datafusion_functions_nested/
lib.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
18#![doc(
19    html_logo_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg",
20    html_favicon_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg"
21)]
22#![cfg_attr(docsrs, feature(doc_cfg))]
23// Make sure fast / cheap clones on Arc are explicit:
24// https://github.com/apache/datafusion/issues/11143
25#![deny(clippy::clone_on_ref_ptr)]
26#![cfg_attr(test, allow(clippy::needless_pass_by_value))]
27
28//! Nested type Functions for [DataFusion].
29//!
30//! This crate contains a collection of nested type functions implemented using the
31//! extension API.
32//!
33//! [DataFusion]: https://crates.io/crates/datafusion
34//!
35//! You can register the functions in this crate using the [`register_all`] function.
36
37#[macro_use]
38pub mod macros;
39
40pub mod array_has;
41pub mod arrays_zip;
42pub mod cardinality;
43pub mod concat;
44pub mod dimension;
45pub mod distance;
46pub mod empty;
47pub mod except;
48pub mod expr_ext;
49pub mod extract;
50pub mod flatten;
51pub mod length;
52pub mod make_array;
53pub mod map;
54pub mod map_entries;
55pub mod map_extract;
56pub mod map_keys;
57pub mod map_values;
58pub mod min_max;
59pub mod planner;
60pub mod position;
61pub mod range;
62pub mod remove;
63pub mod repeat;
64pub mod replace;
65pub mod resize;
66pub mod reverse;
67pub mod set_ops;
68pub mod sort;
69pub mod string;
70pub mod utils;
71
72use datafusion_common::Result;
73use datafusion_execution::FunctionRegistry;
74use datafusion_expr::ScalarUDF;
75use log::debug;
76use std::sync::Arc;
77
78/// Fluent-style API for creating `Expr`s
79pub mod expr_fn {
80    pub use super::array_has::array_has;
81    pub use super::array_has::array_has_all;
82    pub use super::array_has::array_has_any;
83    pub use super::arrays_zip::arrays_zip;
84    pub use super::cardinality::cardinality;
85    pub use super::concat::array_append;
86    pub use super::concat::array_concat;
87    pub use super::concat::array_prepend;
88    pub use super::dimension::array_dims;
89    pub use super::dimension::array_ndims;
90    pub use super::distance::array_distance;
91    pub use super::empty::array_empty;
92    pub use super::except::array_except;
93    pub use super::extract::array_any_value;
94    pub use super::extract::array_element;
95    pub use super::extract::array_pop_back;
96    pub use super::extract::array_pop_front;
97    pub use super::extract::array_slice;
98    pub use super::flatten::flatten;
99    pub use super::length::array_length;
100    pub use super::make_array::make_array;
101    pub use super::map_entries::map_entries;
102    pub use super::map_extract::map_extract;
103    pub use super::map_keys::map_keys;
104    pub use super::map_values::map_values;
105    pub use super::min_max::array_max;
106    pub use super::min_max::array_min;
107    pub use super::position::array_position;
108    pub use super::position::array_positions;
109    pub use super::range::gen_series;
110    pub use super::range::range;
111    pub use super::remove::array_remove;
112    pub use super::remove::array_remove_all;
113    pub use super::remove::array_remove_n;
114    pub use super::repeat::array_repeat;
115    pub use super::replace::array_replace;
116    pub use super::replace::array_replace_all;
117    pub use super::replace::array_replace_n;
118    pub use super::resize::array_resize;
119    pub use super::reverse::array_reverse;
120    pub use super::set_ops::array_distinct;
121    pub use super::set_ops::array_intersect;
122    pub use super::set_ops::array_union;
123    pub use super::sort::array_sort;
124    pub use super::string::array_to_string;
125    pub use super::string::string_to_array;
126}
127
128/// Return all default nested type functions
129pub fn all_default_nested_functions() -> Vec<Arc<ScalarUDF>> {
130    vec![
131        string::array_to_string_udf(),
132        string::string_to_array_udf(),
133        range::range_udf(),
134        range::gen_series_udf(),
135        dimension::array_dims_udf(),
136        cardinality::cardinality_udf(),
137        dimension::array_ndims_udf(),
138        concat::array_append_udf(),
139        concat::array_prepend_udf(),
140        concat::array_concat_udf(),
141        except::array_except_udf(),
142        extract::array_element_udf(),
143        extract::array_pop_back_udf(),
144        extract::array_pop_front_udf(),
145        extract::array_slice_udf(),
146        extract::array_any_value_udf(),
147        make_array::make_array_udf(),
148        array_has::array_has_udf(),
149        array_has::array_has_all_udf(),
150        array_has::array_has_any_udf(),
151        empty::array_empty_udf(),
152        length::array_length_udf(),
153        distance::array_distance_udf(),
154        flatten::flatten_udf(),
155        min_max::array_max_udf(),
156        min_max::array_min_udf(),
157        sort::array_sort_udf(),
158        repeat::array_repeat_udf(),
159        resize::array_resize_udf(),
160        reverse::array_reverse_udf(),
161        set_ops::array_distinct_udf(),
162        set_ops::array_intersect_udf(),
163        set_ops::array_union_udf(),
164        arrays_zip::arrays_zip_udf(),
165        position::array_position_udf(),
166        position::array_positions_udf(),
167        remove::array_remove_udf(),
168        remove::array_remove_all_udf(),
169        remove::array_remove_n_udf(),
170        replace::array_replace_n_udf(),
171        replace::array_replace_all_udf(),
172        replace::array_replace_udf(),
173        map::map_udf(),
174        map_entries::map_entries_udf(),
175        map_extract::map_extract_udf(),
176        map_keys::map_keys_udf(),
177        map_values::map_values_udf(),
178    ]
179}
180
181/// Registers all enabled packages with a [`FunctionRegistry`]
182pub fn register_all(registry: &mut dyn FunctionRegistry) -> Result<()> {
183    let functions: Vec<Arc<ScalarUDF>> = all_default_nested_functions();
184    functions.into_iter().try_for_each(|udf| {
185        let existing_udf = registry.register_udf(udf)?;
186        if let Some(existing_udf) = existing_udf {
187            debug!("Overwrite existing UDF: {}", existing_udf.name());
188        }
189        Ok(()) as Result<()>
190    })?;
191
192    Ok(())
193}
194
195#[cfg(test)]
196mod tests {
197    use crate::all_default_nested_functions;
198    use datafusion_common::Result;
199    use std::collections::HashSet;
200
201    #[test]
202    fn test_no_duplicate_name() -> Result<()> {
203        let mut names = HashSet::new();
204        for func in all_default_nested_functions() {
205            assert!(
206                names.insert(func.name().to_string().to_lowercase()),
207                "duplicate function name: {}",
208                func.name()
209            );
210            for alias in func.aliases() {
211                assert!(
212                    names.insert(alias.to_string().to_lowercase()),
213                    "duplicate function name: {alias}"
214                );
215            }
216        }
217        Ok(())
218    }
219}
datafusion_functions_nested/lib.rs

datafusion_functions_nested/
lib.rs
