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

datafusion_functions_nested/
lib.rs
