refactor: remove uses of arrow_schema and use reexport in arrow instead

apache · Feb 11, 2025 · be33471 · be33471
1 parent 3550758
commit be33471
Show file tree

Hide file tree

Showing 198 changed files with 327 additions and 357 deletions.
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/datafusion/catalog-listing/Cargo.toml b/datafusion/catalog-listing/Cargo.toml
@@ -29,7 +29,6 @@ version.workspace = true
 
 [dependencies]
 arrow = { workspace = true }
-arrow-schema = { workspace = true }
 async-compression = { version = "0.4.0", features = [
     "bzip2",
     "gzip",

diff --git a/datafusion/catalog-listing/src/helpers.rs b/datafusion/catalog-listing/src/helpers.rs
@@ -30,10 +30,9 @@ use datafusion_expr::{BinaryExpr, Operator};
 use arrow::{
     array::{Array, ArrayRef, AsArray, StringBuilder},
     compute::{and, cast, prep_null_mask_filter},
-    datatypes::{DataType, Field, Schema},
+    datatypes::{DataType, Field, Fields, Schema},
     record_batch::RecordBatch,
 };
-use arrow_schema::Fields;
 use datafusion_expr::execution_props::ExecutionProps;
 use futures::stream::FuturesUnordered;
 use futures::{stream::BoxStream, StreamExt, TryStreamExt};

diff --git a/datafusion/common/Cargo.toml b/datafusion/common/Cargo.toml
@@ -52,7 +52,6 @@ apache-avro = { version = "0.17", default-features = false, features = [
 ], optional = true }
 arrow = { workspace = true }
 arrow-ipc = { workspace = true }
-arrow-schema = { workspace = true }
 base64 = "0.22.1"
 half = { workspace = true }
 hashbrown = { workspace = true }

diff --git a/datafusion/common/src/column.rs b/datafusion/common/src/column.rs
@@ -20,7 +20,7 @@
 use crate::error::_schema_err;
 use crate::utils::{parse_identifiers_normalized, quote_identifier};
 use crate::{DFSchema, Diagnostic, Result, SchemaError, Spans, TableReference};
-use arrow_schema::{Field, FieldRef};
+use arrow::datatypes::{Field, FieldRef};
 use std::collections::HashSet;
 use std::convert::Infallible;
 use std::fmt;
@@ -380,8 +380,7 @@ impl fmt::Display for Column {
 #[cfg(test)]
 mod tests {
     use super::*;
-    use arrow::datatypes::DataType;
-    use arrow_schema::SchemaBuilder;
+    use arrow::datatypes::{DataType, SchemaBuilder};
     use std::sync::Arc;
 
     fn create_qualified_schema(qualifier: &str, names: Vec<&str>) -> Result<DFSchema> {

diff --git a/datafusion/common/src/dfschema.rs b/datafusion/common/src/dfschema.rs
@@ -30,8 +30,9 @@ use crate::{
 };
 
 use arrow::compute::can_cast_types;
-use arrow::datatypes::{DataType, Field, FieldRef, Fields, Schema, SchemaRef};
-use arrow_schema::SchemaBuilder;
+use arrow::datatypes::{
+    DataType, Field, FieldRef, Fields, Schema, SchemaBuilder, SchemaRef,
+};
 
 /// A reference-counted reference to a [DFSchema].
 pub type DFSchemaRef = Arc<DFSchema>;
@@ -56,7 +57,7 @@ pub type DFSchemaRef = Arc<DFSchema>;
 ///
 /// ```rust
 /// use datafusion_common::{DFSchema, Column};
-/// use arrow_schema::{DataType, Field, Schema};
+/// use arrow::datatypes::{DataType, Field, Schema};
 ///
 /// let arrow_schema = Schema::new(vec![
 ///    Field::new("c1", DataType::Int32, false),
@@ -77,7 +78,7 @@ pub type DFSchemaRef = Arc<DFSchema>;
 ///
 /// ```rust
 /// use datafusion_common::{DFSchema, Column};
-/// use arrow_schema::{DataType, Field, Schema};
+/// use arrow::datatypes::{DataType, Field, Schema};
 ///
 /// let arrow_schema = Schema::new(vec![
 ///    Field::new("c1", DataType::Int32, false),
@@ -94,8 +95,7 @@ pub type DFSchemaRef = Arc<DFSchema>;
 ///
 /// ```rust
 /// use datafusion_common::DFSchema;
-/// use arrow_schema::Schema;
-/// use arrow::datatypes::Field;
+/// use arrow::datatypes::{Schema, Field};
 /// use std::collections::HashMap;
 ///
 /// let df_schema = DFSchema::from_unqualified_fields(vec![

diff --git a/datafusion/common/src/file_options/parquet_writer.rs b/datafusion/common/src/file_options/parquet_writer.rs
@@ -25,7 +25,7 @@ use crate::{
     DataFusionError, Result, _internal_datafusion_err,
 };
 
-use arrow_schema::Schema;
+use arrow::datatypes::Schema;
 // TODO: handle once deprecated
 #[allow(deprecated)]
 use parquet::{

diff --git a/datafusion/common/src/param_value.rs b/datafusion/common/src/param_value.rs
@@ -17,7 +17,7 @@
 
 use crate::error::{_plan_datafusion_err, _plan_err};
 use crate::{Result, ScalarValue};
-use arrow_schema::DataType;
+use arrow::datatypes::DataType;
 use std::collections::HashMap;
 
 /// The parameter value corresponding to the placeholder

diff --git a/datafusion/common/src/scalar/mod.rs b/datafusion/common/src/scalar/mod.rs
@@ -55,10 +55,9 @@ use arrow::datatypes::{
     Int8Type, IntervalDayTimeType, IntervalMonthDayNanoType, IntervalUnit,
     IntervalYearMonthType, TimeUnit, TimestampMicrosecondType, TimestampMillisecondType,
     TimestampNanosecondType, TimestampSecondType, UInt16Type, UInt32Type, UInt64Type,
-    UInt8Type, DECIMAL128_MAX_PRECISION,
+    UInt8Type, UnionFields, UnionMode, DECIMAL128_MAX_PRECISION,
 };
 use arrow::util::display::{array_value_to_string, ArrayFormatter, FormatOptions};
-use arrow_schema::{UnionFields, UnionMode};
 
 use crate::format::DEFAULT_CAST_OPTIONS;
 use half::f16;
@@ -3964,9 +3963,9 @@ mod tests {
     use arrow::array::{types::Float64Type, NullBufferBuilder};
     use arrow::buffer::{Buffer, OffsetBuffer};
     use arrow::compute::{is_null, kernels};
+    use arrow::datatypes::Fields;
     use arrow::error::ArrowError;
     use arrow::util::pretty::pretty_format_columns;
-    use arrow_schema::Fields;
     use chrono::NaiveDate;
     use rand::Rng;
 

diff --git a/datafusion/common/src/scalar/struct_builder.rs b/datafusion/common/src/scalar/struct_builder.rs
@@ -20,8 +20,7 @@
 use crate::error::_internal_err;
 use crate::{Result, ScalarValue};
 use arrow::array::{ArrayRef, StructArray};
-use arrow::datatypes::{DataType, FieldRef, Fields};
-use arrow_schema::Field;
+use arrow::datatypes::{DataType, Field, FieldRef, Fields};
 use std::sync::Arc;
 
 /// Builder for [`ScalarValue::Struct`].

diff --git a/datafusion/common/src/stats.rs b/datafusion/common/src/stats.rs
@@ -21,7 +21,7 @@ use std::fmt::{self, Debug, Display};
 
 use crate::{Result, ScalarValue};
 
-use arrow_schema::{DataType, Schema, SchemaRef};
+use arrow::datatypes::{DataType, Schema, SchemaRef};
 
 /// Represents a value with a degree of certainty. `Precision` is used to
 /// propagate information the precision of statistical values.

diff --git a/datafusion/common/src/test_util.rs b/datafusion/common/src/test_util.rs
@@ -338,9 +338,9 @@ macro_rules! create_array {
 macro_rules! record_batch {
     ($(($name: expr, $type: ident, $values: expr)),*) => {
         {
-            let schema = std::sync::Arc::new(arrow_schema::Schema::new(vec![
+            let schema = std::sync::Arc::new(arrow::datatypes::Schema::new(vec![
                 $(
-                    arrow_schema::Field::new($name, arrow_schema::DataType::$type, true),
+                    arrow::datatypes::Field::new($name, arrow::datatypes::DataType::$type, true),
                 )*
             ]));
 

diff --git a/datafusion/common/src/types/field.rs b/datafusion/common/src/types/field.rs
@@ -15,7 +15,7 @@
 // specific language governing permissions and limitations
 // under the License.
 
-use arrow_schema::{Field, Fields, UnionFields};
+use arrow::datatypes::{Field, Fields, UnionFields};
 use std::hash::{Hash, Hasher};
 use std::{ops::Deref, sync::Arc};
 

diff --git a/datafusion/common/src/types/logical.rs b/datafusion/common/src/types/logical.rs
@@ -17,7 +17,7 @@
 
 use super::NativeType;
 use crate::error::Result;
-use arrow_schema::DataType;
+use arrow::datatypes::DataType;
 use core::fmt;
 use std::{cmp::Ordering, hash::Hash, sync::Arc};
 

diff --git a/datafusion/common/src/types/native.rs b/datafusion/common/src/types/native.rs
@@ -21,7 +21,7 @@ use super::{
 };
 use crate::error::{Result, _internal_err};
 use arrow::compute::can_cast_types;
-use arrow_schema::{
+use arrow::datatypes::{
     DataType, Field, FieldRef, Fields, IntervalUnit, TimeUnit, UnionFields,
 };
 use std::{fmt::Display, sync::Arc};
@@ -126,7 +126,7 @@ pub enum NativeType {
     /// nevertheless correct).
     ///
     /// ```
-    /// # use arrow_schema::{DataType, TimeUnit};
+    /// # use arrow::datatypes::{DataType, TimeUnit};
     /// DataType::Timestamp(TimeUnit::Second, None);
     /// DataType::Timestamp(TimeUnit::Second, Some("literal".into()));
     /// DataType::Timestamp(TimeUnit::Second, Some("string".to_string().into()));

diff --git a/datafusion/common/src/utils/mod.rs b/datafusion/common/src/utils/mod.rs
@@ -30,8 +30,7 @@ use arrow::array::{
 };
 use arrow::buffer::OffsetBuffer;
 use arrow::compute::{partition, SortColumn, SortOptions};
-use arrow::datatypes::{Field, SchemaRef};
-use arrow_schema::DataType;
+use arrow::datatypes::{DataType, Field, SchemaRef};
 use sqlparser::ast::Ident;
 use sqlparser::dialect::GenericDialect;
 use sqlparser::parser::Parser;

diff --git a/datafusion/core/Cargo.toml b/datafusion/core/Cargo.toml
@@ -74,7 +74,7 @@ recursive_protection = [
     "datafusion-physical-optimizer/recursive_protection",
     "datafusion-sql/recursive_protection",
 ]
-serde = ["arrow-schema/serde"]
+serde = ["dep:serde"]
 string_expressions = ["datafusion-functions/string_expressions"]
 unicode_expressions = [
     "datafusion-sql/unicode_expressions",
@@ -126,6 +126,7 @@ parking_lot = { workspace = true }
 parquet = { workspace = true, optional = true, default-features = true }
 rand = { workspace = true }
 regex = { workspace = true }
+serde = { version = "1.0", default-features = false, features = ["derive"], optional = true }
 sqlparser = { workspace = true }
 tempfile = { workspace = true }
 tokio = { workspace = true }

diff --git a/datafusion/core/benches/sql_query_with_io.rs b/datafusion/core/benches/sql_query_with_io.rs
@@ -18,7 +18,7 @@
 use std::{fmt::Write, sync::Arc, time::Duration};
 
 use arrow::array::{Int64Builder, RecordBatch, UInt64Builder};
-use arrow_schema::{DataType, Field, Schema, SchemaRef};
+use arrow::datatypes::{DataType, Field, Schema, SchemaRef};
 use bytes::Bytes;
 use criterion::{criterion_group, criterion_main, Criterion, SamplingMode};
 use datafusion::{

diff --git a/datafusion/core/src/dataframe/mod.rs b/datafusion/core/src/dataframe/mod.rs
@@ -45,8 +45,7 @@ use std::sync::Arc;
 
 use arrow::array::{Array, ArrayRef, Int64Array, StringArray};
 use arrow::compute::{cast, concat};
-use arrow::datatypes::{DataType, Field};
-use arrow_schema::{Schema, SchemaRef};
+use arrow::datatypes::{DataType, Field, Schema, SchemaRef};
 use datafusion_common::config::{CsvOptions, JsonOptions};
 use datafusion_common::{
     exec_err, not_impl_err, plan_err, Column, DFSchema, DataFusionError, ParamValues,

diff --git a/datafusion/core/src/datasource/data_source.rs b/datafusion/core/src/datasource/data_source.rs
@@ -24,7 +24,7 @@ use std::sync::Arc;
 
 use crate::datasource::physical_plan::{FileOpener, FileScanConfig};
 
-use arrow_schema::SchemaRef;
+use arrow::datatypes::SchemaRef;
 use datafusion_common::Statistics;
 use datafusion_physical_plan::metrics::ExecutionPlanMetricsSet;
 use datafusion_physical_plan::DisplayFormatType;