| OLD | NEW |
| (Empty) |
| 1 // Copyright (c) 2012 The Chromium Authors. All rights reserved. | |
| 2 // Use of this source code is governed by a BSD-style license that can be | |
| 3 // found in the LICENSE file. | |
| 4 | |
| 5 #ifndef BASE_JSON_JSON_PARSER_H_ | |
| 6 #define BASE_JSON_JSON_PARSER_H_ | |
| 7 | |
| 8 #include <string> | |
| 9 | |
| 10 #include "base/base_export.h" | |
| 11 #include "base/basictypes.h" | |
| 12 #include "base/compiler_specific.h" | |
| 13 #include "base/json/json_reader.h" | |
| 14 #include "base/strings/string_piece.h" | |
| 15 | |
| 16 #if !defined(OS_CHROMEOS) | |
| 17 #include "base/gtest_prod_util.h" | |
| 18 #endif | |
| 19 | |
| 20 namespace base { | |
| 21 class Value; | |
| 22 } | |
| 23 | |
| 24 #if defined(OS_CHROMEOS) | |
| 25 // Chromium and Chromium OS check out gtest to different places, so this is | |
| 26 // unable to compile on both if gtest_prod.h is included here. Instead, include | |
| 27 // its only contents -- this will need to be updated if the macro ever changes. | |
| 28 #define FRIEND_TEST(test_case_name, test_name)\ | |
| 29 friend class test_case_name##_##test_name##_Test | |
| 30 | |
| 31 #define FRIEND_TEST_ALL_PREFIXES(test_case_name, test_name) \ | |
| 32 FRIEND_TEST(test_case_name, test_name); \ | |
| 33 FRIEND_TEST(test_case_name, DISABLED_##test_name); \ | |
| 34 FRIEND_TEST(test_case_name, FLAKY_##test_name) | |
| 35 #endif // OS_CHROMEOS | |
| 36 | |
| 37 namespace base { | |
| 38 namespace internal { | |
| 39 | |
| 40 class JSONParserTest; | |
| 41 | |
| 42 // The implementation behind the JSONReader interface. This class is not meant | |
| 43 // to be used directly; it encapsulates logic that need not be exposed publicly. | |
| 44 // | |
| 45 // This parser guarantees O(n) time through the input string. It also optimizes | |
| 46 // base::StringValue by using StringPiece where possible when returning Value | |
| 47 // objects by using "hidden roots," discussed in the implementation. | |
| 48 // | |
| 49 // Iteration happens on the byte level, with the functions CanConsume and | |
| 50 // NextChar. The conversion from byte to JSON token happens without advancing | |
| 51 // the parser in GetNextToken/ParseToken, that is tokenization operates on | |
| 52 // the current parser position without advancing. | |
| 53 // | |
| 54 // Built on top of these are a family of Consume functions that iterate | |
| 55 // internally. Invariant: on entry of a Consume function, the parser is wound | |
| 56 // to the first byte of a valid JSON token. On exit, it is on the last byte | |
| 57 // of a token, such that the next iteration of the parser will be at the byte | |
| 58 // immediately following the token, which would likely be the first byte of the | |
| 59 // next token. | |
| 60 class BASE_EXPORT_PRIVATE JSONParser { | |
| 61 public: | |
| 62 explicit JSONParser(int options); | |
| 63 ~JSONParser(); | |
| 64 | |
| 65 // Parses the input string according to the set options and returns the | |
| 66 // result as a Value owned by the caller. | |
| 67 Value* Parse(const StringPiece& input); | |
| 68 | |
| 69 // Returns the error code. | |
| 70 JSONReader::JsonParseError error_code() const; | |
| 71 | |
| 72 // Returns the human-friendly error message. | |
| 73 std::string GetErrorMessage() const; | |
| 74 | |
| 75 private: | |
| 76 enum Token { | |
| 77 T_OBJECT_BEGIN, // { | |
| 78 T_OBJECT_END, // } | |
| 79 T_ARRAY_BEGIN, // [ | |
| 80 T_ARRAY_END, // ] | |
| 81 T_STRING, | |
| 82 T_NUMBER, | |
| 83 T_BOOL_TRUE, // true | |
| 84 T_BOOL_FALSE, // false | |
| 85 T_NULL, // null | |
| 86 T_LIST_SEPARATOR, // , | |
| 87 T_OBJECT_PAIR_SEPARATOR, // : | |
| 88 T_END_OF_INPUT, | |
| 89 T_INVALID_TOKEN, | |
| 90 }; | |
| 91 | |
| 92 // A helper class used for parsing strings. One optimization performed is to | |
| 93 // create base::Value with a StringPiece to avoid unnecessary std::string | |
| 94 // copies. This is not possible if the input string needs to be decoded from | |
| 95 // UTF-16 to UTF-8, or if an escape sequence causes characters to be skipped. | |
| 96 // This class centralizes that logic. | |
| 97 class StringBuilder { | |
| 98 public: | |
| 99 // Empty constructor. Used for creating a builder with which to Swap(). | |
| 100 StringBuilder(); | |
| 101 | |
| 102 // |pos| is the beginning of an input string, excluding the |"|. | |
| 103 explicit StringBuilder(const char* pos); | |
| 104 | |
| 105 ~StringBuilder(); | |
| 106 | |
| 107 // Swaps the contents of |other| with this. | |
| 108 void Swap(StringBuilder* other); | |
| 109 | |
| 110 // Either increases the |length_| of the string or copies the character if | |
| 111 // the StringBuilder has been converted. |c| must be in the basic ASCII | |
| 112 // plane; all other characters need to be in UTF-8 units, appended with | |
| 113 // AppendString below. | |
| 114 void Append(const char& c); | |
| 115 | |
| 116 // Appends a string to the std::string. Must be Convert()ed to use. | |
| 117 void AppendString(const std::string& str); | |
| 118 | |
| 119 // Converts the builder from its default StringPiece to a full std::string, | |
| 120 // performing a copy. Once a builder is converted, it cannot be made a | |
| 121 // StringPiece again. | |
| 122 void Convert(); | |
| 123 | |
| 124 // Returns whether the builder can be converted to a StringPiece. | |
| 125 bool CanBeStringPiece() const; | |
| 126 | |
| 127 // Returns the StringPiece representation. Returns an empty piece if it | |
| 128 // cannot be converted. | |
| 129 StringPiece AsStringPiece(); | |
| 130 | |
| 131 // Returns the builder as a std::string. | |
| 132 const std::string& AsString(); | |
| 133 | |
| 134 private: | |
| 135 // The beginning of the input string. | |
| 136 const char* pos_; | |
| 137 | |
| 138 // Number of bytes in |pos_| that make up the string being built. | |
| 139 size_t length_; | |
| 140 | |
| 141 // The copied string representation. NULL until Convert() is called. | |
| 142 // Strong. scoped_ptr<T> has too much of an overhead here. | |
| 143 std::string* string_; | |
| 144 }; | |
| 145 | |
| 146 // Quick check that the stream has capacity to consume |length| more bytes. | |
| 147 bool CanConsume(int length); | |
| 148 | |
| 149 // The basic way to consume a single character in the stream. Consumes one | |
| 150 // byte of the input stream and returns a pointer to the rest of it. | |
| 151 const char* NextChar(); | |
| 152 | |
| 153 // Performs the equivalent of NextChar N times. | |
| 154 void NextNChars(int n); | |
| 155 | |
| 156 // Skips over whitespace and comments to find the next token in the stream. | |
| 157 // This does not advance the parser for non-whitespace or comment chars. | |
| 158 Token GetNextToken(); | |
| 159 | |
| 160 // Consumes whitespace characters and comments until the next non-that is | |
| 161 // encountered. | |
| 162 void EatWhitespaceAndComments(); | |
| 163 // Helper function that consumes a comment, assuming that the parser is | |
| 164 // currently wound to a '/'. | |
| 165 bool EatComment(); | |
| 166 | |
| 167 // Calls GetNextToken() and then ParseToken(). Caller owns the result. | |
| 168 Value* ParseNextToken(); | |
| 169 | |
| 170 // Takes a token that represents the start of a Value ("a structural token" | |
| 171 // in RFC terms) and consumes it, returning the result as an object the | |
| 172 // caller owns. | |
| 173 Value* ParseToken(Token token); | |
| 174 | |
| 175 // Assuming that the parser is currently wound to '{', this parses a JSON | |
| 176 // object into a DictionaryValue. | |
| 177 Value* ConsumeDictionary(); | |
| 178 | |
| 179 // Assuming that the parser is wound to '[', this parses a JSON list into a | |
| 180 // ListValue. | |
| 181 Value* ConsumeList(); | |
| 182 | |
| 183 // Calls through ConsumeStringRaw and wraps it in a value. | |
| 184 Value* ConsumeString(); | |
| 185 | |
| 186 // Assuming that the parser is wound to a double quote, this parses a string, | |
| 187 // decoding any escape sequences and converts UTF-16 to UTF-8. Returns true on | |
| 188 // success and Swap()s the result into |out|. Returns false on failure with | |
| 189 // error information set. | |
| 190 bool ConsumeStringRaw(StringBuilder* out); | |
| 191 // Helper function for ConsumeStringRaw() that consumes the next four or 10 | |
| 192 // bytes (parser is wound to the first character of a HEX sequence, with the | |
| 193 // potential for consuming another \uXXXX for a surrogate). Returns true on | |
| 194 // success and places the UTF8 code units in |dest_string|, and false on | |
| 195 // failure. | |
| 196 bool DecodeUTF16(std::string* dest_string); | |
| 197 // Helper function for ConsumeStringRaw() that takes a single code point, | |
| 198 // decodes it into UTF-8 units, and appends it to the given builder. The | |
| 199 // point must be valid. | |
| 200 void DecodeUTF8(const int32& point, StringBuilder* dest); | |
| 201 | |
| 202 // Assuming that the parser is wound to the start of a valid JSON number, | |
| 203 // this parses and converts it to either an int or double value. | |
| 204 Value* ConsumeNumber(); | |
| 205 // Helper that reads characters that are ints. Returns true if a number was | |
| 206 // read and false on error. | |
| 207 bool ReadInt(bool allow_leading_zeros); | |
| 208 | |
| 209 // Consumes the literal values of |true|, |false|, and |null|, assuming the | |
| 210 // parser is wound to the first character of any of those. | |
| 211 Value* ConsumeLiteral(); | |
| 212 | |
| 213 // Compares two string buffers of a given length. | |
| 214 static bool StringsAreEqual(const char* left, const char* right, size_t len); | |
| 215 | |
| 216 // Sets the error information to |code| at the current column, based on | |
| 217 // |index_| and |index_last_line_|, with an optional positive/negative | |
| 218 // adjustment by |column_adjust|. | |
| 219 void ReportError(JSONReader::JsonParseError code, int column_adjust); | |
| 220 | |
| 221 // Given the line and column number of an error, formats one of the error | |
| 222 // message contants from json_reader.h for human display. | |
| 223 static std::string FormatErrorMessage(int line, int column, | |
| 224 const std::string& description); | |
| 225 | |
| 226 // base::JSONParserOptions that control parsing. | |
| 227 int options_; | |
| 228 | |
| 229 // Pointer to the start of the input data. | |
| 230 const char* start_pos_; | |
| 231 | |
| 232 // Pointer to the current position in the input data. Equivalent to | |
| 233 // |start_pos_ + index_|. | |
| 234 const char* pos_; | |
| 235 | |
| 236 // Pointer to the last character of the input data. | |
| 237 const char* end_pos_; | |
| 238 | |
| 239 // The index in the input stream to which the parser is wound. | |
| 240 int index_; | |
| 241 | |
| 242 // The number of times the parser has recursed (current stack depth). | |
| 243 int stack_depth_; | |
| 244 | |
| 245 // The line number that the parser is at currently. | |
| 246 int line_number_; | |
| 247 | |
| 248 // The last value of |index_| on the previous line. | |
| 249 int index_last_line_; | |
| 250 | |
| 251 // Error information. | |
| 252 JSONReader::JsonParseError error_code_; | |
| 253 int error_line_; | |
| 254 int error_column_; | |
| 255 | |
| 256 friend class JSONParserTest; | |
| 257 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, NextChar); | |
| 258 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ConsumeDictionary); | |
| 259 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ConsumeList); | |
| 260 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ConsumeString); | |
| 261 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ConsumeLiterals); | |
| 262 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ConsumeNumbers); | |
| 263 FRIEND_TEST_ALL_PREFIXES(JSONParserTest, ErrorMessages); | |
| 264 | |
| 265 DISALLOW_COPY_AND_ASSIGN(JSONParser); | |
| 266 }; | |
| 267 | |
| 268 } // namespace internal | |
| 269 } // namespace base | |
| 270 | |
| 271 #endif // BASE_JSON_JSON_PARSER_H_ | |
| OLD | NEW |