2000-02-05 00:19:15 +00:00
|
|
|
/*
|
|
|
|
**********************************************************************
|
2001-03-21 20:44:20 +00:00
|
|
|
* Copyright (C) 2000-2001, International Business Machines
|
2000-02-05 00:19:15 +00:00
|
|
|
* Corporation and others. All Rights Reserved.
|
|
|
|
**********************************************************************
|
|
|
|
* file name: ucnv_utf.cpp
|
|
|
|
* encoding: US-ASCII
|
|
|
|
* tab size: 8 (not used)
|
|
|
|
* indentation:4
|
|
|
|
*
|
|
|
|
* created on: 2000feb03
|
|
|
|
* created by: Markus W. Scherer
|
2000-06-27 20:47:56 +00:00
|
|
|
*
|
|
|
|
* Change history:
|
|
|
|
*
|
|
|
|
* 06/29/2000 helena Major rewrite of the callback APIs.
|
2000-07-21 01:25:30 +00:00
|
|
|
* 07/20/2000 george Change the coding style to conform to the coding guidelines,
|
|
|
|
* and a few miscellaneous bug fixes.
|
2000-11-16 17:20:03 +00:00
|
|
|
* 11/15/2000 george Added UTF-32
|
2000-02-05 00:19:15 +00:00
|
|
|
*/
|
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
#include "cmemory.h"
|
2000-02-05 00:19:15 +00:00
|
|
|
#include "unicode/utypes.h"
|
2000-06-27 20:47:56 +00:00
|
|
|
#include "unicode/ucnv_err.h"
|
|
|
|
#include "ucnv_bld.h"
|
2000-02-05 00:19:15 +00:00
|
|
|
#include "unicode/ucnv.h"
|
|
|
|
#include "ucnv_cnv.h"
|
|
|
|
|
2001-02-01 17:49:08 +00:00
|
|
|
/* Prototypes --------------------------------------------------------------- */
|
|
|
|
|
|
|
|
/* Keep these here to make finicky compilers happy */
|
|
|
|
|
|
|
|
U_CFUNC void T_UConverter_toUnicode_UTF8(UConverterToUnicodeArgs *args,
|
|
|
|
UErrorCode *err);
|
|
|
|
U_CFUNC void T_UConverter_toUnicode_UTF8_OFFSETS_LOGIC(UConverterToUnicodeArgs *args,
|
|
|
|
UErrorCode *err);
|
|
|
|
U_CFUNC void T_UConverter_fromUnicode_UTF8(UConverterFromUnicodeArgs *args,
|
|
|
|
UErrorCode *err);
|
|
|
|
U_CFUNC void T_UConverter_fromUnicode_UTF8_OFFSETS_LOGIC(UConverterFromUnicodeArgs *args,
|
|
|
|
UErrorCode *err);
|
|
|
|
U_CFUNC UChar32 T_UConverter_getNextUChar_UTF8(UConverterToUnicodeArgs *args,
|
|
|
|
UErrorCode *err);
|
|
|
|
|
|
|
|
|
2000-02-05 00:19:15 +00:00
|
|
|
/* UTF-8 -------------------------------------------------------------------- */
|
|
|
|
|
|
|
|
/* UTF-8 Conversion DATA
|
|
|
|
* for more information see Unicode Strandard 2.0 , Transformation Formats Appendix A-9
|
|
|
|
*/
|
2000-07-21 01:25:30 +00:00
|
|
|
/*static const uint32_t REPLACEMENT_CHARACTER = 0x0000FFFD;*/
|
2001-01-19 20:29:08 +00:00
|
|
|
#define MAXIMUM_UCS2 0x0000FFFF
|
|
|
|
#define MAXIMUM_UTF 0x0010FFFF
|
|
|
|
#define MAXIMUM_UCS4 0x7FFFFFFF
|
|
|
|
#define HALF_SHIFT 10
|
|
|
|
#define HALF_BASE 0x0010000
|
|
|
|
#define HALF_MASK 0x3FF
|
|
|
|
#define SURROGATE_HIGH_START 0xD800
|
|
|
|
#define SURROGATE_HIGH_END 0xDBFF
|
|
|
|
#define SURROGATE_LOW_START 0xDC00
|
|
|
|
#define SURROGATE_LOW_END 0xDFFF
|
|
|
|
|
|
|
|
/* -SURROGATE_LOW_START + HALF_BASE */
|
|
|
|
#define SURROGATE_LOW_BASE 9216
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-20 05:17:49 +00:00
|
|
|
static const uint32_t offsetsFromUTF8[7] = {0,
|
2000-02-05 00:19:15 +00:00
|
|
|
(uint32_t) 0x00000000, (uint32_t) 0x00003080, (uint32_t) 0x000E2080,
|
|
|
|
(uint32_t) 0x03C82080, (uint32_t) 0xFA082080, (uint32_t) 0x82082080
|
|
|
|
};
|
|
|
|
|
|
|
|
/* END OF UTF-8 Conversion DATA */
|
|
|
|
|
2000-07-20 05:17:49 +00:00
|
|
|
static const int8_t bytesFromUTF8[256] = {
|
2000-02-05 00:19:15 +00:00
|
|
|
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,
|
|
|
|
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,
|
|
|
|
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,
|
|
|
|
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,
|
|
|
|
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,
|
|
|
|
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
|
|
|
|
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
|
|
|
|
};
|
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/*
|
|
|
|
* Starting with Unicode 3.0.1:
|
|
|
|
* UTF-8 byte sequences of length N _must_ encode code points of or above utf8_minChar32[N];
|
|
|
|
* byte sequences with more than 4 bytes are illegal in UTF-8,
|
|
|
|
* which is tested with impossible values for them
|
|
|
|
*/
|
|
|
|
static const uint32_t
|
|
|
|
utf8_minChar32[7]={ 0, 0, 0x80, 0x800, 0x10000, 0xffffffff, 0xffffffff };
|
2000-07-22 03:21:16 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Calls invalid char callback when an invalid character sequence is encountered.
|
|
|
|
* It presumes that the converter has a callback to call.
|
|
|
|
*
|
2000-07-25 19:33:02 +00:00
|
|
|
* @returns true when callback fails
|
2000-07-22 03:21:16 +00:00
|
|
|
*/
|
2000-11-16 17:20:03 +00:00
|
|
|
static UBool
|
|
|
|
T_UConverter_toUnicode_InvalidChar_Callback(UConverterToUnicodeArgs * args,
|
2001-10-12 18:54:09 +00:00
|
|
|
UConverterCallbackReason reason,
|
|
|
|
UErrorCode *err)
|
2000-07-22 03:21:16 +00:00
|
|
|
{
|
|
|
|
UConverter *converter = args->converter;
|
|
|
|
|
2000-07-25 19:33:02 +00:00
|
|
|
if (U_SUCCESS(*err))
|
|
|
|
{
|
2001-10-12 18:54:09 +00:00
|
|
|
if (reason == UCNV_ILLEGAL) {
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
} else {
|
|
|
|
*err = U_INVALID_CHAR_FOUND;
|
|
|
|
}
|
2000-07-25 19:33:02 +00:00
|
|
|
}
|
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
/* copy the toUBytes to the invalidCharBuffer */
|
2000-07-22 03:21:16 +00:00
|
|
|
uprv_memcpy(converter->invalidCharBuffer,
|
|
|
|
converter->toUBytes,
|
|
|
|
converter->invalidCharLength);
|
|
|
|
|
|
|
|
/* Call the ErrorFunction */
|
|
|
|
args->converter->fromCharErrorBehaviour(converter->toUContext,
|
|
|
|
args,
|
|
|
|
converter->invalidCharBuffer,
|
|
|
|
converter->invalidCharLength,
|
2001-10-12 18:54:09 +00:00
|
|
|
reason,
|
2000-07-22 03:21:16 +00:00
|
|
|
err);
|
|
|
|
|
2000-07-26 00:55:30 +00:00
|
|
|
return (UBool)U_FAILURE(*err);
|
2000-07-22 03:21:16 +00:00
|
|
|
}
|
|
|
|
|
2000-11-16 17:20:03 +00:00
|
|
|
static UBool
|
|
|
|
T_UConverter_toUnicode_InvalidChar_OffsetCallback(UConverterToUnicodeArgs * args,
|
2001-10-12 18:54:09 +00:00
|
|
|
int32_t currentOffset,
|
|
|
|
UConverterCallbackReason reason,
|
|
|
|
UErrorCode *err)
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
|
|
|
int32_t *saveOffsets = args->offsets;
|
|
|
|
UBool result;
|
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
result = T_UConverter_toUnicode_InvalidChar_Callback(args, reason, err);
|
2000-07-25 19:33:02 +00:00
|
|
|
|
|
|
|
while (saveOffsets < args->offsets)
|
|
|
|
{
|
|
|
|
*(saveOffsets++) = currentOffset;
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
2000-07-17 23:36:52 +00:00
|
|
|
U_CFUNC void T_UConverter_toUnicode_UTF8 (UConverterToUnicodeArgs * args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode * err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
2000-07-25 19:33:02 +00:00
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
2000-07-28 03:11:38 +00:00
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
2000-07-25 19:33:02 +00:00
|
|
|
UBool invalidTailChar = FALSE;
|
2000-07-28 03:11:38 +00:00
|
|
|
uint32_t ch, ch2 = 0, i;
|
|
|
|
uint32_t inBytes; /* Total number of bytes in the current UTF8 sequence */
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-25 19:33:02 +00:00
|
|
|
/* Restore size of current sequence */
|
2000-07-28 03:11:38 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
inBytes = args->converter->toULength; /* restore # of bytes to consume */
|
|
|
|
i = args->converter->invalidCharLength; /* restore # of bytes consumed */
|
2000-07-25 19:33:02 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = args->converter->toUnicodeStatus;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
2000-07-21 01:25:30 +00:00
|
|
|
goto morebytes;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = *(mySource++);
|
|
|
|
if (ch < 0x80) /* Simple case */
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* store the first char */
|
|
|
|
toUBytes[0] = (char)ch;
|
|
|
|
inBytes = bytesFromUTF8[ch]; /* lookup current sequence length */
|
|
|
|
i = 1;
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
morebytes:
|
2000-07-28 03:11:38 +00:00
|
|
|
while (i < inBytes)
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
toUBytes[i] = (char) (ch2 = *(mySource++));
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(ch2))
|
2000-07-21 01:25:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
invalidTailChar = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
ch = (ch << 6) + ch2;
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if (args->flush)
|
|
|
|
{
|
|
|
|
if (U_SUCCESS(*err))
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
|
|
|
else
|
2000-07-28 03:11:38 +00:00
|
|
|
{ /* stores a partially calculated target*/
|
|
|
|
args->converter->toUnicodeStatus = ch;
|
|
|
|
args->converter->toULength = (int8_t) inBytes;
|
|
|
|
args->converter->invalidCharLength = (int8_t) i;
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
goto donefornow;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/* Remove the accumulated high bits */
|
2000-07-28 03:11:38 +00:00
|
|
|
ch -= offsetsFromUTF8[inBytes];
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/*
|
|
|
|
* Legal UTF-8 byte sequences in Unicode 3.0.1 and up:
|
|
|
|
* - use only trail bytes after a lead byte (checked above)
|
|
|
|
* - use the right number of trail bytes for a given lead byte
|
|
|
|
* - encode a code point <= U+10ffff
|
|
|
|
* - use the fewest possible number of bytes for their code points
|
|
|
|
* - use at most 4 bytes (for i>=4 it is 0x10ffff<utf8_minChar32[])
|
|
|
|
* - single surrogate code points are legal but irregular (also cause a callback)
|
|
|
|
*/
|
|
|
|
if (i == inBytes && ch <= MAXIMUM_UTF && ch >= utf8_minChar32[i] && !UTF_IS_SURROGATE(ch))
|
2000-07-28 03:11:38 +00:00
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
2000-07-26 00:55:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
2000-07-26 00:55:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar)ch;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
break;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
2001-10-12 18:54:09 +00:00
|
|
|
UConverterCallbackReason reason =
|
|
|
|
i == inBytes && i == 3 && UTF_IS_SURROGATE(ch) ? UCNV_IRREGULAR : UCNV_ILLEGAL;
|
2000-07-28 03:11:38 +00:00
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (T_UConverter_toUnicode_InvalidChar_Callback(args, reason, err))
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
|
|
|
if (invalidTailChar)
|
|
|
|
{
|
|
|
|
/* Treat the tail as ASCII*/
|
|
|
|
if (myTarget < targetLimit)
|
2000-07-21 01:25:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar) ch2;
|
|
|
|
invalidTailChar = FALSE;
|
2000-07-25 19:33:02 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch2;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
break;
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
/* End of target buffer */
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
2000-07-25 19:33:02 +00:00
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-17 23:36:52 +00:00
|
|
|
U_CFUNC void T_UConverter_toUnicode_UTF8_OFFSETS_LOGIC (UConverterToUnicodeArgs * args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode * err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
2000-07-21 20:42:04 +00:00
|
|
|
int32_t *myOffsets = args->offsets;
|
2000-07-25 19:33:02 +00:00
|
|
|
int32_t offsetNum = 0;
|
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
2000-07-28 03:11:38 +00:00
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
2000-07-25 19:33:02 +00:00
|
|
|
UBool invalidTailChar = FALSE;
|
|
|
|
uint32_t ch, ch2 = 0, i;
|
2000-07-28 03:11:38 +00:00
|
|
|
uint32_t inBytes;
|
2000-07-25 19:33:02 +00:00
|
|
|
|
|
|
|
/* Restore size of current sequence */
|
2000-07-28 03:11:38 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
inBytes = args->converter->toULength; /* restore # of bytes to consume */
|
|
|
|
i = args->converter->invalidCharLength; /* restore # of bytes consumed */
|
2000-07-22 03:21:16 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = args->converter->toUnicodeStatus;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
2000-07-21 01:25:30 +00:00
|
|
|
goto morebytes;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = *(mySource++);
|
|
|
|
if (ch < 0x80) /* Simple case */
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
*(myOffsets++) = offsetNum++;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
toUBytes[0] = (char)ch;
|
|
|
|
inBytes = bytesFromUTF8[ch];
|
|
|
|
i = 1;
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
morebytes:
|
2000-07-28 03:11:38 +00:00
|
|
|
while (i < inBytes)
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
toUBytes[i] = (char) (ch2 = *(mySource++));
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(ch2))
|
2000-07-28 03:11:38 +00:00
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
invalidTailChar = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
ch = (ch << 6) + ch2;
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
if (args->flush)
|
2000-07-21 01:25:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
if (U_SUCCESS(*err))
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->toUnicodeStatus = 0;
|
2000-07-25 19:33:02 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2000-07-22 03:21:16 +00:00
|
|
|
else
|
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->toUnicodeStatus = ch;
|
|
|
|
args->converter->toULength = (int8_t)inBytes;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
goto donefornow;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/* Remove the accumulated high bits */
|
2000-07-28 03:11:38 +00:00
|
|
|
ch -= offsetsFromUTF8[inBytes];
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/*
|
|
|
|
* Legal UTF-8 byte sequences in Unicode 3.0.1 and up:
|
|
|
|
* - use only trail bytes after a lead byte (checked above)
|
|
|
|
* - use the right number of trail bytes for a given lead byte
|
|
|
|
* - encode a code point <= U+10ffff
|
|
|
|
* - use the fewest possible number of bytes for their code points
|
|
|
|
* - use at most 4 bytes (for i>=4 it is 0x10ffff<utf8_minChar32[])
|
|
|
|
* - single surrogate code points are legal but irregular (also cause a callback)
|
|
|
|
*/
|
|
|
|
if (i == inBytes && ch <= MAXIMUM_UTF && ch >= utf8_minChar32[i] && !UTF_IS_SURROGATE(ch))
|
2000-07-28 03:11:38 +00:00
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
2000-07-21 01:25:30 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
2002-06-11 22:49:05 +00:00
|
|
|
*(myOffsets++) = offsetNum;
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar)ch;
|
2000-07-26 00:55:30 +00:00
|
|
|
*(myOffsets++) = offsetNum;
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
offsetNum += i;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
2001-10-12 18:54:09 +00:00
|
|
|
UConverterCallbackReason reason =
|
|
|
|
i == inBytes && i == 3 && UTF_IS_SURROGATE(ch) ? UCNV_IRREGULAR : UCNV_ILLEGAL;
|
2000-07-28 03:11:38 +00:00
|
|
|
UBool useOffset;
|
|
|
|
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->offsets = myOffsets;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
|
|
|
if (T_UConverter_toUnicode_InvalidChar_OffsetCallback(args,
|
2001-10-12 18:54:09 +00:00
|
|
|
offsetNum, reason, err))
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
2000-07-26 00:55:30 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
2000-07-26 00:55:30 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
useOffset = (UBool)(myOffsets != args->offsets);
|
|
|
|
myOffsets = args->offsets;
|
|
|
|
offsetNum += i;
|
2000-07-26 00:55:30 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (invalidTailChar)
|
|
|
|
{
|
|
|
|
/* Treat the tail as ASCII*/
|
|
|
|
if (myTarget < targetLimit)
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (UChar) ch2;
|
|
|
|
*myOffsets = offsetNum++;
|
|
|
|
if (useOffset)
|
2000-07-25 19:33:02 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* Increment when the target was consumed */
|
|
|
|
myOffsets++;
|
2000-07-25 19:33:02 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
invalidTailChar = FALSE;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch2;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
break;
|
2000-07-25 19:33:02 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{ /* End of target buffer */
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
2000-07-25 19:33:02 +00:00
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->offsets = myOffsets;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-19 00:21:29 +00:00
|
|
|
U_CFUNC void T_UConverter_fromUnicode_UTF8 (UConverterFromUnicodeArgs * args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode * err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
2000-07-28 03:11:38 +00:00
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
uint32_t ch, ch2;
|
|
|
|
int16_t indexToWrite;
|
2000-07-21 01:25:30 +00:00
|
|
|
char temp[4];
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (args->converter->fromUnicodeStatus && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = *(mySource++);
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (ch < 0x80) /* Single byte */
|
|
|
|
{
|
|
|
|
*(myTarget++) = (char) ch;
|
|
|
|
}
|
|
|
|
else if (ch < 0x800) /* Double byte */
|
|
|
|
{
|
|
|
|
*(myTarget++) = (char) ((ch >> 6) | 0xc0);
|
|
|
|
if (myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = (char) ((ch & 0x3f) | 0x80);
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->charErrorBuffer[0] = (char) ((ch & 0x3f) | 0x80);
|
|
|
|
args->converter->charErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
/* Check for surogates */
|
|
|
|
{
|
|
|
|
if ((ch >= SURROGATE_HIGH_START) && (ch <= SURROGATE_HIGH_END))
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
lowsurogate:
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if ((ch2 >= SURROGATE_LOW_START) && (ch2 <= SURROGATE_LOW_END))
|
2000-07-20 00:10:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* If there were two surrogates, combine them otherwise treat them normally */
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
2000-07-20 00:10:15 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else if (!args->flush)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (ch < 0x10000)
|
|
|
|
{
|
|
|
|
indexToWrite = 2;
|
|
|
|
temp[2] = (char) ((ch >> 12) | 0xe0);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
indexToWrite = 3;
|
|
|
|
temp[3] = (char) ((ch >> 18) | 0xf0);
|
2000-11-28 19:30:04 +00:00
|
|
|
temp[2] = (char) (((ch >> 12) & 0x3f) | 0x80);
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
2000-11-28 19:30:04 +00:00
|
|
|
temp[1] = (char) (((ch >> 6) & 0x3f) | 0x80);
|
|
|
|
temp[0] = (char) ((ch & 0x3f) | 0x80);
|
2000-07-28 03:11:38 +00:00
|
|
|
|
|
|
|
for (; indexToWrite >= 0; indexToWrite--)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myTarget++) = temp[indexToWrite];
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-19 00:21:29 +00:00
|
|
|
U_CFUNC void T_UConverter_fromUnicode_UTF8_OFFSETS_LOGIC (UConverterFromUnicodeArgs * args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode * err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
2000-07-21 20:42:04 +00:00
|
|
|
int32_t *myOffsets = args->offsets;
|
2000-07-28 03:11:38 +00:00
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
uint32_t ch, ch2;
|
|
|
|
int32_t offsetNum = 0;
|
|
|
|
int16_t indexToWrite;
|
2000-07-21 01:25:30 +00:00
|
|
|
char temp[4];
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (args->converter->fromUnicodeStatus && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
ch = *(mySource++);
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (ch < 0x80) /* Single byte */
|
|
|
|
{
|
|
|
|
*(myOffsets++) = offsetNum++;
|
|
|
|
*(myTarget++) = (char) ch;
|
|
|
|
}
|
|
|
|
else if (ch < 0x800) /* Double byte */
|
|
|
|
{
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
*(myTarget++) = (char) ((ch >> 6) | 0xc0);
|
|
|
|
if (myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myOffsets++) = offsetNum++;
|
|
|
|
*(myTarget++) = (char) ((ch & 0x3f) | 0x80);
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->charErrorBuffer[0] = (char) ((ch & 0x3f) | 0x80);
|
|
|
|
args->converter->charErrorBufferLength = 1;
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
/* Check for surogates */
|
|
|
|
{
|
|
|
|
if ((ch >= SURROGATE_HIGH_START) && (ch <= SURROGATE_HIGH_END))
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
lowsurogate:
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if ((ch2 >= SURROGATE_LOW_START) && (ch2 <= SURROGATE_LOW_END))
|
2000-07-20 00:10:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
/* If there were two surrogates, combine them otherwise treat them normally */
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
2000-07-20 00:10:15 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else if (!args->flush)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (ch < 0x10000)
|
|
|
|
{
|
|
|
|
indexToWrite = 2;
|
|
|
|
temp[2] = (char) ((ch >> 12) | 0xe0);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
indexToWrite = 3;
|
|
|
|
temp[3] = (char) ((ch >> 18) | 0xf0);
|
2000-11-28 19:30:04 +00:00
|
|
|
temp[2] = (char) (((ch >> 12) & 0x3f) | 0x80);
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
2000-11-28 19:30:04 +00:00
|
|
|
temp[1] = (char) (((ch >> 6) & 0x3f) | 0x80);
|
|
|
|
temp[0] = (char) ((ch & 0x3f) | 0x80);
|
2000-07-28 03:11:38 +00:00
|
|
|
|
|
|
|
for (; indexToWrite >= 0; indexToWrite--)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-28 03:11:38 +00:00
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
2000-07-28 03:11:38 +00:00
|
|
|
offsetNum += (ch >= 0x10000) + 1;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2000-07-28 03:11:38 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
2000-08-11 19:23:50 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-07-28 03:11:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
2002-06-11 22:49:05 +00:00
|
|
|
args->offsets = myOffsets;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-19 00:21:29 +00:00
|
|
|
U_CFUNC UChar32 T_UConverter_getNextUChar_UTF8(UConverterToUnicodeArgs *args,
|
2001-01-10 02:19:33 +00:00
|
|
|
UErrorCode *err) {
|
|
|
|
UChar buffer[2];
|
|
|
|
char const *sourceInitial;
|
|
|
|
UChar* myUCharPtr;
|
2001-10-12 18:54:09 +00:00
|
|
|
UConverterCallbackReason reason;
|
2000-07-21 01:25:30 +00:00
|
|
|
uint16_t extraBytesToWrite;
|
|
|
|
uint8_t myByte;
|
|
|
|
UChar32 ch;
|
2001-01-10 02:19:33 +00:00
|
|
|
int8_t isLegalSequence;
|
2000-07-21 01:25:30 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
while (args->source < args->sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2001-01-10 02:19:33 +00:00
|
|
|
sourceInitial = args->source;
|
|
|
|
myByte = (uint8_t)*(args->source++);
|
|
|
|
if (myByte < 0x80)
|
|
|
|
{
|
|
|
|
return (UChar32)myByte;
|
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
extraBytesToWrite = (uint16_t)bytesFromUTF8[myByte];
|
|
|
|
if (extraBytesToWrite == 0) {
|
2001-10-15 21:44:27 +00:00
|
|
|
isLegalSequence = FALSE;
|
|
|
|
ch = 0;
|
2001-01-10 02:19:33 +00:00
|
|
|
goto CALL_ERROR_FUNCTION;
|
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
/*The byte sequence is longer than the buffer area passed*/
|
|
|
|
if ((args->source + extraBytesToWrite - 1) > args->sourceLimit)
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
isLegalSequence = 1;
|
|
|
|
ch = myByte << 6;
|
|
|
|
switch(extraBytesToWrite)
|
|
|
|
{
|
|
|
|
/* note: code falls through cases! (sic)*/
|
|
|
|
case 6:
|
|
|
|
ch += (myByte = (uint8_t)*(args->source++));
|
|
|
|
ch <<= 6;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(myByte))
|
2001-01-10 02:19:33 +00:00
|
|
|
{
|
|
|
|
isLegalSequence = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case 5:
|
|
|
|
ch += (myByte = *(args->source++));
|
|
|
|
ch <<= 6;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(myByte))
|
2001-01-10 02:19:33 +00:00
|
|
|
{
|
|
|
|
isLegalSequence = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case 4:
|
|
|
|
ch += (myByte = *(args->source++));
|
|
|
|
ch <<= 6;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(myByte))
|
2001-01-10 02:19:33 +00:00
|
|
|
{
|
|
|
|
isLegalSequence = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case 3:
|
|
|
|
ch += (myByte = *(args->source++));
|
|
|
|
ch <<= 6;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(myByte))
|
2001-01-10 02:19:33 +00:00
|
|
|
{
|
|
|
|
isLegalSequence = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case 2:
|
|
|
|
ch += (myByte = *(args->source++));
|
2001-10-12 18:54:09 +00:00
|
|
|
if (!UTF8_IS_TRAIL(myByte))
|
2001-01-10 02:19:33 +00:00
|
|
|
{
|
|
|
|
isLegalSequence = 0;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
}
|
|
|
|
ch -= offsetsFromUTF8[extraBytesToWrite];
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2001-10-12 18:54:09 +00:00
|
|
|
/*
|
|
|
|
* Legal UTF-8 byte sequences in Unicode 3.0.1 and up:
|
|
|
|
* - use only trail bytes after a lead byte (checked above)
|
|
|
|
* - use the right number of trail bytes for a given lead byte
|
|
|
|
* - encode a code point <= U+10ffff
|
|
|
|
* - use the fewest possible number of bytes for their code points
|
|
|
|
* - use at most 4 bytes (for i>=4 it is 0x10ffff<utf8_minChar32[])
|
|
|
|
* - single surrogate code points are legal but irregular (also cause a callback)
|
|
|
|
*/
|
|
|
|
if (isLegalSequence && (uint32_t)ch <= MAXIMUM_UTF && (uint32_t)ch >= utf8_minChar32[extraBytesToWrite] && !UTF_IS_SURROGATE(ch)) {
|
2001-01-10 02:19:33 +00:00
|
|
|
return ch; /* return the code point */
|
|
|
|
}
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
CALL_ERROR_FUNCTION:
|
2001-02-02 23:19:57 +00:00
|
|
|
extraBytesToWrite = (uint16_t)(args->source - sourceInitial);
|
2001-01-10 02:19:33 +00:00
|
|
|
args->converter->invalidCharLength = (uint8_t)extraBytesToWrite;
|
|
|
|
uprv_memcpy(args->converter->invalidCharBuffer, sourceInitial, extraBytesToWrite);
|
|
|
|
|
|
|
|
myUCharPtr = buffer;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (isLegalSequence && extraBytesToWrite == 3 && UTF_IS_SURROGATE(ch)) {
|
|
|
|
reason = UCNV_IRREGULAR;
|
|
|
|
*err = U_INVALID_CHAR_FOUND;
|
|
|
|
} else {
|
|
|
|
reason = UCNV_ILLEGAL;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
}
|
2001-01-10 02:19:33 +00:00
|
|
|
args->target = myUCharPtr;
|
|
|
|
args->targetLimit = buffer + 2;
|
|
|
|
args->converter->fromCharErrorBehaviour(args->converter->toUContext,
|
|
|
|
args,
|
|
|
|
sourceInitial,
|
|
|
|
extraBytesToWrite,
|
2001-10-12 18:54:09 +00:00
|
|
|
reason,
|
2001-01-10 02:19:33 +00:00
|
|
|
err);
|
|
|
|
|
|
|
|
if(U_SUCCESS(*err)) {
|
|
|
|
extraBytesToWrite = (uint16_t)(args->target - buffer);
|
|
|
|
if(extraBytesToWrite > 0) {
|
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, buffer, extraBytesToWrite);
|
|
|
|
}
|
|
|
|
/* else (callback did not write anything) continue */
|
|
|
|
} else if(*err == U_BUFFER_OVERFLOW_ERROR) {
|
|
|
|
*err = U_ZERO_ERROR;
|
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, buffer, 2);
|
|
|
|
} else {
|
|
|
|
/* break on error */
|
|
|
|
/* ### what if a callback set an error but _also_ generated output?! */
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
}
|
2001-01-10 02:19:33 +00:00
|
|
|
|
|
|
|
/* no input or only skipping callback calls */
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
|
|
|
return 0xffff;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-02-09 19:15:17 +00:00
|
|
|
static const UConverterImpl _UTF8Impl={
|
2000-02-05 00:19:15 +00:00
|
|
|
UCNV_UTF8,
|
|
|
|
|
2000-02-08 23:41:16 +00:00
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
2000-02-05 00:19:15 +00:00
|
|
|
T_UConverter_toUnicode_UTF8,
|
|
|
|
T_UConverter_toUnicode_UTF8_OFFSETS_LOGIC,
|
|
|
|
T_UConverter_fromUnicode_UTF8,
|
|
|
|
T_UConverter_fromUnicode_UTF8_OFFSETS_LOGIC,
|
2000-02-09 19:15:17 +00:00
|
|
|
T_UConverter_getNextUChar_UTF8,
|
|
|
|
|
2000-11-27 17:40:41 +00:00
|
|
|
NULL,
|
2000-02-09 19:15:17 +00:00
|
|
|
NULL
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
|
|
|
|
2001-12-27 18:23:04 +00:00
|
|
|
/* The 1208 CCSID refers to any version of Unicode of UTF-8 */
|
2001-09-06 18:34:57 +00:00
|
|
|
static const UConverterStaticData _UTF8StaticData={
|
2001-01-26 02:54:57 +00:00
|
|
|
sizeof(UConverterStaticData),
|
2001-09-28 00:29:40 +00:00
|
|
|
"UTF-8",
|
2000-04-19 23:05:27 +00:00
|
|
|
1208, UCNV_IBM, UCNV_UTF8, 1, 4,
|
2000-12-19 23:07:50 +00:00
|
|
|
{ 0xef, 0xbf, 0xbd, 0 },3,FALSE,FALSE,
|
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2000-04-19 23:05:27 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
2000-04-13 23:00:43 +00:00
|
|
|
const UConverterSharedData _UTF8Data={
|
2000-03-29 21:41:11 +00:00
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
2000-04-19 23:05:27 +00:00
|
|
|
NULL, NULL, &_UTF8StaticData, FALSE, &_UTF8Impl,
|
|
|
|
0
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
|
|
|
|
2001-01-03 20:42:13 +00:00
|
|
|
/* UTF-16 Platform Endian --------------------------------------------------- */
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-03 20:42:13 +00:00
|
|
|
_UTF16PEToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
2001-01-04 22:00:02 +00:00
|
|
|
UConverter *cnv = pArgs->converter;
|
|
|
|
const uint8_t *source = (const uint8_t *)pArgs->source;
|
|
|
|
UChar *target = pArgs->target;
|
|
|
|
int32_t *offsets = pArgs->offsets;
|
|
|
|
int32_t targetCapacity = pArgs->targetLimit - pArgs->target;
|
|
|
|
int32_t length = (const uint8_t *)pArgs->sourceLimit - source;
|
|
|
|
int32_t count;
|
|
|
|
int32_t sourceIndex = 0;
|
|
|
|
|
2002-06-11 20:44:36 +00:00
|
|
|
if(length <= 0 && cnv->toUnicodeStatus == 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(targetCapacity <= 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* complete a partial UChar from the last call */
|
2002-06-11 20:44:36 +00:00
|
|
|
if(length != 0 && cnv->toUnicodeStatus != 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/*
|
|
|
|
* copy the byte from the last call and the first one here into the target,
|
|
|
|
* byte-wise to keep the platform endianness
|
|
|
|
*/
|
2001-01-04 22:00:02 +00:00
|
|
|
uint8_t *p = (uint8_t *)target++;
|
|
|
|
*p++ = (uint8_t)cnv->toUnicodeStatus;
|
|
|
|
cnv->toUnicodeStatus = 0;
|
|
|
|
*p = *source++;
|
2001-01-03 20:42:13 +00:00
|
|
|
--length;
|
|
|
|
--targetCapacity;
|
2001-01-04 22:00:02 +00:00
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = -1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* copy an even number of bytes for complete UChars */
|
2001-01-04 22:00:02 +00:00
|
|
|
count = 2 * targetCapacity;
|
|
|
|
if(count > length) {
|
|
|
|
count = length & ~1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
if(count > 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
uprv_memcpy(target, source, count);
|
2001-01-04 22:00:02 +00:00
|
|
|
source += count;
|
|
|
|
length -= count;
|
|
|
|
count >>= 1;
|
|
|
|
target += count;
|
|
|
|
targetCapacity -= count;
|
|
|
|
if(offsets != NULL) {
|
|
|
|
while(count > 0) {
|
|
|
|
*offsets++ = sourceIndex;
|
|
|
|
sourceIndex += 2;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* check for a remaining source byte and store the status */
|
2001-01-04 22:00:02 +00:00
|
|
|
if(length >= 2) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* it must be targetCapacity==0 because otherwise the above would have copied more */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
} else if(length == 1) {
|
2001-01-03 20:42:13 +00:00
|
|
|
if(pArgs->flush) {
|
|
|
|
/* a UChar remains incomplete */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_TRUNCATED_CHAR_FOUND;
|
2001-01-03 20:42:13 +00:00
|
|
|
} else {
|
|
|
|
/* consume the last byte and store it, making sure that it will never set the status to 0 */
|
2001-01-04 22:00:02 +00:00
|
|
|
cnv->toUnicodeStatus = *source++ | 0x100;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2002-06-11 20:44:36 +00:00
|
|
|
} else /* length==0 */ if(cnv->toUnicodeStatus!=0 && pArgs->flush) {
|
|
|
|
/* a UChar remains incomplete */
|
|
|
|
*pErrorCode = U_TRUNCATED_CHAR_FOUND;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* write back the updated pointers */
|
2001-01-04 22:00:02 +00:00
|
|
|
pArgs->source = (const char *)source;
|
|
|
|
pArgs->target = target;
|
|
|
|
pArgs->offsets = offsets;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-03 20:42:13 +00:00
|
|
|
_UTF16PEFromUnicodeWithOffsets(UConverterFromUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
2001-01-04 22:00:02 +00:00
|
|
|
UConverter *cnv = pArgs->converter;
|
|
|
|
const UChar *source = pArgs->source;
|
|
|
|
uint8_t *target = (uint8_t *)pArgs->target;
|
|
|
|
int32_t *offsets = pArgs->offsets;
|
|
|
|
int32_t targetCapacity = pArgs->targetLimit - pArgs->target;
|
|
|
|
int32_t length = pArgs->sourceLimit - source;
|
|
|
|
int32_t count;
|
|
|
|
int32_t sourceIndex = 0;
|
|
|
|
|
|
|
|
if(length <= 0 && cnv->fromUnicodeStatus == 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(targetCapacity <= 0) {
|
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
2001-01-03 20:42:13 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* complete a partial UChar from the last call */
|
2001-01-04 22:00:02 +00:00
|
|
|
if(cnv->fromUnicodeStatus != 0) {
|
|
|
|
*target++ = (uint8_t)cnv->fromUnicodeStatus;
|
|
|
|
cnv->fromUnicodeStatus = 0;
|
2001-01-03 20:42:13 +00:00
|
|
|
--targetCapacity;
|
2001-01-04 22:00:02 +00:00
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = -1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* copy an even number of bytes for complete UChars */
|
2001-01-04 22:00:02 +00:00
|
|
|
count = 2 * length;
|
|
|
|
if(count > targetCapacity) {
|
|
|
|
count = targetCapacity & ~1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
if(count>0) {
|
|
|
|
uprv_memcpy(target, source, count);
|
2001-01-04 22:00:02 +00:00
|
|
|
target += count;
|
|
|
|
targetCapacity -= count;
|
|
|
|
count >>= 1;
|
|
|
|
source += count;
|
|
|
|
length -= count;
|
|
|
|
if(offsets != NULL) {
|
|
|
|
while(count > 0) {
|
|
|
|
*offsets++ = sourceIndex;
|
|
|
|
*offsets++ = sourceIndex++;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(length > 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* it must be targetCapacity<=1 because otherwise the above would have copied more */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
if(targetCapacity > 0) /* targetCapacity==1 */ {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* copy one byte and keep the other in the status */
|
2001-01-04 22:00:02 +00:00
|
|
|
const uint8_t *p = (const uint8_t *)source++;
|
|
|
|
*target++ = *p++;
|
|
|
|
cnv->fromUnicodeStatus = *p | 0x100;
|
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = sourceIndex;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2001-01-03 20:42:13 +00:00
|
|
|
/* write back the updated pointers */
|
2001-01-04 22:00:02 +00:00
|
|
|
pArgs->source = source;
|
|
|
|
pArgs->target = (char *)target;
|
|
|
|
pArgs->offsets = offsets;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* UTF-16 Opposite Endian --------------------------------------------------- */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* For opposite-endian UTF-16, we keep a byte pointer to the UChars
|
|
|
|
* and copy two bytes at a time and reverse them.
|
|
|
|
*/
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-03 20:42:13 +00:00
|
|
|
_UTF16OEToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
2001-01-04 22:00:02 +00:00
|
|
|
UConverter *cnv = pArgs->converter;
|
|
|
|
const uint8_t *source = (const uint8_t *)pArgs->source;
|
|
|
|
UChar *target = pArgs->target;
|
|
|
|
uint8_t *target8 = (uint8_t *)target; /* byte pointer to the target */
|
|
|
|
int32_t *offsets = pArgs->offsets;
|
|
|
|
int32_t targetCapacity = pArgs->targetLimit - pArgs->target;
|
|
|
|
int32_t length = (const uint8_t *)pArgs->sourceLimit - source;
|
|
|
|
int32_t count;
|
|
|
|
int32_t sourceIndex = 0;
|
|
|
|
|
2002-06-11 20:44:36 +00:00
|
|
|
if(length <= 0 && cnv->toUnicodeStatus == 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(targetCapacity <= 0) {
|
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
2001-01-03 20:42:13 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* complete a partial UChar from the last call */
|
2002-06-11 20:44:36 +00:00
|
|
|
if(length != 0 && cnv->toUnicodeStatus != 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/*
|
|
|
|
* copy the byte from the last call and the first one here into the target,
|
|
|
|
* byte-wise, reversing the platform endianness
|
|
|
|
*/
|
2001-01-04 22:00:02 +00:00
|
|
|
*target8++ = *source++;
|
|
|
|
*target8++ = (uint8_t)cnv->toUnicodeStatus;
|
|
|
|
cnv->toUnicodeStatus = 0;
|
2001-01-03 20:42:13 +00:00
|
|
|
++target;
|
|
|
|
--length;
|
|
|
|
--targetCapacity;
|
2001-01-04 22:00:02 +00:00
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = -1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* copy an even number of bytes for complete UChars */
|
2001-01-04 22:00:02 +00:00
|
|
|
count = 2 * targetCapacity;
|
|
|
|
if(count > length) {
|
|
|
|
count = length & ~1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
if(count>0) {
|
2001-01-04 22:00:02 +00:00
|
|
|
length -= count;
|
|
|
|
count >>= 1;
|
|
|
|
targetCapacity -= count;
|
|
|
|
if(offsets == NULL) {
|
|
|
|
while(count > 0) {
|
|
|
|
target8[1] = *source++;
|
|
|
|
target8[0] = *source++;
|
|
|
|
target8 += 2;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
while(count>0) {
|
2001-01-04 22:00:02 +00:00
|
|
|
target8[1] = *source++;
|
|
|
|
target8[0] = *source++;
|
|
|
|
target8 += 2;
|
|
|
|
*offsets++ = sourceIndex;
|
|
|
|
sourceIndex += 2;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2001-01-03 20:42:13 +00:00
|
|
|
target=(UChar *)target8;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
2001-01-03 20:42:13 +00:00
|
|
|
/* check for a remaining source byte and store the status */
|
2001-01-04 22:00:02 +00:00
|
|
|
if(length >= 2) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* it must be targetCapacity==0 because otherwise the above would have copied more */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
} else if(length == 1) {
|
2001-01-03 20:42:13 +00:00
|
|
|
if(pArgs->flush) {
|
|
|
|
/* a UChar remains incomplete */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_TRUNCATED_CHAR_FOUND;
|
2001-01-03 20:42:13 +00:00
|
|
|
} else {
|
|
|
|
/* consume the last byte and store it, making sure that it will never set the status to 0 */
|
2001-01-04 22:00:02 +00:00
|
|
|
cnv->toUnicodeStatus = *source++ | 0x100;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2002-06-11 20:44:36 +00:00
|
|
|
} else /* length==0 */ if(cnv->toUnicodeStatus!=0 && pArgs->flush) {
|
|
|
|
/* a UChar remains incomplete */
|
|
|
|
*pErrorCode = U_TRUNCATED_CHAR_FOUND;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* write back the updated pointers */
|
2001-01-04 22:00:02 +00:00
|
|
|
pArgs->source = (const char *)source;
|
|
|
|
pArgs->target = target;
|
|
|
|
pArgs->offsets = offsets;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-03 20:42:13 +00:00
|
|
|
_UTF16OEFromUnicodeWithOffsets(UConverterFromUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
2001-01-04 22:00:02 +00:00
|
|
|
UConverter *cnv = pArgs->converter;
|
|
|
|
const UChar *source = pArgs->source;
|
|
|
|
const uint8_t *source8 = (const uint8_t *)source; /* byte pointer to the source */
|
|
|
|
uint8_t *target = (uint8_t *)pArgs->target;
|
|
|
|
int32_t *offsets = pArgs->offsets;
|
|
|
|
int32_t targetCapacity = pArgs->targetLimit - pArgs->target;
|
|
|
|
int32_t length = pArgs->sourceLimit - source;
|
|
|
|
int32_t count;
|
|
|
|
int32_t sourceIndex = 0;
|
|
|
|
|
|
|
|
if(length <= 0 && cnv->fromUnicodeStatus == 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(targetCapacity <= 0) {
|
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
2001-01-03 20:42:13 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* complete a partial UChar from the last call */
|
2001-01-04 22:00:02 +00:00
|
|
|
if(cnv->fromUnicodeStatus != 0) {
|
|
|
|
*target++ = (uint8_t)cnv->fromUnicodeStatus;
|
|
|
|
cnv->fromUnicodeStatus = 0;
|
2001-01-03 20:42:13 +00:00
|
|
|
--targetCapacity;
|
2001-01-04 22:00:02 +00:00
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = -1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* copy an even number of bytes for complete UChars */
|
2001-01-04 22:00:02 +00:00
|
|
|
count = 2 * length;
|
|
|
|
if(count > targetCapacity) {
|
|
|
|
count = targetCapacity & ~1;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
if(count > 0) {
|
|
|
|
targetCapacity -= count;
|
|
|
|
count >>= 1;
|
|
|
|
length -= count;
|
|
|
|
if(offsets == NULL) {
|
|
|
|
while(count > 0) {
|
|
|
|
target[1] = *source8++;
|
|
|
|
target[0] = *source8++;
|
|
|
|
target += 2;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2001-01-03 20:42:13 +00:00
|
|
|
} else {
|
|
|
|
while(count>0) {
|
2001-01-04 22:00:02 +00:00
|
|
|
target[1] = *source8++;
|
|
|
|
target[0] = *source8++;
|
|
|
|
target += 2;
|
|
|
|
*offsets++ = sourceIndex;
|
|
|
|
*offsets++ = sourceIndex++;
|
2001-01-03 20:42:13 +00:00
|
|
|
--count;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
2001-01-03 20:42:13 +00:00
|
|
|
source=(const UChar *)source8;
|
|
|
|
}
|
|
|
|
|
2001-01-04 22:00:02 +00:00
|
|
|
if(length > 0) {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* it must be targetCapacity<=1 because otherwise the above would have copied more */
|
2001-01-04 22:00:02 +00:00
|
|
|
*pErrorCode = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
if(targetCapacity > 0) /* targetCapacity==1 */ {
|
2001-01-03 20:42:13 +00:00
|
|
|
/* copy one byte and keep the other in the status */
|
2001-01-04 22:00:02 +00:00
|
|
|
cnv->fromUnicodeStatus = *source8++ | 0x100;
|
|
|
|
*target++ = *source8;
|
2001-01-03 20:42:13 +00:00
|
|
|
++source;
|
2001-01-04 22:00:02 +00:00
|
|
|
if(offsets != NULL) {
|
|
|
|
*offsets++ = sourceIndex;
|
2001-01-03 20:42:13 +00:00
|
|
|
}
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2001-01-03 20:42:13 +00:00
|
|
|
/* write back the updated pointers */
|
2001-01-04 22:00:02 +00:00
|
|
|
pArgs->source = source;
|
|
|
|
pArgs->target = (char *)target;
|
|
|
|
pArgs->offsets = offsets;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2001-01-03 20:42:13 +00:00
|
|
|
/* UTF-16BE ----------------------------------------------------------------- */
|
|
|
|
|
2002-06-11 04:43:45 +00:00
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
# define _UTF16BEToUnicodeWithOffsets _UTF16PEToUnicodeWithOffsets
|
|
|
|
# define _UTF16LEToUnicodeWithOffsets _UTF16OEToUnicodeWithOffsets
|
|
|
|
# define _UTF16BEFromUnicodeWithOffsets _UTF16PEFromUnicodeWithOffsets
|
|
|
|
# define _UTF16LEFromUnicodeWithOffsets _UTF16OEFromUnicodeWithOffsets
|
|
|
|
#else
|
|
|
|
# define _UTF16BEToUnicodeWithOffsets _UTF16OEToUnicodeWithOffsets
|
|
|
|
# define _UTF16LEToUnicodeWithOffsets _UTF16PEToUnicodeWithOffsets
|
|
|
|
# define _UTF16BEFromUnicodeWithOffsets _UTF16OEFromUnicodeWithOffsets
|
|
|
|
# define _UTF16LEFromUnicodeWithOffsets _UTF16PEFromUnicodeWithOffsets
|
|
|
|
#endif
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static UChar32 T_UConverter_getNextUChar_UTF16_BE(UConverterToUnicodeArgs* args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode* err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
UChar32 myUChar;
|
|
|
|
uint16_t first;
|
|
|
|
/*Checks boundaries and set appropriate error codes*/
|
|
|
|
if (args->source+2 > args->sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
if (args->source >= args->sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
/*Either caller has reached the end of the byte stream*/
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
/* a character was cut in half*/
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
return 0xffff;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/*Gets the corresponding codepoint*/
|
2000-08-11 02:08:59 +00:00
|
|
|
first = (uint16_t)(((uint16_t)(*(args->source)) << 8) |((uint8_t)*((args->source)+1)));
|
2000-07-21 01:25:30 +00:00
|
|
|
myUChar = first;
|
|
|
|
args->source += 2;
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
if(UTF_IS_FIRST_SURROGATE(first)) {
|
|
|
|
uint16_t second;
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
if (args->source+2 > args->sourceLimit) {
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/* get the second surrogate and assemble the code point */
|
2000-08-11 02:08:59 +00:00
|
|
|
second = (uint16_t)(((uint16_t)(*(args->source)) << 8) |((uint8_t)*(args->source+1)));
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/* ignore unmatched surrogates and just deliver the first one in such a case */
|
|
|
|
if(UTF_IS_SECOND_SURROGATE(second)) {
|
|
|
|
/* matched pair, get pair value */
|
|
|
|
myUChar = UTF16_GET_PAIR_VALUE(first, second);
|
|
|
|
args->source += 2;
|
|
|
|
}
|
2000-04-13 17:27:35 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
|
|
|
return myUChar;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-02-09 19:15:17 +00:00
|
|
|
static const UConverterImpl _UTF16BEImpl={
|
2000-02-05 00:19:15 +00:00
|
|
|
UCNV_UTF16_BigEndian,
|
|
|
|
|
2000-02-08 23:41:16 +00:00
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
2002-06-11 04:43:45 +00:00
|
|
|
_UTF16BEToUnicodeWithOffsets,
|
|
|
|
_UTF16BEToUnicodeWithOffsets,
|
|
|
|
_UTF16BEFromUnicodeWithOffsets,
|
|
|
|
_UTF16BEFromUnicodeWithOffsets,
|
2000-02-09 19:15:17 +00:00
|
|
|
T_UConverter_getNextUChar_UTF16_BE,
|
|
|
|
|
2000-11-27 17:40:41 +00:00
|
|
|
NULL,
|
2000-02-09 19:15:17 +00:00
|
|
|
NULL
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
|
|
|
|
2001-12-27 18:23:04 +00:00
|
|
|
/* The 1200 CCSID refers to any version of Unicode with any endianess of UTF-16 */
|
2001-09-06 18:34:57 +00:00
|
|
|
static const UConverterStaticData _UTF16BEStaticData={
|
2001-01-03 20:42:13 +00:00
|
|
|
sizeof(UConverterStaticData),
|
2001-09-28 00:29:40 +00:00
|
|
|
"UTF-16BE",
|
2000-04-19 23:05:27 +00:00
|
|
|
1200, UCNV_IBM, UCNV_UTF16_BigEndian, 2, 2,
|
2000-12-19 23:07:50 +00:00
|
|
|
{ 0xff, 0xfd, 0, 0 },2,FALSE,FALSE,
|
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2000-04-19 23:05:27 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
2000-04-13 23:00:43 +00:00
|
|
|
const UConverterSharedData _UTF16BEData={
|
2000-03-29 21:41:11 +00:00
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
2000-04-19 23:05:27 +00:00
|
|
|
NULL, NULL, &_UTF16BEStaticData, FALSE, &_UTF16BEImpl,
|
|
|
|
0
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/* UTF-16LE ----------------------------------------------------------------- */
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static UChar32 T_UConverter_getNextUChar_UTF16_LE(UConverterToUnicodeArgs* args,
|
2000-02-05 00:19:15 +00:00
|
|
|
UErrorCode* err)
|
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
UChar32 myUChar;
|
|
|
|
uint16_t first;
|
|
|
|
/*Checks boundaries and set appropriate error codes*/
|
|
|
|
if (args->source+2 > args->sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
if (args->source >= args->sourceLimit)
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
/*Either caller has reached the end of the byte stream*/
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
else
|
2000-02-05 00:19:15 +00:00
|
|
|
{
|
2000-07-21 01:25:30 +00:00
|
|
|
/* a character was cut in half*/
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
|
|
|
return 0xffff;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/*Gets the corresponding codepoint*/
|
2000-08-11 02:08:59 +00:00
|
|
|
first = (uint16_t)(((uint16_t)*((args->source)+1) << 8) | ((uint8_t)(*(args->source))));
|
2000-07-21 01:25:30 +00:00
|
|
|
myUChar=first;
|
|
|
|
/*updates the source*/
|
|
|
|
args->source += 2;
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
if (UTF_IS_FIRST_SURROGATE(first))
|
|
|
|
{
|
|
|
|
uint16_t second;
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
if (args->source+2 > args->sourceLimit)
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/* get the second surrogate and assemble the code point */
|
2000-08-11 02:08:59 +00:00
|
|
|
second = (uint16_t)(((uint16_t)*(args->source+1) << 8) |((uint8_t)(*(args->source))));
|
2000-04-13 17:27:35 +00:00
|
|
|
|
2000-07-21 01:25:30 +00:00
|
|
|
/* ignore unmatched surrogates and just deliver the first one in such a case */
|
|
|
|
if(UTF_IS_SECOND_SURROGATE(second))
|
|
|
|
{
|
|
|
|
/* matched pair, get pair value */
|
|
|
|
myUChar = UTF16_GET_PAIR_VALUE(first, second);
|
|
|
|
args->source += 2;
|
|
|
|
}
|
2000-04-13 17:27:35 +00:00
|
|
|
}
|
2000-07-21 01:25:30 +00:00
|
|
|
|
|
|
|
return myUChar;
|
2000-02-05 00:19:15 +00:00
|
|
|
}
|
|
|
|
|
2000-02-09 19:15:17 +00:00
|
|
|
static const UConverterImpl _UTF16LEImpl={
|
2000-02-05 00:19:15 +00:00
|
|
|
UCNV_UTF16_LittleEndian,
|
|
|
|
|
2000-02-08 23:41:16 +00:00
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
2002-06-11 04:43:45 +00:00
|
|
|
_UTF16LEToUnicodeWithOffsets,
|
|
|
|
_UTF16LEToUnicodeWithOffsets,
|
|
|
|
_UTF16LEFromUnicodeWithOffsets,
|
|
|
|
_UTF16LEFromUnicodeWithOffsets,
|
2000-02-09 19:15:17 +00:00
|
|
|
T_UConverter_getNextUChar_UTF16_LE,
|
|
|
|
|
2000-11-27 17:40:41 +00:00
|
|
|
NULL,
|
2000-02-09 19:15:17 +00:00
|
|
|
NULL
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
|
|
|
|
2000-04-19 23:05:27 +00:00
|
|
|
|
2001-12-27 18:23:04 +00:00
|
|
|
/* The 1200 CCSID refers to any version of Unicode with any endianess of UTF-16 */
|
2001-09-06 18:34:57 +00:00
|
|
|
static const UConverterStaticData _UTF16LEStaticData={
|
2000-07-21 01:25:30 +00:00
|
|
|
sizeof(UConverterStaticData),
|
2001-09-28 00:29:40 +00:00
|
|
|
"UTF-16LE",
|
2000-07-21 01:25:30 +00:00
|
|
|
1200, UCNV_IBM, UCNV_UTF16_LittleEndian, 2, 2,
|
2001-01-26 02:54:57 +00:00
|
|
|
{ 0xfd, 0xff, 0, 0 },2,FALSE,FALSE,
|
2000-12-19 23:07:50 +00:00
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2000-04-19 23:05:27 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
2000-04-13 23:00:43 +00:00
|
|
|
const UConverterSharedData _UTF16LEData={
|
2000-03-29 21:41:11 +00:00
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
2000-04-19 23:05:27 +00:00
|
|
|
NULL, NULL, &_UTF16LEStaticData, FALSE, &_UTF16LEImpl,
|
|
|
|
0
|
2000-02-05 00:19:15 +00:00
|
|
|
};
|
2000-11-16 17:20:03 +00:00
|
|
|
|
|
|
|
/* UTF-32BE ----------------------------------------------------------------- */
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_BE(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
|
|
|
uint32_t ch, i;
|
|
|
|
|
|
|
|
/* UTF-8 returns here for only non-offset, this needs to change.*/
|
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
|
|
|
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
|
|
|
while (i < sizeof(uint32_t))
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch = (ch << 8) | (uint8_t)(*mySource);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if (args->flush)
|
|
|
|
{
|
|
|
|
if (U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
args->converter->toUnicodeStatus = MAXIMUM_UCS4;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{ /* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
|
|
|
}
|
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ch <= MAXIMUM_UTF)
|
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
|
|
|
{
|
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (T_UConverter_toUnicode_InvalidChar_Callback(args, UCNV_ILLEGAL, err))
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
/* End of target buffer */
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_BE_OFFSET_LOGIC(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
|
|
|
int32_t *myOffsets = args->offsets;
|
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
|
|
|
uint32_t ch, i;
|
|
|
|
int32_t offsetNum = 0;
|
|
|
|
|
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
|
|
|
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
|
|
|
while (i < sizeof(uint32_t))
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch = (ch << 8) | (uint8_t)(*mySource);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if (args->flush)
|
|
|
|
{
|
|
|
|
if (U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
args->converter->toUnicodeStatus = MAXIMUM_UCS4;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{ /* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
|
|
|
}
|
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ch <= MAXIMUM_UTF)
|
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
|
|
|
{
|
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
|
|
|
*myOffsets++ = offsetNum;
|
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2001-01-26 02:54:57 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (T_UConverter_toUnicode_InvalidChar_OffsetCallback(args, offsetNum, UCNV_ILLEGAL, err))
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
2001-01-26 02:54:57 +00:00
|
|
|
myOffsets = args->offsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
2001-01-26 02:54:57 +00:00
|
|
|
offsetNum += i;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
/* End of target buffer */
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
2002-06-11 22:49:05 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_BE(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
UChar32 ch, ch2;
|
|
|
|
unsigned int indexToWrite;
|
|
|
|
unsigned char temp[sizeof(uint32_t)];
|
|
|
|
|
|
|
|
temp[0] = 0;
|
|
|
|
|
|
|
|
if (args->converter->fromUnicodeStatus)
|
|
|
|
{
|
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
|
|
|
if (SURROGATE_HIGH_START <= ch && ch < SURROGATE_LOW_START)
|
|
|
|
{
|
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (SURROGATE_LOW_START <= ch2 && ch2 <= SURROGATE_LOW_END)
|
|
|
|
{
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (!args->flush)
|
|
|
|
{
|
2000-11-20 18:09:58 +00:00
|
|
|
/* ran out of source */
|
2000-11-16 17:20:03 +00:00
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[1] = (uint8_t) (ch >> 16 & 0x1F);
|
2001-01-26 02:54:57 +00:00
|
|
|
temp[2] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[3] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
2000-11-16 17:20:03 +00:00
|
|
|
|
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2001-01-26 02:54:57 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-11-16 17:20:03 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_BE_OFFSET_LOGIC(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
|
|
|
int32_t *myOffsets = args->offsets;
|
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
UChar32 ch, ch2;
|
|
|
|
int32_t offsetNum = 0;
|
|
|
|
unsigned int indexToWrite;
|
|
|
|
unsigned char temp[sizeof(uint32_t)];
|
|
|
|
|
|
|
|
temp[0] = 0;
|
|
|
|
|
|
|
|
if (args->converter->fromUnicodeStatus)
|
|
|
|
{
|
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
|
|
|
if (SURROGATE_HIGH_START <= ch && ch < SURROGATE_LOW_START)
|
|
|
|
{
|
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (SURROGATE_LOW_START <= ch2 && ch2 <= SURROGATE_LOW_END)
|
|
|
|
{
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (!args->flush)
|
|
|
|
{
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[1] = (uint8_t) (ch >> 16 & 0x1F);
|
2001-01-26 02:54:57 +00:00
|
|
|
temp[2] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[3] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
2001-01-04 22:00:02 +00:00
|
|
|
|
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2001-01-26 02:54:57 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
offsetNum++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
2002-06-11 22:49:05 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static UChar32
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_getNextUChar_UTF32_BE(UConverterToUnicodeArgs* args,
|
|
|
|
UErrorCode* err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
2001-01-10 02:19:33 +00:00
|
|
|
UChar myUCharBuf[2];
|
|
|
|
UChar *myUCharPtr;
|
|
|
|
const unsigned char *mySource;
|
2001-01-04 22:00:02 +00:00
|
|
|
UChar32 myUChar;
|
2001-01-10 02:19:33 +00:00
|
|
|
int32_t length;
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
while (args->source < args->sourceLimit)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
2001-01-10 02:19:33 +00:00
|
|
|
if (args->source + 4 > args->sourceLimit)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
/* got a partial character */
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
2001-01-10 02:19:33 +00:00
|
|
|
return 0xffff;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
/* Don't even try to do a direct cast because the value may be on an odd address. */
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myUChar = (mySource[0] << 24)
|
|
|
|
| (mySource[1] << 16)
|
|
|
|
| (mySource[2] << 8)
|
|
|
|
| (mySource[3]);
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
args->source = (const char *)(mySource + 4);
|
2001-01-30 00:52:24 +00:00
|
|
|
if (myUChar <= MAXIMUM_UTF && myUChar >= 0) {
|
2001-01-10 02:19:33 +00:00
|
|
|
return myUChar;
|
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
uprv_memcpy(args->converter->invalidCharBuffer, mySource, 4);
|
|
|
|
args->converter->invalidCharLength = 4;
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
myUCharPtr = myUCharBuf;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
args->target = myUCharPtr;
|
|
|
|
args->targetLimit = myUCharBuf + 2;
|
2001-01-04 22:00:02 +00:00
|
|
|
args->converter->fromCharErrorBehaviour(args->converter->toUContext,
|
|
|
|
args,
|
2001-01-10 02:19:33 +00:00
|
|
|
(const char *)mySource,
|
|
|
|
4,
|
2001-01-04 22:00:02 +00:00
|
|
|
UCNV_ILLEGAL,
|
|
|
|
err);
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
if(U_SUCCESS(*err)) {
|
|
|
|
length = (uint16_t)(args->target - myUCharBuf);
|
|
|
|
if(length > 0) {
|
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, myUCharBuf, length);
|
|
|
|
}
|
|
|
|
/* else (callback did not write anything) continue */
|
|
|
|
} else if(*err == U_BUFFER_OVERFLOW_ERROR) {
|
2001-01-04 22:00:02 +00:00
|
|
|
*err = U_ZERO_ERROR;
|
2001-01-10 02:19:33 +00:00
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, myUCharBuf, 2);
|
|
|
|
} else {
|
|
|
|
/* break on error */
|
|
|
|
/* ### what if a callback set an error but _also_ generated output?! */
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
/* no input or only skipping callbacks */
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
|
|
|
return 0xffff;
|
2000-11-16 17:20:03 +00:00
|
|
|
}
|
2000-11-16 18:21:01 +00:00
|
|
|
|
2000-11-16 17:20:03 +00:00
|
|
|
static const UConverterImpl _UTF32BEImpl = {
|
|
|
|
UCNV_UTF32_BigEndian,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
T_UConverter_toUnicode_UTF32_BE,
|
2001-01-04 22:00:02 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_BE_OFFSET_LOGIC,
|
2000-11-16 17:20:03 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_BE,
|
2001-01-04 22:00:02 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_BE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_getNextUChar_UTF32_BE,
|
2000-11-16 17:20:03 +00:00
|
|
|
|
2000-11-27 17:40:41 +00:00
|
|
|
NULL,
|
2000-11-16 17:20:03 +00:00
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
2001-12-27 18:23:04 +00:00
|
|
|
/* The 1232 CCSID refers to any version of Unicode with any endianess of UTF-32 */
|
2001-09-06 18:34:57 +00:00
|
|
|
static const UConverterStaticData _UTF32BEStaticData = {
|
2000-12-11 23:45:39 +00:00
|
|
|
sizeof(UConverterStaticData),
|
2001-09-28 00:29:40 +00:00
|
|
|
"UTF-32BE",
|
|
|
|
1232,
|
2000-12-11 23:45:39 +00:00
|
|
|
UCNV_IBM, UCNV_UTF32_BigEndian, 4, 4,
|
2000-11-16 17:20:03 +00:00
|
|
|
{ 0, 0, 0xff, 0xfd }, 4, FALSE, FALSE,
|
2000-12-19 23:07:50 +00:00
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2000-11-16 17:20:03 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
const UConverterSharedData _UTF32BEData = {
|
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
|
|
|
NULL, NULL, &_UTF32BEStaticData, FALSE, &_UTF32BEImpl,
|
|
|
|
0
|
|
|
|
};
|
|
|
|
|
|
|
|
/* UTF-32LE ---------------------------------------------------------- */
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_LE(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
|
|
|
uint32_t ch, i;
|
|
|
|
|
|
|
|
/* UTF-8 returns here for only non-offset, this needs to change.*/
|
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
|
|
|
|
2000-11-20 18:09:58 +00:00
|
|
|
/* Stores the previously calculated ch from a previous call*/
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;
|
2000-11-16 17:20:03 +00:00
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
|
|
|
while (i < sizeof(uint32_t))
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch |= ((uint8_t)(*mySource)) << (i * 8);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if (args->flush)
|
|
|
|
{
|
|
|
|
if (U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
2000-11-20 18:09:58 +00:00
|
|
|
{ /* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
2000-11-16 17:20:03 +00:00
|
|
|
args->converter->toULength = (int8_t) i;
|
|
|
|
}
|
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ch <= MAXIMUM_UTF)
|
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
|
|
|
{
|
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (T_UConverter_toUnicode_InvalidChar_Callback(args, UCNV_ILLEGAL, err))
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
/* End of target buffer */
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_LE_OFFSET_LOGIC(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
const unsigned char *mySource = (unsigned char *) args->source;
|
|
|
|
UChar *myTarget = args->target;
|
|
|
|
int32_t *myOffsets = args->offsets;
|
|
|
|
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
|
|
|
|
const UChar *targetLimit = args->targetLimit;
|
|
|
|
unsigned char *toUBytes = args->converter->toUBytes;
|
|
|
|
uint32_t ch, i;
|
|
|
|
int32_t offsetNum = 0;
|
|
|
|
|
|
|
|
/* UTF-8 returns here for only non-offset, this needs to change.*/
|
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
|
|
|
|
|
|
|
/* Stores the previously calculated ch from a previous call*/
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
|
|
|
while (i < sizeof(uint32_t))
|
|
|
|
{
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch |= ((uint8_t)(*mySource)) << (i * 8);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if (args->flush)
|
|
|
|
{
|
|
|
|
if (U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{ /* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
|
|
|
}
|
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ch <= MAXIMUM_UTF)
|
|
|
|
{
|
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
|
|
|
{
|
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* write out the surrogates */
|
|
|
|
ch -= HALF_BASE;
|
|
|
|
*(myTarget++) = (UChar) ((ch >> HALF_SHIFT) + SURROGATE_HIGH_START);
|
2002-06-11 22:49:05 +00:00
|
|
|
*(myOffsets++) = offsetNum;
|
2001-01-04 22:00:02 +00:00
|
|
|
ch = (ch & HALF_MASK) + SURROGATE_LOW_START;
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
args->target = myTarget;
|
|
|
|
args->converter->invalidCharLength = (int8_t)i;
|
2001-01-26 02:54:57 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-10-12 18:54:09 +00:00
|
|
|
if (T_UConverter_toUnicode_InvalidChar_OffsetCallback(args, offsetNum, UCNV_ILLEGAL, err))
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
/* Stop if the error wasn't handled */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
args->converter->invalidCharLength = 0;
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myTarget = args->target;
|
2001-01-26 02:54:57 +00:00
|
|
|
myOffsets = args->offsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
2001-01-26 02:54:57 +00:00
|
|
|
offsetNum += i;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
/* End of target buffer */
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
2002-06-11 22:49:05 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_LE(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
UChar32 ch, ch2;
|
|
|
|
unsigned int indexToWrite;
|
|
|
|
unsigned char temp[sizeof(uint32_t)];
|
|
|
|
|
|
|
|
temp[3] = 0;
|
|
|
|
|
|
|
|
if (args->converter->fromUnicodeStatus)
|
|
|
|
{
|
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
|
|
|
if (SURROGATE_HIGH_START <= ch && ch < SURROGATE_LOW_START)
|
|
|
|
{
|
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (SURROGATE_LOW_START <= ch2 && ch2 <= SURROGATE_LOW_END)
|
|
|
|
{
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (!args->flush)
|
|
|
|
{
|
2000-11-20 18:09:58 +00:00
|
|
|
/* ran out of source */
|
2000-11-16 17:20:03 +00:00
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[2] = (uint8_t) (ch >> 16 & 0x1F);
|
2001-01-26 02:54:57 +00:00
|
|
|
temp[1] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[0] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
2000-11-16 17:20:03 +00:00
|
|
|
|
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2001-01-26 02:54:57 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2000-11-16 17:20:03 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_LE_OFFSET_LOGIC(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
|
|
|
unsigned char *myTarget = (unsigned char *) args->target;
|
|
|
|
int32_t *myOffsets = args->offsets;
|
|
|
|
const UChar *sourceLimit = args->sourceLimit;
|
|
|
|
const unsigned char *targetLimit = (unsigned char *) args->targetLimit;
|
|
|
|
UChar32 ch, ch2;
|
|
|
|
unsigned int indexToWrite;
|
|
|
|
unsigned char temp[sizeof(uint32_t)];
|
|
|
|
int32_t offsetNum = 0;
|
|
|
|
|
|
|
|
temp[3] = 0;
|
|
|
|
|
|
|
|
if (args->converter->fromUnicodeStatus)
|
|
|
|
{
|
|
|
|
ch = args->converter->fromUnicodeStatus;
|
|
|
|
args->converter->fromUnicodeStatus = 0;
|
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
|
|
|
if (SURROGATE_HIGH_START <= ch && ch < SURROGATE_LOW_START)
|
|
|
|
{
|
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
|
|
|
{
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (SURROGATE_LOW_START <= ch2 && ch2 <= SURROGATE_LOW_END)
|
|
|
|
{
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (!args->flush)
|
|
|
|
{
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUnicodeStatus = ch;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[2] = (uint8_t) (ch >> 16 & 0x1F);
|
2001-01-26 02:54:57 +00:00
|
|
|
temp[1] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[0] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
2001-01-04 22:00:02 +00:00
|
|
|
|
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
2001-01-26 02:54:57 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
offsetNum++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
2002-06-11 22:49:05 +00:00
|
|
|
args->offsets = myOffsets;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static UChar32
|
2001-02-01 17:49:08 +00:00
|
|
|
T_UConverter_getNextUChar_UTF32_LE(UConverterToUnicodeArgs* args,
|
|
|
|
UErrorCode* err)
|
2000-11-16 17:20:03 +00:00
|
|
|
{
|
2001-01-10 02:19:33 +00:00
|
|
|
UChar myUCharBuf[2];
|
|
|
|
UChar *myUCharPtr;
|
|
|
|
const unsigned char *mySource;
|
2001-01-04 22:00:02 +00:00
|
|
|
UChar32 myUChar;
|
2001-01-10 02:19:33 +00:00
|
|
|
int32_t length;
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
while (args->source < args->sourceLimit)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
2001-01-10 02:19:33 +00:00
|
|
|
if (args->source + 4 > args->sourceLimit)
|
2001-01-04 22:00:02 +00:00
|
|
|
{
|
|
|
|
/* got a partial character */
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
2001-01-10 02:19:33 +00:00
|
|
|
return 0xffff;
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
/* Don't even try to do a direct cast because the value may be on an odd address. */
|
|
|
|
mySource = (unsigned char *) args->source;
|
|
|
|
myUChar = (mySource[0])
|
|
|
|
| (mySource[1] << 8)
|
|
|
|
| (mySource[2] << 16)
|
|
|
|
| (mySource[3] << 24);
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
args->source = (const char *)(mySource + 4);
|
2001-01-30 00:52:24 +00:00
|
|
|
if (myUChar <= MAXIMUM_UTF && myUChar >= 0) {
|
2001-01-10 02:19:33 +00:00
|
|
|
return myUChar;
|
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
uprv_memcpy(args->converter->invalidCharBuffer, mySource, 4);
|
|
|
|
args->converter->invalidCharLength = 4;
|
2001-01-04 22:00:02 +00:00
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
myUCharPtr = myUCharBuf;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
args->target = myUCharPtr;
|
|
|
|
args->targetLimit = myUCharBuf + 2;
|
2001-01-04 22:00:02 +00:00
|
|
|
args->converter->fromCharErrorBehaviour(args->converter->toUContext,
|
|
|
|
args,
|
2001-01-10 02:19:33 +00:00
|
|
|
(const char *)mySource,
|
|
|
|
4,
|
2001-01-04 22:00:02 +00:00
|
|
|
UCNV_ILLEGAL,
|
|
|
|
err);
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
if(U_SUCCESS(*err)) {
|
|
|
|
length = (uint16_t)(args->target - myUCharBuf);
|
|
|
|
if(length > 0) {
|
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, myUCharBuf, length);
|
|
|
|
}
|
|
|
|
/* else (callback did not write anything) continue */
|
|
|
|
} else if(*err == U_BUFFER_OVERFLOW_ERROR) {
|
2001-01-04 22:00:02 +00:00
|
|
|
*err = U_ZERO_ERROR;
|
2001-01-10 02:19:33 +00:00
|
|
|
return ucnv_getUChar32KeepOverflow(args->converter, myUCharBuf, 2);
|
|
|
|
} else {
|
|
|
|
/* break on error */
|
|
|
|
/* ### what if a callback set an error but _also_ generated output?! */
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2001-01-04 22:00:02 +00:00
|
|
|
}
|
|
|
|
|
2001-01-10 02:19:33 +00:00
|
|
|
/* no input or only skipping callbacks */
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
|
|
|
return 0xffff;
|
2000-11-16 17:20:03 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF32LEImpl = {
|
|
|
|
UCNV_UTF32_LittleEndian,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
T_UConverter_toUnicode_UTF32_LE,
|
2001-01-04 22:00:02 +00:00
|
|
|
T_UConverter_toUnicode_UTF32_LE_OFFSET_LOGIC,
|
2000-11-16 17:20:03 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_LE,
|
2001-01-04 22:00:02 +00:00
|
|
|
T_UConverter_fromUnicode_UTF32_LE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_getNextUChar_UTF32_LE,
|
2000-11-16 17:20:03 +00:00
|
|
|
|
2000-11-27 17:40:41 +00:00
|
|
|
NULL,
|
2000-11-16 17:20:03 +00:00
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
2001-12-27 18:23:04 +00:00
|
|
|
/* The 1232 CCSID refers to any version of Unicode with any endianess of UTF-32 */
|
2001-09-06 18:34:57 +00:00
|
|
|
static const UConverterStaticData _UTF32LEStaticData = {
|
2000-12-11 23:45:39 +00:00
|
|
|
sizeof(UConverterStaticData),
|
2001-09-28 00:29:40 +00:00
|
|
|
"UTF-32LE",
|
|
|
|
1232,
|
2002-03-27 23:24:15 +00:00
|
|
|
UCNV_IBM, UCNV_UTF32_LittleEndian, 4, 4,
|
2000-11-16 17:20:03 +00:00
|
|
|
{ 0xfd, 0xff, 0, 0 }, 4, FALSE, FALSE,
|
2000-12-19 23:07:50 +00:00
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2000-11-16 17:20:03 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
const UConverterSharedData _UTF32LEData = {
|
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
|
|
|
NULL, NULL, &_UTF32LEStaticData, FALSE, &_UTF32LEImpl,
|
|
|
|
0
|
|
|
|
};
|
2001-01-09 03:33:07 +00:00
|
|
|
|
2002-06-11 04:43:45 +00:00
|
|
|
/* UTF-16 (Detect BOM) ------------------------------------------------------ */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Detect a BOM at the beginning of the stream and select UTF-16BE or UTF-16LE
|
|
|
|
* accordingly.
|
|
|
|
* This is a simpler version of the UTF-32 converter below, with
|
|
|
|
* fewer states for shorter BOMs.
|
|
|
|
*
|
|
|
|
* State values:
|
|
|
|
* 0 initial state
|
|
|
|
* 1 saw FE
|
|
|
|
* 2..4 -
|
|
|
|
* 5 saw FF
|
|
|
|
* 6..7 -
|
|
|
|
* 8 UTF-16BE mode
|
|
|
|
* 9 UTF-16LE mode
|
|
|
|
*
|
|
|
|
* During detection: state&3==number of matching bytes so far.
|
|
|
|
*
|
|
|
|
* On output, emit U+FEFF as the first code point.
|
|
|
|
*/
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF16Reset(UConverter *cnv, UConverterResetChoice choice) {
|
|
|
|
if(choice<=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset toUnicode: state=0 */
|
|
|
|
cnv->mode=0;
|
|
|
|
}
|
|
|
|
if(choice!=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset fromUnicode: prepare to output the UTF-16PE BOM */
|
|
|
|
cnv->charErrorBufferLength=2;
|
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
cnv->charErrorBuffer[0]=0xfe;
|
|
|
|
cnv->charErrorBuffer[1]=0xff;
|
|
|
|
#else
|
|
|
|
cnv->charErrorBuffer[0]=0xff;
|
|
|
|
cnv->charErrorBuffer[1]=0xfe;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF16Open(UConverter *cnv,
|
|
|
|
const char *name,
|
|
|
|
const char *locale,
|
|
|
|
uint32_t options,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
_UTF16Reset(cnv, UCNV_RESET_BOTH);
|
|
|
|
}
|
|
|
|
|
|
|
|
static const char utf16BOM[8]={ (char)0xfe, (char)0xff, 0, 0, (char)0xff, (char)0xfe, 0, 0 };
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF16ToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
UConverter *cnv=pArgs->converter;
|
|
|
|
const char *source=pArgs->source;
|
|
|
|
const char *sourceLimit=pArgs->sourceLimit;
|
|
|
|
int32_t *offsets=pArgs->offsets;
|
|
|
|
|
|
|
|
int32_t state, offsetDelta;
|
|
|
|
char b;
|
|
|
|
|
|
|
|
state=cnv->mode;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we detect a BOM in this buffer, then we must add the BOM size to the
|
|
|
|
* offsets because the actual converter function will not see and count the BOM.
|
|
|
|
* offsetDelta will have the number of the BOM bytes that are in the current buffer.
|
|
|
|
*/
|
|
|
|
offsetDelta=0;
|
|
|
|
|
|
|
|
while(source<sourceLimit && U_SUCCESS(*pErrorCode)) {
|
|
|
|
switch(state) {
|
|
|
|
case 0:
|
|
|
|
b=*source;
|
2002-06-11 18:00:41 +00:00
|
|
|
if(b==(char)0xfe) {
|
2002-06-11 04:43:45 +00:00
|
|
|
state=1; /* could be FE FF */
|
|
|
|
} else if(b==(char)0xff) {
|
|
|
|
state=5; /* could be FF FE */
|
|
|
|
} else {
|
|
|
|
state=8; /* default to UTF-16BE */
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
++source;
|
|
|
|
break;
|
|
|
|
case 1:
|
|
|
|
case 5:
|
|
|
|
if(*source==utf16BOM[state]) {
|
|
|
|
++source;
|
|
|
|
if(state==1) {
|
|
|
|
state=8; /* detect UTF-16BE */
|
|
|
|
offsetDelta=source-pArgs->source;
|
|
|
|
} else if(state==5) {
|
|
|
|
state=9; /* detect UTF-16LE */
|
|
|
|
offsetDelta=source-pArgs->source;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* switch to UTF-16BE and pass the previous bytes */
|
|
|
|
if(source!=pArgs->source) {
|
|
|
|
/* just reset the source */
|
|
|
|
source=pArgs->source;
|
|
|
|
} else {
|
|
|
|
UBool oldFlush=pArgs->flush;
|
|
|
|
|
|
|
|
/* the first byte is from a previous buffer, replay it first */
|
|
|
|
pArgs->source=utf16BOM+(state&4); /* select the correct BOM */
|
|
|
|
pArgs->sourceLimit=pArgs->source+1; /* replay previous byte */
|
|
|
|
pArgs->flush=FALSE; /* this sourceLimit is not the real source stream limit */
|
|
|
|
|
|
|
|
_UTF16BEToUnicodeWithOffsets(pArgs, pErrorCode);
|
|
|
|
|
|
|
|
/* restore real pointers; pArgs->source will be set in case 8/9 */
|
|
|
|
pArgs->sourceLimit=sourceLimit;
|
|
|
|
pArgs->flush=oldFlush;
|
|
|
|
}
|
2002-06-11 18:00:41 +00:00
|
|
|
state=8;
|
2002-06-11 04:43:45 +00:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case 8:
|
|
|
|
/* call UTF-16BE */
|
|
|
|
pArgs->source=source;
|
|
|
|
_UTF16BEToUnicodeWithOffsets(pArgs, pErrorCode);
|
|
|
|
source=pArgs->source;
|
|
|
|
break;
|
|
|
|
case 9:
|
|
|
|
/* call UTF-16LE */
|
|
|
|
pArgs->source=source;
|
|
|
|
_UTF16LEToUnicodeWithOffsets(pArgs, pErrorCode);
|
|
|
|
source=pArgs->source;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break; /* does not occur */
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* add BOM size to offsets - see comment at offsetDelta declaration */
|
|
|
|
if(offsets!=NULL && offsetDelta!=0) {
|
|
|
|
int32_t *offsetsLimit=pArgs->offsets;
|
|
|
|
while(offsets<offsetsLimit) {
|
|
|
|
*offsets++ += offsetDelta;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2002-06-11 20:44:36 +00:00
|
|
|
if(source==sourceLimit && pArgs->flush) {
|
|
|
|
/* handle truncated input */
|
|
|
|
switch(state) {
|
|
|
|
case 0:
|
|
|
|
break; /* no input at all, nothing to do */
|
|
|
|
case 8:
|
|
|
|
_UTF16BEToUnicodeWithOffsets(pArgs, pErrorCode);
|
|
|
|
break;
|
|
|
|
case 9:
|
|
|
|
_UTF16LEToUnicodeWithOffsets(pArgs, pErrorCode);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
/* handle 0<state<8: call UTF-16BE with too-short input */
|
2002-06-11 04:43:45 +00:00
|
|
|
pArgs->source=utf16BOM+(state&4); /* select the correct BOM */
|
|
|
|
pArgs->sourceLimit=pArgs->source+(state&3); /* replay bytes */
|
|
|
|
|
|
|
|
/* no offsets: not enough for output */
|
|
|
|
_UTF16BEToUnicodeWithOffsets(pArgs, pErrorCode);
|
2002-06-11 20:44:36 +00:00
|
|
|
/* pArgs->source restored below */
|
|
|
|
pArgs->sourceLimit=sourceLimit;
|
|
|
|
break;
|
2002-06-11 04:43:45 +00:00
|
|
|
}
|
|
|
|
cnv->mode=0; /* reset */
|
|
|
|
} else {
|
|
|
|
cnv->mode=state;
|
|
|
|
}
|
|
|
|
|
|
|
|
pArgs->source=source;
|
|
|
|
}
|
|
|
|
|
|
|
|
static UChar32
|
|
|
|
_UTF16GetNextUChar(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
switch(pArgs->converter->mode) {
|
|
|
|
case 8:
|
|
|
|
return T_UConverter_getNextUChar_UTF16_BE(pArgs, pErrorCode);
|
|
|
|
case 9:
|
|
|
|
return T_UConverter_getNextUChar_UTF16_LE(pArgs, pErrorCode);
|
|
|
|
default:
|
|
|
|
return ucnv_getNextUCharFromToUImpl(pArgs, _UTF16ToUnicodeWithOffsets, TRUE, pErrorCode);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF16Impl = {
|
|
|
|
UCNV_UTF16,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
_UTF16Open,
|
|
|
|
NULL,
|
|
|
|
_UTF16Reset,
|
|
|
|
|
|
|
|
_UTF16ToUnicodeWithOffsets,
|
|
|
|
_UTF16ToUnicodeWithOffsets,
|
|
|
|
_UTF16PEFromUnicodeWithOffsets,
|
|
|
|
_UTF16PEFromUnicodeWithOffsets,
|
|
|
|
_UTF16GetNextUChar,
|
|
|
|
|
|
|
|
NULL, /* ### TODO implement getStarters for all Unicode encodings?! */
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static const UConverterStaticData _UTF16StaticData = {
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-16",
|
|
|
|
1200, /* ### TODO review correctness of all Unicode CCSIDs */
|
|
|
|
UCNV_IBM, UCNV_UTF16, 2, 2,
|
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
{ 0xff, 0xfd, 0, 0 }, 2,
|
|
|
|
#else
|
|
|
|
{ 0xfd, 0xff, 0, 0 }, 2,
|
|
|
|
#endif
|
|
|
|
FALSE, FALSE,
|
|
|
|
0,
|
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
|
|
|
};
|
|
|
|
|
|
|
|
const UConverterSharedData _UTF16Data = {
|
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
|
|
|
NULL, NULL, &_UTF16StaticData, FALSE, &_UTF16Impl,
|
|
|
|
0
|
|
|
|
};
|
|
|
|
|
|
|
|
/* UTF-32 (Detect BOM) ------------------------------------------------------ */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Detect a BOM at the beginning of the stream and select UTF-32BE or UTF-32LE
|
|
|
|
* accordingly.
|
|
|
|
*
|
|
|
|
* State values:
|
|
|
|
* 0 initial state
|
|
|
|
* 1 saw 00
|
|
|
|
* 2 saw 00 00
|
|
|
|
* 3 saw 00 00 FE
|
|
|
|
* 4 -
|
|
|
|
* 5 saw FF
|
|
|
|
* 6 saw FF FE
|
|
|
|
* 7 saw FF FE 00
|
|
|
|
* 8 UTF-32BE mode
|
|
|
|
* 9 UTF-32LE mode
|
|
|
|
*
|
|
|
|
* During detection: state&3==number of matching bytes so far.
|
|
|
|
*
|
|
|
|
* On output, emit U+FEFF as the first code point.
|
|
|
|
*/
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF32Reset(UConverter *cnv, UConverterResetChoice choice) {
|
|
|
|
if(choice<=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset toUnicode: state=0 */
|
|
|
|
cnv->mode=0;
|
|
|
|
}
|
|
|
|
if(choice!=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset fromUnicode: prepare to output the UTF-32PE BOM */
|
|
|
|
cnv->charErrorBufferLength=4;
|
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
cnv->charErrorBuffer[0]=0;
|
|
|
|
cnv->charErrorBuffer[1]=0;
|
|
|
|
cnv->charErrorBuffer[2]=0xfe;
|
|
|
|
cnv->charErrorBuffer[3]=0xff;
|
|
|
|
#else
|
|
|
|
cnv->charErrorBuffer[0]=0xff;
|
|
|
|
cnv->charErrorBuffer[1]=0xfe;
|
|
|
|
cnv->charErrorBuffer[2]=0;
|
|
|
|
cnv->charErrorBuffer[3]=0;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF32Open(UConverter *cnv,
|
|
|
|
const char *name,
|
|
|
|
const char *locale,
|
|
|
|
uint32_t options,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
_UTF32Reset(cnv, UCNV_RESET_BOTH);
|
|
|
|
}
|
|
|
|
|
|
|
|
static const char utf32BOM[8]={ 0, 0, (char)0xfe, (char)0xff, (char)0xff, (char)0xfe, 0, 0 };
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF32ToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
UConverter *cnv=pArgs->converter;
|
|
|
|
const char *source=pArgs->source;
|
|
|
|
const char *sourceLimit=pArgs->sourceLimit;
|
|
|
|
int32_t *offsets=pArgs->offsets;
|
|
|
|
|
|
|
|
int32_t state, offsetDelta;
|
|
|
|
char b;
|
|
|
|
|
|
|
|
state=cnv->mode;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we detect a BOM in this buffer, then we must add the BOM size to the
|
|
|
|
* offsets because the actual converter function will not see and count the BOM.
|
|
|
|
* offsetDelta will have the number of the BOM bytes that are in the current buffer.
|
|
|
|
*/
|
|
|
|
offsetDelta=0;
|
|
|
|
|
|
|
|
while(source<sourceLimit && U_SUCCESS(*pErrorCode)) {
|
|
|
|
switch(state) {
|
|
|
|
case 0:
|
|
|
|
b=*source;
|
|
|
|
if(b==0) {
|
|
|
|
state=1; /* could be 00 00 FE FF */
|
|
|
|
} else if(b==(char)0xff) {
|
|
|
|
state=5; /* could be FF FE 00 00 */
|
|
|
|
} else {
|
|
|
|
state=8; /* default to UTF-32BE */
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
++source;
|
|
|
|
break;
|
|
|
|
case 1:
|
|
|
|
case 2:
|
|
|
|
case 3:
|
|
|
|
case 5:
|
|
|
|
case 6:
|
|
|
|
case 7:
|
|
|
|
if(*source==utf32BOM[state]) {
|
|
|
|
++state;
|
|
|
|
++source;
|
|
|
|
if(state==4) {
|
|
|
|
state=8; /* detect UTF-32BE */
|
|
|
|
offsetDelta=source-pArgs->source;
|
|
|
|
} else if(state==8) {
|
|
|
|
state=9; /* detect UTF-32LE */
|
|
|
|
offsetDelta=source-pArgs->source;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* switch to UTF-32BE and pass the previous bytes */
|
|
|
|
int32_t count=source-pArgs->source; /* number of bytes from this buffer */
|
|
|
|
|
|
|
|
/* reset the source */
|
|
|
|
source=pArgs->source;
|
|
|
|
|
|
|
|
if(count==(state&3)) {
|
|
|
|
/* simple: all in the same buffer, just reset source */
|
|
|
|
} else {
|
|
|
|
UBool oldFlush=pArgs->flush;
|
|
|
|
|
|
|
|
/* some of the bytes are from a previous buffer, replay those first */
|
|
|
|
pArgs->source=utf32BOM+(state&4); /* select the correct BOM */
|
|
|
|
pArgs->sourceLimit=pArgs->source+((state&3)-count); /* replay previous bytes */
|
|
|
|
pArgs->flush=FALSE; /* this sourceLimit is not the real source stream limit */
|
|
|
|
|
|
|
|
/* no offsets: bytes from previous buffer, and not enough for output */
|
|
|
|
T_UConverter_toUnicode_UTF32_BE(pArgs, pErrorCode);
|
|
|
|
|
|
|
|
/* restore real pointers; pArgs->source will be set in case 8/9 */
|
|
|
|
pArgs->sourceLimit=sourceLimit;
|
|
|
|
pArgs->flush=oldFlush;
|
|
|
|
}
|
2002-06-11 18:00:41 +00:00
|
|
|
state=8;
|
2002-06-11 04:43:45 +00:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case 8:
|
|
|
|
/* call UTF-32BE */
|
|
|
|
pArgs->source=source;
|
|
|
|
if(offsets==NULL) {
|
|
|
|
T_UConverter_toUnicode_UTF32_BE(pArgs, pErrorCode);
|
|
|
|
} else {
|
|
|
|
T_UConverter_toUnicode_UTF32_BE_OFFSET_LOGIC(pArgs, pErrorCode);
|
|
|
|
}
|
|
|
|
source=pArgs->source;
|
|
|
|
break;
|
|
|
|
case 9:
|
|
|
|
/* call UTF-32LE */
|
|
|
|
pArgs->source=source;
|
|
|
|
if(offsets==NULL) {
|
|
|
|
T_UConverter_toUnicode_UTF32_LE(pArgs, pErrorCode);
|
|
|
|
} else {
|
|
|
|
T_UConverter_toUnicode_UTF32_LE_OFFSET_LOGIC(pArgs, pErrorCode);
|
|
|
|
}
|
|
|
|
source=pArgs->source;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break; /* does not occur */
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* add BOM size to offsets - see comment at offsetDelta declaration */
|
|
|
|
if(offsets!=NULL && offsetDelta!=0) {
|
|
|
|
int32_t *offsetsLimit=pArgs->offsets;
|
|
|
|
while(offsets<offsetsLimit) {
|
|
|
|
*offsets++ += offsetDelta;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2002-06-11 20:44:36 +00:00
|
|
|
if(source==sourceLimit && pArgs->flush) {
|
|
|
|
/* handle truncated input */
|
|
|
|
switch(state) {
|
|
|
|
case 0:
|
|
|
|
break; /* no input at all, nothing to do */
|
|
|
|
case 8:
|
|
|
|
T_UConverter_toUnicode_UTF32_BE(pArgs, pErrorCode);
|
|
|
|
break;
|
|
|
|
case 9:
|
|
|
|
T_UConverter_toUnicode_UTF32_LE(pArgs, pErrorCode);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
/* handle 0<state<8: call UTF-32BE with too-short input */
|
2002-06-11 04:43:45 +00:00
|
|
|
pArgs->source=utf32BOM+(state&4); /* select the correct BOM */
|
|
|
|
pArgs->sourceLimit=pArgs->source+(state&3); /* replay bytes */
|
|
|
|
|
|
|
|
/* no offsets: not enough for output */
|
|
|
|
T_UConverter_toUnicode_UTF32_BE(pArgs, pErrorCode);
|
2002-06-11 20:44:36 +00:00
|
|
|
/* pArgs->source restored below */
|
|
|
|
pArgs->sourceLimit=sourceLimit;
|
|
|
|
break;
|
2002-06-11 04:43:45 +00:00
|
|
|
}
|
|
|
|
cnv->mode=0; /* reset */
|
|
|
|
} else {
|
|
|
|
cnv->mode=state;
|
|
|
|
}
|
|
|
|
|
|
|
|
pArgs->source=source;
|
|
|
|
}
|
|
|
|
|
|
|
|
static UChar32
|
|
|
|
_UTF32GetNextUChar(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
switch(pArgs->converter->mode) {
|
|
|
|
case 8:
|
|
|
|
return T_UConverter_getNextUChar_UTF32_BE(pArgs, pErrorCode);
|
|
|
|
case 9:
|
|
|
|
return T_UConverter_getNextUChar_UTF32_LE(pArgs, pErrorCode);
|
|
|
|
default:
|
|
|
|
return ucnv_getNextUCharFromToUImpl(pArgs, _UTF32ToUnicodeWithOffsets, FALSE, pErrorCode);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF32Impl = {
|
|
|
|
UCNV_UTF32,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
_UTF32Open,
|
|
|
|
NULL,
|
|
|
|
_UTF32Reset,
|
|
|
|
|
|
|
|
_UTF32ToUnicodeWithOffsets,
|
|
|
|
_UTF32ToUnicodeWithOffsets,
|
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE,
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE_OFFSET_LOGIC,
|
|
|
|
#else
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE,
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE_OFFSET_LOGIC,
|
|
|
|
#endif
|
|
|
|
_UTF32GetNextUChar,
|
|
|
|
|
|
|
|
NULL, /* ### TODO implement getStarters for all Unicode encodings?! */
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static const UConverterStaticData _UTF32StaticData = {
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-32",
|
|
|
|
1232, /* ### TODO review correctness of all Unicode CCSIDs */
|
|
|
|
UCNV_IBM, UCNV_UTF32, 4, 4,
|
|
|
|
#if U_IS_BIG_ENDIAN
|
|
|
|
{ 0, 0, 0xff, 0xfd }, 4,
|
|
|
|
#else
|
|
|
|
{ 0xfd, 0xff, 0, 0 }, 4,
|
|
|
|
#endif
|
|
|
|
FALSE, FALSE,
|
|
|
|
0,
|
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
|
|
|
};
|
|
|
|
|
|
|
|
const UConverterSharedData _UTF32Data = {
|
|
|
|
sizeof(UConverterSharedData), ~((uint32_t) 0),
|
|
|
|
NULL, NULL, &_UTF32StaticData, FALSE, &_UTF32Impl,
|
|
|
|
0
|
|
|
|
};
|
|
|
|
|
2001-01-09 03:33:07 +00:00
|
|
|
/* UTF-7 -------------------------------------------------------------------- */
|
|
|
|
|
2002-06-11 04:43:45 +00:00
|
|
|
/* ### TODO: in convrtrs.txt and user guide, document version option (=1 for escaping set O characters) */
|
|
|
|
/* TODO: version=1 is not really for IMAP, fix documentation; consider version=2 for IMAP */
|
2001-01-09 03:33:07 +00:00
|
|
|
/*
|
|
|
|
* UTF-7 is a stateful encoding of Unicode, somewhat like UTF7.
|
|
|
|
* It is defined in RFC 2152 http://www.imc.org/rfc2152 .
|
|
|
|
* It was intended for use in Internet email systems, using in its bytewise
|
|
|
|
* encoding only a subset of 7-bit US-ASCII.
|
|
|
|
* UTF-7 is deprecated in favor of UTF-8/16/32 and UTF7, but still
|
|
|
|
* occasionally used.
|
2001-01-09 22:55:37 +00:00
|
|
|
*
|
|
|
|
* For converting Unicode to UTF-7, the RFC allows to encode some US-ASCII
|
|
|
|
* characters directly or in base64. Especially, the characters in set O
|
|
|
|
* as defined in the RFC (see below) may be encoded directly but are not
|
|
|
|
* allowed in, e.g., email headers.
|
|
|
|
* By default, the ICU UTF-7 converter encodes set O directly.
|
|
|
|
* By choosing the option "version=1", set O will be escaped instead.
|
|
|
|
* For example:
|
|
|
|
* utf7Converter=ucnv_open("UTF-7,version=1");
|
2001-01-09 03:33:07 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Tests for US-ASCII characters belonging to character classes
|
|
|
|
* defined in UTF-7.
|
|
|
|
*
|
|
|
|
* Set D (directly encoded characters) consists of the following
|
|
|
|
* characters: the upper and lower case letters A through Z
|
|
|
|
* and a through z, the 10 digits 0-9, and the following nine special
|
|
|
|
* characters (note that "+" and "=" are omitted):
|
|
|
|
* '(),-./:?
|
|
|
|
*
|
|
|
|
* Set O (optional direct characters) consists of the following
|
|
|
|
* characters (note that "\" and "~" are omitted):
|
|
|
|
* !"#$%&*;<=>@[]^_`{|}
|
2001-01-09 22:55:37 +00:00
|
|
|
*
|
|
|
|
* According to the rules in RFC 2152, the byte values for the following
|
|
|
|
* US-ASCII characters are not used in UTF-7 and are therefore illegal:
|
|
|
|
* - all C0 control codes except for CR LF TAB
|
|
|
|
* - BACKSLASH
|
|
|
|
* - TILDE
|
|
|
|
* - DEL
|
|
|
|
* - all codes beyond US-ASCII, i.e. all >127
|
2001-01-09 03:33:07 +00:00
|
|
|
*/
|
|
|
|
#define inSetD(c) \
|
|
|
|
((uint8_t)((c)-97)<26 || (uint8_t)((c)-65)<26 || /* letters */ \
|
2001-01-09 22:55:37 +00:00
|
|
|
(uint8_t)((c)-48)<10 || /* digits */ \
|
2001-01-09 03:33:07 +00:00
|
|
|
(uint8_t)((c)-39)<3 || /* '() */ \
|
|
|
|
(uint8_t)((c)-44)<4 || /* ,-./ */ \
|
|
|
|
(c)==58 || (c)==63 /* :? */ \
|
|
|
|
)
|
|
|
|
|
|
|
|
#define inSetO(c) \
|
|
|
|
((uint8_t)((c)-33)<6 || /* !"#$%& */ \
|
|
|
|
(uint8_t)((c)-59)<4 || /* ;<=> */ \
|
|
|
|
(uint8_t)((c)-93)<4 || /* ]^_` */ \
|
|
|
|
(uint8_t)((c)-123)<3 || /* {|} */ \
|
|
|
|
(c)==42 || (c)==64 || (c)==91 /* *@[ */ \
|
|
|
|
)
|
|
|
|
|
|
|
|
#define isCRLFTAB(c) ((c)==13 || (c)==10 || (c)==9)
|
|
|
|
#define isCRLFSPTAB(c) ((c)==32 || (c)==13 || (c)==10 || (c)==9)
|
|
|
|
|
|
|
|
#define PLUS 43
|
|
|
|
#define MINUS 45
|
|
|
|
#define BACKSLASH 92
|
|
|
|
#define TILDE 126
|
|
|
|
|
2001-01-09 22:55:37 +00:00
|
|
|
/* legal byte values: all US-ASCII graphic characters from space to before tilde, and CR LF TAB */
|
|
|
|
#define isLegalUTF7(c) (((uint8_t)((c)-32)<94 && (c)!=BACKSLASH) || isCRLFTAB(c))
|
|
|
|
|
|
|
|
/* encode directly sets D and O and CR LF SP TAB */
|
|
|
|
static const UBool encodeDirectlyMaximum[128]={
|
|
|
|
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
|
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 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, 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, 1, 1, 1, 1,
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 1, 0, 0
|
|
|
|
};
|
|
|
|
|
|
|
|
/* encode directly set D and CR LF SP TAB but not set O */
|
|
|
|
static const UBool encodeDirectlyRestricted[128]={
|
|
|
|
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
|
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0,
|
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
|
|
|
|
|
|
|
1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1,
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1,
|
|
|
|
|
|
|
|
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, 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, 0, 0, 0, 0, 0
|
|
|
|
};
|
|
|
|
|
2001-01-09 03:33:07 +00:00
|
|
|
static const uint8_t
|
|
|
|
toBase64[64]={
|
|
|
|
/* A-Z */
|
|
|
|
65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77,
|
|
|
|
78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90,
|
|
|
|
/* a-z */
|
|
|
|
97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109,
|
|
|
|
110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122,
|
|
|
|
/* 0-9 */
|
|
|
|
48, 49, 50, 51, 52, 53, 54, 55, 56, 57,
|
|
|
|
/* +/ */
|
|
|
|
43, 47
|
|
|
|
};
|
|
|
|
|
|
|
|
static const int8_t
|
|
|
|
fromBase64[128]={
|
2001-01-09 22:55:37 +00:00
|
|
|
/* C0 controls, -1 for legal ones (CR LF TAB), -3 for illegal ones */
|
|
|
|
-3, -3, -3, -3, -3, -3, -3, -3, -3, -1, -1, -3, -3, -1, -3, -3,
|
|
|
|
-3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3, -3,
|
2001-01-09 03:33:07 +00:00
|
|
|
|
|
|
|
/* general punctuation with + and / and a special value (-2) for - */
|
|
|
|
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -2, -1, 63,
|
|
|
|
/* digits */
|
|
|
|
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1,
|
|
|
|
|
|
|
|
/* A-Z */
|
|
|
|
-1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
|
2001-01-09 22:55:37 +00:00
|
|
|
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -3, -1, -1, -1,
|
2001-01-09 03:33:07 +00:00
|
|
|
|
|
|
|
/* a-z */
|
|
|
|
-1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
|
2001-01-09 22:55:37 +00:00
|
|
|
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -3, -3
|
2001-01-09 03:33:07 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* converter status values:
|
|
|
|
*
|
|
|
|
* toUnicodeStatus:
|
|
|
|
* 24 inDirectMode (boolean)
|
|
|
|
* 23..16 base64Counter (-1..7)
|
|
|
|
* 15..0 bits (up to 14 bits incoming base64)
|
|
|
|
*
|
|
|
|
* fromUnicodeStatus:
|
|
|
|
* 31..28 version (0: set O direct 1: set O escaped)
|
|
|
|
* 24 inDirectMode (boolean)
|
|
|
|
* 23..16 base64Counter (0..2)
|
|
|
|
* 7..0 bits (6 bits outgoing base64)
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-09 03:33:07 +00:00
|
|
|
_UTF7Reset(UConverter *cnv, UConverterResetChoice choice) {
|
|
|
|
if(choice<=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset toUnicode */
|
|
|
|
cnv->toUnicodeStatus=0x1000000; /* inDirectMode=TRUE */
|
2001-01-09 22:55:37 +00:00
|
|
|
cnv->toULength=0;
|
2001-01-09 03:33:07 +00:00
|
|
|
}
|
|
|
|
if(choice!=UCNV_RESET_TO_UNICODE) {
|
|
|
|
/* reset fromUnicode */
|
|
|
|
cnv->fromUnicodeStatus=(cnv->fromUnicodeStatus&0xf0000000)|0x1000000; /* keep version, inDirectMode=TRUE */
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-09 03:33:07 +00:00
|
|
|
_UTF7Open(UConverter *cnv,
|
|
|
|
const char *name,
|
|
|
|
const char *locale,
|
|
|
|
uint32_t options,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
if((options&0xf)<=1) {
|
|
|
|
cnv->fromUnicodeStatus=(options&0xf)<<28;
|
|
|
|
_UTF7Reset(cnv, UCNV_RESET_BOTH);
|
|
|
|
} else {
|
|
|
|
*pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-09 03:33:07 +00:00
|
|
|
_UTF7ToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
UConverter *cnv;
|
|
|
|
const uint8_t *source, *sourceLimit;
|
|
|
|
UChar *target;
|
|
|
|
const UChar *targetLimit;
|
|
|
|
int32_t *offsets;
|
|
|
|
|
|
|
|
uint8_t *bytes;
|
|
|
|
uint8_t byteIndex;
|
|
|
|
|
|
|
|
int32_t length, targetCapacity;
|
|
|
|
|
|
|
|
/* UTF-7 state */
|
|
|
|
uint16_t bits;
|
|
|
|
int8_t base64Counter;
|
|
|
|
UBool inDirectMode;
|
|
|
|
|
|
|
|
int8_t base64Value;
|
|
|
|
|
|
|
|
int32_t sourceIndex, nextSourceIndex;
|
|
|
|
|
|
|
|
uint8_t b;
|
|
|
|
|
|
|
|
/* set up the local pointers */
|
|
|
|
cnv=pArgs->converter;
|
|
|
|
|
|
|
|
source=(const uint8_t *)pArgs->source;
|
|
|
|
sourceLimit=(const uint8_t *)pArgs->sourceLimit;
|
|
|
|
target=pArgs->target;
|
|
|
|
targetLimit=pArgs->targetLimit;
|
|
|
|
offsets=pArgs->offsets;
|
|
|
|
|
|
|
|
/* get the state machine state */
|
|
|
|
{
|
|
|
|
uint32_t status=cnv->toUnicodeStatus;
|
|
|
|
inDirectMode=(UBool)((status>>24)&1);
|
|
|
|
base64Counter=(int8_t)(status>>16);
|
|
|
|
bits=(uint16_t)status;
|
|
|
|
}
|
|
|
|
bytes=cnv->toUBytes;
|
|
|
|
byteIndex=cnv->toULength;
|
|
|
|
|
|
|
|
/* sourceIndex=-1 if the current character began in the previous buffer */
|
|
|
|
sourceIndex=byteIndex==0 ? 0 : -1;
|
|
|
|
nextSourceIndex=0;
|
|
|
|
|
|
|
|
loop:
|
|
|
|
if(inDirectMode) {
|
|
|
|
directMode:
|
|
|
|
/*
|
|
|
|
* In Direct Mode, most US-ASCII characters are encoded directly, i.e.,
|
|
|
|
* with their US-ASCII byte values.
|
|
|
|
* Backslash and Tilde and most control characters are not allowed in UTF-7.
|
|
|
|
* A plus sign starts Unicode (or "escape") Mode.
|
|
|
|
*
|
|
|
|
* In Direct Mode, only the sourceIndex is used.
|
|
|
|
*/
|
|
|
|
byteIndex=0;
|
|
|
|
length=sourceLimit-source;
|
|
|
|
targetCapacity=targetLimit-target;
|
|
|
|
if(length>targetCapacity) {
|
|
|
|
length=targetCapacity;
|
|
|
|
}
|
|
|
|
while(length>0) {
|
|
|
|
b=*source++;
|
2001-01-09 22:55:37 +00:00
|
|
|
if(!isLegalUTF7(b)) {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* illegal */
|
|
|
|
bytes[0]=b;
|
|
|
|
byteIndex=1;
|
|
|
|
nextSourceIndex=sourceIndex+1;
|
|
|
|
goto callback;
|
|
|
|
} else if(b!=PLUS) {
|
|
|
|
/* write directly encoded character */
|
|
|
|
*target++=b;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
2001-01-09 22:55:37 +00:00
|
|
|
} else /* PLUS */ {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* switch to Unicode mode */
|
|
|
|
nextSourceIndex=++sourceIndex;
|
|
|
|
inDirectMode=FALSE;
|
|
|
|
byteIndex=0;
|
|
|
|
bits=0;
|
|
|
|
base64Counter=-1;
|
|
|
|
goto unicodeMode;
|
|
|
|
}
|
|
|
|
--length;
|
|
|
|
}
|
|
|
|
if(source<sourceLimit && target>=targetLimit) {
|
|
|
|
/* target is full */
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
unicodeMode:
|
|
|
|
/*
|
|
|
|
* In Unicode (or "escape") Mode, UTF-16BE is base64-encoded.
|
|
|
|
* The base64 sequence ends with any character that is not in the base64 alphabet.
|
|
|
|
* A terminating minus sign is consumed.
|
|
|
|
*
|
|
|
|
* In Unicode Mode, the sourceIndex has the index to the start of the current
|
|
|
|
* base64 bytes, while nextSourceIndex is precisely parallel to source,
|
|
|
|
* keeping the index to the following byte.
|
|
|
|
* Note that in 2 out of 3 cases, UChars overlap within a base64 byte.
|
|
|
|
*/
|
|
|
|
while(source<sourceLimit) {
|
|
|
|
if(target<targetLimit) {
|
|
|
|
bytes[byteIndex++]=b=*source++;
|
|
|
|
++nextSourceIndex;
|
2001-01-09 22:55:37 +00:00
|
|
|
if(b>=126) {
|
|
|
|
/* illegal - test other illegal US-ASCII values by base64Value==-3 */
|
2001-01-09 03:33:07 +00:00
|
|
|
inDirectMode=TRUE;
|
|
|
|
goto callback;
|
|
|
|
} else if((base64Value=fromBase64[b])>=0) {
|
|
|
|
/* collect base64 bytes into UChars */
|
|
|
|
switch(base64Counter) {
|
|
|
|
case -1: /* -1 is immediately after the + */
|
|
|
|
case 0:
|
|
|
|
bits=base64Value;
|
|
|
|
base64Counter=1;
|
|
|
|
break;
|
|
|
|
case 1:
|
|
|
|
case 3:
|
|
|
|
case 4:
|
|
|
|
case 6:
|
2001-02-02 23:19:57 +00:00
|
|
|
bits=(uint16_t)((bits<<6)|base64Value);
|
2001-01-09 03:33:07 +00:00
|
|
|
++base64Counter;
|
|
|
|
break;
|
|
|
|
case 2:
|
2001-02-02 23:19:57 +00:00
|
|
|
*target++=(UChar)((bits<<4)|(base64Value>>2));
|
2001-01-09 03:33:07 +00:00
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
sourceIndex=nextSourceIndex-1;
|
|
|
|
}
|
|
|
|
bytes[0]=b; /* keep this byte in case an error occurs */
|
|
|
|
byteIndex=1;
|
2001-02-02 23:19:57 +00:00
|
|
|
bits=(uint16_t)(base64Value&3);
|
2001-01-09 03:33:07 +00:00
|
|
|
base64Counter=3;
|
|
|
|
break;
|
|
|
|
case 5:
|
2001-02-02 23:19:57 +00:00
|
|
|
*target++=(UChar)((bits<<2)|(base64Value>>4));
|
2001-01-09 03:33:07 +00:00
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
sourceIndex=nextSourceIndex-1;
|
|
|
|
}
|
|
|
|
bytes[0]=b; /* keep this byte in case an error occurs */
|
|
|
|
byteIndex=1;
|
2001-02-02 23:19:57 +00:00
|
|
|
bits=(uint16_t)(base64Value&15);
|
2001-01-09 03:33:07 +00:00
|
|
|
base64Counter=6;
|
|
|
|
break;
|
|
|
|
case 7:
|
2001-02-02 23:19:57 +00:00
|
|
|
*target++=(UChar)((bits<<6)|base64Value);
|
2001-01-09 03:33:07 +00:00
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
sourceIndex=nextSourceIndex;
|
|
|
|
}
|
|
|
|
byteIndex=0;
|
|
|
|
bits=0;
|
|
|
|
base64Counter=0;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
/* will never occur */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} else if(base64Value==-2) {
|
|
|
|
/* minus sign terminates the base64 sequence */
|
|
|
|
inDirectMode=TRUE;
|
|
|
|
if(base64Counter==-1) {
|
|
|
|
/* +- i.e. a minus immediately following a plus */
|
|
|
|
*target++=PLUS;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex-1;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* absorb the minus and leave the Unicode Mode */
|
|
|
|
if(bits!=0) {
|
|
|
|
/* bits are illegally left over, a UChar is incomplete */
|
|
|
|
goto callback;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
sourceIndex=nextSourceIndex;
|
|
|
|
goto directMode;
|
2001-01-09 22:55:37 +00:00
|
|
|
} else if(base64Value==-1) /* for any legal character except base64 and minus sign */ {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* leave the Unicode Mode */
|
|
|
|
inDirectMode=TRUE;
|
2001-01-09 22:55:37 +00:00
|
|
|
if(base64Counter==-1) {
|
|
|
|
/* illegal: + immediately followed by something other than base64 or minus sign */
|
|
|
|
/* include the plus sign in the reported sequence */
|
|
|
|
--sourceIndex;
|
|
|
|
bytes[0]=PLUS;
|
|
|
|
bytes[1]=b;
|
|
|
|
byteIndex=2;
|
|
|
|
goto callback;
|
|
|
|
} else if(bits==0) {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* un-read the character in case it is a plus sign */
|
|
|
|
--source;
|
|
|
|
sourceIndex=nextSourceIndex-1;
|
|
|
|
goto directMode;
|
|
|
|
} else {
|
|
|
|
/* bits are illegally left over, a UChar is incomplete */
|
|
|
|
goto callback;
|
|
|
|
}
|
2001-01-09 22:55:37 +00:00
|
|
|
} else /* base64Value==-3 for illegal characters */ {
|
|
|
|
/* illegal */
|
|
|
|
inDirectMode=TRUE;
|
|
|
|
goto callback;
|
2001-01-09 03:33:07 +00:00
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* target is full */
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
endloop:
|
|
|
|
|
|
|
|
if(pArgs->flush && source>=sourceLimit) {
|
|
|
|
/* reset the state for the next conversion */
|
|
|
|
if(!inDirectMode && bits!=0 && U_SUCCESS(*pErrorCode)) {
|
|
|
|
/* a character byte sequence remains incomplete */
|
|
|
|
*pErrorCode=U_TRUNCATED_CHAR_FOUND;
|
|
|
|
}
|
|
|
|
cnv->toUnicodeStatus=0x1000000; /* inDirectMode=TRUE */
|
2001-01-09 22:55:37 +00:00
|
|
|
cnv->toULength=0;
|
2001-01-09 03:33:07 +00:00
|
|
|
} else {
|
|
|
|
/* set the converter state back into UConverter */
|
|
|
|
cnv->toUnicodeStatus=((uint32_t)inDirectMode<<24)|((uint32_t)((uint8_t)base64Counter)<<16)|(uint32_t)bits;
|
2001-01-09 22:55:37 +00:00
|
|
|
cnv->toULength=byteIndex;
|
2001-01-09 03:33:07 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
finish:
|
|
|
|
/* write back the updated pointers */
|
|
|
|
pArgs->source=(const char *)source;
|
|
|
|
pArgs->target=target;
|
|
|
|
pArgs->offsets=offsets;
|
|
|
|
return;
|
|
|
|
|
|
|
|
callback:
|
|
|
|
/* call the callback function with all the preparations and post-processing */
|
|
|
|
/* update the arguments structure */
|
|
|
|
pArgs->source=(const char *)source;
|
|
|
|
pArgs->target=target;
|
|
|
|
pArgs->offsets=offsets;
|
|
|
|
|
|
|
|
/* copy the current bytes to invalidCharBuffer */
|
|
|
|
for(b=0; b<(uint8_t)byteIndex; ++b) {
|
|
|
|
cnv->invalidCharBuffer[b]=(char)bytes[b];
|
|
|
|
}
|
|
|
|
cnv->invalidCharLength=byteIndex;
|
|
|
|
|
|
|
|
/* set the converter state in UConverter to deal with the next character */
|
|
|
|
cnv->toUnicodeStatus=(uint32_t)inDirectMode<<24;
|
|
|
|
cnv->toULength=0;
|
|
|
|
|
|
|
|
/* call the callback function */
|
|
|
|
*pErrorCode=U_ILLEGAL_CHAR_FOUND;
|
|
|
|
cnv->fromCharErrorBehaviour(cnv->toUContext, pArgs, cnv->invalidCharBuffer, cnv->invalidCharLength, UCNV_ILLEGAL, pErrorCode);
|
|
|
|
|
|
|
|
/* get the converter state from UConverter */
|
|
|
|
{
|
|
|
|
uint32_t status=cnv->toUnicodeStatus;
|
|
|
|
inDirectMode=(UBool)((status>>24)&1);
|
|
|
|
base64Counter=(int8_t)(status>>16);
|
|
|
|
bits=(uint16_t)status;
|
|
|
|
}
|
|
|
|
byteIndex=cnv->toULength;
|
|
|
|
|
|
|
|
/* update target and deal with offsets if necessary */
|
|
|
|
offsets=ucnv_updateCallbackOffsets(offsets, pArgs->target-target, sourceIndex);
|
|
|
|
target=pArgs->target;
|
|
|
|
|
|
|
|
/* update the source pointer and index */
|
|
|
|
sourceIndex=nextSourceIndex+((const uint8_t *)pArgs->source-source);
|
|
|
|
source=(const uint8_t *)pArgs->source;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the callback overflowed the target, then we need to
|
|
|
|
* stop here with an overflow indication.
|
|
|
|
*/
|
|
|
|
if(*pErrorCode==U_BUFFER_OVERFLOW_ERROR) {
|
|
|
|
goto endloop;
|
|
|
|
} else if(cnv->UCharErrorBufferLength>0) {
|
|
|
|
/* target is full */
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
goto endloop;
|
|
|
|
} else if(U_FAILURE(*pErrorCode)) {
|
|
|
|
/* break on error */
|
|
|
|
cnv->toUnicodeStatus=0x1000000; /* inDirectMode=TRUE */
|
2001-01-09 22:55:37 +00:00
|
|
|
cnv->toULength=0;
|
2001-01-09 03:33:07 +00:00
|
|
|
goto finish;
|
|
|
|
} else {
|
|
|
|
goto loop;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static UChar32
|
2001-01-09 22:55:37 +00:00
|
|
|
_UTF7GetNextUChar(UConverterToUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
return ucnv_getNextUCharFromToUImpl(pArgs, _UTF7ToUnicodeWithOffsets, TRUE, pErrorCode);
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static void
|
2001-01-09 03:33:07 +00:00
|
|
|
_UTF7FromUnicodeWithOffsets(UConverterFromUnicodeArgs *pArgs,
|
|
|
|
UErrorCode *pErrorCode) {
|
|
|
|
UConverter *cnv;
|
|
|
|
const UChar *source, *sourceLimit;
|
|
|
|
uint8_t *target, *targetLimit;
|
|
|
|
int32_t *offsets;
|
|
|
|
|
|
|
|
int32_t length, targetCapacity, sourceIndex;
|
|
|
|
UChar c;
|
|
|
|
|
|
|
|
/* UTF-7 state */
|
2001-01-09 22:55:37 +00:00
|
|
|
const UBool *encodeDirectly;
|
2001-01-09 03:33:07 +00:00
|
|
|
uint8_t bits;
|
|
|
|
int8_t base64Counter;
|
2001-01-09 22:55:37 +00:00
|
|
|
UBool inDirectMode;
|
2001-01-09 03:33:07 +00:00
|
|
|
|
|
|
|
/* set up the local pointers */
|
|
|
|
cnv=pArgs->converter;
|
|
|
|
|
|
|
|
/* set up the local pointers */
|
|
|
|
source=pArgs->source;
|
|
|
|
sourceLimit=pArgs->sourceLimit;
|
|
|
|
target=(uint8_t *)pArgs->target;
|
|
|
|
targetLimit=(uint8_t *)pArgs->targetLimit;
|
|
|
|
offsets=pArgs->offsets;
|
|
|
|
|
|
|
|
/* get the state machine state */
|
|
|
|
{
|
|
|
|
uint32_t status=cnv->fromUnicodeStatus;
|
2001-01-09 22:55:37 +00:00
|
|
|
encodeDirectly= status<0x10000000 ? encodeDirectlyMaximum : encodeDirectlyRestricted;
|
2001-01-09 03:33:07 +00:00
|
|
|
inDirectMode=(UBool)((status>>24)&1);
|
|
|
|
base64Counter=(int8_t)(status>>16);
|
|
|
|
bits=(uint8_t)status;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* UTF-7 always encodes UTF-16 code units, therefore we need only a simple sourceIndex */
|
|
|
|
sourceIndex=0;
|
|
|
|
|
|
|
|
if(inDirectMode) {
|
|
|
|
directMode:
|
|
|
|
length=sourceLimit-source;
|
|
|
|
targetCapacity=targetLimit-target;
|
|
|
|
if(length>targetCapacity) {
|
|
|
|
length=targetCapacity;
|
|
|
|
}
|
|
|
|
while(length>0) {
|
|
|
|
c=*source++;
|
|
|
|
/* currently always encode CR LF SP TAB directly */
|
2001-01-09 22:55:37 +00:00
|
|
|
if(c<=127 && encodeDirectly[c]) {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* encode directly */
|
|
|
|
*target++=(uint8_t)c;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
} else if(c==PLUS) {
|
|
|
|
/* output +- for + */
|
|
|
|
*target++=PLUS;
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=MINUS;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
/* realign length and targetCapacity */
|
|
|
|
goto directMode;
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=MINUS;
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* un-read this character and switch to Unicode Mode */
|
|
|
|
--source;
|
|
|
|
*target++=PLUS;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
}
|
|
|
|
inDirectMode=FALSE;
|
|
|
|
base64Counter=0;
|
|
|
|
goto unicodeMode;
|
|
|
|
}
|
|
|
|
--length;
|
|
|
|
}
|
|
|
|
if(source<sourceLimit && target>=targetLimit) {
|
|
|
|
/* target is full */
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
unicodeMode:
|
|
|
|
while(source<sourceLimit) {
|
|
|
|
if(target<targetLimit) {
|
|
|
|
c=*source++;
|
2001-01-09 22:55:37 +00:00
|
|
|
if(c<=127 && encodeDirectly[c]) {
|
2001-01-09 03:33:07 +00:00
|
|
|
/* encode directly */
|
|
|
|
inDirectMode=TRUE;
|
|
|
|
|
|
|
|
/* trick: back out this character to make this easier */
|
|
|
|
--source;
|
|
|
|
|
|
|
|
/* terminate the base64 sequence */
|
|
|
|
if(base64Counter!=0) {
|
|
|
|
/* write remaining bits for the previous character */
|
|
|
|
*target++=toBase64[bits];
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex-1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if(fromBase64[c]!=-1) {
|
|
|
|
/* need to terminate with a minus */
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=MINUS;
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
cnv->charErrorBuffer[0]=MINUS;
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
goto directMode;
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* base64 this character:
|
|
|
|
* Output 2 or 3 base64 bytes for the remaining bits of the previous character
|
|
|
|
* and the bits of this character, each implicitly in UTF-16BE.
|
|
|
|
*
|
|
|
|
* Here, bits is an 8-bit variable because only 6 bits need to be kept from one
|
|
|
|
* character to the next. The actual 2 or 4 bits are shifted to the left edge
|
|
|
|
* of the 6-bits field 5..0 to make the termination of the base64 sequence easier.
|
|
|
|
*/
|
|
|
|
switch(base64Counter) {
|
|
|
|
case 0:
|
|
|
|
*target++=toBase64[c>>10];
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[(c>>4)&0x3f];
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[(c>>4)&0x3f];
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
bits=(uint8_t)((c&15)<<2);
|
|
|
|
base64Counter=1;
|
|
|
|
break;
|
|
|
|
case 1:
|
|
|
|
*target++=toBase64[bits|(c>>14)];
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[(c>>8)&0x3f];
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[(c>>2)&0x3f];
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[(c>>2)&0x3f];
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[(c>>8)&0x3f];
|
|
|
|
cnv->charErrorBuffer[1]=toBase64[(c>>2)&0x3f];
|
|
|
|
cnv->charErrorBufferLength=2;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
bits=(uint8_t)((c&3)<<4);
|
|
|
|
base64Counter=2;
|
|
|
|
break;
|
|
|
|
case 2:
|
|
|
|
*target++=toBase64[bits|(c>>12)];
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[(c>>6)&0x3f];
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[c&0x3f];
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex;
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[c&0x3f];
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex++;
|
|
|
|
}
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[(c>>6)&0x3f];
|
|
|
|
cnv->charErrorBuffer[1]=toBase64[c&0x3f];
|
|
|
|
cnv->charErrorBufferLength=2;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
bits=0;
|
|
|
|
base64Counter=0;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
/* will never occur */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* target is full */
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if(pArgs->flush && source>=sourceLimit) {
|
|
|
|
/* flush remaining bits to the target */
|
|
|
|
if(!inDirectMode && base64Counter!=0) {
|
|
|
|
if(target<targetLimit) {
|
|
|
|
*target++=toBase64[bits];
|
|
|
|
if(offsets!=NULL) {
|
|
|
|
*offsets++=sourceIndex-1;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
cnv->charErrorBuffer[0]=toBase64[bits];
|
|
|
|
cnv->charErrorBufferLength=1;
|
|
|
|
*pErrorCode=U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* reset the state for the next conversion */
|
|
|
|
cnv->fromUnicodeStatus=(cnv->fromUnicodeStatus&0xf0000000)|0x1000000; /* keep version, inDirectMode=TRUE */
|
|
|
|
} else {
|
|
|
|
/* set the converter state back into UConverter */
|
|
|
|
cnv->fromUnicodeStatus=
|
|
|
|
(cnv->fromUnicodeStatus&0xf0000000)| /* keep version*/
|
|
|
|
((uint32_t)inDirectMode<<24)|((uint32_t)base64Counter<<16)|(uint32_t)bits;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write back the updated pointers */
|
|
|
|
pArgs->source=source;
|
|
|
|
pArgs->target=(char *)target;
|
|
|
|
pArgs->offsets=offsets;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2001-09-06 18:34:57 +00:00
|
|
|
static const char *
|
2001-01-09 03:33:07 +00:00
|
|
|
_UTF7GetName(const UConverter *cnv) {
|
|
|
|
switch(cnv->fromUnicodeStatus>>28) {
|
|
|
|
case 1:
|
|
|
|
return "UTF-7,version=1";
|
|
|
|
default:
|
|
|
|
return "UTF-7";
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF7Impl={
|
|
|
|
UCNV_UTF7,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
_UTF7Open,
|
|
|
|
NULL,
|
|
|
|
_UTF7Reset,
|
|
|
|
|
|
|
|
_UTF7ToUnicodeWithOffsets,
|
|
|
|
_UTF7ToUnicodeWithOffsets,
|
|
|
|
_UTF7FromUnicodeWithOffsets,
|
|
|
|
_UTF7FromUnicodeWithOffsets,
|
2001-01-09 22:55:37 +00:00
|
|
|
_UTF7GetNextUChar,
|
2001-01-09 03:33:07 +00:00
|
|
|
|
|
|
|
NULL,
|
|
|
|
_UTF7GetName,
|
|
|
|
NULL /* we don't need writeSub() because we never call a callback at fromUnicode() */
|
|
|
|
};
|
|
|
|
|
|
|
|
static const UConverterStaticData _UTF7StaticData={
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-7",
|
2001-09-28 00:29:40 +00:00
|
|
|
0, /* TODO CCSID for UTF-7 */
|
2001-01-09 03:33:07 +00:00
|
|
|
UCNV_IBM, UCNV_UTF7,
|
|
|
|
1, 4,
|
|
|
|
{ 0x3f, 0, 0, 0 }, 1, /* the subchar is not used */
|
|
|
|
FALSE, FALSE,
|
|
|
|
0,
|
2001-02-26 19:51:20 +00:00
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
2001-01-09 03:33:07 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
const UConverterSharedData _UTF7Data={
|
|
|
|
sizeof(UConverterSharedData), ~((uint32_t)0),
|
|
|
|
NULL, NULL, &_UTF7StaticData, FALSE, &_UTF7Impl,
|
|
|
|
0
|
|
|
|
};
|