/root/doris/be/src/exprs/json_functions.h
Line | Count | Source (jump to first uncovered line) |
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 | | #pragma once |
19 | | |
20 | | #include <fmt/format.h> |
21 | | #include <rapidjson/document.h> |
22 | | #include <simdjson.h> // IWYU pragma: keep |
23 | | |
24 | | #include <sstream> |
25 | | #include <string> |
26 | | #include <utility> |
27 | | #include <vector> |
28 | | |
29 | | #include "common/status.h" |
30 | | |
31 | | namespace simdjson { |
32 | | namespace fallback { |
33 | | namespace ondemand { |
34 | | class object; |
35 | | class value; |
36 | | } // namespace ondemand |
37 | | } // namespace fallback |
38 | | } // namespace simdjson |
39 | | |
40 | | namespace doris { |
41 | | |
42 | | enum JsonFunctionType { |
43 | | JSON_FUN_INT = 0, |
44 | | JSON_FUN_DOUBLE, |
45 | | JSON_FUN_STRING, |
46 | | |
47 | | JSON_FUN_UNKNOWN //The last |
48 | | }; |
49 | | |
50 | | struct JsonPath { |
51 | | std::string key; // key of a json object |
52 | | int idx; // array index of a json array, -1 means not set, -2 means * |
53 | | bool is_valid; // true if the path is successfully parsed |
54 | | |
55 | | JsonPath(const std::string& key_, int idx_, bool is_valid_) |
56 | 18 | : key(key_), idx(idx_), is_valid(is_valid_) {} |
57 | | |
58 | | JsonPath(std::string&& key_, int idx_, bool is_valid_) |
59 | 36 | : key(std::move(key_)), idx(idx_), is_valid(is_valid_) {} |
60 | | |
61 | 0 | std::string to_string() const { |
62 | 0 | std::stringstream ss; |
63 | 0 | if (!is_valid) { |
64 | 0 | return "INVALID"; |
65 | 0 | } |
66 | 0 | if (!key.empty()) { |
67 | 0 | ss << key; |
68 | 0 | } |
69 | 0 | if (idx == -2) { |
70 | 0 | ss << "[*]"; |
71 | 0 | } else if (idx > -1) { |
72 | 0 | ss << "[" << idx << "]"; |
73 | 0 | } |
74 | 0 | return ss.str(); |
75 | 0 | } |
76 | | |
77 | 0 | std::string debug_string() const { |
78 | 0 | return fmt::format("key:{}, idx:{}, valid:{}", key, idx, is_valid); |
79 | 0 | } |
80 | | }; |
81 | | |
82 | | class JsonFunctions { |
83 | | public: |
84 | | /** |
85 | | * The `document` parameter must be has parsed. |
86 | | * return Value Is Array object |
87 | | * wrap_explicitly is set to true when the returned Array is wrapped actively. |
88 | | */ |
89 | | static rapidjson::Value* get_json_array_from_parsed_json( |
90 | | const std::vector<JsonPath>& parsed_paths, rapidjson::Value* document, |
91 | | rapidjson::Document::AllocatorType& mem_allocator, bool* wrap_explicitly); |
92 | | |
93 | | // this is only for test, it will parse the json path inside, |
94 | | // so that we can easily pass a json path as string. |
95 | | static rapidjson::Value* get_json_array_from_parsed_json( |
96 | | const std::string& jsonpath, rapidjson::Value* document, |
97 | | rapidjson::Document::AllocatorType& mem_allocator, bool* wrap_explicitly); |
98 | | |
99 | | static rapidjson::Value* get_json_object_from_parsed_json( |
100 | | const std::vector<JsonPath>& parsed_paths, rapidjson::Value* document, |
101 | | rapidjson::Document::AllocatorType& mem_allocator); |
102 | | |
103 | | static void parse_json_paths(const std::string& path_strings, |
104 | | std::vector<JsonPath>* parsed_paths); |
105 | | // extract_from_object extracts value from object according to the json path. |
106 | | // Now, we do not support complete functions of json path. |
107 | | // Eg. city[*].id is not supported in this function |
108 | | static Status extract_from_object(simdjson::ondemand::object& obj, |
109 | | const std::vector<JsonPath>& jsonpath, |
110 | | simdjson::ondemand::value* value) noexcept; |
111 | | // src: {"a" : "b" {"c" : 1}, "e" : 123} |
112 | | // dst: {"a" : "b" {"d" : 1}} |
113 | | // merged: {"a" : "b" : {"c" : 1, "d" : 1}, "e" : 123} |
114 | | static void merge_objects(rapidjson::Value& dst_object, rapidjson::Value& src_object, |
115 | | rapidjson::Document::AllocatorType& allocator); |
116 | | |
117 | | static std::string print_json_value(const rapidjson::Value& value); |
118 | | |
119 | | static bool is_root_path(const std::vector<JsonPath>& json_path); |
120 | | |
121 | | private: |
122 | | static rapidjson::Value* match_value(const std::vector<JsonPath>& parsed_paths, |
123 | | rapidjson::Value* document, |
124 | | rapidjson::Document::AllocatorType& mem_allocator, |
125 | | bool is_insert_null = false); |
126 | | static void get_parsed_paths(const std::vector<std::string>& path_exprs, |
127 | | std::vector<JsonPath>* parsed_paths); |
128 | | }; |
129 | | } // namespace doris |