1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
//! Deprecated module. Add new feature in scalar_function.rs
//!
//! This module contains built-in functions' enumeration and metadata.
//!
//! Generally, a function has:
//! * a signature
//! * a return type, that is a function of the incoming argument's types
//! * the computation, that must accept each valid signature
//!
//! * Signature: see `Signature`
//! * Return type: a function `(arg_types) -> return_type`. E.g. for sqrt, ([f32]) -> f32, ([f64]) -> f64.
//!
//! This module also supports coercion to improve user experience: if
//! an argument i32 is passed to a function that supports f64, the
//! argument is automatically is coerced to f64.
use std::sync::Arc;
use arrow::array::ArrayRef;
use arrow_array::Array;
pub use crate::scalar_function::create_physical_expr;
use datafusion_common::{Result, ScalarValue};
pub use datafusion_expr::FuncMonotonicity;
use datafusion_expr::{ColumnarValue, ScalarFunctionImplementation};
#[derive(Debug, Clone, Copy)]
pub enum Hint {
/// Indicates the argument needs to be padded if it is scalar
Pad,
/// Indicates the argument can be converted to an array of length 1
AcceptsSingular,
}
#[deprecated(since = "36.0.0", note = "Use ColumarValue::values_to_arrays instead")]
pub fn columnar_values_to_array(args: &[ColumnarValue]) -> Result<Vec<ArrayRef>> {
ColumnarValue::values_to_arrays(args)
}
/// Decorates a function to handle [`ScalarValue`]s by converting them to arrays before calling the function
/// and vice-versa after evaluation.
/// Note that this function makes a scalar function with no arguments or all scalar inputs return a scalar.
/// That's said its output will be same for all input rows in a batch.
#[deprecated(
since = "36.0.0",
note = "Implement your function directly in terms of ColumnarValue or use `ScalarUDF` instead"
)]
pub fn make_scalar_function<F>(inner: F) -> ScalarFunctionImplementation
where
F: Fn(&[ArrayRef]) -> Result<ArrayRef> + Sync + Send + 'static,
{
make_scalar_function_inner(inner)
}
/// Internal implementation, see comments on `make_scalar_function` for caveats
pub(crate) fn make_scalar_function_inner<F>(inner: F) -> ScalarFunctionImplementation
where
F: Fn(&[ArrayRef]) -> Result<ArrayRef> + Sync + Send + 'static,
{
make_scalar_function_with_hints(inner, vec![])
}
/// Just like [`make_scalar_function`], decorates the given function to handle both [`ScalarValue`]s and arrays.
/// Additionally can receive a `hints` vector which can be used to control the output arrays when generating them
/// from [`ScalarValue`]s.
///
/// Each element of the `hints` vector gets mapped to the corresponding argument of the function. The number of hints
/// can be less or greater than the number of arguments (for functions with variable number of arguments). Each unmapped
/// argument will assume the default hint (for padding, it is [`Hint::Pad`]).
pub(crate) fn make_scalar_function_with_hints<F>(
inner: F,
hints: Vec<Hint>,
) -> ScalarFunctionImplementation
where
F: Fn(&[ArrayRef]) -> Result<ArrayRef> + Sync + Send + 'static,
{
Arc::new(move |args: &[ColumnarValue]| {
// first, identify if any of the arguments is an Array. If yes, store its `len`,
// as any scalar will need to be converted to an array of len `len`.
let len = args
.iter()
.fold(Option::<usize>::None, |acc, arg| match arg {
ColumnarValue::Scalar(_) => acc,
ColumnarValue::Array(a) => Some(a.len()),
});
let is_scalar = len.is_none();
let inferred_length = len.unwrap_or(1);
let args = args
.iter()
.zip(hints.iter().chain(std::iter::repeat(&Hint::Pad)))
.map(|(arg, hint)| {
// Decide on the length to expand this scalar to depending
// on the given hints.
let expansion_len = match hint {
Hint::AcceptsSingular => 1,
Hint::Pad => inferred_length,
};
arg.clone().into_array(expansion_len)
})
.collect::<Result<Vec<_>>>()?;
let result = (inner)(&args);
if is_scalar {
// If all inputs are scalar, keeps output as scalar
let result = result.and_then(|arr| ScalarValue::try_from_array(&arr, 0));
result.map(ColumnarValue::Scalar)
} else {
result.map(ColumnarValue::Array)
}
})
}