataTypeLayout::new_empty(), // all 
in child data
-        DataType::LargeList(_) => 
DataTypeLayout::new_fixed_width(size_of::<i32>()),
+        DataType::LargeList(_) => 
DataTypeLayout::new_fixed_width(size_of::<i64>()),
         DataType::Struct(_) => DataTypeLayout::new_empty(), // all in child 
data,
         DataType::Union(_, _, mode) => {
             let type_ids = BufferSpec::FixedWidth {