Merge branch 'main' into friendlymatthew/compare-union-fields-eq

Author: friendlymatthew

.github/pull_request_template.md

@@ -1,28 +1,38 @@
 # Which issue does this PR close?
 
+<!--
 We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax.
+-->
 
 - Closes #NNN.
 
 # Rationale for this change
 
+<!--
 Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed.
 Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes.
+-->
 
 # What changes are included in this PR?
 
+<!--
 There is no need to duplicate the description in the issue here but it is sometimes worth providing a summary of the individual changes in this PR.
+-->
 
 # Are these changes tested?
 
+<!--
 We typically require tests for all PRs in order to:
 1. Prevent the code from being accidentally broken by subsequent changes
 2. Serve as another way to document the expected behavior of the code
 
 If tests are not included in your PR, please explain why (for example, are they covered by existing tests)?
+-->
 
 # Are there any user-facing changes?
 
+<!--
 If there are user-facing changes then we may require documentation to be updated before approving the PR.
 
 If there are any breaking changes to public APIs, please call them out.
+-->

.github/workflows/docs.yml

@@ -71,7 +71,7 @@ jobs:
     steps:
       - uses: actions/checkout@v6
       - name: Download crate docs
-        uses: actions/download-artifact@v6
+        uses: actions/download-artifact@v7
         with:
           name: crate-docs
           path: website/build

.github/workflows/integration.yml

@@ -149,12 +149,12 @@ jobs:
           rustup default ${{ matrix.rust }}
           rustup component add rustfmt clippy
       - name: Cache Cargo
-        uses: actions/cache@v4
+        uses: actions/cache@v5
         with:
           path: /home/runner/.cargo
           key: cargo-maturin-cache-
       - name: Cache Rust dependencies
-        uses: actions/cache@v4
+        uses: actions/cache@v5
         with:
           path: /home/runner/target
           # this key is not equal because maturin uses different compilation flags.

arrow-array/benches/union_array.rs

@@ -54,10 +54,10 @@ fn criterion_benchmark(c: &mut Criterion) {
                 |b| {
                     let type_ids = 0..with_nulls+without_nulls;
 
-                    let fields = UnionFields::new(
+                    let fields = UnionFields::try_new(
                         type_ids.clone(),
                         type_ids.clone().map(|i| Field::new(format!("f{i}"), DataType::Int32, true)),
-                    );
+                    ).unwrap();
 
                     let array = UnionArray::try_new(
                         fields,

arrow-array/src/array/list_view_array.rs

@@ -26,8 +26,8 @@ use crate::array::{make_array, print_long_array};
 use crate::builder::{GenericListViewBuilder, PrimitiveBuilder};
 use crate::iterator::GenericListViewArrayIter;
 use crate::{
-    Array, ArrayAccessor, ArrayRef, ArrowPrimitiveType, FixedSizeListArray, OffsetSizeTrait,
-    new_empty_array,
+    Array, ArrayAccessor, ArrayRef, ArrowPrimitiveType, FixedSizeListArray, GenericListArray,
+    OffsetSizeTrait, new_empty_array,
 };
 
 /// A [`GenericListViewArray`] of variable size lists, storing offsets as `i32`.
@@ -498,6 +498,29 @@ impl<OffsetSize: OffsetSizeTrait> std::fmt::Debug for GenericListViewArray<Offse
     }
 }
 
+impl<OffsetSize: OffsetSizeTrait> From<GenericListArray<OffsetSize>>
+    for GenericListViewArray<OffsetSize>
+{
+    fn from(value: GenericListArray<OffsetSize>) -> Self {
+        let (field, offsets, values, nulls) = value.into_parts();
+        let len = offsets.len() - 1;
+        let mut sizes = Vec::with_capacity(len);
+        let mut view_offsets = Vec::with_capacity(len);
+        for (i, offset) in offsets.iter().enumerate().take(len) {
+            view_offsets.push(*offset);
+            sizes.push(offsets[i + 1] - offsets[i]);
+        }
+
+        Self::new(
+            field,
+            ScalarBuffer::from(view_offsets),
+            ScalarBuffer::from(sizes),
+            values,
+            nulls,
+        )
+    }
+}
+
 impl<OffsetSize: OffsetSizeTrait> From<GenericListViewArray<OffsetSize>> for ArrayData {
     fn from(array: GenericListViewArray<OffsetSize>) -> Self {
         let len = array.len();

arrow-array/src/array/mod.rs

@@ -620,10 +620,11 @@ impl<'a> StringArrayType<'a> for &'a StringViewArray {
     }
 }
 
-/// A trait for Arrow String Arrays, currently three types are supported:
+/// A trait for Arrow Binary Arrays, currently four types are supported:
 /// - `BinaryArray`
 /// - `LargeBinaryArray`
 /// - `BinaryViewArray`
+/// - `FixedSizeBinaryArray`
 ///
 /// This trait helps to abstract over the different types of binary arrays
 /// so that we don't need to duplicate the implementation for each type.
@@ -642,6 +643,11 @@ impl<'a> BinaryArrayType<'a> for &'a BinaryViewArray {
         BinaryViewArray::iter(self)
     }
 }
+impl<'a> BinaryArrayType<'a> for &'a FixedSizeBinaryArray {
+    fn iter(&self) -> ArrayIter<Self> {
+        FixedSizeBinaryArray::iter(self)
+    }
+}
 
 impl PartialEq for dyn Array + '_ {
     fn eq(&self, other: &Self) -> bool {
@@ -1067,13 +1073,14 @@ mod tests {
     fn test_null_union() {
         for mode in [UnionMode::Sparse, UnionMode::Dense] {
             let data_type = DataType::Union(
-                UnionFields::new(
+                UnionFields::try_new(
                     vec![2, 1],
                     vec![
                         Field::new("foo", DataType::Int32, true),
                         Field::new("bar", DataType::Int64, true),
                     ],
-                ),
+                )
+                .unwrap(),
                 mode,
             );
             let array = new_null_array(&data_type, 4);

arrow-array/src/array/union_array.rs

@@ -1682,14 +1682,15 @@ mod tests {
     #[test]
     fn test_custom_type_ids() {
         let data_type = DataType::Union(
-            UnionFields::new(
+            UnionFields::try_new(
                 vec![8, 4, 9],
                 vec![
                     Field::new("strings", DataType::Utf8, false),
                     Field::new("integers", DataType::Int32, false),
                     Field::new("floats", DataType::Float64, false),
                 ],
-            ),
+            )
+            .unwrap(),
             UnionMode::Dense,
         );
 
@@ -1796,14 +1797,15 @@ mod tests {
     fn into_parts_custom_type_ids() {
         let set_field_type_ids: [i8; 3] = [8, 4, 9];
         let data_type = DataType::Union(
-            UnionFields::new(
+            UnionFields::try_new(
                 set_field_type_ids,
                 [
                     Field::new("strings", DataType::Utf8, false),
                     Field::new("integers", DataType::Int32, false),
                     Field::new("floats", DataType::Float64, false),
                 ],
-            ),
+            )
+            .unwrap(),
             UnionMode::Dense,
         );
         let string_array = StringArray::from(vec!["foo", "bar", "baz"]);
@@ -1836,13 +1838,14 @@ mod tests {
 
     #[test]
     fn test_invalid() {
-        let fields = UnionFields::new(
+        let fields = UnionFields::try_new(
             [3, 2],
             [
                 Field::new("a", DataType::Utf8, false),
                 Field::new("b", DataType::Utf8, false),
             ],
-        );
+        )
+        .unwrap();
         let children = vec![
             Arc::new(StringArray::from_iter_values(["a", "b"])) as _,
             Arc::new(StringArray::from_iter_values(["c", "d"])) as _,
@@ -1912,13 +1915,14 @@ mod tests {
 
         assert_eq!(array.logical_nulls(), None);
 
-        let fields = UnionFields::new(
+        let fields = UnionFields::try_new(
             [1, 3],
             [
                 Field::new("a", DataType::Int8, false), // non nullable
                 Field::new("b", DataType::Int8, false), // non nullable
             ],
-        );
+        )
+        .unwrap();
         let array = UnionArray::try_new(
             fields,
             vec![1].into(),
@@ -1932,13 +1936,14 @@ mod tests {
 
         assert_eq!(array.logical_nulls(), None);
 
-        let nullable_fields = UnionFields::new(
+        let nullable_fields = UnionFields::try_new(
             [1, 3],
             [
                 Field::new("a", DataType::Int8, true), // nullable but without nulls
                 Field::new("b", DataType::Int8, true), // nullable but without nulls
             ],
-        );
+        )
+        .unwrap();
         let array = UnionArray::try_new(
             nullable_fields.clone(),
             vec![1, 1].into(),

arrow-avro/benches/avro_writer.rs

@@ -688,14 +688,15 @@ static ENUM_DATA: Lazy<Vec<RecordBatch>> = Lazy::new(|| {
 
 static UNION_DATA: Lazy<Vec<RecordBatch>> = Lazy::new(|| {
     // Basic Dense Union of three types: Utf8, Int32, Float64
-    let union_fields = UnionFields::new(
+    let union_fields = UnionFields::try_new(
         vec![0, 1, 2],
         vec![
             Field::new("u_str", DataType::Utf8, true),
             Field::new("u_int", DataType::Int32, true),
             Field::new("u_f64", DataType::Float64, true),
         ],
-    );
+    )
+    .expect("UnionFields should be valid");
     let union_dt = DataType::Union(union_fields.clone(), UnionMode::Dense);
     let schema = schema_single("field1", union_dt);
 

arrow-avro/src/codec.rs

@@ -993,13 +993,13 @@ fn union_branch_name(dt: &AvroDataType) -> String {
     dt.codec.union_field_name()
 }
 
-fn build_union_fields(encodings: &[AvroDataType]) -> UnionFields {
+fn build_union_fields(encodings: &[AvroDataType]) -> Result<UnionFields, ArrowError> {
     let arrow_fields: Vec<Field> = encodings
         .iter()
         .map(|encoding| encoding.field_with_name(&union_branch_name(encoding)))
         .collect();
     let type_ids: Vec<i8> = (0..arrow_fields.len()).map(|i| i as i8).collect();
-    UnionFields::new(type_ids, arrow_fields)
+    UnionFields::try_new(type_ids, arrow_fields)
 }
 
 /// Resolves Avro type names to [`AvroDataType`]
@@ -1267,7 +1267,7 @@ impl<'a> Maker<'a> {
                     .map(|s| self.parse_type(s, namespace))
                     .collect::<Result<_, _>>()?;
                 // Build Arrow layout once here
-                let union_fields = build_union_fields(&children);
+                let union_fields = build_union_fields(&children)?;
                 Ok(AvroDataType::new(
                     Codec::Union(Arc::from(children), union_fields, UnionMode::Dense),
                     Default::default(),
@@ -1620,7 +1620,7 @@ impl<'a> Maker<'a> {
         for writer in writer_variants {
             writer_to_reader.push(self.find_best_promotion(writer, reader_variants, namespace));
         }
-        let union_fields = build_union_fields(&reader_encodings);
+        let union_fields = build_union_fields(&reader_encodings)?;
         let mut dt = AvroDataType::new(
             Codec::Union(reader_encodings.into(), union_fields, UnionMode::Dense),
             Default::default(),