refactor: use LazyLock in the user_doc macro (#13684)

jonahgao · comphead · web-flow · commit 6c9355d5be8b · 2024-12-08T23:37:53.000+08:00
* refactor: use `LazyLock` in the `user_doc` macro

* Fix cargo doc

* Update datafusion/macros/src/lib.rs

* Fix doc comment

---------

Co-authored-by: Oleks V &lt;comphead@users.noreply.github.com&gt;
diff --git a/datafusion/functions-aggregate/src/approx_distinct.rs b/datafusion/functions-aggregate/src/approx_distinct.rs
@@ -31,7 +31,6 @@ use datafusion_common::ScalarValue;
 use datafusion_common::{
     downcast_value, internal_err, not_impl_err, DataFusionError, Result,
 };
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::utils::format_state_name;
 use datafusion_expr::{
@@ -42,7 +41,6 @@ use std::any::Any;
 use std::fmt::{Debug, Formatter};
 use std::hash::Hash;
 use std::marker::PhantomData;
-use std::sync::OnceLock;
 
 make_udaf_expr_and_func!(
     ApproxDistinct,
diff --git a/datafusion/functions-aggregate/src/approx_median.rs b/datafusion/functions-aggregate/src/approx_median.rs
@@ -19,13 +19,11 @@
 
 use std::any::Any;
 use std::fmt::Debug;
-use std::sync::OnceLock;
 
 use arrow::{datatypes::DataType, datatypes::Field};
 use arrow_schema::DataType::{Float64, UInt64};
 
 use datafusion_common::{not_impl_err, plan_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::type_coercion::aggregates::NUMERICS;
 use datafusion_expr::utils::format_state_name;
diff --git a/datafusion/functions-aggregate/src/approx_percentile_cont.rs b/datafusion/functions-aggregate/src/approx_percentile_cont.rs
@@ -18,7 +18,7 @@
 use std::any::Any;
 use std::fmt::{Debug, Formatter};
 use std::mem::size_of_val;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::{Array, RecordBatch};
 use arrow::compute::{filter, is_not_null};
@@ -35,7 +35,6 @@ use datafusion_common::{
     downcast_value, internal_err, not_impl_datafusion_err, not_impl_err, plan_err,
     DataFusionError, Result, ScalarValue,
 };
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::type_coercion::aggregates::{INTEGERS, NUMERICS};
 use datafusion_expr::utils::format_state_name;
diff --git a/datafusion/functions-aggregate/src/approx_percentile_cont_with_weight.rs b/datafusion/functions-aggregate/src/approx_percentile_cont_with_weight.rs
@@ -18,7 +18,7 @@
 use std::any::Any;
 use std::fmt::{Debug, Formatter};
 use std::mem::size_of_val;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::{
     array::ArrayRef,
@@ -27,7 +27,6 @@ use arrow::{
 
 use datafusion_common::ScalarValue;
 use datafusion_common::{not_impl_err, plan_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::type_coercion::aggregates::NUMERICS;
 use datafusion_expr::Volatility::Immutable;
diff --git a/datafusion/functions-aggregate/src/array_agg.rs b/datafusion/functions-aggregate/src/array_agg.rs
@@ -25,7 +25,6 @@ use datafusion_common::cast::as_list_array;
 use datafusion_common::utils::{get_row_at_idx, SingleRowListArrayBuilder};
 use datafusion_common::{exec_err, ScalarValue};
 use datafusion_common::{internal_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::utils::format_state_name;
 use datafusion_expr::{Accumulator, Signature, Volatility};
@@ -36,7 +35,7 @@ use datafusion_macros::user_doc;
 use datafusion_physical_expr_common::sort_expr::{LexOrdering, PhysicalSortExpr};
 use std::collections::{HashSet, VecDeque};
 use std::mem::{size_of, size_of_val};
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 make_udaf_expr_and_func!(
     ArrayAgg,
diff --git a/datafusion/functions-aggregate/src/average.rs b/datafusion/functions-aggregate/src/average.rs
@@ -42,14 +42,13 @@ use datafusion_functions_aggregate_common::aggregate::groups_accumulator::nulls:
     filtered_null_mask, set_nulls,
 };
 
-use datafusion_doc::DocSection;
 use datafusion_functions_aggregate_common::utils::DecimalAverager;
 use datafusion_macros::user_doc;
 use log::debug;
 use std::any::Any;
 use std::fmt::Debug;
 use std::mem::{size_of, size_of_val};
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 make_udaf_expr_and_func!(
     Avg,
diff --git a/datafusion/functions-aggregate/src/bool_and_or.rs b/datafusion/functions-aggregate/src/bool_and_or.rs
@@ -19,7 +19,6 @@
 
 use std::any::Any;
 use std::mem::size_of_val;
-use std::sync::OnceLock;
 
 use arrow::array::ArrayRef;
 use arrow::array::BooleanArray;
@@ -38,7 +37,6 @@ use datafusion_expr::{
     Signature, Volatility,
 };
 
-use datafusion_doc::DocSection;
 use datafusion_functions_aggregate_common::aggregate::groups_accumulator::bool_op::BooleanGroupsAccumulator;
 use datafusion_macros::user_doc;
 
diff --git a/datafusion/functions-aggregate/src/correlation.rs b/datafusion/functions-aggregate/src/correlation.rs
@@ -20,7 +20,7 @@
 use std::any::Any;
 use std::fmt::Debug;
 use std::mem::size_of_val;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::compute::{and, filter, is_not_null};
 use arrow::{
@@ -31,7 +31,6 @@ use arrow::{
 use crate::covariance::CovarianceAccumulator;
 use crate::stddev::StddevAccumulator;
 use datafusion_common::{plan_err, Result, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::{
     function::{AccumulatorArgs, StateFieldsArgs},
     type_coercion::aggregates::NUMERICS,
diff --git a/datafusion/functions-aggregate/src/count.rs b/datafusion/functions-aggregate/src/count.rs
@@ -17,15 +17,14 @@
 
 use ahash::RandomState;
 use datafusion_common::stats::Precision;
-use datafusion_doc::DocSection;
 use datafusion_functions_aggregate_common::aggregate::count_distinct::BytesViewDistinctCountAccumulator;
 use datafusion_macros::user_doc;
 use datafusion_physical_expr::expressions;
 use std::collections::HashSet;
 use std::fmt::Debug;
 use std::mem::{size_of, size_of_val};
 use std::ops::BitAnd;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::{
     array::{ArrayRef, AsArray},
diff --git a/datafusion/functions-aggregate/src/covariance.rs b/datafusion/functions-aggregate/src/covariance.rs
@@ -19,7 +19,6 @@
 
 use std::fmt::Debug;
 use std::mem::size_of_val;
-use std::sync::OnceLock;
 
 use arrow::{
     array::{ArrayRef, Float64Array, UInt64Array},
@@ -31,7 +30,6 @@ use datafusion_common::{
     downcast_value, plan_err, unwrap_or_internal_err, DataFusionError, Result,
     ScalarValue,
 };
-use datafusion_doc::DocSection;
 use datafusion_expr::{
     function::{AccumulatorArgs, StateFieldsArgs},
     type_coercion::aggregates::NUMERICS,
diff --git a/datafusion/functions-aggregate/src/first_last.rs b/datafusion/functions-aggregate/src/first_last.rs
@@ -20,7 +20,7 @@
 use std::any::Any;
 use std::fmt::Debug;
 use std::mem::size_of_val;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::{ArrayRef, AsArray, BooleanArray};
 use arrow::compute::{self, lexsort_to_indices, take_arrays, SortColumn};
@@ -29,7 +29,6 @@ use datafusion_common::utils::{compare_rows, get_row_at_idx};
 use datafusion_common::{
     arrow_datafusion_err, internal_err, DataFusionError, Result, ScalarValue,
 };
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::utils::{format_state_name, AggregateOrderSensitivity};
 use datafusion_expr::{
diff --git a/datafusion/functions-aggregate/src/grouping.rs b/datafusion/functions-aggregate/src/grouping.rs
@@ -19,12 +19,10 @@
 
 use std::any::Any;
 use std::fmt;
-use std::sync::OnceLock;
 
 use arrow::datatypes::DataType;
 use arrow::datatypes::Field;
 use datafusion_common::{not_impl_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::AccumulatorArgs;
 use datafusion_expr::function::StateFieldsArgs;
 use datafusion_expr::utils::format_state_name;
diff --git a/datafusion/functions-aggregate/src/median.rs b/datafusion/functions-aggregate/src/median.rs
@@ -18,7 +18,7 @@
 use std::cmp::Ordering;
 use std::fmt::{Debug, Formatter};
 use std::mem::{size_of, size_of_val};
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::{downcast_integer, ArrowNumericType};
 use arrow::{
@@ -34,7 +34,6 @@ use arrow::array::ArrowNativeTypeOp;
 use arrow::datatypes::{ArrowNativeType, ArrowPrimitiveType};
 
 use datafusion_common::{DataFusionError, HashSet, Result, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::StateFieldsArgs;
 use datafusion_expr::{
     function::AccumulatorArgs, utils::format_state_name, Accumulator, AggregateUDFImpl,
diff --git a/datafusion/functions-aggregate/src/min_max.rs b/datafusion/functions-aggregate/src/min_max.rs
@@ -55,7 +55,6 @@ use arrow::datatypes::{
 
 use crate::min_max::min_max_bytes::MinMaxBytesAccumulator;
 use datafusion_common::ScalarValue;
-use datafusion_doc::DocSection;
 use datafusion_expr::{
     function::AccumulatorArgs, Accumulator, AggregateUDFImpl, Documentation, Signature,
     Volatility,
@@ -65,7 +64,6 @@ use datafusion_macros::user_doc;
 use half::f16;
 use std::mem::size_of_val;
 use std::ops::Deref;
-use std::sync::OnceLock;
 
 fn get_min_max_result_type(input_types: &[DataType]) -> Result<Vec<DataType>> {
     // make sure that the input types only has one element.
diff --git a/datafusion/functions-aggregate/src/nth_value.rs b/datafusion/functions-aggregate/src/nth_value.rs
@@ -21,14 +21,13 @@
 use std::any::Any;
 use std::collections::VecDeque;
 use std::mem::{size_of, size_of_val};
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::{new_empty_array, ArrayRef, AsArray, StructArray};
 use arrow_schema::{DataType, Field, Fields};
 
 use datafusion_common::utils::{get_row_at_idx, SingleRowListArrayBuilder};
 use datafusion_common::{exec_err, internal_err, not_impl_err, Result, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::utils::format_state_name;
 use datafusion_expr::{
diff --git a/datafusion/functions-aggregate/src/stddev.rs b/datafusion/functions-aggregate/src/stddev.rs
@@ -20,14 +20,13 @@
 use std::any::Any;
 use std::fmt::{Debug, Formatter};
 use std::mem::align_of_val;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::Float64Array;
 use arrow::{array::ArrayRef, datatypes::DataType, datatypes::Field};
 
 use datafusion_common::{internal_err, not_impl_err, Result};
 use datafusion_common::{plan_err, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::utils::format_state_name;
 use datafusion_expr::{
diff --git a/datafusion/functions-aggregate/src/string_agg.rs b/datafusion/functions-aggregate/src/string_agg.rs
@@ -22,7 +22,6 @@ use arrow_schema::DataType;
 use datafusion_common::cast::as_generic_string_array;
 use datafusion_common::Result;
 use datafusion_common::{not_impl_err, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::AccumulatorArgs;
 use datafusion_expr::{
     Accumulator, AggregateUDFImpl, Documentation, Signature, TypeSignature, Volatility,
@@ -31,7 +30,6 @@ use datafusion_macros::user_doc;
 use datafusion_physical_expr::expressions::Literal;
 use std::any::Any;
 use std::mem::size_of_val;
-use std::sync::OnceLock;
 
 make_udaf_expr_and_func!(
     StringAgg,
diff --git a/datafusion/functions-aggregate/src/sum.rs b/datafusion/functions-aggregate/src/sum.rs
@@ -22,7 +22,6 @@ use datafusion_expr::utils::AggregateOrderSensitivity;
 use std::any::Any;
 use std::collections::HashSet;
 use std::mem::{size_of, size_of_val};
-use std::sync::OnceLock;
 
 use arrow::array::Array;
 use arrow::array::ArrowNativeTypeOp;
@@ -35,7 +34,6 @@ use arrow::datatypes::{
 };
 use arrow::{array::ArrayRef, datatypes::Field};
 use datafusion_common::{exec_err, not_impl_err, Result, ScalarValue};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::AccumulatorArgs;
 use datafusion_expr::function::StateFieldsArgs;
 use datafusion_expr::utils::format_state_name;
diff --git a/datafusion/functions-aggregate/src/variance.rs b/datafusion/functions-aggregate/src/variance.rs
@@ -25,13 +25,11 @@ use arrow::{
     datatypes::{DataType, Field},
 };
 use std::mem::{size_of, size_of_val};
-use std::sync::OnceLock;
 use std::{fmt::Debug, sync::Arc};
 
 use datafusion_common::{
     downcast_value, not_impl_err, plan_err, DataFusionError, Result, ScalarValue,
 };
-use datafusion_doc::DocSection;
 use datafusion_expr::{
     function::{AccumulatorArgs, StateFieldsArgs},
     utils::format_state_name,
diff --git a/datafusion/functions/src/datetime/to_date.rs b/datafusion/functions/src/datetime/to_date.rs
@@ -22,13 +22,11 @@ use arrow::error::ArrowError::ParseError;
 use arrow::{array::types::Date32Type, compute::kernels::cast_utils::Parser};
 use datafusion_common::error::DataFusionError;
 use datafusion_common::{arrow_err, exec_err, internal_datafusion_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::{
     ColumnarValue, Documentation, ScalarUDFImpl, Signature, Volatility,
 };
 use datafusion_macros::user_doc;
 use std::any::Any;
-use std::sync::OnceLock;
 
 #[user_doc(
     doc_section(label = "Time and Date Functions"),
diff --git a/datafusion/functions/src/math/abs.rs b/datafusion/functions/src/math/abs.rs
@@ -18,7 +18,7 @@
 //! math expressions
 
 use std::any::Any;
-use std::sync::{Arc, OnceLock};
+use std::sync::Arc;
 
 use arrow::array::{
     ArrayRef, Decimal128Array, Decimal256Array, Float32Array, Float64Array, Int16Array,
@@ -27,7 +27,6 @@ use arrow::array::{
 use arrow::datatypes::DataType;
 use arrow::error::ArrowError;
 use datafusion_common::{exec_err, not_impl_err, DataFusionError, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::interval_arithmetic::Interval;
 use datafusion_expr::sort_properties::{ExprProperties, SortProperties};
 use datafusion_expr::{
diff --git a/datafusion/functions/src/string/ltrim.rs b/datafusion/functions/src/string/ltrim.rs
@@ -22,12 +22,10 @@ use std::any::Any;
 use crate::string::common::*;
 use crate::utils::{make_scalar_function, utf8_to_str_type};
 use datafusion_common::{exec_err, Result};
-use datafusion_doc::DocSection;
 use datafusion_expr::function::Hint;
 use datafusion_expr::{ColumnarValue, Documentation, TypeSignature, Volatility};
 use datafusion_expr::{ScalarUDFImpl, Signature};
 use datafusion_macros::user_doc;
-use std::sync::OnceLock;
 
 /// Returns the longest string  with leading characters removed. If the characters are not specified, whitespace is removed.
 /// ltrim('zzzytest', 'xyz') = 'test'
diff --git a/datafusion/macros/src/user_doc.rs b/datafusion/macros/src/user_doc.rs
