|
| 1 | +// Copyright 2023 Greptime Team |
| 2 | +// |
| 3 | +// Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | +// you may not use this file except in compliance with the License. |
| 5 | +// You may obtain a copy of the License at |
| 6 | +// |
| 7 | +// http://www.apache.org/licenses/LICENSE-2.0 |
| 8 | +// |
| 9 | +// Unless required by applicable law or agreed to in writing, software |
| 10 | +// distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | +// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | +// See the License for the specific language governing permissions and |
| 13 | +// limitations under the License. |
| 14 | + |
| 15 | +use std::fmt::{self, Display}; |
| 16 | + |
| 17 | +use common_query::error::{InvalidFuncArgsSnafu, Result, UnsupportedInputDataTypeSnafu}; |
| 18 | +use common_query::prelude::Signature; |
| 19 | +use datafusion::logical_expr::Volatility; |
| 20 | +use datatypes::data_type::ConcreteDataType; |
| 21 | +use datatypes::prelude::VectorRef; |
| 22 | +use datatypes::scalars::ScalarVectorBuilder; |
| 23 | +use datatypes::vectors::{BooleanVectorBuilder, MutableVector}; |
| 24 | +use snafu::ensure; |
| 25 | + |
| 26 | +use crate::function::{Function, FunctionContext}; |
| 27 | + |
| 28 | +/// Checks if the input is a JSON object of the given type. |
| 29 | +macro_rules! json_is { |
| 30 | + ($name:ident, $json_type:ident, $doc:expr) => { |
| 31 | + paste::paste! { |
| 32 | + #[derive(Clone, Debug, Default)] |
| 33 | + pub struct $name; |
| 34 | + |
| 35 | + impl Function for $name { |
| 36 | + fn name(&self) -> &str { |
| 37 | + stringify!([<$name:snake>]) |
| 38 | + } |
| 39 | + |
| 40 | + fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> { |
| 41 | + Ok(ConcreteDataType::boolean_datatype()) |
| 42 | + } |
| 43 | + |
| 44 | + fn signature(&self) -> Signature { |
| 45 | + Signature::exact(vec![ConcreteDataType::json_datatype()], Volatility::Immutable) |
| 46 | + } |
| 47 | + |
| 48 | + fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> { |
| 49 | + ensure!( |
| 50 | + columns.len() == 1, |
| 51 | + InvalidFuncArgsSnafu { |
| 52 | + err_msg: format!( |
| 53 | + "The length of the args is not correct, expect exactly one, have: {}", |
| 54 | + columns.len() |
| 55 | + ), |
| 56 | + } |
| 57 | + ); |
| 58 | + |
| 59 | + let jsons = &columns[0]; |
| 60 | + let size = jsons.len(); |
| 61 | + let datatype = jsons.data_type(); |
| 62 | + let mut results = BooleanVectorBuilder::with_capacity(size); |
| 63 | + |
| 64 | + match datatype { |
| 65 | + // JSON data type uses binary vector |
| 66 | + ConcreteDataType::Binary(_) => { |
| 67 | + for i in 0..size { |
| 68 | + let json = jsons.get_ref(i); |
| 69 | + let json = json.as_binary(); |
| 70 | + let result = match json { |
| 71 | + Ok(Some(json)) => { |
| 72 | + Some(jsonb::[<is_ $json_type>](json)) |
| 73 | + } |
| 74 | + _ => None, |
| 75 | + }; |
| 76 | + results.push(result); |
| 77 | + } |
| 78 | + } |
| 79 | + _ => { |
| 80 | + return UnsupportedInputDataTypeSnafu { |
| 81 | + function: stringify!([<$name:snake>]), |
| 82 | + datatypes: columns.iter().map(|c| c.data_type()).collect::<Vec<_>>(), |
| 83 | + } |
| 84 | + .fail(); |
| 85 | + } |
| 86 | + } |
| 87 | + |
| 88 | + Ok(results.to_vector()) |
| 89 | + } |
| 90 | + } |
| 91 | + |
| 92 | + impl Display for $name { |
| 93 | + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { |
| 94 | + write!(f, "{}", stringify!([<$name:snake>]).to_ascii_uppercase()) |
| 95 | + } |
| 96 | + } |
| 97 | + } |
| 98 | + } |
| 99 | +} |
| 100 | + |
| 101 | +json_is!(JsonIsNull, null, "Checks if the input JSONB is null"); |
| 102 | +json_is!( |
| 103 | + JsonIsBool, |
| 104 | + boolean, |
| 105 | + "Checks if the input JSONB is a boolean type JSON value" |
| 106 | +); |
| 107 | +json_is!( |
| 108 | + JsonIsInt, |
| 109 | + i64, |
| 110 | + "Checks if the input JSONB is a integer type JSON value" |
| 111 | +); |
| 112 | +json_is!( |
| 113 | + JsonIsFloat, |
| 114 | + number, |
| 115 | + "Checks if the input JSONB is a JSON float" |
| 116 | +); |
| 117 | +json_is!( |
| 118 | + JsonIsString, |
| 119 | + string, |
| 120 | + "Checks if the input JSONB is a JSON string" |
| 121 | +); |
| 122 | +json_is!( |
| 123 | + JsonIsArray, |
| 124 | + array, |
| 125 | + "Checks if the input JSONB is a JSON array" |
| 126 | +); |
| 127 | +json_is!( |
| 128 | + JsonIsObject, |
| 129 | + object, |
| 130 | + "Checks if the input JSONB is a JSON object" |
| 131 | +); |
| 132 | + |
| 133 | +#[cfg(test)] |
| 134 | +mod tests { |
| 135 | + use std::sync::Arc; |
| 136 | + |
| 137 | + use datatypes::scalars::ScalarVector; |
| 138 | + use datatypes::vectors::BinaryVector; |
| 139 | + |
| 140 | + use super::*; |
| 141 | + |
| 142 | + #[test] |
| 143 | + fn test_json_is_functions() { |
| 144 | + let json_is_functions: [&dyn Function; 6] = [ |
| 145 | + &JsonIsBool, |
| 146 | + &JsonIsInt, |
| 147 | + &JsonIsFloat, |
| 148 | + &JsonIsString, |
| 149 | + &JsonIsArray, |
| 150 | + &JsonIsObject, |
| 151 | + ]; |
| 152 | + let expected_names = [ |
| 153 | + "json_is_bool", |
| 154 | + "json_is_int", |
| 155 | + "json_is_float", |
| 156 | + "json_is_string", |
| 157 | + "json_is_array", |
| 158 | + "json_is_object", |
| 159 | + ]; |
| 160 | + for (func, expected_name) in json_is_functions.iter().zip(expected_names.iter()) { |
| 161 | + assert_eq!(func.name(), *expected_name); |
| 162 | + assert_eq!( |
| 163 | + func.return_type(&[ConcreteDataType::json_datatype()]) |
| 164 | + .unwrap(), |
| 165 | + ConcreteDataType::boolean_datatype() |
| 166 | + ); |
| 167 | + assert_eq!( |
| 168 | + func.signature(), |
| 169 | + Signature::exact( |
| 170 | + vec![ConcreteDataType::json_datatype()], |
| 171 | + Volatility::Immutable |
| 172 | + ) |
| 173 | + ); |
| 174 | + } |
| 175 | + |
| 176 | + let json_strings = [ |
| 177 | + r#"true"#, |
| 178 | + r#"1"#, |
| 179 | + r#"1.0"#, |
| 180 | + r#""The pig fly through a castle, and has been attracted by the princess.""#, |
| 181 | + r#"[1, 2]"#, |
| 182 | + r#"{"a": 1}"#, |
| 183 | + ]; |
| 184 | + let expected_results = [ |
| 185 | + [true, false, false, false, false, false], |
| 186 | + [false, true, false, false, false, false], |
| 187 | + // Integers are also floats |
| 188 | + [false, true, true, false, false, false], |
| 189 | + [false, false, false, true, false, false], |
| 190 | + [false, false, false, false, true, false], |
| 191 | + [false, false, false, false, false, true], |
| 192 | + ]; |
| 193 | + |
| 194 | + let jsonbs = json_strings |
| 195 | + .iter() |
| 196 | + .map(|s| { |
| 197 | + let value = jsonb::parse_value(s.as_bytes()).unwrap(); |
| 198 | + value.to_vec() |
| 199 | + }) |
| 200 | + .collect::<Vec<_>>(); |
| 201 | + let json_vector = BinaryVector::from_vec(jsonbs); |
| 202 | + let args: Vec<VectorRef> = vec![Arc::new(json_vector)]; |
| 203 | + |
| 204 | + for (func, expected_result) in json_is_functions.iter().zip(expected_results.iter()) { |
| 205 | + let vector = func.eval(FunctionContext::default(), &args).unwrap(); |
| 206 | + assert_eq!(vector.len(), json_strings.len()); |
| 207 | + |
| 208 | + for (i, expected) in expected_result.iter().enumerate() { |
| 209 | + let result = vector.get_ref(i); |
| 210 | + let result = result.as_boolean().unwrap().unwrap(); |
| 211 | + assert_eq!(result, *expected); |
| 212 | + } |
| 213 | + } |
| 214 | + } |
| 215 | +} |
0 commit comments