ci: use official action to setup rust toolchain

.github/dependabot.yaml

@@ -0,0 +1,14 @@
+# Please see the documentation for all configuration options:
+# https://help.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
+
+version: 2
+updates:
+  - package-ecosystem: "cargo"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "weekly"

.github/workflows/build.yml

@@ -10,27 +10,20 @@ jobs:
   format:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install minimal stable with clippy and rustfmt
-        uses: actions-rs/toolchain@v1
+        uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+          components: rustfmt
       - name: format
         run: cargo fmt -- --check
+
   msrv:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install minimal stable and cargo msrv
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Install cargo-msrv
         shell: bash
         run: |
@@ -41,21 +34,18 @@ jobs:
           cargo msrv --path derive-macros/ verify --all-features
           cargo msrv --path ffi/ verify --all-features
           cargo msrv --path ffi-proc-macros/ verify --all-features
+
   docs:
     runs-on: ubuntu-latest
     env:
       RUSTDOCFLAGS: -D warnings
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install minimal stable
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: build docs
         run: cargo doc
+
   build:
     runs-on: ${{ matrix.os }}
     strategy:
@@ -65,18 +55,16 @@ jobs:
           - ubuntu-latest
           - windows-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install minimal stable with clippy and rustfmt
-        uses: actions-rs/toolchain@v1
+        uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+          components: clippy
       - name: build and lint with clippy
         run: cargo clippy --benches --tests --all-features -- -D warnings
       - name: lint without default features
         run: cargo clippy --no-default-features -- -D warnings
+
   test:
     runs-on: ${{ matrix.os }}
     strategy:
@@ -86,14 +74,9 @@ jobs:
           - ubuntu-latest
           - windows-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Install minimal stable with clippy and rustfmt
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: test
         run: cargo test --workspace --verbose --all-features -- --skip read_table_version_hdfs
 
@@ -109,7 +92,7 @@ jobs:
       - name: Setup cmake
         uses: jwlawson/actions-setup-cmake@v2
         with:
-          cmake-version: '3.30.x'
+          cmake-version: "3.30.x"
       - name: Install arrow-glib
         run: |
           if [ "$RUNNER_OS" == "Linux" ]; then
@@ -186,14 +169,9 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - name: Install rust
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Install cargo-llvm-cov
         uses: taiki-e/install-action@cargo-llvm-cov
-      - uses: Swatinem/rust-cache@v2
       - name: Generate code coverage
         run: cargo llvm-cov --all-features --workspace --codecov --output-path codecov.json -- --skip read_table_version_hdfs
       - name: Upload coverage to Codecov

.github/workflows/run_integration_test.yml

@@ -18,17 +18,11 @@ jobs:
       - name: Skip job for pull requests on Windows
         if: ${{ matrix.skip }}
         run: echo "Skipping job for pull requests on Windows."
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         if: ${{ !matrix.skip }}
-      - name: Install minimal stable rust
-        if: ${{ !matrix.skip }}
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+      - name: Setup rust toolchain
         if: ${{ !matrix.skip }}
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Run integration tests
         if: ${{ !matrix.skip }}
         shell: bash

.github/workflows/semver-checks.yml

@@ -25,12 +25,7 @@ jobs:
           fetch-depth: 0
           ref: ${{ github.event.pull_request.head.sha }}
       - name: Install minimal stable
-        uses: actions-rs/toolchain@v1
-        with:
-          profile: default
-          toolchain: stable
-          override: true
-      - uses: Swatinem/rust-cache@v2
+        uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Install cargo-semver-checks
         shell: bash
         run: |

kernel/src/engine/arrow_data.rs

@@ -4,9 +4,11 @@ use crate::{DeltaResult, Error};
 
 use arrow_array::cast::AsArray;
 use arrow_array::types::{Int32Type, Int64Type};
-use arrow_array::{Array, ArrayRef, GenericListArray, MapArray, OffsetSizeTrait, RecordBatch, StructArray};
-use arrow_schema::{FieldRef, DataType as ArrowDataType};
-use tracing::{debug};
+use arrow_array::{
+    Array, ArrayRef, GenericListArray, MapArray, OffsetSizeTrait, RecordBatch, StructArray,
+};
+use arrow_schema::{DataType as ArrowDataType, FieldRef};
+use tracing::debug;
 
 use std::collections::{HashMap, HashSet};
 
@@ -138,14 +140,20 @@ impl EngineData for ArrowEngineData {
         self.data.num_rows()
     }
 
-    fn visit_rows(&self, leaf_columns: &[ColumnName], visitor: &mut dyn RowVisitor) -> DeltaResult<()> {
+    fn visit_rows(
+        &self,
+        leaf_columns: &[ColumnName],
+        visitor: &mut dyn RowVisitor,
+    ) -> DeltaResult<()> {
         // Make sure the caller passed the correct number of column names
         let leaf_types = visitor.selected_column_names_and_types().1;
         if leaf_types.len() != leaf_columns.len() {
             return Err(Error::MissingColumn(format!(
                 "Visitor expected {} column names, but caller passed {}",
-                leaf_types.len(), leaf_columns.len()
-            )).with_backtrace());
+                leaf_types.len(),
+                leaf_columns.len()
+            ))
+            .with_backtrace());
         }
 
         // Collect the names of all leaf columns we want to extract, along with their parents, to
@@ -154,20 +162,19 @@ impl EngineData for ArrowEngineData {
         let mut mask = HashSet::new();
         for column in leaf_columns {
             for i in 0..column.len() {
-                mask.insert(&column[..i+1]);
+                mask.insert(&column[..i + 1]);
             }
         }
         debug!("Column mask for selected columns {leaf_columns:?} is {mask:#?}");
 
         let mut getters = vec![];
         Self::extract_columns(&mut vec![], &mut getters, leaf_types, &mask, &self.data)?;
         if getters.len() != leaf_columns.len() {
-            return Err(Error::MissingColumn(
-                format!(
-                    "Visitor expected {} leaf columns, but only {} were found in the data",
-                    leaf_columns.len(), getters.len()
-                )
-            ));
+            return Err(Error::MissingColumn(format!(
+                "Visitor expected {} leaf columns, but only {} were found in the data",
+                leaf_columns.len(),
+                getters.len()
+            )));
         }
         visitor.visit(self.len(), &getters)
     }
@@ -185,14 +192,11 @@ impl ArrowEngineData {
             path.push(field.name().to_string());
             if column_mask.contains(&path[..]) {
                 if let Some(struct_array) = column.as_struct_opt() {
-                    debug!("Recurse into a struct array for {}", ColumnName::new(path.iter()));
-                    Self::extract_columns(
-                        path,
-                        getters,
-                        leaf_types,
-                        column_mask,
-                        struct_array,
-                    )?;
+                    debug!(
+                        "Recurse into a struct array for {}",
+                        ColumnName::new(path.iter())
+                    );
+                    Self::extract_columns(path, getters, leaf_types, column_mask, struct_array)?;
                 } else if column.data_type() == &ArrowDataType::Null {
                     debug!("Pushing a null array for {}", ColumnName::new(path.iter()));
                     getters.push(&());
@@ -215,16 +219,20 @@ impl ArrowEngineData {
         col: &'a dyn Array,
     ) -> DeltaResult<&'a dyn GetData<'a>> {
         use ArrowDataType::Utf8;
-        let col_as_list = || if let Some(array) = col.as_list_opt::<i32>() {
-            (array.value_type() == Utf8).then_some(array as _)
-        } else if let Some(array) = col.as_list_opt::<i64>() {
-            (array.value_type() == Utf8).then_some(array as _)
-        } else {
-            None
+        let col_as_list = || {
+            if let Some(array) = col.as_list_opt::<i32>() {
+                (array.value_type() == Utf8).then_some(array as _)
+            } else if let Some(array) = col.as_list_opt::<i64>() {
+                (array.value_type() == Utf8).then_some(array as _)
+            } else {
+                None
+            }
+        };
+        let col_as_map = || {
+            col.as_map_opt().and_then(|array| {
+                (array.key_type() == &Utf8 && array.value_type() == &Utf8).then_some(array as _)
+            })
         };
-        let col_as_map = || col.as_map_opt().and_then(|array| {
-            (array.key_type() == &Utf8 && array.value_type() == &Utf8).then_some(array as _)
-        });
         let result: Result<&'a dyn GetData<'a>, _> = match data_type {
             &DataType::BOOLEAN => {
                 debug!("Pushing boolean array for {}", ColumnName::new(path));
@@ -236,11 +244,15 @@ impl ArrowEngineData {
             }
             &DataType::INTEGER => {
                 debug!("Pushing int32 array for {}", ColumnName::new(path));
-                col.as_primitive_opt::<Int32Type>().map(|a| a as _).ok_or("int")
+                col.as_primitive_opt::<Int32Type>()
+                    .map(|a| a as _)
+                    .ok_or("int")
             }
             &DataType::LONG => {
                 debug!("Pushing int64 array for {}", ColumnName::new(path));
-                col.as_primitive_opt::<Int64Type>().map(|a| a as _).ok_or("long")
+                col.as_primitive_opt::<Int64Type>()
+                    .map(|a| a as _)
+                    .ok_or("long")
             }
             DataType::Array(_) => {
                 debug!("Pushing list for {}", ColumnName::new(path));
@@ -252,14 +264,17 @@ impl ArrowEngineData {
             }
             data_type => {
                 return Err(Error::UnexpectedColumnType(format!(
-                    "On {}: Unsupported type {data_type}", ColumnName::new(path)
+                    "On {}: Unsupported type {data_type}",
+                    ColumnName::new(path)
                 )));
             }
         };
         result.map_err(|type_name| {
             Error::UnexpectedColumnType(format!(
                 "Type mismatch on {}: expected {}, got {}",
-                ColumnName::new(path), type_name, col.data_type()
+                ColumnName::new(path),
+                type_name,
+                col.data_type()
             ))
         })
     }