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, PartialEq, Eq, Hash)]
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 arrow::datatypes::DataType::Utf8;
102    use arrow::datatypes::Field;
103    use datafusion_common::config::ConfigOptions;
104    use std::sync::Arc;
105
106    fn to_lower(input: ArrayRef, expected: ArrayRef) -> Result<()> {
107        let func = LowerFunc::new();
108        let arg_fields = vec![Field::new("a", input.data_type().clone(), true).into()];
109
110        let args = ScalarFunctionArgs {
111            number_rows: input.len(),
112            args: vec![ColumnarValue::Array(input)],
113            arg_fields,
114            return_field: Field::new("f", Utf8, true).into(),
115            config_options: Arc::new(ConfigOptions::default()),
116        };
117
118        let result = match func.invoke_with_args(args)? {
119            ColumnarValue::Array(result) => result,
120            _ => unreachable!("lower"),
121        };
122        assert_eq!(&expected, &result);
123        Ok(())
124    }
125
126    #[test]
127    fn lower_maybe_optimization() -> Result<()> {
128        let input = Arc::new(StringArray::from(vec![
129            Some("农历新年"),
130            None,
131            Some("DATAFUSION"),
132            Some("0123456789"),
133            Some(""),
134        ])) as ArrayRef;
135
136        let expected = Arc::new(StringArray::from(vec![
137            Some("农历新年"),
138            None,
139            Some("datafusion"),
140            Some("0123456789"),
141            Some(""),
142        ])) as ArrayRef;
143
144        to_lower(input, expected)
145    }
146
147    #[test]
148    fn lower_full_optimization() -> Result<()> {
149        let input = Arc::new(StringArray::from(vec![
150            Some("ARROW"),
151            None,
152            Some("DATAFUSION"),
153            Some("0123456789"),
154            Some(""),
155        ])) as ArrayRef;
156
157        let expected = Arc::new(StringArray::from(vec![
158            Some("arrow"),
159            None,
160            Some("datafusion"),
161            Some("0123456789"),
162            Some(""),
163        ])) as ArrayRef;
164
165        to_lower(input, expected)
166    }
167
168    #[test]
169    fn lower_partial_optimization() -> Result<()> {
170        let input = Arc::new(StringArray::from(vec![
171            Some("ARROW"),
172            None,
173            Some("DATAFUSION"),
174            Some("@_"),
175            Some("0123456789"),
176            Some(""),
177            Some("\t\n"),
178            Some("ὈΔΥΣΣΕΎΣ"),
179            Some("TSCHÜSS"),
180            Some("Ⱦ"), // ⱦ: length change
181            Some("农历新年"),
182        ])) as ArrayRef;
183
184        let expected = Arc::new(StringArray::from(vec![
185            Some("arrow"),
186            None,
187            Some("datafusion"),
188            Some("@_"),
189            Some("0123456789"),
190            Some(""),
191            Some("\t\n"),
192            Some("ὀδυσσεύς"),
193            Some("tschüss"),
194            Some("ⱦ"),
195            Some("农历新年"),
196        ])) as ArrayRef;
197
198        to_lower(input, expected)
199    }
200}
datafusion_functions/string/lower.rs

datafusion_functions/string/
lower.rs
