Merge 5270354 into 85d5363

.github/workflows/rust.yml

@@ -82,6 +82,16 @@ jobs:
       - uses: actions/checkout@v2
         with:
           submodules: true
+      - name: Install protobuf compiler
+        shell: bash
+        run: |
+          mkdir -p $HOME/d/protoc
+          cd $HOME/d/protoc
+          export PROTO_ZIP="protoc-21.4-linux-x86_64.zip"
+          curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v21.4/$PROTO_ZIP
+          unzip $PROTO_ZIP
+          export PATH=$PATH:$HOME/d/protoc/bin
+          protoc --version
       - name: Cache Cargo
         uses: actions/cache@v3
         with:
@@ -94,6 +104,7 @@ jobs:
           rust-version: ${{ matrix.rust }}
       - name: Run tests
         run: |
+          export PATH=$PATH:$HOME/d/protoc/bin
           cargo test --features avro,jit,scheduler,json
           # test datafusion-sql examples
           cargo run --example sql
@@ -159,17 +170,65 @@ jobs:
           POSTGRES_USER: postgres
           POSTGRES_PASSWORD: postgres
 
-  windows-and-macos:
-    name: Test on ${{ matrix.os }} Rust ${{ matrix.rust }}
+  windows:
+    name: Test on Windows Rust ${{ matrix.rust }}
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [windows-latest, macos-latest]
+        os: [windows-latest]
         rust: [stable]
     steps:
       - uses: actions/checkout@v2
         with:
           submodules: true
+      - name: Install protobuf compiler
+        shell: bash
+        run: |
+          mkdir -p $HOME/d/protoc
+          cd $HOME/d/protoc
+          export PROTO_ZIP="protoc-21.4-win64.zip"
+          curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v21.4/$PROTO_ZIP
+          unzip $PROTO_ZIP
+          export PATH=$PATH:$HOME/d/protoc/bin
+          protoc.exe --version
+      # TODO: this won't cache anything, which is expensive. Setup this action
+      # with a OS-dependent path.
+      - name: Setup Rust toolchain
+        run: |
+          rustup toolchain install ${{ matrix.rust }}
+          rustup default ${{ matrix.rust }}
+          rustup component add rustfmt
+      - name: Run tests
+        shell: bash
+        run: |
+          export PATH=$PATH:$HOME/d/protoc/bin
+          cargo test
+        env:
+          # do not produce debug symbols to keep memory usage down
+          RUSTFLAGS: "-C debuginfo=0"
+
+  macos:
+    name: Test on MacOS Rust ${{ matrix.rust }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [macos-latest]
+        rust: [stable]
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          submodules: true
+      - name: Install protobuf compiler
+        shell: bash
+        run: |
+          mkdir -p $HOME/d/protoc
+          cd $HOME/d/protoc
+          export PROTO_ZIP="protoc-21.4-osx-x86_64.zip"
+          curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v21.4/$PROTO_ZIP
+          unzip $PROTO_ZIP
+          echo "$HOME/d/protoc/bin" >> $GITHUB_PATH
+          export PATH=$PATH:$HOME/d/protoc/bin
+          protoc --version
       # TODO: this won't cache anything, which is expensive. Setup this action
       # with a OS-dependent path.
       - name: Setup Rust toolchain
@@ -250,6 +309,16 @@ jobs:
       - uses: actions/checkout@v2
         with:
           submodules: true
+      - name: Install protobuf compiler
+        shell: bash
+        run: |
+          mkdir -p $HOME/d/protoc
+          cd $HOME/d/protoc
+          export PROTO_ZIP="protoc-21.4-linux-x86_64.zip"
+          curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v21.4/$PROTO_ZIP
+          unzip $PROTO_ZIP
+          export PATH=$PATH:$HOME/d/protoc/bin
+          protoc --version
       - name: Setup Rust toolchain
         run: |
           rustup toolchain install ${{ matrix.rust }}
@@ -263,6 +332,7 @@ jobs:
           key: cargo-coverage-cache3-
       - name: Run coverage
         run: |
+          export PATH=$PATH:$HOME/d/protoc/bin
           rustup toolchain install stable
           rustup default stable
           cargo install --version 0.20.1 cargo-tarpaulin

benchmarks/Cargo.toml

@@ -32,7 +32,7 @@ simd = ["datafusion/simd"]
 snmalloc = ["snmalloc-rs"]
 
 [dependencies]
-datafusion = { path = "../datafusion/core" }
+datafusion = { path = "../datafusion/core", features = [], optional = false }
 env_logger = "0.9"
 futures = "0.3"
 mimalloc = { version = "0.1", optional = true, default-features = false }

datafusion-cli/Cargo.toml

@@ -29,9 +29,9 @@ rust-version = "1.59"
 readme = "README.md"
 
 [dependencies]
-arrow = { version = "19.0.0" }
+arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = [], optional = false }
 clap = { version = "3", features = ["derive", "cargo"] }
-datafusion = { path = "../datafusion/core", version = "10.0.0" }
+datafusion = { path = "../datafusion/core", features = [], optional = false }
 dirs = "4.0.0"
 env_logger = "0.9"
 mimalloc = { version = "0.1", default-features = false }

datafusion-examples/Cargo.toml

@@ -34,13 +34,13 @@ path = "examples/avro_sql.rs"
 required-features = ["datafusion/avro"]
 
 [dev-dependencies]
-arrow-flight = { version = "19.0.0" }
+arrow-flight = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = [], optional = false }
 async-trait = "0.1.41"
-datafusion = { path = "../datafusion/core" }
+datafusion = { path = "../datafusion/core", features = [], optional = false }
 futures = "0.3"
 num_cpus = "1.13.0"
-prost = "0.10"
+prost = "0.11.0"
 serde = { version = "1.0.136", features = ["derive"] }
 serde_json = "1.0.82"
 tokio = { version = "1.0", features = ["macros", "rt", "rt-multi-thread", "sync", "parking_lot"] }
-tonic = "0.7"
+tonic = "0.8"

datafusion/common/Cargo.toml

@@ -39,11 +39,12 @@ pyarrow = ["pyo3"]
 
 [dependencies]
 apache-avro = { version = "0.14", features = ["snappy"], optional = true }
-arrow = { version = "19.0.0", features = ["prettyprint"] }
+arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = ["prettyprint"], optional = false }
+avro-rs = { version = "0.13", features = ["snappy"], optional = true }
 cranelift-module = { version = "0.86.1", optional = true }
-object_store = { version = "0.3", optional = true }
+object_store = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = [], optional = true }
 ordered-float = "3.0"
-parquet = { version = "19.0.0", features = ["arrow"], optional = true }
+parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = ["arrow"], optional = true }
 pyo3 = { version = "0.16", optional = true }
 serde_json = "1.0"
 sqlparser = "0.20"

datafusion/common/src/scalar.rs

@@ -27,7 +27,7 @@ use arrow::{
         IntervalMonthDayNanoType, IntervalUnit, IntervalYearMonthType, TimeUnit,
         TimestampMicrosecondType, TimestampMillisecondType, TimestampNanosecondType,
         TimestampSecondType, UInt16Type, UInt32Type, UInt64Type, UInt8Type,
-        DECIMAL_MAX_PRECISION,
+        DECIMAL128_MAX_PRECISION,
     },
     util::decimal::{BasicDecimal, Decimal128},
 };
@@ -611,7 +611,7 @@ impl ScalarValue {
         scale: usize,
     ) -> Result<Self> {
         // make sure the precision and scale is valid
-        if precision <= DECIMAL_MAX_PRECISION && scale <= precision {
+        if precision <= DECIMAL128_MAX_PRECISION && scale <= precision {
             return Ok(ScalarValue::Decimal128(Some(value), precision, scale));
         }
         Err(DataFusionError::Internal(format!(
@@ -654,7 +654,7 @@ impl ScalarValue {
             ScalarValue::Int32(_) => DataType::Int32,
             ScalarValue::Int64(_) => DataType::Int64,
             ScalarValue::Decimal128(_, precision, scale) => {
-                DataType::Decimal(*precision, *scale)
+                DataType::Decimal128(*precision, *scale)
             }
             ScalarValue::TimestampSecond(_, tz_opt) => {
                 DataType::Timestamp(TimeUnit::Second, tz_opt.clone())
@@ -935,7 +935,7 @@ impl ScalarValue {
         }
 
         let array: ArrayRef = match &data_type {
-            DataType::Decimal(precision, scale) => {
+            DataType::Decimal128(precision, scale) => {
                 let decimal_array =
                     ScalarValue::iter_to_decimal_array(scalars, precision, scale)?;
                 Arc::new(decimal_array)
@@ -1448,7 +1448,7 @@ impl ScalarValue {
 
         Ok(match array.data_type() {
             DataType::Null => ScalarValue::Null,
-            DataType::Decimal(precision, scale) => {
+            DataType::Decimal128(precision, scale) => {
                 ScalarValue::get_decimal_value_from_array(array, index, precision, scale)
             }
             DataType::Boolean => typed_cast!(array, index, BooleanArray, Boolean),
@@ -1899,7 +1899,7 @@ impl TryFrom<&DataType> for ScalarValue {
             DataType::UInt16 => ScalarValue::UInt16(None),
             DataType::UInt32 => ScalarValue::UInt32(None),
             DataType::UInt64 => ScalarValue::UInt64(None),
-            DataType::Decimal(precision, scale) => {
+            DataType::Decimal128(precision, scale) => {
                 ScalarValue::Decimal128(None, *precision, *scale)
             }
             DataType::Utf8 => ScalarValue::Utf8(None),
@@ -2145,7 +2145,7 @@ mod tests {
     #[test]
     fn scalar_decimal_test() {
         let decimal_value = ScalarValue::Decimal128(Some(123), 10, 1);
-        assert_eq!(DataType::Decimal(10, 1), decimal_value.get_datatype());
+        assert_eq!(DataType::Decimal128(10, 1), decimal_value.get_datatype());
         let try_into_value: i128 = decimal_value.clone().try_into().unwrap();
         assert_eq!(123_i128, try_into_value);
         assert!(!decimal_value.is_null());
@@ -2163,14 +2163,14 @@ mod tests {
         let array = decimal_value.to_array();
         let array = array.as_any().downcast_ref::<Decimal128Array>().unwrap();
         assert_eq!(1, array.len());
-        assert_eq!(DataType::Decimal(10, 1), array.data_type().clone());
+        assert_eq!(DataType::Decimal128(10, 1), array.data_type().clone());
         assert_eq!(123i128, array.value(0).as_i128());
 
         // decimal scalar to array with size
         let array = decimal_value.to_array_of_size(10);
         let array_decimal = array.as_any().downcast_ref::<Decimal128Array>().unwrap();
         assert_eq!(10, array.len());
-        assert_eq!(DataType::Decimal(10, 1), array.data_type().clone());
+        assert_eq!(DataType::Decimal128(10, 1), array.data_type().clone());
         assert_eq!(123i128, array_decimal.value(0).as_i128());
         assert_eq!(123i128, array_decimal.value(9).as_i128());
         // test eq array
@@ -2208,7 +2208,7 @@ mod tests {
         // convert the vec to decimal array and check the result
         let array = ScalarValue::iter_to_array(decimal_vec.into_iter()).unwrap();
         assert_eq!(3, array.len());
-        assert_eq!(DataType::Decimal(10, 2), array.data_type().clone());
+        assert_eq!(DataType::Decimal128(10, 2), array.data_type().clone());
 
         let decimal_vec = vec![
             ScalarValue::Decimal128(Some(1), 10, 2),
@@ -2218,7 +2218,7 @@ mod tests {
         ];
         let array = ScalarValue::iter_to_array(decimal_vec.into_iter()).unwrap();
         assert_eq!(4, array.len());
-        assert_eq!(DataType::Decimal(10, 2), array.data_type().clone());
+        assert_eq!(DataType::Decimal128(10, 2), array.data_type().clone());
 
         assert!(ScalarValue::try_new_decimal128(1, 10, 2)
             .unwrap()

datafusion/core/Cargo.toml

@@ -56,17 +56,17 @@ unicode_expressions = ["datafusion-physical-expr/regex_expressions", "datafusion
 [dependencies]
 ahash = { version = "0.7", default-features = false }
 apache-avro = { version = "0.14", optional = true }
-arrow = { version = "19.0.0", features = ["prettyprint"] }
+arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = ["prettyprint"], optional = false }
 async-trait = "0.1.41"
 bytes = "1.1"
 chrono = { version = "0.4", default-features = false }
-datafusion-common = { path = "../common", version = "10.0.0", features = ["parquet", "object_store"] }
-datafusion-expr = { path = "../expr", version = "10.0.0" }
-datafusion-jit = { path = "../jit", version = "10.0.0", optional = true }
-datafusion-optimizer = { path = "../optimizer", version = "10.0.0" }
-datafusion-physical-expr = { path = "../physical-expr", version = "10.0.0" }
-datafusion-row = { path = "../row", version = "10.0.0" }
-datafusion-sql = { path = "../sql", version = "10.0.0" }
+datafusion-common = { path = "../common", features = ["parquet", "object_store"], optional = false }
+datafusion-expr = { path = "../expr", features = [], optional = false }
+datafusion-jit = { path = "../jit", features = [], optional = true }
+datafusion-optimizer = { path = "../optimizer", features = [], optional = false }
+datafusion-physical-expr = { path = "../physical-expr", features = [], optional = false }
+datafusion-row = { path = "../row", features = [], optional = false }
+datafusion-sql = { path = "../sql", features = [], optional = false }
 futures = "0.3"
 glob = "0.3.0"
 hashbrown = { version = "0.12", features = ["raw"] }
@@ -75,10 +75,10 @@ lazy_static = { version = "^1.4.0" }
 log = "^0.4"
 num-traits = { version = "0.2", optional = true }
 num_cpus = "1.13.0"
-object_store = "0.3.0"
+object_store = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = [], optional = false }
 ordered-float = "3.0"
 parking_lot = "0.12"
-parquet = { version = "19.0.0", features = ["arrow", "async"] }
+parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = ["arrow", "async"], optional = false }
 paste = "^1.0"
 pin-project-lite = "^0.2.7"
 pyo3 = { version = "0.16", optional = true }
@@ -98,7 +98,7 @@ csv = "1.1.6"
 ctor = "0.1.22"
 doc-comment = "0.3"
 env_logger = "0.9"
-fuzz-utils = { path = "fuzz-utils" }
+fuzz-utils = { path = "fuzz-utils", features = [], optional = false }
 
 [[bench]]
 harness = false

datafusion/core/fuzz-utils/Cargo.toml

@@ -23,6 +23,6 @@ edition = "2021"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-arrow = { version = "19.0.0", features = ["prettyprint"] }
+arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "30c94dbf1c422f81f8520b9956e96ab7b53c3f47", features = ["prettyprint"], optional = false }
 env_logger = "0.9.0"
 rand = "0.8"

datafusion/core/src/avro_to_arrow/arrow_array_reader.rs

@@ -101,12 +101,10 @@ impl<'a, R: Read> AvroArrowArrayReader<'a, R> {
                     "Failed to parse avro value: {:?}",
                     e
                 ))),
-                other => {
-                    return Err(ArrowError::ParseError(format!(
-                        "Row needs to be of type object, got: {:?}",
-                        other
-                    )))
-                }
+                other => Err(ArrowError::ParseError(format!(
+                    "Row needs to be of type object, got: {:?}",
+                    other
+                ))),
             })
             .collect::<ArrowResult<Vec<Vec<(String, Value)>>>>()?;
         if rows.is_empty() {

datafusion/core/src/avro_to_arrow/schema.rs

@@ -141,7 +141,7 @@ fn schema_to_field_with_props(
         AvroSchema::Fixed { size, .. } => DataType::FixedSizeBinary(*size as i32),
         AvroSchema::Decimal {
             precision, scale, ..
-        } => DataType::Decimal(*precision, *scale),
+        } => DataType::Decimal128(*precision, *scale),
         AvroSchema::Uuid => DataType::FixedSizeBinary(16),
         AvroSchema::Date => DataType::Date32,
         AvroSchema::TimeMillis => DataType::Time32(TimeUnit::Millisecond),
@@ -217,7 +217,7 @@ fn default_field_name(dt: &DataType) -> &str {
         DataType::Union(_, _, _) => "union",
         DataType::Dictionary(_, _) => "map",
         DataType::Map(_, _) => unimplemented!("Map support not implemented"),
-        DataType::Decimal(_, _) => "decimal",
+        DataType::Decimal128(_, _) => "decimal",
         DataType::Decimal256(_, _) => "decimal",
     }
 }

datafusion/core/src/catalog/information_schema.rs

@@ -508,7 +508,7 @@ impl InformationSchemaColumnsBuilder {
             Float32 => (Some(24), Some(2), None),
             // Numbers from postgres `double` type
             Float64 => (Some(24), Some(2), None),
-            Decimal(precision, scale) => {
+            Decimal128(precision, scale) => {
                 (Some(*precision as u64), Some(10), Some(*scale as u64))
             }
             _ => (None, None, None),

datafusion/core/src/datasource/file_format/json.rs

@@ -231,17 +231,21 @@ mod tests {
         projection: Option<Vec<usize>>,
         limit: Option<usize>,
     ) -> Result<Arc<dyn ExecutionPlan>> {
+        let store_root = env!("CARGO_MANIFEST_DIR");
         let filename = "tests/jsons/2.json";
         let format = JsonFormat::default();
-        scan_format(&format, ".", filename, projection, limit).await
+        scan_format(&format, store_root, filename, projection, limit).await
     }
 
     #[tokio::test]
     async fn infer_schema_with_limit() {
         let store = Arc::new(LocalFileSystem::new()) as _;
-        let filename = "tests/jsons/schema_infer_limit.json";
         let format = JsonFormat::default().with_schema_infer_max_rec(Some(3));
 
+        let store_root = std::path::Path::new(env!("CARGO_MANIFEST_DIR"));
+        let filename = store_root.join("tests/jsons/schema_infer_limit.json");
+        let filename = filename.to_str().expect("Unable to get path!");
+
         let file_schema = format
             .infer_schema(&store, &[local_unpartitioned_file(filename)])
             .await