datafusion_functions/math/
trunc.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 std::any::Any;
19use std::sync::Arc;
20
21use crate::utils::make_scalar_function;
22
23use arrow::array::{ArrayRef, AsArray, PrimitiveArray};
24use arrow::datatypes::DataType::{Float32, Float64};
25use arrow::datatypes::{DataType, Float32Type, Float64Type, Int64Type};
26use datafusion_common::ScalarValue::Int64;
27use datafusion_common::{exec_err, Result};
28use datafusion_expr::sort_properties::{ExprProperties, SortProperties};
29use datafusion_expr::TypeSignature::Exact;
30use datafusion_expr::{
31    ColumnarValue, Documentation, ScalarFunctionArgs, ScalarUDFImpl, Signature,
32    Volatility,
33};
34use datafusion_macros::user_doc;
35
36#[user_doc(
37    doc_section(label = "Math Functions"),
38    description = "Truncates a number to a whole number or truncated to the specified decimal places.",
39    syntax_example = "trunc(numeric_expression[, decimal_places])",
40    standard_argument(name = "numeric_expression", prefix = "Numeric"),
41    argument(
42        name = "decimal_places",
43        description = r#"Optional. The number of decimal places to
44  truncate to. Defaults to 0 (truncate to a whole number). If
45  `decimal_places` is a positive integer, truncates digits to the
46  right of the decimal point. If `decimal_places` is a negative
47  integer, replaces digits to the left of the decimal point with `0`."#
48    ),
49    sql_example = r#"
50  ```sql
51  > SELECT trunc(42.738);
52  +----------------+
53  | trunc(42.738)  |
54  +----------------+
55  | 42             |
56  +----------------+
57  ```"#
58)]
59#[derive(Debug, PartialEq, Eq, Hash)]
60pub struct TruncFunc {
61    signature: Signature,
62}
63
64impl Default for TruncFunc {
65    fn default() -> Self {
66        TruncFunc::new()
67    }
68}
69
70impl TruncFunc {
71    pub fn new() -> Self {
72        use DataType::*;
73        Self {
74            // math expressions expect 1 argument of type f64 or f32
75            // priority is given to f64 because e.g. `sqrt(1i32)` is in IR (real numbers) and thus we
76            // return the best approximation for it (in f64).
77            // We accept f32 because in this case it is clear that the best approximation
78            // will be as good as the number of digits in the number
79            signature: Signature::one_of(
80                vec![
81                    Exact(vec![Float32, Int64]),
82                    Exact(vec![Float64, Int64]),
83                    Exact(vec![Float64]),
84                    Exact(vec![Float32]),
85                ],
86                Volatility::Immutable,
87            ),
88        }
89    }
90}
91
92impl ScalarUDFImpl for TruncFunc {
93    fn as_any(&self) -> &dyn Any {
94        self
95    }
96
97    fn name(&self) -> &str {
98        "trunc"
99    }
100
101    fn signature(&self) -> &Signature {
102        &self.signature
103    }
104
105    fn return_type(&self, arg_types: &[DataType]) -> Result<DataType> {
106        match arg_types[0] {
107            Float32 => Ok(Float32),
108            _ => Ok(Float64),
109        }
110    }
111
112    fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
113        make_scalar_function(trunc, vec![])(&args.args)
114    }
115
116    fn output_ordering(&self, input: &[ExprProperties]) -> Result<SortProperties> {
117        // trunc preserves the order of the first argument
118        let value = &input[0];
119        let precision = input.get(1);
120
121        if precision
122            .map(|r| r.sort_properties.eq(&SortProperties::Singleton))
123            .unwrap_or(true)
124        {
125            Ok(value.sort_properties)
126        } else {
127            Ok(SortProperties::Unordered)
128        }
129    }
130
131    fn documentation(&self) -> Option<&Documentation> {
132        self.doc()
133    }
134}
135
136/// Truncate(numeric, decimalPrecision) and trunc(numeric) SQL function
137fn trunc(args: &[ArrayRef]) -> Result<ArrayRef> {
138    if args.len() != 1 && args.len() != 2 {
139        return exec_err!(
140            "truncate function requires one or two arguments, got {}",
141            args.len()
142        );
143    }
144
145    // If only one arg then invoke toolchain trunc(num) and precision = 0 by default
146    // or then invoke the compute_truncate method to process precision
147    let num = &args[0];
148    let precision = if args.len() == 1 {
149        ColumnarValue::Scalar(Int64(Some(0)))
150    } else {
151        ColumnarValue::Array(Arc::clone(&args[1]))
152    };
153
154    match num.data_type() {
155        Float64 => match precision {
156            ColumnarValue::Scalar(Int64(Some(0))) => {
157                Ok(Arc::new(
158                    args[0]
159                        .as_primitive::<Float64Type>()
160                        .unary::<_, Float64Type>(|x: f64| {
161                            if x == 0_f64 {
162                                0_f64
163                            } else {
164                                x.trunc()
165                            }
166                        }),
167                ) as ArrayRef)
168            }
169            ColumnarValue::Array(precision) => {
170                let num_array = num.as_primitive::<Float64Type>();
171                let precision_array = precision.as_primitive::<Int64Type>();
172                let result: PrimitiveArray<Float64Type> =
173                    arrow::compute::binary(num_array, precision_array, |x, y| {
174                        compute_truncate64(x, y)
175                    })?;
176
177                Ok(Arc::new(result) as ArrayRef)
178            }
179            _ => exec_err!("trunc function requires a scalar or array for precision"),
180        },
181        Float32 => match precision {
182            ColumnarValue::Scalar(Int64(Some(0))) => {
183                Ok(Arc::new(
184                    args[0]
185                        .as_primitive::<Float32Type>()
186                        .unary::<_, Float32Type>(|x: f32| {
187                            if x == 0_f32 {
188                                0_f32
189                            } else {
190                                x.trunc()
191                            }
192                        }),
193                ) as ArrayRef)
194            }
195            ColumnarValue::Array(precision) => {
196                let num_array = num.as_primitive::<Float32Type>();
197                let precision_array = precision.as_primitive::<Int64Type>();
198                let result: PrimitiveArray<Float32Type> =
199                    arrow::compute::binary(num_array, precision_array, |x, y| {
200                        compute_truncate32(x, y)
201                    })?;
202
203                Ok(Arc::new(result) as ArrayRef)
204            }
205            _ => exec_err!("trunc function requires a scalar or array for precision"),
206        },
207        other => exec_err!("Unsupported data type {other:?} for function trunc"),
208    }
209}
210
211fn compute_truncate32(x: f32, y: i64) -> f32 {
212    let factor = 10.0_f32.powi(y as i32);
213    (x * factor).round() / factor
214}
215
216fn compute_truncate64(x: f64, y: i64) -> f64 {
217    let factor = 10.0_f64.powi(y as i32);
218    (x * factor).round() / factor
219}
220
221#[cfg(test)]
222mod test {
223    use std::sync::Arc;
224
225    use crate::math::trunc::trunc;
226
227    use arrow::array::{ArrayRef, Float32Array, Float64Array, Int64Array};
228    use datafusion_common::cast::{as_float32_array, as_float64_array};
229
230    #[test]
231    fn test_truncate_32() {
232        let args: Vec<ArrayRef> = vec![
233            Arc::new(Float32Array::from(vec![
234                15.0,
235                1_234.267_8,
236                1_233.123_4,
237                3.312_979_2,
238                -21.123_4,
239            ])),
240            Arc::new(Int64Array::from(vec![0, 3, 2, 5, 6])),
241        ];
242
243        let result = trunc(&args).expect("failed to initialize function truncate");
244        let floats =
245            as_float32_array(&result).expect("failed to initialize function truncate");
246
247        assert_eq!(floats.len(), 5);
248        assert_eq!(floats.value(0), 15.0);
249        assert_eq!(floats.value(1), 1_234.268);
250        assert_eq!(floats.value(2), 1_233.12);
251        assert_eq!(floats.value(3), 3.312_98);
252        assert_eq!(floats.value(4), -21.123_4);
253    }
254
255    #[test]
256    fn test_truncate_64() {
257        let args: Vec<ArrayRef> = vec![
258            Arc::new(Float64Array::from(vec![
259                5.0,
260                234.267_812_176,
261                123.123_456_789,
262                123.312_979_313_2,
263                -321.123_1,
264            ])),
265            Arc::new(Int64Array::from(vec![0, 3, 2, 5, 6])),
266        ];
267
268        let result = trunc(&args).expect("failed to initialize function truncate");
269        let floats =
270            as_float64_array(&result).expect("failed to initialize function truncate");
271
272        assert_eq!(floats.len(), 5);
273        assert_eq!(floats.value(0), 5.0);
274        assert_eq!(floats.value(1), 234.268);
275        assert_eq!(floats.value(2), 123.12);
276        assert_eq!(floats.value(3), 123.312_98);
277        assert_eq!(floats.value(4), -321.123_1);
278    }
279
280    #[test]
281    fn test_truncate_64_one_arg() {
282        let args: Vec<ArrayRef> = vec![Arc::new(Float64Array::from(vec![
283            5.0,
284            234.267_812,
285            123.123_45,
286            123.312_979_313_2,
287            -321.123,
288        ]))];
289
290        let result = trunc(&args).expect("failed to initialize function truncate");
291        let floats =
292            as_float64_array(&result).expect("failed to initialize function truncate");
293
294        assert_eq!(floats.len(), 5);
295        assert_eq!(floats.value(0), 5.0);
296        assert_eq!(floats.value(1), 234.0);
297        assert_eq!(floats.value(2), 123.0);
298        assert_eq!(floats.value(3), 123.0);
299        assert_eq!(floats.value(4), -321.0);
300    }
301}