add json-to-benc parser
This commit is contained in:
parent
c48c9025c8
commit
e607f26893
|
@ -0,0 +1,539 @@
|
|||
/*
|
||||
* Copyright 2001-2004 Unicode, Inc.
|
||||
*
|
||||
* Disclaimer
|
||||
*
|
||||
* This source code is provided as is by Unicode, Inc. No claims are
|
||||
* made as to fitness for any particular purpose. No warranties of any
|
||||
* kind are expressed or implied. The recipient agrees to determine
|
||||
* applicability of information provided. If this file has been
|
||||
* purchased on magnetic or optical media from Unicode, Inc., the
|
||||
* sole remedy for any claim will be exchange of defective media
|
||||
* within 90 days of receipt.
|
||||
*
|
||||
* Limitations on Rights to Redistribute This Code
|
||||
*
|
||||
* Unicode, Inc. hereby grants the right to freely use the information
|
||||
* supplied in this file in the creation of products supporting the
|
||||
* Unicode Standard, and to make copies of this file in any form
|
||||
* for internal or external distribution as long as this notice
|
||||
* remains attached.
|
||||
*/
|
||||
|
||||
/* ---------------------------------------------------------------------
|
||||
|
||||
Conversions between UTF32, UTF-16, and UTF-8. Source code file.
|
||||
Author: Mark E. Davis, 1994.
|
||||
Rev History: Rick McGowan, fixes & updates May 2001.
|
||||
Sept 2001: fixed const & error conditions per
|
||||
mods suggested by S. Parent & A. Lillich.
|
||||
June 2002: Tim Dodd added detection and handling of incomplete
|
||||
source sequences, enhanced error detection, added casts
|
||||
to eliminate compiler warnings.
|
||||
July 2003: slight mods to back out aggressive FFFE detection.
|
||||
Jan 2004: updated switches in from-UTF8 conversions.
|
||||
Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions.
|
||||
|
||||
See the header file "ConvertUTF.h" for complete documentation.
|
||||
|
||||
------------------------------------------------------------------------ */
|
||||
|
||||
|
||||
#include "ConvertUTF.h"
|
||||
#ifdef CVTUTF_DEBUG
|
||||
#include <stdio.h>
|
||||
#endif
|
||||
|
||||
static const int halfShift = 10; /* used for shifting by 10 bits */
|
||||
|
||||
static const UTF32 halfBase = 0x0010000UL;
|
||||
static const UTF32 halfMask = 0x3FFUL;
|
||||
|
||||
#define UNI_SUR_HIGH_START (UTF32)0xD800
|
||||
#define UNI_SUR_HIGH_END (UTF32)0xDBFF
|
||||
#define UNI_SUR_LOW_START (UTF32)0xDC00
|
||||
#define UNI_SUR_LOW_END (UTF32)0xDFFF
|
||||
#define false 0
|
||||
#define true 1
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF32toUTF16 (
|
||||
const UTF32** sourceStart, const UTF32* sourceEnd,
|
||||
UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF32* source = *sourceStart;
|
||||
UTF16* target = *targetStart;
|
||||
while (source < sourceEnd) {
|
||||
UTF32 ch;
|
||||
if (target >= targetEnd) {
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
ch = *source++;
|
||||
if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
|
||||
/* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
|
||||
if (flags == strictConversion) {
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
} else {
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
} else {
|
||||
*target++ = (UTF16)ch; /* normal case */
|
||||
}
|
||||
} else if (ch > UNI_MAX_LEGAL_UTF32) {
|
||||
if (flags == strictConversion) {
|
||||
result = sourceIllegal;
|
||||
} else {
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
} else {
|
||||
/* target is a character in range 0xFFFF - 0x10FFFF. */
|
||||
if (target + 1 >= targetEnd) {
|
||||
--source; /* Back up source pointer! */
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
ch -= halfBase;
|
||||
*target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
|
||||
*target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
|
||||
}
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
return result;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF16toUTF32 (
|
||||
const UTF16** sourceStart, const UTF16* sourceEnd,
|
||||
UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF16* source = *sourceStart;
|
||||
UTF32* target = *targetStart;
|
||||
UTF32 ch, ch2;
|
||||
while (source < sourceEnd) {
|
||||
const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
|
||||
ch = *source++;
|
||||
/* If we have a surrogate pair, convert to UTF32 first. */
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
|
||||
/* If the 16 bits following the high surrogate are in the source buffer... */
|
||||
if (source < sourceEnd) {
|
||||
ch2 = *source;
|
||||
/* If it's a low surrogate, convert to UTF32. */
|
||||
if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
|
||||
ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
|
||||
+ (ch2 - UNI_SUR_LOW_START) + halfBase;
|
||||
++source;
|
||||
} else if (flags == strictConversion) { /* it's an unpaired high surrogate */
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
} else { /* We don't have the 16 bits following the high surrogate. */
|
||||
--source; /* return to the high surrogate */
|
||||
result = sourceExhausted;
|
||||
break;
|
||||
}
|
||||
} else if (flags == strictConversion) {
|
||||
/* UTF-16 surrogate values are illegal in UTF-32 */
|
||||
if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (target >= targetEnd) {
|
||||
source = oldSource; /* Back up source pointer! */
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
*target++ = ch;
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
#ifdef CVTUTF_DEBUG
|
||||
if (result == sourceIllegal) {
|
||||
fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2);
|
||||
fflush(stderr);
|
||||
}
|
||||
#endif
|
||||
return result;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
/*
|
||||
* Index into the table below with the first byte of a UTF-8 sequence to
|
||||
* get the number of trailing bytes that are supposed to follow it.
|
||||
* Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
|
||||
* left as-is for anyone who may want to do such conversion, which was
|
||||
* allowed in earlier algorithms.
|
||||
*/
|
||||
static const char trailingBytesForUTF8[256] = {
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
|
||||
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5
|
||||
};
|
||||
|
||||
/*
|
||||
* Magic values subtracted from a buffer value during UTF8 conversion.
|
||||
* This table contains as many values as there might be trailing bytes
|
||||
* in a UTF-8 sequence.
|
||||
*/
|
||||
static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL,
|
||||
0x03C82080UL, 0xFA082080UL, 0x82082080UL };
|
||||
|
||||
/*
|
||||
* Once the bits are split out into bytes of UTF-8, this is a mask OR-ed
|
||||
* into the first byte, depending on how many bytes follow. There are
|
||||
* as many entries in this table as there are UTF-8 sequence types.
|
||||
* (I.e., one byte sequence, two byte... etc.). Remember that sequencs
|
||||
* for *legal* UTF-8 will be 4 or fewer bytes total.
|
||||
*/
|
||||
static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
/* The interface converts a whole buffer to avoid function-call overhead.
|
||||
* Constants have been gathered. Loops & conditionals have been removed as
|
||||
* much as possible for efficiency, in favor of drop-through switches.
|
||||
* (See "Note A" at the bottom of the file for equivalent code.)
|
||||
* If your compiler supports it, the "isLegalUTF8" call can be turned
|
||||
* into an inline function.
|
||||
*/
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF16toUTF8 (
|
||||
const UTF16** sourceStart, const UTF16* sourceEnd,
|
||||
UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF16* source = *sourceStart;
|
||||
UTF8* target = *targetStart;
|
||||
while (source < sourceEnd) {
|
||||
UTF32 ch;
|
||||
unsigned short bytesToWrite = 0;
|
||||
const UTF32 byteMask = 0xBF;
|
||||
const UTF32 byteMark = 0x80;
|
||||
const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
|
||||
ch = *source++;
|
||||
/* If we have a surrogate pair, convert to UTF32 first. */
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
|
||||
/* If the 16 bits following the high surrogate are in the source buffer... */
|
||||
if (source < sourceEnd) {
|
||||
UTF32 ch2 = *source;
|
||||
/* If it's a low surrogate, convert to UTF32. */
|
||||
if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
|
||||
ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
|
||||
+ (ch2 - UNI_SUR_LOW_START) + halfBase;
|
||||
++source;
|
||||
} else if (flags == strictConversion) { /* it's an unpaired high surrogate */
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
} else { /* We don't have the 16 bits following the high surrogate. */
|
||||
--source; /* return to the high surrogate */
|
||||
result = sourceExhausted;
|
||||
break;
|
||||
}
|
||||
} else if (flags == strictConversion) {
|
||||
/* UTF-16 surrogate values are illegal in UTF-32 */
|
||||
if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
}
|
||||
/* Figure out how many bytes the result will require */
|
||||
if (ch < (UTF32)0x80) { bytesToWrite = 1;
|
||||
} else if (ch < (UTF32)0x800) { bytesToWrite = 2;
|
||||
} else if (ch < (UTF32)0x10000) { bytesToWrite = 3;
|
||||
} else if (ch < (UTF32)0x110000) { bytesToWrite = 4;
|
||||
} else { bytesToWrite = 3;
|
||||
ch = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
|
||||
target += bytesToWrite;
|
||||
if (target > targetEnd) {
|
||||
source = oldSource; /* Back up source pointer! */
|
||||
target -= bytesToWrite; result = targetExhausted; break;
|
||||
}
|
||||
switch (bytesToWrite) { /* note: everything falls through. */
|
||||
case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 1: *--target = (UTF8)(ch | firstByteMark[bytesToWrite]);
|
||||
}
|
||||
target += bytesToWrite;
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
return result;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
/*
|
||||
* Utility routine to tell whether a sequence of bytes is legal UTF-8.
|
||||
* This must be called with the length pre-determined by the first byte.
|
||||
* If not calling this from ConvertUTF8to*, then the length can be set by:
|
||||
* length = trailingBytesForUTF8[*source]+1;
|
||||
* and the sequence is illegal right away if there aren't that many bytes
|
||||
* available.
|
||||
* If presented with a length > 4, this returns false. The Unicode
|
||||
* definition of UTF-8 goes up to 4-byte sequences.
|
||||
*/
|
||||
|
||||
static Boolean isLegalUTF8(const UTF8 *source, int length) {
|
||||
UTF8 a;
|
||||
const UTF8 *srcptr = source+length;
|
||||
switch (length) {
|
||||
default: return false;
|
||||
/* Everything else falls through when "true"... */
|
||||
case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
|
||||
case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
|
||||
case 2: if ((a = (*--srcptr)) > 0xBF) return false;
|
||||
|
||||
switch (*source) {
|
||||
/* no fall-through in this inner switch */
|
||||
case 0xE0: if (a < 0xA0) return false; break;
|
||||
case 0xED: if (a > 0x9F) return false; break;
|
||||
case 0xF0: if (a < 0x90) return false; break;
|
||||
case 0xF4: if (a > 0x8F) return false; break;
|
||||
default: if (a < 0x80) return false;
|
||||
}
|
||||
|
||||
case 1: if (*source >= 0x80 && *source < 0xC2) return false;
|
||||
}
|
||||
if (*source > 0xF4) return false;
|
||||
return true;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
/*
|
||||
* Exported function to return whether a UTF-8 sequence is legal or not.
|
||||
* This is not used here; it's just exported.
|
||||
*/
|
||||
Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) {
|
||||
int length = trailingBytesForUTF8[*source]+1;
|
||||
if (source+length > sourceEnd) {
|
||||
return false;
|
||||
}
|
||||
return isLegalUTF8(source, length);
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF8toUTF16 (
|
||||
const UTF8** sourceStart, const UTF8* sourceEnd,
|
||||
UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF8* source = *sourceStart;
|
||||
UTF16* target = *targetStart;
|
||||
while (source < sourceEnd) {
|
||||
UTF32 ch = 0;
|
||||
unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
|
||||
if (source + extraBytesToRead >= sourceEnd) {
|
||||
result = sourceExhausted; break;
|
||||
}
|
||||
/* Do this check whether lenient or strict */
|
||||
if (! isLegalUTF8(source, extraBytesToRead+1)) {
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
/*
|
||||
* The cases all fall through. See "Note A" below.
|
||||
*/
|
||||
switch (extraBytesToRead) {
|
||||
case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
|
||||
case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
|
||||
case 3: ch += *source++; ch <<= 6;
|
||||
case 2: ch += *source++; ch <<= 6;
|
||||
case 1: ch += *source++; ch <<= 6;
|
||||
case 0: ch += *source++;
|
||||
}
|
||||
ch -= offsetsFromUTF8[extraBytesToRead];
|
||||
|
||||
if (target >= targetEnd) {
|
||||
source -= (extraBytesToRead+1); /* Back up source pointer! */
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
|
||||
/* UTF-16 surrogate values are illegal in UTF-32 */
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
|
||||
if (flags == strictConversion) {
|
||||
source -= (extraBytesToRead+1); /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
} else {
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
} else {
|
||||
*target++ = (UTF16)ch; /* normal case */
|
||||
}
|
||||
} else if (ch > UNI_MAX_UTF16) {
|
||||
if (flags == strictConversion) {
|
||||
result = sourceIllegal;
|
||||
source -= (extraBytesToRead+1); /* return to the start */
|
||||
break; /* Bail out; shouldn't continue */
|
||||
} else {
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
} else {
|
||||
/* target is a character in range 0xFFFF - 0x10FFFF. */
|
||||
if (target + 1 >= targetEnd) {
|
||||
source -= (extraBytesToRead+1); /* Back up source pointer! */
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
ch -= halfBase;
|
||||
*target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
|
||||
*target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
|
||||
}
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
return result;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF32toUTF8 (
|
||||
const UTF32** sourceStart, const UTF32* sourceEnd,
|
||||
UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF32* source = *sourceStart;
|
||||
UTF8* target = *targetStart;
|
||||
while (source < sourceEnd) {
|
||||
UTF32 ch;
|
||||
unsigned short bytesToWrite = 0;
|
||||
const UTF32 byteMask = 0xBF;
|
||||
const UTF32 byteMark = 0x80;
|
||||
ch = *source++;
|
||||
if (flags == strictConversion ) {
|
||||
/* UTF-16 surrogate values are illegal in UTF-32 */
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
|
||||
--source; /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
}
|
||||
/*
|
||||
* Figure out how many bytes the result will require. Turn any
|
||||
* illegally large UTF32 things (> Plane 17) into replacement chars.
|
||||
*/
|
||||
if (ch < (UTF32)0x80) { bytesToWrite = 1;
|
||||
} else if (ch < (UTF32)0x800) { bytesToWrite = 2;
|
||||
} else if (ch < (UTF32)0x10000) { bytesToWrite = 3;
|
||||
} else if (ch <= UNI_MAX_LEGAL_UTF32) { bytesToWrite = 4;
|
||||
} else { bytesToWrite = 3;
|
||||
ch = UNI_REPLACEMENT_CHAR;
|
||||
result = sourceIllegal;
|
||||
}
|
||||
|
||||
target += bytesToWrite;
|
||||
if (target > targetEnd) {
|
||||
--source; /* Back up source pointer! */
|
||||
target -= bytesToWrite; result = targetExhausted; break;
|
||||
}
|
||||
switch (bytesToWrite) { /* note: everything falls through. */
|
||||
case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
|
||||
case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]);
|
||||
}
|
||||
target += bytesToWrite;
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
return result;
|
||||
}
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
ConversionResult ConvertUTF8toUTF32 (
|
||||
const UTF8** sourceStart, const UTF8* sourceEnd,
|
||||
UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
|
||||
ConversionResult result = conversionOK;
|
||||
const UTF8* source = *sourceStart;
|
||||
UTF32* target = *targetStart;
|
||||
while (source < sourceEnd) {
|
||||
UTF32 ch = 0;
|
||||
unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
|
||||
if (source + extraBytesToRead >= sourceEnd) {
|
||||
result = sourceExhausted; break;
|
||||
}
|
||||
/* Do this check whether lenient or strict */
|
||||
if (! isLegalUTF8(source, extraBytesToRead+1)) {
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
}
|
||||
/*
|
||||
* The cases all fall through. See "Note A" below.
|
||||
*/
|
||||
switch (extraBytesToRead) {
|
||||
case 5: ch += *source++; ch <<= 6;
|
||||
case 4: ch += *source++; ch <<= 6;
|
||||
case 3: ch += *source++; ch <<= 6;
|
||||
case 2: ch += *source++; ch <<= 6;
|
||||
case 1: ch += *source++; ch <<= 6;
|
||||
case 0: ch += *source++;
|
||||
}
|
||||
ch -= offsetsFromUTF8[extraBytesToRead];
|
||||
|
||||
if (target >= targetEnd) {
|
||||
source -= (extraBytesToRead+1); /* Back up the source pointer! */
|
||||
result = targetExhausted; break;
|
||||
}
|
||||
if (ch <= UNI_MAX_LEGAL_UTF32) {
|
||||
/*
|
||||
* UTF-16 surrogate values are illegal in UTF-32, and anything
|
||||
* over Plane 17 (> 0x10FFFF) is illegal.
|
||||
*/
|
||||
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
|
||||
if (flags == strictConversion) {
|
||||
source -= (extraBytesToRead+1); /* return to the illegal value itself */
|
||||
result = sourceIllegal;
|
||||
break;
|
||||
} else {
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
} else {
|
||||
*target++ = ch;
|
||||
}
|
||||
} else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
|
||||
result = sourceIllegal;
|
||||
*target++ = UNI_REPLACEMENT_CHAR;
|
||||
}
|
||||
}
|
||||
*sourceStart = source;
|
||||
*targetStart = target;
|
||||
return result;
|
||||
}
|
||||
|
||||
/* ---------------------------------------------------------------------
|
||||
|
||||
Note A.
|
||||
The fall-through switches in UTF-8 reading code save a
|
||||
temp variable, some decrements & conditionals. The switches
|
||||
are equivalent to the following loop:
|
||||
{
|
||||
int tmpBytesToRead = extraBytesToRead+1;
|
||||
do {
|
||||
ch += *source++;
|
||||
--tmpBytesToRead;
|
||||
if (tmpBytesToRead) ch <<= 6;
|
||||
} while (tmpBytesToRead > 0);
|
||||
}
|
||||
In UTF-8 writing code, the switches on "bytesToWrite" are
|
||||
similarly unrolled loops.
|
||||
|
||||
--------------------------------------------------------------------- */
|
|
@ -0,0 +1,155 @@
|
|||
#ifndef CONVERT_UNICODE_H
|
||||
#define CONVERT_UNICODE_H
|
||||
|
||||
/*
|
||||
* Copyright 2001-2004 Unicode, Inc.
|
||||
*
|
||||
* Disclaimer
|
||||
*
|
||||
* This source code is provided as is by Unicode, Inc. No claims are
|
||||
* made as to fitness for any particular purpose. No warranties of any
|
||||
* kind are expressed or implied. The recipient agrees to determine
|
||||
* applicability of information provided. If this file has been
|
||||
* purchased on magnetic or optical media from Unicode, Inc., the
|
||||
* sole remedy for any claim will be exchange of defective media
|
||||
* within 90 days of receipt.
|
||||
*
|
||||
* Limitations on Rights to Redistribute This Code
|
||||
*
|
||||
* Unicode, Inc. hereby grants the right to freely use the information
|
||||
* supplied in this file in the creation of products supporting the
|
||||
* Unicode Standard, and to make copies of this file in any form
|
||||
* for internal or external distribution as long as this notice
|
||||
* remains attached.
|
||||
*/
|
||||
|
||||
/* ---------------------------------------------------------------------
|
||||
|
||||
Conversions between UTF32, UTF-16, and UTF-8. Header file.
|
||||
|
||||
Several funtions are included here, forming a complete set of
|
||||
conversions between the three formats. UTF-7 is not included
|
||||
here, but is handled in a separate source file.
|
||||
|
||||
Each of these routines takes pointers to input buffers and output
|
||||
buffers. The input buffers are const.
|
||||
|
||||
Each routine converts the text between *sourceStart and sourceEnd,
|
||||
putting the result into the buffer between *targetStart and
|
||||
targetEnd. Note: the end pointers are *after* the last item: e.g.
|
||||
*(sourceEnd - 1) is the last item.
|
||||
|
||||
The return result indicates whether the conversion was successful,
|
||||
and if not, whether the problem was in the source or target buffers.
|
||||
(Only the first encountered problem is indicated.)
|
||||
|
||||
After the conversion, *sourceStart and *targetStart are both
|
||||
updated to point to the end of last text successfully converted in
|
||||
the respective buffers.
|
||||
|
||||
Input parameters:
|
||||
sourceStart - pointer to a pointer to the source buffer.
|
||||
The contents of this are modified on return so that
|
||||
it points at the next thing to be converted.
|
||||
targetStart - similarly, pointer to pointer to the target buffer.
|
||||
sourceEnd, targetEnd - respectively pointers to the ends of the
|
||||
two buffers, for overflow checking only.
|
||||
|
||||
These conversion functions take a ConversionFlags argument. When this
|
||||
flag is set to strict, both irregular sequences and isolated surrogates
|
||||
will cause an error. When the flag is set to lenient, both irregular
|
||||
sequences and isolated surrogates are converted.
|
||||
|
||||
Whether the flag is strict or lenient, all illegal sequences will cause
|
||||
an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
|
||||
or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
|
||||
must check for illegal sequences.
|
||||
|
||||
When the flag is set to lenient, characters over 0x10FFFF are converted
|
||||
to the replacement character; otherwise (when the flag is set to strict)
|
||||
they constitute an error.
|
||||
|
||||
Output parameters:
|
||||
The value "sourceIllegal" is returned from some routines if the input
|
||||
sequence is malformed. When "sourceIllegal" is returned, the source
|
||||
value will point to the illegal value that caused the problem. E.g.,
|
||||
in UTF-8 when a sequence is malformed, it points to the start of the
|
||||
malformed sequence.
|
||||
|
||||
Author: Mark E. Davis, 1994.
|
||||
Rev History: Rick McGowan, fixes & updates May 2001.
|
||||
Fixes & updates, Sept 2001.
|
||||
|
||||
------------------------------------------------------------------------ */
|
||||
|
||||
/* ---------------------------------------------------------------------
|
||||
The following 4 definitions are compiler-specific.
|
||||
The C standard does not guarantee that wchar_t has at least
|
||||
16 bits, so wchar_t is no less portable than unsigned short!
|
||||
All should be unsigned values to avoid sign extension during
|
||||
bit mask & shift operations.
|
||||
------------------------------------------------------------------------ */
|
||||
|
||||
|
||||
typedef unsigned int UTF32; /* at least 32 bits */
|
||||
typedef unsigned short UTF16; /* at least 16 bits */
|
||||
typedef unsigned char UTF8; /* typically 8 bits */
|
||||
typedef unsigned char Boolean; /* 0 or 1 */
|
||||
|
||||
/* Some fundamental constants */
|
||||
#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
|
||||
#define UNI_MAX_BMP (UTF32)0x0000FFFF
|
||||
#define UNI_MAX_UTF16 (UTF32)0x0010FFFF
|
||||
#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
|
||||
#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF
|
||||
|
||||
typedef enum {
|
||||
conversionOK, /* conversion successful */
|
||||
sourceExhausted, /* partial character in source, but hit end */
|
||||
targetExhausted, /* insuff. room in target for conversion */
|
||||
sourceIllegal /* source sequence is illegal/malformed */
|
||||
} ConversionResult;
|
||||
|
||||
typedef enum {
|
||||
strictConversion = 0,
|
||||
lenientConversion
|
||||
} ConversionFlags;
|
||||
|
||||
/* This is for C++ and does no harm in C */
|
||||
#ifdef __cplusplus
|
||||
extern "C" {
|
||||
#endif
|
||||
|
||||
ConversionResult ConvertUTF8toUTF16 (
|
||||
const UTF8** sourceStart, const UTF8* sourceEnd,
|
||||
UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
|
||||
|
||||
ConversionResult ConvertUTF16toUTF8 (
|
||||
const UTF16** sourceStart, const UTF16* sourceEnd,
|
||||
UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
|
||||
|
||||
ConversionResult ConvertUTF8toUTF32 (
|
||||
const UTF8** sourceStart, const UTF8* sourceEnd,
|
||||
UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
|
||||
|
||||
ConversionResult ConvertUTF32toUTF8 (
|
||||
const UTF32** sourceStart, const UTF32* sourceEnd,
|
||||
UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
|
||||
|
||||
ConversionResult ConvertUTF16toUTF32 (
|
||||
const UTF16** sourceStart, const UTF16* sourceEnd,
|
||||
UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
|
||||
|
||||
ConversionResult ConvertUTF32toUTF16 (
|
||||
const UTF32** sourceStart, const UTF32* sourceEnd,
|
||||
UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
|
||||
|
||||
Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd);
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
/* --------------------------------------------------------------------- */
|
||||
|
||||
#endif // CONVERT_UNICODE_H
|
|
@ -0,0 +1,905 @@
|
|||
/* JSON_checker.c */
|
||||
|
||||
/* 2007-08-24 */
|
||||
|
||||
/*
|
||||
Copyright (c) 2005 JSON.org
|
||||
|
||||
Permission is hereby granted, free of charge, to any person obtaining a copy
|
||||
of this software and associated documentation files (the "Software"), to deal
|
||||
in the Software without restriction, including without limitation the rights
|
||||
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
||||
copies of the Software, and to permit persons to whom the Software is
|
||||
furnished to do so, subject to the following conditions:
|
||||
|
||||
The above copyright notice and this permission notice shall be included in all
|
||||
copies or substantial portions of the Software.
|
||||
|
||||
The Software shall be used for Good, not Evil.
|
||||
|
||||
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
||||
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
||||
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
||||
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
||||
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
||||
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
||||
SOFTWARE.
|
||||
*/
|
||||
|
||||
/*
|
||||
Callbacks, comments, and Unicode handling by Jean Gressmann (jean@0x42.de), 2007.
|
||||
|
||||
For the added features the license above applies also.
|
||||
*/
|
||||
|
||||
|
||||
|
||||
#include <assert.h>
|
||||
#include <ctype.h>
|
||||
#include <float.h>
|
||||
#include <stddef.h>
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "JSON_checker.h"
|
||||
#include "ConvertUTF.h"
|
||||
|
||||
#if _MSC_VER >= 1400 /* Visual Studio 2005 and up */
|
||||
# pragma warning(disable:4996) // unsecure sscanf
|
||||
#endif
|
||||
|
||||
|
||||
#define true 1
|
||||
#define false 0
|
||||
#define __ -1 /* the universal error code */
|
||||
|
||||
/* values chosen so that the object approx. fits into a page (4K) */
|
||||
#ifndef JSON_CHECKER_STACK_SIZE
|
||||
# define JSON_CHECKER_STACK_SIZE 128
|
||||
#endif
|
||||
|
||||
#ifndef JSON_CHECKER_PARSE_BUFFER_SIZE
|
||||
# define JSON_CHECKER_PARSE_BUFFER_SIZE 3500
|
||||
#endif
|
||||
|
||||
|
||||
typedef struct JSON_checker_struct {
|
||||
JSON_checker_callback* callback;
|
||||
void* ctx;
|
||||
signed char state, before_comment_state, type, escaped, comment, allow_comments;
|
||||
UTF16 utf16_decode_buffer[2];
|
||||
long depth;
|
||||
long top;
|
||||
signed char* stack;
|
||||
long stack_capacity;
|
||||
signed char static_stack[JSON_CHECKER_STACK_SIZE];
|
||||
char* parse_buffer;
|
||||
size_t parse_buffer_capacity;
|
||||
size_t parse_buffer_count;
|
||||
size_t comment_begin_offset;
|
||||
char static_parse_buffer[JSON_CHECKER_PARSE_BUFFER_SIZE];
|
||||
} * JSON_checker;
|
||||
|
||||
#define COUNTOF(x) (sizeof(x)/sizeof(x[0]))
|
||||
|
||||
/*
|
||||
Characters are mapped into these 31 character classes. This allows for
|
||||
a significant reduction in the size of the state transition table.
|
||||
*/
|
||||
|
||||
|
||||
|
||||
enum classes {
|
||||
C_SPACE, /* space */
|
||||
C_WHITE, /* other whitespace */
|
||||
C_LCURB, /* { */
|
||||
C_RCURB, /* } */
|
||||
C_LSQRB, /* [ */
|
||||
C_RSQRB, /* ] */
|
||||
C_COLON, /* : */
|
||||
C_COMMA, /* , */
|
||||
C_QUOTE, /* " */
|
||||
C_BACKS, /* \ */
|
||||
C_SLASH, /* / */
|
||||
C_PLUS, /* + */
|
||||
C_MINUS, /* - */
|
||||
C_POINT, /* . */
|
||||
C_ZERO , /* 0 */
|
||||
C_DIGIT, /* 123456789 */
|
||||
C_LOW_A, /* a */
|
||||
C_LOW_B, /* b */
|
||||
C_LOW_C, /* c */
|
||||
C_LOW_D, /* d */
|
||||
C_LOW_E, /* e */
|
||||
C_LOW_F, /* f */
|
||||
C_LOW_L, /* l */
|
||||
C_LOW_N, /* n */
|
||||
C_LOW_R, /* r */
|
||||
C_LOW_S, /* s */
|
||||
C_LOW_T, /* t */
|
||||
C_LOW_U, /* u */
|
||||
C_ABCDF, /* ABCDF */
|
||||
C_E, /* E */
|
||||
C_ETC, /* everything else */
|
||||
C_STAR, /* * */
|
||||
NR_CLASSES
|
||||
};
|
||||
|
||||
static int ascii_class[128] = {
|
||||
/*
|
||||
This array maps the 128 ASCII characters into character classes.
|
||||
The remaining Unicode characters should be mapped to C_ETC.
|
||||
Non-whitespace control characters are errors.
|
||||
*/
|
||||
__, __, __, __, __, __, __, __,
|
||||
__, C_WHITE, C_WHITE, __, __, C_WHITE, __, __,
|
||||
__, __, __, __, __, __, __, __,
|
||||
__, __, __, __, __, __, __, __,
|
||||
|
||||
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
|
||||
};
|
||||
|
||||
|
||||
/*
|
||||
The state codes.
|
||||
*/
|
||||
enum states {
|
||||
GO, /* start */
|
||||
OK, /* ok */
|
||||
OB, /* object */
|
||||
KE, /* key */
|
||||
CO, /* colon */
|
||||
VA, /* value */
|
||||
AR, /* array */
|
||||
ST, /* string */
|
||||
ES, /* escape */
|
||||
U1, /* u1 */
|
||||
U2, /* u2 */
|
||||
U3, /* u3 */
|
||||
U4, /* u4 */
|
||||
MI, /* minus */
|
||||
ZE, /* zero */
|
||||
IN, /* integer */
|
||||
FR, /* fraction */
|
||||
E1, /* e */
|
||||
E2, /* ex */
|
||||
E3, /* exp */
|
||||
T1, /* tr */
|
||||
T2, /* tru */
|
||||
T3, /* true */
|
||||
F1, /* fa */
|
||||
F2, /* fal */
|
||||
F3, /* fals */
|
||||
F4, /* false */
|
||||
N1, /* nu */
|
||||
N2, /* nul */
|
||||
N3, /* null */
|
||||
C1, /* / */
|
||||
C2, /* / * */
|
||||
C3, /* * */
|
||||
FX, /* *.* *eE* */
|
||||
D1, /* second UTF-16 character decoding started by \ */
|
||||
D2, /* second UTF-16 character proceeded by u */
|
||||
NR_STATES
|
||||
};
|
||||
|
||||
enum actions
|
||||
{
|
||||
CB = -10, /* comment begin */
|
||||
CE = -11, /* comment end */
|
||||
FA = -12, /* false */
|
||||
TR = -13, /* false */
|
||||
NU = -14, /* null */
|
||||
DE = -15, /* double detected by exponent e E */
|
||||
DF = -16, /* double detected by fraction . */
|
||||
SB = -17, /* string begin */
|
||||
MX = -18, /* integer detected by minus */
|
||||
ZX = -19, /* integer detected by zero */
|
||||
IX = -20, /* integer detected by 1-9 */
|
||||
EX = -21, /* next char is escaped */
|
||||
UC = -22, /* Unicode character read */
|
||||
};
|
||||
|
||||
|
||||
static int state_transition_table[NR_STATES][NR_CLASSES] = {
|
||||
/*
|
||||
The state transition table takes the current state and the current symbol,
|
||||
and returns either a new state or an action. An action is represented as a
|
||||
negative number. A JSON text is accepted if at the end of the text the
|
||||
state is OK and if the mode is MODE_DONE.
|
||||
|
||||
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,__,-5,__,__,__,__,__,CB,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*ok OK*/ {OK,OK,__,-8,__,-7,__,-3,__,__,CB,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*object OB*/ {OB,OB,__,-9,__,__,__,__,SB,__,CB,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*key KE*/ {KE,KE,__,__,__,__,__,__,SB,__,CB,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*colon CO*/ {CO,CO,__,__,__,__,-2,__,__,__,CB,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*value VA*/ {VA,VA,-6,__,-5,__,__,__,SB,__,CB,__,MX,__,ZX,IX,__,__,__,__,__,FA,__,NU,__,__,TR,__,__,__,__,__},
|
||||
/*array AR*/ {AR,AR,-6,__,-5,-7,__,__,SB,__,CB,__,MX,__,ZX,IX,__,__,__,__,__,FA,__,NU,__,__,TR,__,__,__,__,__},
|
||||
/*string ST*/ {ST,__,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},
|
||||
/*escape ES*/ {__,__,__,__,__,__,__,__,ST,ST,ST,__,__,__,__,__,__,ST,__,__,__,ST,__,ST,ST,__,ST,U1,__,__,__,__},
|
||||
/*u1 U1*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,U2,U2,U2,U2,U2,U2,U2,U2,__,__,__,__,__,__,U2,U2,__,__},
|
||||
/*u2 U2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,U3,U3,U3,U3,U3,U3,U3,U3,__,__,__,__,__,__,U3,U3,__,__},
|
||||
/*u3 U3*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,U4,U4,U4,U4,U4,U4,U4,U4,__,__,__,__,__,__,U4,U4,__,__},
|
||||
/*u4 U4*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,UC,UC,UC,UC,UC,UC,UC,UC,__,__,__,__,__,__,UC,UC,__,__},
|
||||
/*minus MI*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,ZE,IN,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*zero ZE*/ {OK,OK,__,-8,__,-7,__,-3,__,__,CB,__,__,DF,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*int IN*/ {OK,OK,__,-8,__,-7,__,-3,__,__,CB,__,__,DF,IN,IN,__,__,__,__,DE,__,__,__,__,__,__,__,__,DE,__,__},
|
||||
/*frac FR*/ {OK,OK,__,-8,__,-7,__,-3,__,__,CB,__,__,__,FR,FR,__,__,__,__,E1,__,__,__,__,__,__,__,__,E1,__,__},
|
||||
/*e E1*/ {__,__,__,__,__,__,__,__,__,__,__,E2,E2,__,E3,E3,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*ex E2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,E3,E3,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*exp E3*/ {OK,OK,__,-8,__,-7,__,-3,__,__,__,__,__,__,E3,E3,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*tr T1*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,T2,__,__,__,__,__,__,__},
|
||||
/*tru T2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,T3,__,__,__,__},
|
||||
/*true T3*/ {__,__,__,__,__,__,__,__,__,__,CB,__,__,__,__,__,__,__,__,__,OK,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*fa F1*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,F2,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*fal F2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,F3,__,__,__,__,__,__,__,__,__},
|
||||
/*fals F3*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,F4,__,__,__,__,__,__},
|
||||
/*false F4*/ {__,__,__,__,__,__,__,__,__,__,CB,__,__,__,__,__,__,__,__,__,OK,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*nu N1*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,N2,__,__,__,__},
|
||||
/*nul N2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,N3,__,__,__,__,__,__,__,__,__},
|
||||
/*null N3*/ {__,__,__,__,__,__,__,__,__,__,CB,__,__,__,__,__,__,__,__,__,__,__,OK,__,__,__,__,__,__,__,__,__},
|
||||
/*/ C1*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,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,__,-8,__,-7,__,-3,__,__,__,__,__,__,FR,FR,__,__,__,__,E1,__,__,__,__,__,__,__,__,E1,__,__},
|
||||
/*\ D1*/ {__,__,__,__,__,__,__,__,__,D2,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__},
|
||||
/*\ D2*/ {__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,__,U1,__,__,__,__},
|
||||
};
|
||||
|
||||
|
||||
/*
|
||||
These modes can be pushed on the stack.
|
||||
*/
|
||||
enum modes {
|
||||
MODE_ARRAY = 1,
|
||||
MODE_DONE = 2,
|
||||
MODE_KEY = 3,
|
||||
MODE_OBJECT = 4
|
||||
};
|
||||
|
||||
static int
|
||||
push(JSON_checker jc, int mode)
|
||||
{
|
||||
/*
|
||||
Push a mode onto the stack. Return false if there is overflow.
|
||||
*/
|
||||
jc->top += 1;
|
||||
if (jc->depth < 0) {
|
||||
if (jc->top >= jc->stack_capacity) {
|
||||
jc->stack_capacity *= 2;
|
||||
if (jc->stack == &jc->static_stack[0]) {
|
||||
jc->stack = (signed char*)malloc(jc->stack_capacity * sizeof(jc->static_stack[0]));
|
||||
memcpy(jc->stack, jc->static_stack, sizeof(jc->static_stack));
|
||||
} else {
|
||||
jc->stack = (signed char*)realloc(jc->stack, jc->stack_capacity * sizeof(jc->static_stack[0]));
|
||||
}
|
||||
}
|
||||
} else {
|
||||
if (jc->top >= jc->depth) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
|
||||
jc->stack[jc->top] = mode;
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
static int
|
||||
pop(JSON_checker jc, int mode)
|
||||
{
|
||||
/*
|
||||
Pop the stack, assuring that the current mode matches the expectation.
|
||||
Return false if there is underflow or if the modes mismatch.
|
||||
*/
|
||||
if (jc->top < 0 || jc->stack[jc->top] != mode) {
|
||||
return false;
|
||||
}
|
||||
jc->top -= 1;
|
||||
return true;
|
||||
}
|
||||
|
||||
static void parse_buffer_clear(JSON_checker jc)
|
||||
{
|
||||
jc->parse_buffer_count = 0;
|
||||
jc->parse_buffer[0] = 0;
|
||||
}
|
||||
|
||||
void delete_JSON_checker(JSON_checker jc)
|
||||
{
|
||||
if (jc) {
|
||||
if (jc->stack != &jc->static_stack[0]) {
|
||||
free((void*)jc->stack);
|
||||
}
|
||||
if (jc->parse_buffer != &jc->static_parse_buffer[0]) {
|
||||
free((void*)jc->parse_buffer);
|
||||
}
|
||||
free((void*)jc);
|
||||
}
|
||||
}
|
||||
|
||||
JSON_checker
|
||||
new_JSON_checker(int depth, JSON_checker_callback* callback, void* ctx, int allowComments)
|
||||
{
|
||||
/*
|
||||
new_JSON_checker starts the checking process by constructing a JSON_checker
|
||||
object. It takes a depth parameter that restricts the level of maximum
|
||||
nesting.
|
||||
|
||||
To continue the process, call JSON_checker_char for each character in the
|
||||
JSON text, and then call JSON_checker_done to obtain the final result.
|
||||
These functions are fully reentrant.
|
||||
*/
|
||||
JSON_checker jc = (JSON_checker)malloc(sizeof(struct JSON_checker_struct));
|
||||
memset(jc, 0, sizeof(struct JSON_checker_struct));
|
||||
|
||||
/* We need to be able to push at least one object */
|
||||
if (depth == 0) {
|
||||
depth = 1;
|
||||
}
|
||||
|
||||
jc->state = GO;
|
||||
jc->top = -1;
|
||||
if (depth > 0) {
|
||||
jc->stack_capacity = depth;
|
||||
jc->depth = depth;
|
||||
if (depth <= (int)COUNTOF(jc->static_stack)) {
|
||||
jc->stack = &jc->static_stack[0];
|
||||
} else {
|
||||
jc->stack = (signed char*)malloc(jc->stack_capacity * sizeof(jc->static_stack[0]));
|
||||
}
|
||||
} else {
|
||||
jc->stack_capacity = COUNTOF(jc->static_stack);
|
||||
jc->depth = -1;
|
||||
jc->stack = &jc->static_stack[0];
|
||||
}
|
||||
push(jc, MODE_DONE);
|
||||
jc->parse_buffer = &jc->static_parse_buffer[0];
|
||||
jc->parse_buffer_capacity = COUNTOF(jc->static_parse_buffer);
|
||||
parse_buffer_clear(jc);
|
||||
jc->callback = callback;
|
||||
jc->ctx = ctx;
|
||||
jc->allow_comments = allowComments != 0;
|
||||
return jc;
|
||||
}
|
||||
|
||||
static void parse_buffer_push_back_char(JSON_checker jc, char c)
|
||||
{
|
||||
if (jc->parse_buffer_count + 1 >= jc->parse_buffer_capacity) {
|
||||
jc->parse_buffer_capacity *= 2;
|
||||
if (jc->parse_buffer == &jc->static_parse_buffer[0]) {
|
||||
jc->parse_buffer = (char*)malloc(jc->parse_buffer_capacity * sizeof(jc->parse_buffer[0]));
|
||||
} else {
|
||||
jc->parse_buffer = (char*)realloc(jc->parse_buffer, jc->parse_buffer_capacity * sizeof(jc->parse_buffer[0]));
|
||||
}
|
||||
}
|
||||
|
||||
jc->parse_buffer[jc->parse_buffer_count++] = c;
|
||||
jc->parse_buffer[jc->parse_buffer_count] = 0;
|
||||
}
|
||||
|
||||
static void parse_buffer_pop_back_chars(JSON_checker jc, size_t count)
|
||||
{
|
||||
assert(jc->parse_buffer_count >= count);
|
||||
jc->parse_buffer_count -= count;
|
||||
jc->parse_buffer[jc->parse_buffer_count] = 0;
|
||||
}
|
||||
|
||||
|
||||
|
||||
static int parse_parse_buffer(JSON_checker jc)
|
||||
{
|
||||
if (jc->callback) {
|
||||
int result = 1;
|
||||
JSON_value value, *arg = NULL;
|
||||
|
||||
if (jc->type != JSON_T_NONE) {
|
||||
assert(
|
||||
jc->type == JSON_T_NULL ||
|
||||
jc->type == JSON_T_FALSE ||
|
||||
jc->type == JSON_T_TRUE ||
|
||||
jc->type == JSON_T_FLOAT ||
|
||||
jc->type == JSON_T_INTEGER ||
|
||||
jc->type == JSON_T_STRING);
|
||||
|
||||
|
||||
switch(jc->type) {
|
||||
case JSON_T_FLOAT:
|
||||
arg = &value;
|
||||
result = sscanf(jc->parse_buffer, "%Lf", &value.float_value);
|
||||
break;
|
||||
case JSON_T_INTEGER:
|
||||
arg = &value;
|
||||
result = sscanf(jc->parse_buffer, JSON_CHECKER_INTEGER_SSCANF_TOKEN, &value.integer_value);
|
||||
break;
|
||||
case JSON_T_STRING:
|
||||
arg = &value;
|
||||
value.string_value = jc->parse_buffer;
|
||||
value.string_length = jc->parse_buffer_count;
|
||||
break;
|
||||
}
|
||||
|
||||
if (!(*jc->callback)(jc->ctx, jc->type, arg)) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
jc->parse_buffer_count = 0;
|
||||
jc->parse_buffer[0] = 0;
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
static int decode_unicode_char(JSON_checker jc)
|
||||
{
|
||||
const unsigned chars = jc->utf16_decode_buffer[0] ? 2 : 1;
|
||||
int i;
|
||||
UTF16 *uc = chars == 1 ? &jc->utf16_decode_buffer[0] : &jc->utf16_decode_buffer[1];
|
||||
UTF16 x;
|
||||
char* p;
|
||||
|
||||
assert(jc->parse_buffer_count >= 6);
|
||||
|
||||
p = &jc->parse_buffer[jc->parse_buffer_count - 4];
|
||||
|
||||
for (i = 0; i < 4; ++i, ++p) {
|
||||
x = *p;
|
||||
|
||||
if (x >= 'a') {
|
||||
x -= ('a' - 10);
|
||||
} else if (x >= 'A') {
|
||||
x -= ('A' - 10);
|
||||
} else {
|
||||
x &= ~((UTF16) 0x30);
|
||||
}
|
||||
|
||||
assert(x < 16);
|
||||
|
||||
*uc |= x << ((3u - i) << 2);
|
||||
}
|
||||
|
||||
/* clear UTF-16 char form buffer */
|
||||
jc->parse_buffer_count -= 6;
|
||||
jc->parse_buffer[jc->parse_buffer_count] = 0;
|
||||
|
||||
/* attempt decoding ... */
|
||||
{
|
||||
UTF8* dec_start = (UTF8*)&jc->parse_buffer[jc->parse_buffer_count];
|
||||
UTF8* dec_start_dup = dec_start;
|
||||
UTF8* dec_end = dec_start + 6;
|
||||
|
||||
const UTF16* enc_start = &jc->utf16_decode_buffer[0];
|
||||
const UTF16* enc_end = enc_start + chars;
|
||||
|
||||
const ConversionResult result = ConvertUTF16toUTF8(
|
||||
&enc_start, enc_end, &dec_start, dec_end, strictConversion);
|
||||
|
||||
const size_t new_chars = dec_start - dec_start_dup;
|
||||
|
||||
/* was it a surrogate UTF-16 char? */
|
||||
if (chars == 1 && result == sourceExhausted) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (result != conversionOK) {
|
||||
return false;
|
||||
}
|
||||
|
||||
/* NOTE: clear decode buffer to resume string reading,
|
||||
otherwise we continue to read UTF-16 */
|
||||
jc->utf16_decode_buffer[0] = 0;
|
||||
|
||||
assert(new_chars <= 6);
|
||||
|
||||
jc->parse_buffer_count += new_chars;
|
||||
jc->parse_buffer[jc->parse_buffer_count] = 0;
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
int
|
||||
JSON_checker_char(JSON_checker jc, int next_char)
|
||||
{
|
||||
/*
|
||||
After calling new_JSON_checker, call this function for each character (or
|
||||
partial character) in your JSON text. It can accept UTF-8, UTF-16, or
|
||||
UTF-32. It returns true if things are looking ok so far. If it rejects the
|
||||
text, it deletes the JSON_checker object and returns false.
|
||||
*/
|
||||
int next_class, next_state;
|
||||
|
||||
/*
|
||||
Determine the character's class.
|
||||
*/
|
||||
if (next_char < 0) {
|
||||
return false;
|
||||
}
|
||||
if (next_char >= 128) {
|
||||
next_class = C_ETC;
|
||||
} else {
|
||||
next_class = ascii_class[next_char];
|
||||
if (next_class <= __) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
|
||||
if (jc->escaped) {
|
||||
jc->escaped = 0;
|
||||
/* remove the backslash */
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
switch(next_char) {
|
||||
case 'b':
|
||||
parse_buffer_push_back_char(jc, '\b');
|
||||
break;
|
||||
case 'f':
|
||||
parse_buffer_push_back_char(jc, '\f');
|
||||
break;
|
||||
case 'n':
|
||||
parse_buffer_push_back_char(jc, '\n');
|
||||
break;
|
||||
case 'r':
|
||||
parse_buffer_push_back_char(jc, '\r');
|
||||
break;
|
||||
case 't':
|
||||
parse_buffer_push_back_char(jc, '\t');
|
||||
break;
|
||||
case '"':
|
||||
parse_buffer_push_back_char(jc, '"');
|
||||
break;
|
||||
case '\\':
|
||||
parse_buffer_push_back_char(jc, '\\');
|
||||
break;
|
||||
case '/':
|
||||
parse_buffer_push_back_char(jc, '/');
|
||||
break;
|
||||
case 'u':
|
||||
parse_buffer_push_back_char(jc, '\\');
|
||||
parse_buffer_push_back_char(jc, 'u');
|
||||
break;
|
||||
default:
|
||||
return false;
|
||||
}
|
||||
} else if (!jc->comment) {
|
||||
if (jc->type != JSON_T_NONE || !(next_class == C_SPACE || next_class == C_WHITE) /* non-white-space */) {
|
||||
parse_buffer_push_back_char(jc, (char)next_char);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
|
||||
/*
|
||||
Get the next state from the state transition table.
|
||||
*/
|
||||
next_state = state_transition_table[jc->state][next_class];
|
||||
if (next_state >= 0) {
|
||||
/*
|
||||
Change the state.
|
||||
*/
|
||||
jc->state = next_state;
|
||||
} else {
|
||||
/*
|
||||
Or perform one of the actions.
|
||||
*/
|
||||
switch (next_state) {
|
||||
/* Unicode character */
|
||||
case UC:
|
||||
if(!decode_unicode_char(jc)) {
|
||||
return false;
|
||||
}
|
||||
/* check if we need to read a second UTF-16 char */
|
||||
if (jc->utf16_decode_buffer[0]) {
|
||||
jc->state = D1;
|
||||
} else {
|
||||
jc->state = ST;
|
||||
}
|
||||
break;
|
||||
/* escaped char */
|
||||
case EX:
|
||||
jc->escaped = 1;
|
||||
jc->state = ES;
|
||||
break;
|
||||
/* integer detected by minus */
|
||||
case MX:
|
||||
jc->type = JSON_T_INTEGER;
|
||||
jc->state = MI;
|
||||
break;
|
||||
/* integer detected by zero */
|
||||
case ZX:
|
||||
jc->type = JSON_T_INTEGER;
|
||||
jc->state = ZE;
|
||||
break;
|
||||
/* integer detected by 1-9 */
|
||||
case IX:
|
||||
jc->type = JSON_T_INTEGER;
|
||||
jc->state = IN;
|
||||
break;
|
||||
|
||||
/* floating point number detected by exponent*/
|
||||
case DE:
|
||||
assert(jc->type != JSON_T_FALSE);
|
||||
assert(jc->type != JSON_T_TRUE);
|
||||
assert(jc->type != JSON_T_NULL);
|
||||
assert(jc->type != JSON_T_STRING);
|
||||
jc->type = JSON_T_FLOAT;
|
||||
jc->state = E1;
|
||||
break;
|
||||
|
||||
/* floating point number detected by fraction */
|
||||
case DF:
|
||||
assert(jc->type != JSON_T_FALSE);
|
||||
assert(jc->type != JSON_T_TRUE);
|
||||
assert(jc->type != JSON_T_NULL);
|
||||
assert(jc->type != JSON_T_STRING);
|
||||
jc->type = JSON_T_FLOAT;
|
||||
jc->state = FX;
|
||||
break;
|
||||
/* string begin " */
|
||||
case SB:
|
||||
parse_buffer_clear(jc);
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->type = JSON_T_STRING;
|
||||
jc->state = ST;
|
||||
break;
|
||||
|
||||
/* n */
|
||||
case NU:
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->type = JSON_T_NULL;
|
||||
jc->state = N1;
|
||||
break;
|
||||
/* f */
|
||||
case FA:
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->type = JSON_T_FALSE;
|
||||
jc->state = F1;
|
||||
break;
|
||||
/* t */
|
||||
case TR:
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->type = JSON_T_TRUE;
|
||||
jc->state = T1;
|
||||
break;
|
||||
|
||||
/* closing comment */
|
||||
case CE:
|
||||
jc->comment = 0;
|
||||
assert(jc->parse_buffer_count == 0);
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->state = jc->before_comment_state;
|
||||
break;
|
||||
|
||||
/* opening comment */
|
||||
case CB:
|
||||
if (!jc->allow_comments) {
|
||||
return false;
|
||||
}
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
assert(jc->parse_buffer_count == 0);
|
||||
assert(jc->type != JSON_T_STRING);
|
||||
switch (jc->stack[jc->top]) {
|
||||
case MODE_ARRAY:
|
||||
case MODE_OBJECT:
|
||||
switch(jc->state) {
|
||||
case VA:
|
||||
case AR:
|
||||
jc->before_comment_state = jc->state;
|
||||
break;
|
||||
default:
|
||||
jc->before_comment_state = OK;
|
||||
break;
|
||||
}
|
||||
break;
|
||||
default:
|
||||
jc->before_comment_state = jc->state;
|
||||
break;
|
||||
}
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = C1;
|
||||
jc->comment = 1;
|
||||
break;
|
||||
/* empty } */
|
||||
case -9:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
if (jc->callback && !(*jc->callback)(jc->ctx, JSON_T_OBJECT_END, NULL)) {
|
||||
return false;
|
||||
}
|
||||
if (!pop(jc, MODE_KEY)) {
|
||||
return false;
|
||||
}
|
||||
jc->state = OK;
|
||||
break;
|
||||
|
||||
/* } */ case -8:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
if (jc->callback && !(*jc->callback)(jc->ctx, JSON_T_OBJECT_END, NULL)) {
|
||||
return false;
|
||||
}
|
||||
if (!pop(jc, MODE_OBJECT)) {
|
||||
return false;
|
||||
}
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = OK;
|
||||
break;
|
||||
|
||||
/* ] */ case -7:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
if (jc->callback && !(*jc->callback)(jc->ctx, JSON_T_ARRAY_END, NULL)) {
|
||||
return false;
|
||||
}
|
||||
if (!pop(jc, MODE_ARRAY)) {
|
||||
return false;
|
||||
}
|
||||
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = OK;
|
||||
break;
|
||||
|
||||
/* { */ case -6:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (jc->callback && !(*jc->callback)(jc->ctx, JSON_T_OBJECT_BEGIN, NULL)) {
|
||||
return false;
|
||||
}
|
||||
if (!push(jc, MODE_KEY)) {
|
||||
return false;
|
||||
}
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->state = OB;
|
||||
break;
|
||||
|
||||
/* [ */ case -5:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (jc->callback && !(*jc->callback)(jc->ctx, JSON_T_ARRAY_BEGIN, NULL)) {
|
||||
return false;
|
||||
}
|
||||
if (!push(jc, MODE_ARRAY)) {
|
||||
return false;
|
||||
}
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->state = AR;
|
||||
break;
|
||||
|
||||
/* string end " */ case -4:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
switch (jc->stack[jc->top]) {
|
||||
case MODE_KEY:
|
||||
assert(jc->type == JSON_T_STRING);
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = CO;
|
||||
|
||||
if (jc->callback) {
|
||||
JSON_value value;
|
||||
value.string_value = jc->parse_buffer;
|
||||
value.string_length = jc->parse_buffer_count;
|
||||
if (!(*jc->callback)(jc->ctx, JSON_T_KEY, &value)) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
parse_buffer_clear(jc);
|
||||
break;
|
||||
case MODE_ARRAY:
|
||||
case MODE_OBJECT:
|
||||
assert(jc->type == JSON_T_STRING);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = OK;
|
||||
break;
|
||||
default:
|
||||
return false;
|
||||
}
|
||||
break;
|
||||
|
||||
/* , */ case -3:
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!parse_parse_buffer(jc)) {
|
||||
return false;
|
||||
}
|
||||
switch (jc->stack[jc->top]) {
|
||||
case MODE_OBJECT:
|
||||
/*
|
||||
A comma causes a flip from object mode to key mode.
|
||||
*/
|
||||
if (!pop(jc, MODE_OBJECT) || !push(jc, MODE_KEY)) {
|
||||
return false;
|
||||
}
|
||||
assert(jc->type != JSON_T_STRING);
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = KE;
|
||||
break;
|
||||
case MODE_ARRAY:
|
||||
assert(jc->type != JSON_T_STRING);
|
||||
jc->type = JSON_T_NONE;
|
||||
jc->state = VA;
|
||||
break;
|
||||
default:
|
||||
return false;
|
||||
}
|
||||
break;
|
||||
|
||||
/* : */ case -2:
|
||||
/*
|
||||
A colon causes a flip from key mode to object mode.
|
||||
*/
|
||||
parse_buffer_pop_back_chars(jc, 1);
|
||||
if (!pop(jc, MODE_KEY) || !push(jc, MODE_OBJECT)) {
|
||||
return false;
|
||||
}
|
||||
assert(jc->type == JSON_T_NONE);
|
||||
jc->state = VA;
|
||||
break;
|
||||
/*
|
||||
Bad action.
|
||||
*/
|
||||
default:
|
||||
return false;
|
||||
}
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
int
|
||||
JSON_checker_done(JSON_checker jc)
|
||||
{
|
||||
const int result = jc->state == OK && pop(jc, MODE_DONE);
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
|
||||
int JSON_checker_is_legal_white_space_string(const char* s)
|
||||
{
|
||||
int c, char_class;
|
||||
|
||||
if (s == NULL) {
|
||||
return false;
|
||||
}
|
||||
|
||||
for (; *s; ++s) {
|
||||
c = *s;
|
||||
|
||||
if (c < 0 || c >= 128) {
|
||||
return false;
|
||||
}
|
||||
|
||||
char_class = ascii_class[c];
|
||||
|
||||
if (char_class != C_SPACE && char_class != C_WHITE) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
|
@ -0,0 +1,117 @@
|
|||
#ifndef JSON_CHECKER_H
|
||||
#define JSON_CHECKER_H
|
||||
|
||||
/* JSON_checker.h */
|
||||
|
||||
|
||||
#include <stddef.h>
|
||||
|
||||
/* Windows DLL stuff */
|
||||
#ifdef _WIN32
|
||||
# ifdef JSON_CHECKER_DLL_EXPORTS
|
||||
# define JSON_CHECKER_DLL_API __declspec(dllexport)
|
||||
# else
|
||||
# define JSON_CHECKER_DLL_API __declspec(dllimport)
|
||||
# endif
|
||||
#else
|
||||
# define JSON_CHECKER_DLL_API
|
||||
#endif
|
||||
|
||||
/* Determine the integer type use to parse non-floating point numbers */
|
||||
#if __STDC_VERSION__ >= 199901L || HAVE_LONG_LONG == 1
|
||||
typedef long long JSON_int_t;
|
||||
#define JSON_CHECKER_INTEGER_SSCANF_TOKEN "%lld"
|
||||
#define JSON_CHECKER_INTEGER_SPRINTF_TOKEN "%lld"
|
||||
#else
|
||||
typedef long JSON_int_t;
|
||||
#define JSON_CHECKER_INTEGER_SSCANF_TOKEN "%ld"
|
||||
#define JSON_CHECKER_INTEGER_SPRINTF_TOKEN "%ld"
|
||||
#endif
|
||||
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C" {
|
||||
#endif
|
||||
|
||||
typedef enum
|
||||
{
|
||||
JSON_T_NONE = 0,
|
||||
JSON_T_ARRAY_BEGIN,
|
||||
JSON_T_ARRAY_END,
|
||||
JSON_T_OBJECT_BEGIN,
|
||||
JSON_T_OBJECT_END,
|
||||
JSON_T_INTEGER,
|
||||
JSON_T_FLOAT,
|
||||
JSON_T_NULL,
|
||||
JSON_T_TRUE,
|
||||
JSON_T_FALSE,
|
||||
JSON_T_STRING,
|
||||
JSON_T_KEY,
|
||||
} JSON_type;
|
||||
|
||||
typedef struct JSON_value_struct {
|
||||
union {
|
||||
JSON_int_t integer_value;
|
||||
|
||||
long double float_value;
|
||||
|
||||
struct {
|
||||
const char* string_value;
|
||||
size_t string_length;
|
||||
};
|
||||
};
|
||||
} JSON_value;
|
||||
|
||||
/*! \brief JSON parser callback
|
||||
|
||||
\param ctx The pointer passed to new_JSON_checker.
|
||||
\param type An element of JSON_type but not JSON_T_NONE.
|
||||
\param value A representation of the parsed value. This parameter is NULL for
|
||||
JSON_T_ARRAY_BEGIN, JSON_T_ARRAY_END, JSON_T_OBJECT_BEGIN, JSON_T_OBJECT_END,
|
||||
JSON_T_NULL, JSON_T_TRUE, and SON_T_FALSE.
|
||||
|
||||
\return Non-zero if parsing should continue, else zero.
|
||||
*/
|
||||
typedef int (JSON_checker_callback)(void* ctx, int type, const struct JSON_value_struct* value);
|
||||
|
||||
/*! \brief Create a JSON parser object
|
||||
|
||||
\param depth If negative, the parser can parse arbitrary levels of JSON, otherwise
|
||||
the depth is the limit
|
||||
\param Pointer to a callback. This parameter may be NULL.
|
||||
\param Callback context. This parameter may be NULL.
|
||||
|
||||
\return The parser object.
|
||||
*/
|
||||
JSON_CHECKER_DLL_API extern struct JSON_checker_struct* new_JSON_checker(int depth, JSON_checker_callback*, void* ctx, int allowComments);
|
||||
|
||||
/*! \brief Destroy a previously created JSON parser object. */
|
||||
JSON_CHECKER_DLL_API extern void delete_JSON_checker(struct JSON_checker_struct* jc);
|
||||
|
||||
/*! \brief Parse a character.
|
||||
|
||||
\return Non-zero, if all characters passed to this function are part of are valid JSON.
|
||||
*/
|
||||
JSON_CHECKER_DLL_API extern int JSON_checker_char(struct JSON_checker_struct* jc, int next_char);
|
||||
|
||||
/*! \brief Finalize parsing.
|
||||
|
||||
Call this method once after all input characters have been consumed.
|
||||
|
||||
\return Non-zero, if all parsed characters are valid JSON, zero otherwise.
|
||||
*/
|
||||
JSON_CHECKER_DLL_API extern int JSON_checker_done(struct JSON_checker_struct* jc);
|
||||
|
||||
/*! \brief Determine if a given string is valid JSON white space
|
||||
|
||||
\return Non-zero if the string is valid, zero otherwise.
|
||||
*/
|
||||
JSON_CHECKER_DLL_API extern int JSON_checker_is_legal_white_space_string(const char* s);
|
||||
|
||||
|
||||
#ifdef __cplusplus
|
||||
}
|
||||
#endif
|
||||
|
||||
|
||||
#endif /* JSON_CHECKER_H */
|
|
@ -4,6 +4,8 @@ AM_CFLAGS = $(OPENSSL_CFLAGS) $(LIBCURL_CFLAGS) $(PTHREAD_CFLAGS)
|
|||
noinst_LIBRARIES = libtransmission.a
|
||||
|
||||
libtransmission_a_SOURCES = \
|
||||
ConvertUTF.c \
|
||||
JSON_checker.c \
|
||||
bencode.c \
|
||||
blocklist.c \
|
||||
clients.c \
|
||||
|
@ -15,6 +17,7 @@ libtransmission_a_SOURCES = \
|
|||
handshake.c \
|
||||
inout.c \
|
||||
ipcparse.c \
|
||||
json.c \
|
||||
list.c \
|
||||
makemeta.c \
|
||||
metainfo.c \
|
||||
|
@ -41,6 +44,8 @@ libtransmission_a_SOURCES = \
|
|||
web.c
|
||||
|
||||
noinst_HEADERS = \
|
||||
ConvertUTF.h \
|
||||
JSON_checker.h \
|
||||
bencode.h \
|
||||
blocklist.h \
|
||||
clients.h \
|
||||
|
@ -85,6 +90,7 @@ TESTS = \
|
|||
blocklist-test \
|
||||
bencode-test \
|
||||
clients-test \
|
||||
json-test \
|
||||
test-fastset \
|
||||
test-peer-id
|
||||
|
||||
|
@ -109,6 +115,8 @@ blocklist_test_SOURCES = blocklist-test.c
|
|||
blocklist_test_LDADD = $(APPS_LDADD)
|
||||
clients_test_SOURCES = clients-test.c
|
||||
clients_test_LDADD = $(APPS_LDADD)
|
||||
json_test_SOURCES = json-test.c
|
||||
json_test_LDADD = $(APPS_LDADD)
|
||||
test_fastset_SOURCES = test-fastset.c
|
||||
test_fastset_LDADD = $(APPS_LDADD)
|
||||
test_peer_id_SOURCES = test-peer-id.c
|
||||
|
|
|
@ -591,10 +591,12 @@ tr_bencDictAdd( tr_benc * dict, const char * key )
|
|||
tr_benc * keyval, * itemval;
|
||||
|
||||
assert( tr_bencIsDict( dict ) );
|
||||
if( dict->val.l.count + 2 > dict->val.l.alloc )
|
||||
makeroom( dict, 2 );
|
||||
assert( dict->val.l.count + 2 <= dict->val.l.alloc );
|
||||
|
||||
keyval = dict->val.l.vals + dict->val.l.count++;
|
||||
tr_bencInitStr( keyval, (char*)key, -1, 1 );
|
||||
tr_bencInitStrDup( keyval, key );
|
||||
|
||||
itemval = dict->val.l.vals + dict->val.l.count++;
|
||||
tr_bencInit( itemval, TYPE_INT );
|
||||
|
|
|
@ -56,6 +56,11 @@ typedef struct tr_benc
|
|||
/* backwards compatability */
|
||||
typedef tr_benc benc_val_t;
|
||||
|
||||
int tr_jsonParse( const void * buf,
|
||||
const void * bufend,
|
||||
tr_benc * setme_benc,
|
||||
const uint8_t ** setme_end );
|
||||
|
||||
int tr_bencParse( const void * buf,
|
||||
const void * bufend,
|
||||
tr_benc * setme_benc,
|
||||
|
@ -102,7 +107,6 @@ tr_benc * tr_bencListAddInt( tr_benc * list, int64_t val );
|
|||
tr_benc * tr_bencListAddStr( tr_benc * list, const char * val );
|
||||
tr_benc * tr_bencListAddList( tr_benc * list, int reserveCount );
|
||||
tr_benc * tr_bencListAddDict( tr_benc * list, int reserveCount );
|
||||
/* note: key must not be freed or modified while val is in use */
|
||||
tr_benc * tr_bencDictAdd( tr_benc * dict, const char * key );
|
||||
tr_benc * tr_bencDictAddDouble( tr_benc * dict, const char * key, double d );
|
||||
tr_benc * tr_bencDictAddInt( tr_benc * dict, const char * key, int64_t val );
|
||||
|
|
|
@ -0,0 +1,78 @@
|
|||
#include <stdio.h>
|
||||
#include "transmission.h"
|
||||
#include "bencode.h"
|
||||
#include "utils.h" /* tr_free */
|
||||
|
||||
#define VERBOSE 0
|
||||
|
||||
int test = 0;
|
||||
|
||||
#define check(A) { \
|
||||
++test; \
|
||||
if (A) { \
|
||||
if( VERBOSE ) \
|
||||
fprintf( stderr, "PASS test #%d (%s, %d)\n", test, __FILE__, __LINE__ ); \
|
||||
} else { \
|
||||
if( VERBOSE ) \
|
||||
fprintf( stderr, "FAIL test #%d (%s, %d)\n", test, __FILE__, __LINE__ ); \
|
||||
return test; \
|
||||
} \
|
||||
}
|
||||
|
||||
static int
|
||||
test1( void )
|
||||
{
|
||||
const char * in =
|
||||
"{\n"
|
||||
" \"headers\": {\n"
|
||||
" \"type\": \"request\",\n"
|
||||
" \"tag\": 666\n"
|
||||
" },\n"
|
||||
" \"body\": {\n"
|
||||
" \"name\": \"torrent-info\",\n"
|
||||
" \"arguments\": {\n"
|
||||
" \"ids\": [ 7, 10 ]\n"
|
||||
" }\n"
|
||||
" }\n"
|
||||
"}\n";
|
||||
tr_benc top, *headers, *body, *args, *ids;
|
||||
const char * str;
|
||||
int64_t i;
|
||||
const uint8_t * end = NULL;
|
||||
const int err = tr_jsonParse( in, in+strlen(in), &top, &end );
|
||||
|
||||
check( !err );
|
||||
check( tr_bencIsDict( &top ) );
|
||||
check(( headers = tr_bencDictFind( &top, "headers" )));
|
||||
check( tr_bencIsDict( headers ) );
|
||||
check( tr_bencDictFindStr( headers, "type", &str ) );
|
||||
check( !strcmp( str, "request" ) );
|
||||
check( tr_bencDictFindInt( headers, "tag", &i ) );
|
||||
check( i == 666 );
|
||||
check(( body = tr_bencDictFind( &top, "body" )));
|
||||
check( tr_bencDictFindStr( body, "name", &str ) );
|
||||
check( !strcmp( str, "torrent-info" ) );
|
||||
check(( args = tr_bencDictFind( body, "arguments" )));
|
||||
check( tr_bencIsDict( args ) );
|
||||
check(( ids = tr_bencDictFind( args, "ids" )));
|
||||
check( tr_bencIsList( ids ) );
|
||||
check( tr_bencListSize( ids ) == 2 );
|
||||
check( tr_bencGetInt( tr_bencListChild( ids, 0 ), &i ) );
|
||||
check( i == 7 );
|
||||
check( tr_bencGetInt( tr_bencListChild( ids, 1 ), &i ) );
|
||||
check( i == 10 );
|
||||
|
||||
tr_bencFree( &top );
|
||||
return 0;
|
||||
}
|
||||
|
||||
int
|
||||
main( void )
|
||||
{
|
||||
int i;
|
||||
|
||||
if(( i = test1( )))
|
||||
return i;
|
||||
|
||||
return 0;
|
||||
}
|
|
@ -0,0 +1,146 @@
|
|||
/*
|
||||
* This file Copyright (C) 2008 Charles Kerr <charles@rebelbase.com>
|
||||
*
|
||||
* This file is licensed by the GPL version 2. Works owned by the
|
||||
* Transmission project are granted a special exemption to clause 2(b)
|
||||
* so that the bulk of its code can remain under the MIT license.
|
||||
* This exemption does not extend to derived works not owned by
|
||||
* the Transmission project.
|
||||
*
|
||||
* $Id:$
|
||||
*/
|
||||
|
||||
#include <assert.h>
|
||||
//#include <string.h> /* memcpy, memcmp, strstr */
|
||||
//#include <stdlib.h> /* qsort */
|
||||
#include <stdio.h> /* printf */
|
||||
//#include <limits.h> /* INT_MAX */
|
||||
//
|
||||
#include "JSON_checker.h"
|
||||
|
||||
#include "transmission.h"
|
||||
#include "bencode.h"
|
||||
#include "ptrarray.h"
|
||||
#include "utils.h"
|
||||
|
||||
struct json_benc_data
|
||||
{
|
||||
tr_benc * top;
|
||||
tr_ptrArray * stack;
|
||||
char * key;
|
||||
};
|
||||
|
||||
static tr_benc*
|
||||
getNode( struct json_benc_data * data )
|
||||
{
|
||||
tr_benc * parent;
|
||||
tr_benc * node = NULL;
|
||||
|
||||
if( tr_ptrArrayEmpty( data->stack ) )
|
||||
parent = NULL;
|
||||
else
|
||||
parent = tr_ptrArrayBack( data->stack );
|
||||
|
||||
if( !parent )
|
||||
node = data->top;
|
||||
else if( tr_bencIsList( parent ) )
|
||||
node = tr_bencListAdd( parent );
|
||||
else if( tr_bencIsDict( parent ) && data->key ) {
|
||||
node = tr_bencDictAdd( parent, data->key );
|
||||
tr_free( data->key );
|
||||
data->key = NULL;
|
||||
}
|
||||
|
||||
return node;
|
||||
}
|
||||
|
||||
static int
|
||||
callback( void * vdata, int type, const struct JSON_value_struct * value )
|
||||
{
|
||||
struct json_benc_data * data = vdata;
|
||||
tr_benc * node;
|
||||
|
||||
switch( type )
|
||||
{
|
||||
case JSON_T_ARRAY_BEGIN:
|
||||
node = getNode( data );
|
||||
tr_bencInitList( node, 0 );
|
||||
tr_ptrArrayAppend( data->stack, node );
|
||||
break;
|
||||
|
||||
case JSON_T_ARRAY_END:
|
||||
tr_ptrArrayPop( data->stack );
|
||||
break;
|
||||
|
||||
case JSON_T_OBJECT_BEGIN:
|
||||
node = getNode( data );
|
||||
tr_bencInitDict( node, 0 );
|
||||
tr_ptrArrayAppend( data->stack, node );
|
||||
break;
|
||||
|
||||
case JSON_T_OBJECT_END:
|
||||
tr_ptrArrayPop( data->stack );
|
||||
break;
|
||||
|
||||
case JSON_T_FLOAT: {
|
||||
char buf[128];
|
||||
snprintf( buf, sizeof( buf ), "%f", (double)value->float_value );
|
||||
tr_bencInitStrDup( getNode( data ), buf );
|
||||
break;
|
||||
}
|
||||
|
||||
case JSON_T_NULL:
|
||||
break;
|
||||
|
||||
case JSON_T_INTEGER:
|
||||
tr_bencInitInt( getNode( data ), value->integer_value );
|
||||
break;
|
||||
|
||||
case JSON_T_TRUE:
|
||||
tr_bencInitInt( getNode( data ), 1 );
|
||||
break;
|
||||
|
||||
case JSON_T_FALSE:
|
||||
tr_bencInitInt( getNode( data ), 0 );
|
||||
break;
|
||||
|
||||
case JSON_T_STRING:
|
||||
tr_bencInitStrDup( getNode( data ), value->string_value );
|
||||
break;
|
||||
|
||||
case JSON_T_KEY:
|
||||
assert( !data->key );
|
||||
data->key = tr_strdup( value->string_value );
|
||||
break;
|
||||
}
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
int
|
||||
tr_jsonParse( const void * vbuf,
|
||||
const void * bufend,
|
||||
tr_benc * setme_benc,
|
||||
const uint8_t ** setme_end )
|
||||
{
|
||||
int err = 0;
|
||||
const char * buf = vbuf;
|
||||
struct JSON_checker_struct * checker;
|
||||
struct json_benc_data data;
|
||||
|
||||
data.key = NULL;
|
||||
data.top = setme_benc;
|
||||
data.stack = tr_ptrArrayNew( );
|
||||
|
||||
checker = new_JSON_checker( -1, callback, &data, 0 );
|
||||
while( ( buf != bufend ) && JSON_checker_char( checker, *buf ) )
|
||||
++buf;
|
||||
if( buf != bufend )
|
||||
err = TR_ERROR;
|
||||
|
||||
if( setme_end )
|
||||
*setme_end = (const uint8_t*) buf;
|
||||
|
||||
tr_ptrArrayFree( data.stack, NULL );
|
||||
return err;
|
||||
}
|
Loading…
Reference in New Issue