181 lines
5.6 KiB
C++
181 lines
5.6 KiB
C++
#include "parser.h"
|
|
#include <algorithm>
|
|
#include <cassert>
|
|
#include <codecvt>
|
|
#include <deque>
|
|
#include <locale>
|
|
#include <sstream>
|
|
|
|
#include <iostream>
|
|
|
|
namespace {
|
|
double readNumber(std::string in) {
|
|
std::istringstream iss(in);
|
|
double d;
|
|
iss >> d;
|
|
return d;
|
|
}
|
|
|
|
std::string readString(std::string in) {
|
|
std::string result;
|
|
for (std::size_t i = 1; i < in.length() - 1; i++) {
|
|
if (in[i] == '\\' && in[i + 1] != 'u') {
|
|
char c = 0;
|
|
switch (in[i + 1]) {
|
|
case '"':
|
|
case '\\':
|
|
case '/':
|
|
c = in[i + 1];
|
|
break;
|
|
case 'b':
|
|
c = '\b';
|
|
break;
|
|
case 'f':
|
|
c = '\f';
|
|
break;
|
|
case 'n':
|
|
c = '\n';
|
|
break;
|
|
case 'r':
|
|
c = '\r';
|
|
break;
|
|
case 't':
|
|
c = '\t';
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
result.push_back(c);
|
|
i++;
|
|
} else if (in[i] == '\\'){
|
|
char16_t unicode_value;
|
|
std::wstring_convert<std::codecvt_utf8<char16_t>, char16_t > utf8converter;
|
|
unicode_value = 0;
|
|
for (int j = 0; j < 4; j++, i++) {
|
|
char hex = in[i + 2];
|
|
unicode_value *= 16;
|
|
if (hex >= '0' && hex <= '9')
|
|
unicode_value += hex - '0';
|
|
else if (hex >= 'A' && hex <= '9')
|
|
unicode_value += hex - 'A' + 10;
|
|
else
|
|
unicode_value += hex - 'a' + 10;
|
|
}
|
|
result.append(utf8converter.to_bytes(unicode_value));
|
|
i++;
|
|
} else {
|
|
if (iscntrl(in[i])) {
|
|
throw SyntaxError("Control character inside string");
|
|
}
|
|
result.push_back(in[i]);
|
|
}
|
|
}
|
|
return result;
|
|
}
|
|
}
|
|
|
|
namespace json {
|
|
|
|
Parser::Parser(JSONLexer lex) : JSONParser<JSON>(), m_lex(lex)
|
|
{}
|
|
|
|
Parser::Token Parser::lex() {
|
|
try {
|
|
JSONLexer::Token orig = m_lex.nextToken();
|
|
JSONParser_Symbol s;
|
|
switch (orig.type) {
|
|
case JSONLexer::COLON:
|
|
s = JSONParser_Symbol::T_COLON;
|
|
break;
|
|
case JSONLexer::COMMA:
|
|
s = JSONParser_Symbol::T_COMMA;
|
|
break;
|
|
case JSONLexer::LBRACE:
|
|
s = JSONParser_Symbol::T_LBRACE;
|
|
break;
|
|
case JSONLexer::RBRACE:
|
|
s = JSONParser_Symbol::T_RBRACE;
|
|
break;
|
|
case JSONLexer::LBRACKET:
|
|
s = JSONParser_Symbol::T_LBRACKET;
|
|
break;
|
|
case JSONLexer::RBRACKET:
|
|
s = JSONParser_Symbol::T_RBRACKET;
|
|
break;
|
|
case JSONLexer::STRING:
|
|
return Token{JSONParser_Symbol::T_STRING, JSON::string(readString(orig.content))};
|
|
case JSONLexer::NUMBER:
|
|
return Token{JSONParser_Symbol::T_NUMBER, JSON::num(readNumber(orig.content))};
|
|
case JSONLexer::TTRUE:
|
|
return Token{JSONParser_Symbol::T_TTRUE, JSON::boolean(true)};
|
|
case JSONLexer::TFALSE:
|
|
return Token{JSONParser_Symbol::T_TFALSE, JSON::boolean(false)};
|
|
case JSONLexer::TNULL:
|
|
return Token{JSONParser_Symbol::T_TNULL, JSON::null()};
|
|
default:
|
|
//impossible
|
|
break;
|
|
}
|
|
return Token{s, JSON()};
|
|
}
|
|
catch (JSONLexer::NoMoreTokens) {
|
|
return Token{JSONParser_Symbol::T_EOF, JSON()};
|
|
}
|
|
}
|
|
|
|
JSON Parser::reduce_0(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_1(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_2(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_3(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_4(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_5(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_6(std::deque<Token> subparts) {
|
|
return std::move(subparts[0].value);
|
|
}
|
|
JSON Parser::reduce_7(std::deque<Token> subparts) {
|
|
return std::move(subparts[1].value);
|
|
}
|
|
JSON Parser::reduce_8(std::deque<Token>) {
|
|
return JSON::object();
|
|
}
|
|
JSON Parser::reduce_9(std::deque<Token> subparts) {
|
|
JSON obj = JSON::object();
|
|
obj[std::move(subparts[0].value)] = std::move(subparts[2].value);
|
|
return obj;
|
|
}
|
|
JSON Parser::reduce_10(std::deque<Token> subparts) {
|
|
JSON obj = std::move(subparts[4].value);
|
|
obj[std::move(subparts[0].value)] = std::move(subparts[2].value);
|
|
return obj;
|
|
}
|
|
JSON Parser::reduce_11(std::deque<Token> subparts) {
|
|
return std::move(subparts[1].value);
|
|
}
|
|
JSON Parser::reduce_12(std::deque<Token>) {
|
|
return JSON::array();
|
|
}
|
|
JSON Parser::reduce_13(std::deque<Token> subparts) {
|
|
JSON arr = JSON::array();
|
|
arr.push_front(std::move(subparts[0].value));
|
|
return arr;
|
|
}
|
|
JSON Parser::reduce_14(std::deque<Token> subparts) {
|
|
JSON arr = std::move(subparts[2].value);
|
|
arr.push_front(std::move(subparts[0].value));
|
|
return arr;
|
|
}
|
|
|
|
}
|