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
40#[macro_use]
41pub mod macros_lambda;
42
43pub mod array_any_match;
44pub(crate) mod lambda_utils;
45
46pub mod array_compact;
47pub mod array_filter;
48pub mod array_has;
49pub mod array_normalize;
50pub mod array_transform;
51pub mod arrays_zip;
52pub mod cardinality;
53pub mod concat;
54pub mod cosine_distance;
55pub mod dimension;
56pub mod distance;
57pub mod empty;
58pub mod except;
59pub mod expr_ext;
60pub mod extract;
61pub mod flatten;
62pub mod inner_product;
63pub mod length;
64pub mod make_array;
65pub mod map;
66pub mod map_entries;
67pub mod map_extract;
68pub mod map_keys;
69pub mod map_values;
70pub mod min_max;
71pub mod planner;
72pub mod position;
73pub mod range;
74pub mod remove;
75pub mod repeat;
76pub mod replace;
77pub mod resize;
78pub mod reverse;
79pub mod set_ops;
80pub mod sort;
81pub mod string;
82pub mod utils;
83
84use datafusion_common::Result;
85use datafusion_execution::FunctionRegistry;
86use datafusion_expr::{HigherOrderUDF, ScalarUDF};
87use log::debug;
88use std::sync::Arc;
89
90/// Fluent-style API for creating `Expr`s
91pub mod expr_fn {
92    pub use super::array_any_match::array_any_match;
93    pub use super::array_compact::array_compact;
94    pub use super::array_filter::array_filter;
95    pub use super::array_has::array_has;
96    pub use super::array_has::array_has_all;
97    pub use super::array_has::array_has_any;
98    pub use super::array_normalize::array_normalize;
99    pub use super::array_transform::array_transform;
100    pub use super::arrays_zip::arrays_zip;
101    pub use super::cardinality::cardinality;
102    pub use super::concat::array_append;
103    pub use super::concat::array_concat;
104    pub use super::concat::array_prepend;
105    pub use super::cosine_distance::cosine_distance;
106    pub use super::dimension::array_dims;
107    pub use super::dimension::array_ndims;
108    pub use super::distance::array_distance;
109    pub use super::empty::array_empty;
110    pub use super::except::array_except;
111    pub use super::extract::array_any_value;
112    pub use super::extract::array_element;
113    pub use super::extract::array_pop_back;
114    pub use super::extract::array_pop_front;
115    pub use super::extract::array_slice;
116    pub use super::flatten::flatten;
117    pub use super::inner_product::inner_product;
118    pub use super::length::array_length;
119    pub use super::make_array::make_array;
120    pub use super::map_entries::map_entries;
121    pub use super::map_extract::map_extract;
122    pub use super::map_keys::map_keys;
123    pub use super::map_values::map_values;
124    pub use super::min_max::array_max;
125    pub use super::min_max::array_min;
126    pub use super::position::array_position;
127    pub use super::position::array_positions;
128    pub use super::range::gen_series;
129    pub use super::range::range;
130    pub use super::remove::array_remove;
131    pub use super::remove::array_remove_all;
132    pub use super::remove::array_remove_n;
133    pub use super::repeat::array_repeat;
134    pub use super::replace::array_replace;
135    pub use super::replace::array_replace_all;
136    pub use super::replace::array_replace_n;
137    pub use super::resize::array_resize;
138    pub use super::reverse::array_reverse;
139    pub use super::set_ops::array_distinct;
140    pub use super::set_ops::array_intersect;
141    pub use super::set_ops::array_union;
142    pub use super::sort::array_sort;
143    pub use super::string::array_to_string;
144    pub use super::string::string_to_array;
145}
146
147/// Return all default nested type functions
148pub fn all_default_nested_functions() -> Vec<Arc<ScalarUDF>> {
149    vec![
150        array_compact::array_compact_udf(),
151        string::array_to_string_udf(),
152        string::string_to_array_udf(),
153        range::range_udf(),
154        range::gen_series_udf(),
155        dimension::array_dims_udf(),
156        cardinality::cardinality_udf(),
157        dimension::array_ndims_udf(),
158        concat::array_append_udf(),
159        concat::array_prepend_udf(),
160        concat::array_concat_udf(),
161        except::array_except_udf(),
162        extract::array_element_udf(),
163        extract::array_pop_back_udf(),
164        extract::array_pop_front_udf(),
165        extract::array_slice_udf(),
166        extract::array_any_value_udf(),
167        make_array::make_array_udf(),
168        array_has::array_has_udf(),
169        array_has::array_has_all_udf(),
170        array_has::array_has_any_udf(),
171        empty::array_empty_udf(),
172        length::array_length_udf(),
173        array_normalize::array_normalize_udf(),
174        cosine_distance::cosine_distance_udf(),
175        inner_product::inner_product_udf(),
176        distance::array_distance_udf(),
177        flatten::flatten_udf(),
178        min_max::array_max_udf(),
179        min_max::array_min_udf(),
180        sort::array_sort_udf(),
181        repeat::array_repeat_udf(),
182        resize::array_resize_udf(),
183        reverse::array_reverse_udf(),
184        set_ops::array_distinct_udf(),
185        set_ops::array_intersect_udf(),
186        set_ops::array_union_udf(),
187        arrays_zip::arrays_zip_udf(),
188        position::array_position_udf(),
189        position::array_positions_udf(),
190        remove::array_remove_udf(),
191        remove::array_remove_all_udf(),
192        remove::array_remove_n_udf(),
193        replace::array_replace_n_udf(),
194        replace::array_replace_all_udf(),
195        replace::array_replace_udf(),
196        map::map_udf(),
197        map_entries::map_entries_udf(),
198        map_extract::map_extract_udf(),
199        map_keys::map_keys_udf(),
200        map_values::map_values_udf(),
201    ]
202}
203
204pub fn all_default_higher_order_functions() -> Vec<Arc<HigherOrderUDF>> {
205    vec![
206        array_any_match::array_any_match_higher_order_function(),
207        array_filter::array_filter_higher_order_function(),
208        array_transform::array_transform_higher_order_function(),
209    ]
210}
211
212/// Registers all enabled packages with a [`FunctionRegistry`]
213pub fn register_all(registry: &mut dyn FunctionRegistry) -> Result<()> {
214    let functions: Vec<Arc<ScalarUDF>> = all_default_nested_functions();
215    functions.into_iter().try_for_each(|udf| {
216        let existing_udf = registry.register_udf(udf)?;
217        if let Some(existing_udf) = existing_udf {
218            debug!("Overwrite existing UDF: {}", existing_udf.name());
219        }
220        Ok(()) as Result<()>
221    })?;
222
223    let functions: Vec<Arc<HigherOrderUDF>> = all_default_higher_order_functions();
224    functions.into_iter().try_for_each(|function| {
225        let existing_function = registry.register_higher_order_function(function)?;
226        if let Some(existing_function) = existing_function {
227            debug!(
228                "Overwrite existing higher-order function: {}",
229                existing_function.name()
230            );
231        }
232        Ok(()) as Result<()>
233    })?;
234
235    Ok(())
236}
237
238#[cfg(test)]
239mod tests {
240    use crate::{all_default_higher_order_functions, all_default_nested_functions};
241    use datafusion_common::Result;
242    use std::collections::HashSet;
243
244    #[test]
245    fn test_no_duplicate_name() -> Result<()> {
246        let scalars = all_default_nested_functions();
247        let scalars = scalars.iter().map(|s| (s.name(), s.aliases()));
248
249        let lambdas = all_default_higher_order_functions();
250        let lambdas = lambdas.iter().map(|l| (l.name(), l.aliases()));
251
252        let mut names = HashSet::new();
253
254        for (name, aliases) in scalars.chain(lambdas) {
255            assert!(
256                names.insert(name.to_string().to_lowercase()),
257                "duplicate function name: {name}",
258            );
259            for alias in aliases {
260                assert!(
261                    names.insert(alias.to_string().to_lowercase()),
262                    "duplicate function name: {alias}"
263                );
264            }
265        }
266        Ok(())
267    }
268}
datafusion_functions_nested/lib.rs

datafusion_functions_nested/
lib.rs
