poco/JSON/src/Parser.cpp

510 lines
13 KiB
C++
Raw Normal View History

2012-11-11 09:57:01 +01:00
//
// Parser.cpp
//
// $Id$
//
// Library: JSON
// Package: JSON
// Module: Parser
//
// Copyright (c) 2012, Applied Informatics Software Engineering GmbH.
// and Contributors.
//
// SPDX-License-Identifier: BSL-1.0
2012-11-11 09:57:01 +01:00
//
#include "Poco/JSON/Parser.h"
#include "Poco/JSON/JSONException.h"
#include "Poco/Ascii.h"
#include "Poco/Token.h"
#include "Poco/UTF8Encoding.h"
2015-09-09 14:26:26 +02:00
#include "Poco/String.h"
2012-11-11 09:57:01 +01:00
#undef min
#undef max
#include <limits>
2013-05-20 20:37:13 +02:00
#include <clocale>
#include <istream>
2012-11-11 09:57:01 +01:00
namespace Poco {
namespace JSON {
static const unsigned char UTF8_LEAD_BITS[4] = { 0x00, 0xC0, 0xE0, 0xF0 };
2013-05-20 20:37:13 +02:00
const int Parser::_asciiClass[] = {
xx, xx, xx, xx, xx, xx, xx, xx,
xx, C_WHITE, C_WHITE, xx, xx, C_WHITE, xx, xx,
xx, xx, xx, xx, xx, xx, xx, xx,
xx, xx, xx, xx, xx, xx, xx, xx,
C_SPACE, C_ETC, C_QUOTE, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC,
C_ETC, C_ETC, C_STAR, C_PLUS, C_COMMA, C_MINUS, C_POINT, C_SLASH,
C_ZERO, C_DIGIT, C_DIGIT, C_DIGIT, C_DIGIT, C_DIGIT, C_DIGIT, C_DIGIT,
C_DIGIT, C_DIGIT, C_COLON, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC,
C_ETC, C_ABCDF, C_ABCDF, C_ABCDF, C_ABCDF, C_E, C_ABCDF, C_ETC,
C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC,
C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC, C_ETC,
C_ETC, C_ETC, C_ETC, C_LSQRB, C_BACKS, C_RSQRB, C_ETC, C_ETC,
C_ETC, C_LOW_A, C_LOW_B, C_LOW_C, C_LOW_D, C_LOW_E, C_LOW_F, C_ETC,
C_ETC, C_ETC, C_ETC, C_ETC, C_LOW_L, C_ETC, C_LOW_N, C_ETC,
C_ETC, C_ETC, C_LOW_R, C_LOW_S, C_LOW_T, C_LOW_U, C_ETC, C_ETC,
C_ETC, C_ETC, C_ETC, C_LCURB, C_ETC, C_RCURB, C_ETC, C_ETC
};
const int Parser::_stateTransitionTable[NR_STATES][NR_CLASSES] = {
/*
white 1-9 ABCDF etc
space | { } [ ] : , " \ / + - . 0 | a b c d e f l n r s t u | E | * */
/*start GO*/ {GO,GO,-6,xx,-5,xx,xx,xx,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*ok OK*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*object OB*/ {OB,OB,xx,-9,xx,xx,xx,xx,SB,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*key KE*/ {KE,KE,xx,xx,xx,xx,xx,xx,SB,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*colon CO*/ {CO,CO,xx,xx,xx,xx,-2,xx,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*value VA*/ {VA,VA,-6,xx,-5,xx,xx,xx,SB,xx,CB,xx,MX,xx,ZX,IX,xx,xx,xx,xx,xx,FA,xx,NU,xx,xx,TR,xx,xx,xx,xx,xx},
/*array AR*/ {AR,AR,-6,xx,-5,-7,xx,xx,SB,xx,CB,xx,MX,xx,ZX,IX,xx,xx,xx,xx,xx,FA,xx,NU,xx,xx,TR,xx,xx,xx,xx,xx},
/*string ST*/ {ST,xx,ST,ST,ST,ST,ST,ST,-4,EX,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST,ST},
2012-08-18 02:54:54 -05:00
/*escape EC*/ {xx,xx,xx,xx,xx,xx,xx,xx,ST,ST,ST,xx,xx,xx,xx,xx,xx,ST,xx,xx,xx,ST,xx,ST,ST,xx,ST,U1,xx,xx,xx,xx},
2013-05-20 20:37:13 +02:00
/*u1 U1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,U2,U2,U2,U2,U2,U2,U2,U2,xx,xx,xx,xx,xx,xx,U2,U2,xx,xx},
/*u2 U2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,U3,U3,U3,U3,U3,U3,U3,U3,xx,xx,xx,xx,xx,xx,U3,U3,xx,xx},
/*u3 U3*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,U4,U4,U4,U4,U4,U4,U4,U4,xx,xx,xx,xx,xx,xx,U4,U4,xx,xx},
/*u4 U4*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,UC,UC,UC,UC,UC,UC,UC,UC,xx,xx,xx,xx,xx,xx,UC,UC,xx,xx},
/*minus MI*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,ZE,IT,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*zero ZE*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,CB,xx,xx,DF,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*int IT*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,CB,xx,xx,DF,IT,IT,xx,xx,xx,xx,DE,xx,xx,xx,xx,xx,xx,xx,xx,DE,xx,xx},
/*frac FR*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,CB,xx,xx,xx,FR,FR,xx,xx,xx,xx,E1,xx,xx,xx,xx,xx,xx,xx,xx,E1,xx,xx},
/*e E1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,E2,E2,xx,E3,E3,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*ex E2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,E3,E3,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*exp E3*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,xx,xx,xx,xx,E3,E3,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*tr T1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,T2,xx,xx,xx,xx,xx,xx,xx},
/*tru T2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,T3,xx,xx,xx,xx},
/*1 T3*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,OK,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*fa F1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,F2,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*fal F2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,F3,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*fals F3*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,F4,xx,xx,xx,xx,xx,xx},
/*0 F4*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,OK,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*nu N1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,N2,xx,xx,xx,xx},
/*nul N2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,N3,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*null N3*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,CB,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,OK,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*/ C1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,C2},
/*/* C2*/ {C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C3},
/** C3*/ {C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,CE,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C2,C3},
/*_. FX*/ {OK,OK,xx,-8,xx,-7,xx,-3,xx,xx,xx,xx,xx,xx,FR,FR,xx,xx,xx,xx,E1,xx,xx,xx,xx,xx,xx,xx,xx,E1,xx,xx},
/*\ D1*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,D2,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx},
/*\ D2*/ {xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,xx,U1,xx,xx,xx,xx},
};
2013-05-20 23:36:58 -05:00
Parser::Parser(const Handler::Ptr& pHandler, std::size_t bufSize) :
2013-05-20 20:37:13 +02:00
_pHandler(pHandler),
_state(GO),
_beforeCommentState(0),
_type(JSON_T_NONE),
_escaped(0),
_comment(0),
_utf16HighSurrogate(0),
_depth(JSON_UNLIMITED_DEPTH),
2013-05-20 20:37:13 +02:00
_top(-1),
_stack(JSON_PARSER_STACK_SIZE),
2013-05-20 23:36:58 -05:00
_parseBuffer(bufSize),
_decimalPoint('.'),
2013-05-20 20:37:13 +02:00
_allowNullByte(true),
_allowComments(false)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
_parseBuffer.resize(0);
2013-05-20 20:37:13 +02:00
push(MODE_DONE);
}
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
Parser::~Parser()
{
2013-05-20 23:36:58 -05:00
}
void Parser::reset()
{
_state = GO;
_beforeCommentState = 0;
_type = JSON_T_NONE;
_escaped = 0;
_utf16HighSurrogate = 0;
_top = -1;
2013-05-20 20:37:13 +02:00
2013-05-20 23:36:58 -05:00
_stack.clear();
2013-06-15 20:28:09 -05:00
_parseBuffer.resize(0);
2013-05-20 23:36:58 -05:00
push(MODE_DONE);
if (_pHandler) _pHandler->reset();
2013-05-20 20:37:13 +02:00
}
2013-05-20 23:36:58 -05:00
2013-05-20 20:37:13 +02:00
Dynamic::Var Parser::parse(const std::string& json)
{
std::string::const_iterator it = json.begin();
std::string::const_iterator end = json.end();
Source<std::string::const_iterator> source(it, end);
2013-05-20 20:37:13 +02:00
int c = 0;
2013-05-20 23:36:58 -05:00
while(source.nextChar(c))
2012-11-11 09:57:01 +01:00
{
if (0 == parseChar(c, source))
throw SyntaxException("JSON syntax error");
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
if (!done())
throw JSONException("JSON syntax error");
2012-11-11 09:57:01 +01:00
return asVar();
2013-05-20 20:37:13 +02:00
}
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
Dynamic::Var Parser::parse(std::istream& in)
{
std::istreambuf_iterator<char> it(in.rdbuf());
std::istreambuf_iterator<char> end;
Source<std::istreambuf_iterator<char> > source(it, end);
2013-05-20 20:37:13 +02:00
int c = 0;
2013-05-20 23:36:58 -05:00
while(source.nextChar(c))
2012-11-11 09:57:01 +01:00
{
2013-05-20 23:36:58 -05:00
if (0 == parseChar(c, source)) throw JSONException("JSON syntax error");
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
if (!done())
throw JSONException("JSON syntax error");
return asVar();
2013-05-20 20:37:13 +02:00
}
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
bool Parser::push(int mode)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
_top += 1;
if (_depth < 0)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
if (_top >= _stack.size())
_stack.resize(_stack.size() * 2, true);
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
if (_top >= _depth) return false;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
_stack[_top] = mode;
return true;
}
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
bool Parser::pop(int mode)
{
if (_top < 0 || _stack[_top] != mode)
return false;
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
_top -= 1;
return true;
}
2012-11-11 09:57:01 +01:00
2013-05-20 23:36:58 -05:00
2013-05-20 20:37:13 +02:00
void Parser::clearBuffer()
{
2013-06-15 20:28:09 -05:00
_parseBuffer.resize(0);
2013-05-20 20:37:13 +02:00
}
2013-05-20 20:37:13 +02:00
void Parser::parseBufferPopBackChar()
{
2013-06-15 20:28:09 -05:00
poco_assert(_parseBuffer.size() >= 1);
_parseBuffer.resize(_parseBuffer.size() - 1);
2013-05-20 20:37:13 +02:00
}
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
void Parser::parseBufferPushBackChar(char c)
{
2013-06-15 20:28:09 -05:00
if (_parseBuffer.size() + 1 >= _parseBuffer.capacity())
_parseBuffer.setCapacity(_parseBuffer.capacity() * 2);
2013-06-15 20:28:09 -05:00
_parseBuffer.append(c);
2013-05-20 20:37:13 +02:00
}
2012-11-11 09:57:01 +01:00
void Parser::addEscapedCharToParseBuffer(CharIntType nextChar)
2013-05-20 20:37:13 +02:00
{
_escaped = 0;
// remove the backslash
parseBufferPopBackChar();
2013-05-20 20:37:13 +02:00
switch(nextChar)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
case 'b':
parseBufferPushBackChar('\b');
break;
case 'f':
parseBufferPushBackChar('\f');
break;
case 'n':
parseBufferPushBackChar('\n');
break;
case 'r':
parseBufferPushBackChar('\r');
break;
case 't':
parseBufferPushBackChar('\t');
break;
case '"':
parseBufferPushBackChar('"');
break;
case '\\':
parseBufferPushBackChar('\\');
break;
case '/':
parseBufferPushBackChar('/');
break;
case 'u':
parseBufferPushBackChar('\\');
parseBufferPushBackChar('u');
break;
default:
break;
}
}
2012-11-11 09:57:01 +01:00
void Parser::addCharToParseBuffer(CharIntType nextChar, int nextClass)
2013-05-20 20:37:13 +02:00
{
if (_escaped)
{
addEscapedCharToParseBuffer(nextChar);
return;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else if (!_comment)
{
2013-05-20 20:37:13 +02:00
if ((_type != JSON_T_NONE) ||
!((nextClass == C_SPACE) || (nextClass == C_WHITE)))
{
2013-05-20 20:37:13 +02:00
parseBufferPushBackChar((char) nextChar);
}
}
2013-05-20 20:37:13 +02:00
}
2012-11-11 09:57:01 +01:00
Parser::CharIntType Parser::decodeUnicodeChar()
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
int i;
unsigned uc = 0;
char* p;
int trailBytes;
2012-11-11 09:57:01 +01:00
2013-06-15 20:28:09 -05:00
poco_assert(_parseBuffer.size() >= 6);
p = &_parseBuffer[_parseBuffer.size() - 4];
2012-11-11 09:57:01 +01:00
for (i = 12; i >= 0; i -= 4, ++p)
{
2013-05-20 20:37:13 +02:00
unsigned x = *p;
if (x >= 'a') x -= ('a' - 10);
else if (x >= 'A') x -= ('A' - 10);
else x &= ~0x30u;
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
poco_assert(x < 16);
uc |= x << i;
2012-11-11 09:57:01 +01:00
}
if ( !_allowNullByte && uc == 0 ) return 0;
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
// clear UTF-16 char from buffer
2013-06-15 20:28:09 -05:00
_parseBuffer.resize(_parseBuffer.size() - 6);
2012-11-11 09:57:01 +01:00
2013-06-15 20:28:09 -05:00
if (_utf16HighSurrogate)
{
if (isLowSurrogate(uc))
2013-06-15 20:28:09 -05:00
{
uc = decodeSurrogatePair(_utf16HighSurrogate, uc);
trailBytes = 3;
2013-05-20 20:37:13 +02:00
_utf16HighSurrogate = 0;
2013-06-15 20:28:09 -05:00
}
else // high surrogate without a following low surrogate
2013-06-15 20:28:09 -05:00
{
2013-05-20 20:37:13 +02:00
return 0;
2012-11-11 09:57:01 +01:00
}
2013-06-15 20:28:09 -05:00
}
else
{
if (uc < 0x80)
{
trailBytes = 0;
2013-06-15 20:28:09 -05:00
}
else if (uc < 0x800)
{
trailBytes = 1;
2013-06-15 20:28:09 -05:00
}
else if (isHighSurrogate(uc))
2013-06-15 20:28:09 -05:00
{
2013-05-20 20:37:13 +02:00
// save the high surrogate and wait for the low surrogate
_utf16HighSurrogate = uc;
return 1;
2013-06-15 20:28:09 -05:00
}
else if (isLowSurrogate(uc))
2013-06-15 20:28:09 -05:00
{
2013-05-20 20:37:13 +02:00
// low surrogate without a preceding high surrogate
return 0;
2013-06-15 20:28:09 -05:00
}
else
{
trailBytes = 2;
2012-11-11 09:57:01 +01:00
}
}
_parseBuffer.append((char) ((uc >> (trailBytes * 6)) | UTF8_LEAD_BITS[trailBytes]));
2012-11-11 09:57:01 +01:00
for (i = trailBytes * 6 - 6; i >= 0; i -= 6)
2013-06-15 20:28:09 -05:00
{
_parseBuffer.append((char) (((uc >> i) & 0x3F) | 0x80));
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
return 1;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
void Parser::parseBuffer()
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
if (_pHandler)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
int type = _type; // just to silence g++
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
if (type != JSON_T_NONE)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
assertNonContainer();
2012-11-11 09:57:01 +01:00
2013-05-20 20:37:13 +02:00
switch(type)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
case JSON_T_TRUE:
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
_pHandler->value(true);
break;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
case JSON_T_FALSE:
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
_pHandler->value(false);
break;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
case JSON_T_NULL:
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
_pHandler->null();
break;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
case JSON_T_FLOAT:
{
// Float can't end with a dot
2013-06-15 20:28:09 -05:00
if (_parseBuffer[_parseBuffer.size() - 1] == '.' ) throw SyntaxException("JSON syntax error");
2012-11-11 09:57:01 +01:00
2013-06-15 20:28:09 -05:00
double float_value = NumberParser::parseFloat(std::string(_parseBuffer.begin(), _parseBuffer.size()));
2013-05-20 20:37:13 +02:00
_pHandler->value(float_value);
break;
}
2013-05-20 20:37:13 +02:00
case JSON_T_INTEGER:
{
2013-05-20 20:37:13 +02:00
#if defined(POCO_HAVE_INT64)
2015-09-09 14:26:26 +02:00
std::string numStr(_parseBuffer.begin(), _parseBuffer.size());
2013-05-20 20:37:13 +02:00
try
{
2015-09-09 14:26:26 +02:00
Poco::trimInPlace(numStr);
Int64 value = NumberParser::parse64(numStr);
2013-05-20 20:37:13 +02:00
// if number is 32-bit, then handle as such
2013-05-20 23:36:58 -05:00
if (value > std::numeric_limits<int>::max()
|| value < std::numeric_limits<int>::min() )
2013-05-20 20:37:13 +02:00
{
_pHandler->value(value);
}
else
{
_pHandler->value(static_cast<int>(value));
}
}
// try to handle error as unsigned in case of overflow
catch ( const SyntaxException& )
{
2015-09-09 14:26:26 +02:00
UInt64 value = NumberParser::parseUnsigned64(numStr);
2013-05-20 20:37:13 +02:00
// if number is 32-bit, then handle as such
if ( value > std::numeric_limits<unsigned>::max() )
{
_pHandler->value(value);
}
else
{
_pHandler->value(static_cast<unsigned>(value));
}
}
#else
try
{
2015-09-09 14:26:26 +02:00
int value = NumberParser::parse(numStr);
2013-05-20 20:37:13 +02:00
_pHandler->value(value);
}
// try to handle error as unsigned in case of overflow
catch ( const SyntaxException& )
{
2015-09-09 14:26:26 +02:00
unsigned value = NumberParser::parseUnsigned(numStr);
2013-05-20 20:37:13 +02:00
_pHandler->value(value);
}
#endif
}
2013-05-20 20:37:13 +02:00
break;
case JSON_T_STRING:
{
2013-06-15 20:28:09 -05:00
_pHandler->value(std::string(_parseBuffer.begin(), _parseBuffer.size()));
2013-05-20 20:37:13 +02:00
break;
}
}
2012-11-11 09:57:01 +01:00
}
}
2013-05-20 20:37:13 +02:00
clearBuffer();
}
2012-11-11 09:57:01 +01:00
int Parser::utf8CheckFirst(char byte)
2012-11-11 09:57:01 +01:00
{
2013-05-20 20:37:13 +02:00
unsigned char u = (unsigned char) byte;
if(u < 0x80)
return 1;
if (0x80 <= u && u <= 0xBF)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
// second, third or fourth byte of a multi-byte
// sequence, i.e. a "continuation byte"
2013-05-20 20:37:13 +02:00
return 0;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else if(u == 0xC0 || u == 0xC1)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
// overlong encoding of an ASCII byte
2013-05-20 20:37:13 +02:00
return 0;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else if(0xC2 <= u && u <= 0xDF)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
// 2-byte sequence
2013-05-20 20:37:13 +02:00
return 2;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else if(0xE0 <= u && u <= 0xEF)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
// 3-byte sequence
2013-05-20 20:37:13 +02:00
return 3;
2012-11-11 09:57:01 +01:00
}
2013-05-20 20:37:13 +02:00
else if(0xF0 <= u && u <= 0xF4)
2012-11-11 09:57:01 +01:00
{
2013-06-15 20:28:09 -05:00
// 4-byte sequence
2013-05-20 20:37:13 +02:00
return 4;
}
else
{
2013-06-15 20:28:09 -05:00
// u >= 0xF5
// Restricted (start of 4-, 5- or 6-byte sequence) or invalid UTF-8
2013-05-20 20:37:13 +02:00
return 0;
2012-11-11 09:57:01 +01:00
}
}
} } // namespace Poco::JSON