datafusion_functions/string/
lower.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;
19use std::any::Any;
20
21use crate::string::common::to_lower;
22use crate::utils::utf8_to_str_type;
23use datafusion_common::types::logical_string;
24use datafusion_common::Result;
25use datafusion_expr::{
26    Coercion, ColumnarValue, Documentation, ScalarFunctionArgs, ScalarUDFImpl, Signature,
27    TypeSignatureClass, Volatility,
28};
29use datafusion_macros::user_doc;
30
31#[user_doc(
32    doc_section(label = "String Functions"),
33    description = "Converts a string to lower-case.",
34    syntax_example = "lower(str)",
35    sql_example = r#"```sql
36> select lower('Ångström');
37+-------------------------+
38| lower(Utf8("Ångström")) |
39+-------------------------+
40| ångström                |
41+-------------------------+
42```"#,
43    standard_argument(name = "str", prefix = "String"),
44    related_udf(name = "initcap"),
45    related_udf(name = "upper")
46)]
47#[derive(Debug)]
48pub struct LowerFunc {
49    signature: Signature,
50}
51
52impl Default for LowerFunc {
53    fn default() -> Self {
54        Self::new()
55    }
56}
57
58impl LowerFunc {
59    pub fn new() -> Self {
60        Self {
61            signature: Signature::coercible(
62                vec![Coercion::new_exact(TypeSignatureClass::Native(
63                    logical_string(),
64                ))],
65                Volatility::Immutable,
66            ),
67        }
68    }
69}
70
71impl ScalarUDFImpl for LowerFunc {
72    fn as_any(&self) -> &dyn Any {
73        self
74    }
75
76    fn name(&self) -> &str {
77        "lower"
78    }
79
80    fn signature(&self) -> &Signature {
81        &self.signature
82    }
83
84    fn return_type(&self, arg_types: &[DataType]) -> Result<DataType> {
85        utf8_to_str_type(&arg_types[0], "lower")
86    }
87
88    fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
89        to_lower(&args.args, "lower")
90    }
91
92    fn documentation(&self) -> Option<&Documentation> {
93        self.doc()
94    }
95}
96
97#[cfg(test)]
98mod tests {
99    use super::*;
100    use arrow::array::{Array, ArrayRef, StringArray};
101    use std::sync::Arc;
102
103    fn to_lower(input: ArrayRef, expected: ArrayRef) -> Result<()> {
104        let func = LowerFunc::new();
105
106        let args = ScalarFunctionArgs {
107            number_rows: input.len(),
108            args: vec![ColumnarValue::Array(input)],
109            return_type: &DataType::Utf8,
110        };
111
112        let result = match func.invoke_with_args(args)? {
113            ColumnarValue::Array(result) => result,
114            _ => unreachable!("lower"),
115        };
116        assert_eq!(&expected, &result);
117        Ok(())
118    }
119
120    #[test]
121    fn lower_maybe_optimization() -> Result<()> {
122        let input = Arc::new(StringArray::from(vec![
123            Some("农历新年"),
124            None,
125            Some("DATAFUSION"),
126            Some("0123456789"),
127            Some(""),
128        ])) as ArrayRef;
129
130        let expected = Arc::new(StringArray::from(vec![
131            Some("农历新年"),
132            None,
133            Some("datafusion"),
134            Some("0123456789"),
135            Some(""),
136        ])) as ArrayRef;
137
138        to_lower(input, expected)
139    }
140
141    #[test]
142    fn lower_full_optimization() -> Result<()> {
143        let input = Arc::new(StringArray::from(vec![
144            Some("ARROW"),
145            None,
146            Some("DATAFUSION"),
147            Some("0123456789"),
148            Some(""),
149        ])) as ArrayRef;
150
151        let expected = Arc::new(StringArray::from(vec![
152            Some("arrow"),
153            None,
154            Some("datafusion"),
155            Some("0123456789"),
156            Some(""),
157        ])) as ArrayRef;
158
159        to_lower(input, expected)
160    }
161
162    #[test]
163    fn lower_partial_optimization() -> Result<()> {
164        let input = Arc::new(StringArray::from(vec![
165            Some("ARROW"),
166            None,
167            Some("DATAFUSION"),
168            Some("@_"),
169            Some("0123456789"),
170            Some(""),
171            Some("\t\n"),
172            Some("ὈΔΥΣΣΕΎΣ"),
173            Some("TSCHÜSS"),
174            Some("Ⱦ"), // ⱦ: length change
175            Some("农历新年"),
176        ])) as ArrayRef;
177
178        let expected = Arc::new(StringArray::from(vec![
179            Some("arrow"),
180            None,
181            Some("datafusion"),
182            Some("@_"),
183            Some("0123456789"),
184            Some(""),
185            Some("\t\n"),
186            Some("ὀδυσσεύς"),
187            Some("tschüss"),
188            Some("ⱦ"),
189            Some("农历新年"),
190        ])) as ArrayRef;
191
192        to_lower(input, expected)
193    }
194}
datafusion_functions/string/lower.rs

datafusion_functions/string/
lower.rs
