datafusion_functions/crypto/
sha.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 crate::crypto::basic::{DigestAlgorithm, digest_process};
19
20use arrow::datatypes::DataType;
21use datafusion_common::{
22    Result,
23    types::{logical_binary, logical_string},
24    utils::take_function_args,
25};
26use datafusion_expr::{
27    ColumnarValue, Documentation, ScalarFunctionArgs, ScalarUDFImpl, Signature,
28    TypeSignature, Volatility,
29};
30use datafusion_expr_common::signature::{Coercion, TypeSignatureClass};
31use datafusion_macros::user_doc;
32
33#[user_doc(
34    doc_section(label = "Hashing Functions"),
35    description = "Computes the SHA-224 hash of a binary string.",
36    syntax_example = "sha224(expression)",
37    sql_example = r#"```sql
38> select sha224('foo');
39+----------------------------------------------------------+
40| sha224(Utf8("foo"))                                      |
41+----------------------------------------------------------+
42| 0808f64e60d58979fcb676c96ec938270dea42445aeefcd3a4e6f8db |
43+----------------------------------------------------------+
44```"#,
45    standard_argument(name = "expression", prefix = "String")
46)]
47struct SHA224Doc;
48
49#[user_doc(
50    doc_section(label = "Hashing Functions"),
51    description = "Computes the SHA-256 hash of a binary string.",
52    syntax_example = "sha256(expression)",
53    sql_example = r#"```sql
54> select sha256('foo');
55+------------------------------------------------------------------+
56| sha256(Utf8("foo"))                                              |
57+------------------------------------------------------------------+
58| 2c26b46b68ffc68ff99b453c1d30413413422d706483bfa0f98a5e886266e7ae |
59+------------------------------------------------------------------+
60```"#,
61    standard_argument(name = "expression", prefix = "String")
62)]
63struct SHA256Doc;
64
65#[user_doc(
66    doc_section(label = "Hashing Functions"),
67    description = "Computes the SHA-384 hash of a binary string.",
68    syntax_example = "sha384(expression)",
69    sql_example = r#"```sql
70> select sha384('foo');
71+--------------------------------------------------------------------------------------------------+
72| sha384(Utf8("foo"))                                                                              |
73+--------------------------------------------------------------------------------------------------+
74| 98c11ffdfdd540676b1a137cb1a22b2a70350c9a44171d6b1180c6be5cbb2ee3f79d532c8a1dd9ef2e8e08e752a3babb |
75+--------------------------------------------------------------------------------------------------+
76```"#,
77    standard_argument(name = "expression", prefix = "String")
78)]
79struct SHA384Doc;
80
81#[user_doc(
82    doc_section(label = "Hashing Functions"),
83    description = "Computes the SHA-512 hash of a binary string.",
84    syntax_example = "sha512(expression)",
85    sql_example = r#"```sql
86> select sha512('foo');
87+----------------------------------------------------------------------------------------------------------------------------------+
88| sha512(Utf8("foo"))                                                                                                              |
89+----------------------------------------------------------------------------------------------------------------------------------+
90| f7fbba6e0636f890e56fbbf3283e524c6fa3204ae298382d624741d0dc6638326e282c41be5e4254d8820772c5518a2c5a8c0c7f7eda19594a7eb539453e1ed7 |
91+----------------------------------------------------------------------------------------------------------------------------------+
92```"#,
93    standard_argument(name = "expression", prefix = "String")
94)]
95struct SHA512Doc;
96
97#[derive(Debug, PartialEq, Eq, Hash)]
98pub struct SHAFunc {
99    signature: Signature,
100    name: &'static str,
101    algorithm: DigestAlgorithm,
102}
103
104impl SHAFunc {
105    pub fn sha224() -> Self {
106        Self::new("sha224", DigestAlgorithm::Sha224)
107    }
108
109    pub fn sha256() -> Self {
110        Self::new("sha256", DigestAlgorithm::Sha256)
111    }
112
113    pub fn sha384() -> Self {
114        Self::new("sha384", DigestAlgorithm::Sha384)
115    }
116
117    pub fn sha512() -> Self {
118        Self::new("sha512", DigestAlgorithm::Sha512)
119    }
120
121    fn new(name: &'static str, algorithm: DigestAlgorithm) -> Self {
122        Self {
123            signature: Signature::one_of(
124                vec![
125                    TypeSignature::Coercible(vec![Coercion::new_exact(
126                        TypeSignatureClass::Native(logical_string()),
127                    )]),
128                    TypeSignature::Coercible(vec![Coercion::new_exact(
129                        TypeSignatureClass::Native(logical_binary()),
130                    )]),
131                ],
132                Volatility::Immutable,
133            ),
134            name,
135            algorithm,
136        }
137    }
138}
139
140impl ScalarUDFImpl for SHAFunc {
141    fn name(&self) -> &str {
142        self.name
143    }
144
145    fn signature(&self) -> &Signature {
146        &self.signature
147    }
148
149    fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
150        Ok(DataType::Binary)
151    }
152
153    fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
154        let [data] = take_function_args(self.name(), args.args)?;
155        digest_process(&data, self.algorithm)
156    }
157
158    fn documentation(&self) -> Option<&Documentation> {
159        match self.algorithm {
160            DigestAlgorithm::Sha224 => SHA224Doc {}.doc(),
161            DigestAlgorithm::Sha256 => SHA256Doc {}.doc(),
162            DigestAlgorithm::Sha384 => SHA384Doc {}.doc(),
163            DigestAlgorithm::Sha512 => SHA512Doc {}.doc(),
164            DigestAlgorithm::Md5
165            | DigestAlgorithm::Blake2s
166            | DigestAlgorithm::Blake2b
167            | DigestAlgorithm::Blake3 => unreachable!(),
168        }
169    }
170}
datafusion_functions/crypto/sha.rs

datafusion_functions/crypto/
sha.rs
