datafusion_spark/function/datetime/
date_part.rs

1// Licensed to the Apache Software Foundation (ASF) under one
2// or more contributor license agreements.  See the NOTICE file
3// distributed with this work for additional information
4// regarding copyright ownership.  The ASF licenses this file
5// to you under the Apache License, Version 2.0 (the
6// "License"); you may not use this file except in compliance
7// with the License.  You may obtain a copy of the License at
8//
9//   http://www.apache.org/licenses/LICENSE-2.0
10//
11// Unless required by applicable law or agreed to in writing,
12// software distributed under the License is distributed on an
13// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
14// KIND, either express or implied.  See the License for the
15// specific language governing permissions and limitations
16// under the License.
17
18use arrow::datatypes::{DataType, Field, FieldRef};
19use datafusion_common::types::logical_date;
20use datafusion_common::{
21    Result, ScalarValue, internal_err, types::logical_string, utils::take_function_args,
22};
23use datafusion_expr::expr::ScalarFunction;
24use datafusion_expr::simplify::{ExprSimplifyResult, SimplifyContext};
25use datafusion_expr::{
26    Coercion, ColumnarValue, Expr, ReturnFieldArgs, ScalarFunctionArgs, ScalarUDFImpl,
27    Signature, TypeSignature, TypeSignatureClass, Volatility,
28};
29use std::sync::Arc;
30
31/// Wrapper around datafusion date_part function to handle
32/// Spark behavior returning day of the week 1-indexed instead of 0-indexed and different part aliases.
33/// <https://spark.apache.org/docs/latest/api/sql/index.html#date_part>
34#[derive(Debug, PartialEq, Eq, Hash)]
35pub struct SparkDatePart {
36    signature: Signature,
37    aliases: Vec<String>,
38}
39
40impl Default for SparkDatePart {
41    fn default() -> Self {
42        Self::new()
43    }
44}
45
46impl SparkDatePart {
47    pub fn new() -> Self {
48        Self {
49            signature: Signature::one_of(
50                vec![
51                    TypeSignature::Coercible(vec![
52                        Coercion::new_exact(TypeSignatureClass::Native(logical_string())),
53                        Coercion::new_exact(TypeSignatureClass::Timestamp),
54                    ]),
55                    TypeSignature::Coercible(vec![
56                        Coercion::new_exact(TypeSignatureClass::Native(logical_string())),
57                        Coercion::new_exact(TypeSignatureClass::Native(logical_date())),
58                    ]),
59                ],
60                Volatility::Immutable,
61            ),
62            aliases: vec![String::from("datepart")],
63        }
64    }
65}
66
67impl ScalarUDFImpl for SparkDatePart {
68    fn name(&self) -> &str {
69        "date_part"
70    }
71
72    fn aliases(&self) -> &[String] {
73        &self.aliases
74    }
75
76    fn signature(&self) -> &Signature {
77        &self.signature
78    }
79
80    fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
81        internal_err!("Use return_field_from_args in this case instead.")
82    }
83
84    fn return_field_from_args(&self, args: ReturnFieldArgs) -> Result<FieldRef> {
85        let nullable = args.arg_fields.iter().any(|f| f.is_nullable());
86
87        Ok(Arc::new(Field::new(self.name(), DataType::Int32, nullable)))
88    }
89
90    fn invoke_with_args(&self, _args: ScalarFunctionArgs) -> Result<ColumnarValue> {
91        internal_err!("spark date_part should have been simplified to standard date_part")
92    }
93
94    fn simplify(
95        &self,
96        args: Vec<Expr>,
97        _info: &SimplifyContext,
98    ) -> Result<ExprSimplifyResult> {
99        let [part_expr, date_expr] = take_function_args(self.name(), args)?;
100
101        let part = match part_expr.as_literal() {
102            Some(ScalarValue::Utf8(Some(v)))
103            | Some(ScalarValue::Utf8View(Some(v)))
104            | Some(ScalarValue::LargeUtf8(Some(v))) => v.to_lowercase(),
105            _ => {
106                return internal_err!(
107                    "First argument of `DATE_PART` must be non-null scalar Utf8"
108                );
109            }
110        };
111
112        // Map Spark-specific date part aliases to datafusion ones
113        let part = match part.as_str() {
114            "yearofweek" | "year_iso" => "isoyear",
115            "dayofweek" => "dow",
116            "dayofweek_iso" | "dow_iso" => "isodow",
117            other => other,
118        };
119
120        let part_expr = Expr::Literal(ScalarValue::new_utf8(part), None);
121
122        let date_part_expr = Expr::ScalarFunction(ScalarFunction::new_udf(
123            datafusion_functions::datetime::date_part(),
124            vec![part_expr, date_expr],
125        ));
126
127        match part {
128            // Spark's `dayofweek` is 1..=7 (Sun=1) but df's `dow` is 0..=6
129            // (Sun=0); shift by +1. df's `isodow` already returns the
130            // PG-correct 1..=7 (Mon=1), which matches Spark's
131            // `dayofweek_iso`/`dow_iso`, so no shift is needed there.
132            "dow" => Ok(ExprSimplifyResult::Simplified(
133                date_part_expr + Expr::Literal(ScalarValue::Int32(Some(1)), None),
134            )),
135            _ => Ok(ExprSimplifyResult::Simplified(date_part_expr)),
136        }
137    }
138}
datafusion_spark/function/datetime/date_part.rs

datafusion_spark/function/datetime/
date_part.rs
