2002-07-01 19:32:52 +00:00
|
|
|
/*
|
|
|
|
**********************************************************************
|
2015-02-20 19:31:33 +00:00
|
|
|
* Copyright (C) 2002-2015, International Business Machines
|
2002-07-01 19:32:52 +00:00
|
|
|
* Corporation and others. All Rights Reserved.
|
|
|
|
**********************************************************************
|
|
|
|
* file name: ucnv_u32.c
|
|
|
|
* encoding: US-ASCII
|
|
|
|
* tab size: 8 (not used)
|
|
|
|
* indentation:4
|
|
|
|
*
|
|
|
|
* created on: 2002jul01
|
|
|
|
* created by: Markus W. Scherer
|
|
|
|
*
|
|
|
|
* UTF-32 converter implementation. Used to be in ucnv_utf.c.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "unicode/utypes.h"
|
2004-08-26 22:51:40 +00:00
|
|
|
|
2015-02-20 19:31:33 +00:00
|
|
|
#if !UCONFIG_NO_CONVERSION && !UCONFIG_ONLY_HTML_CONVERSION
|
2004-08-26 22:51:40 +00:00
|
|
|
|
2002-07-01 19:32:52 +00:00
|
|
|
#include "unicode/ucnv.h"
|
2011-07-27 05:53:56 +00:00
|
|
|
#include "unicode/utf.h"
|
2002-07-01 19:32:52 +00:00
|
|
|
#include "ucnv_bld.h"
|
|
|
|
#include "ucnv_cnv.h"
|
|
|
|
#include "cmemory.h"
|
|
|
|
|
|
|
|
#define MAXIMUM_UCS2 0x0000FFFF
|
|
|
|
#define MAXIMUM_UTF 0x0010FFFF
|
|
|
|
#define HALF_SHIFT 10
|
|
|
|
#define HALF_BASE 0x0010000
|
|
|
|
#define HALF_MASK 0x3FF
|
|
|
|
#define SURROGATE_HIGH_START 0xD800
|
|
|
|
#define SURROGATE_LOW_START 0xDC00
|
|
|
|
|
|
|
|
/* -SURROGATE_LOW_START + HALF_BASE */
|
|
|
|
#define SURROGATE_LOW_BASE 9216
|
|
|
|
|
2005-06-12 21:15:18 +00:00
|
|
|
enum {
|
|
|
|
UCNV_NEED_TO_WRITE_BOM=1
|
|
|
|
};
|
|
|
|
|
2002-07-01 19:32:52 +00:00
|
|
|
/* UTF-32BE ----------------------------------------------------------------- */
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_toUnicode_UTF32_BE(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
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;
|
|
|
|
|
2006-05-05 07:08:37 +00:00
|
|
|
/* Restore state of current sequence */
|
2003-10-02 17:02:14 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
2006-05-05 07:08:37 +00:00
|
|
|
args->converter->toULength = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
2003-10-02 17:02:14 +00:00
|
|
|
while (i < sizeof(uint32_t)) {
|
|
|
|
if (mySource < sourceLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
ch = (ch << 8) | (uint8_t)(*mySource);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2003-07-22 04:22:57 +00:00
|
|
|
/* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UTF && !U_IS_SURROGATE(ch)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
|
|
|
if (ch <= MAXIMUM_UCS2)
|
|
|
|
{
|
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* write out the surrogates */
|
2003-10-02 17:02:14 +00:00
|
|
|
*(myTarget++) = U16_LEAD(ch);
|
|
|
|
ch = U16_TRAIL(ch);
|
|
|
|
if (myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2003-08-01 14:58:43 +00:00
|
|
|
args->converter->toULength = (int8_t)i;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
donefornow:
|
2003-10-02 17:02:14 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* End of target buffer */
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = myTarget;
|
|
|
|
args->source = (const char *) mySource;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_toUnicode_UTF32_BE_OFFSET_LOGIC(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
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;
|
|
|
|
|
2006-05-05 07:08:37 +00:00
|
|
|
/* Restore state of current sequence */
|
2003-10-02 17:02:14 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
2006-05-05 07:08:37 +00:00
|
|
|
args->converter->toULength = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
ch = args->converter->toUnicodeStatus - 1;/*Stores the previously calculated ch from a previous call*/
|
|
|
|
args->converter->toUnicodeStatus = 0;
|
|
|
|
goto morebytes;
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
i = 0;
|
|
|
|
ch = 0;
|
|
|
|
morebytes:
|
2003-10-02 17:02:14 +00:00
|
|
|
while (i < sizeof(uint32_t)) {
|
|
|
|
if (mySource < sourceLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
ch = (ch << 8) | (uint8_t)(*mySource);
|
|
|
|
toUBytes[i++] = (char) *(mySource++);
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2003-07-22 04:22:57 +00:00
|
|
|
/* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UTF && !U_IS_SURROGATE(ch)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UCS2) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* write out the surrogates */
|
2003-10-02 17:02:14 +00:00
|
|
|
*(myTarget++) = U16_LEAD(ch);
|
2002-07-01 19:32:52 +00:00
|
|
|
*myOffsets++ = offsetNum;
|
2003-10-02 17:02:14 +00:00
|
|
|
ch = U16_TRAIL(ch);
|
2002-07-01 19:32:52 +00:00
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2003-08-01 14:58:43 +00:00
|
|
|
args->converter->toULength = (int8_t)i;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
offsetNum += i;
|
|
|
|
}
|
|
|
|
|
|
|
|
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;
|
|
|
|
args->offsets = myOffsets;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
2005-06-12 21:15:18 +00:00
|
|
|
unsigned char *myTarget;
|
2002-07-01 19:32:52 +00:00
|
|
|
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)];
|
|
|
|
|
2005-06-12 21:15:18 +00:00
|
|
|
if(mySource >= sourceLimit) {
|
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the BOM if necessary */
|
|
|
|
if(args->converter->fromUnicodeStatus==UCNV_NEED_TO_WRITE_BOM) {
|
|
|
|
static const char bom[]={ 0, 0, (char)0xfe, (char)0xff };
|
|
|
|
ucnv_fromUWriteBytes(args->converter,
|
|
|
|
bom, 4,
|
|
|
|
&args->target, args->targetLimit,
|
|
|
|
&args->offsets, -1,
|
|
|
|
err);
|
|
|
|
args->converter->fromUnicodeStatus=0;
|
|
|
|
}
|
|
|
|
|
|
|
|
myTarget = (unsigned char *) args->target;
|
2002-07-01 19:32:52 +00:00
|
|
|
temp[0] = 0;
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (args->converter->fromUChar32) {
|
2003-07-24 00:28:47 +00:00
|
|
|
ch = args->converter->fromUChar32;
|
|
|
|
args->converter->fromUChar32 = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
ch = *(mySource++);
|
|
|
|
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U_IS_SURROGATE(ch)) {
|
2003-10-02 17:02:14 +00:00
|
|
|
if (U_IS_LEAD(ch)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
lowsurogate:
|
2003-10-02 17:02:14 +00:00
|
|
|
if (mySource < sourceLimit) {
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (U_IS_TRAIL(ch2)) {
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
if (args->flush) {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
}
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
2003-07-24 00:28:47 +00:00
|
|
|
args->converter->fromUChar32 = ch;
|
2003-10-02 17:02:14 +00:00
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[1] = (uint8_t) (ch >> 16 & 0x1F);
|
|
|
|
temp[2] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[3] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++) {
|
|
|
|
if (myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE_OFFSET_LOGIC(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
2005-06-12 21:15:18 +00:00
|
|
|
unsigned char *myTarget;
|
|
|
|
int32_t *myOffsets;
|
2002-07-01 19:32:52 +00:00
|
|
|
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)];
|
|
|
|
|
2005-06-12 21:15:18 +00:00
|
|
|
if(mySource >= sourceLimit) {
|
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the BOM if necessary */
|
|
|
|
if(args->converter->fromUnicodeStatus==UCNV_NEED_TO_WRITE_BOM) {
|
|
|
|
static const char bom[]={ 0, 0, (char)0xfe, (char)0xff };
|
|
|
|
ucnv_fromUWriteBytes(args->converter,
|
|
|
|
bom, 4,
|
|
|
|
&args->target, args->targetLimit,
|
|
|
|
&args->offsets, -1,
|
|
|
|
err);
|
|
|
|
args->converter->fromUnicodeStatus=0;
|
|
|
|
}
|
|
|
|
|
|
|
|
myTarget = (unsigned char *) args->target;
|
|
|
|
myOffsets = args->offsets;
|
2002-07-01 19:32:52 +00:00
|
|
|
temp[0] = 0;
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (args->converter->fromUChar32) {
|
2003-07-24 00:28:47 +00:00
|
|
|
ch = args->converter->fromUChar32;
|
|
|
|
args->converter->fromUChar32 = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
ch = *(mySource++);
|
|
|
|
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U_IS_SURROGATE(ch)) {
|
2003-10-02 17:02:14 +00:00
|
|
|
if (U_IS_LEAD(ch)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
lowsurogate:
|
2003-10-02 17:02:14 +00:00
|
|
|
if (mySource < sourceLimit) {
|
|
|
|
ch2 = *mySource;
|
|
|
|
if (U_IS_TRAIL(ch2)) {
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
if (args->flush) {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
}
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
2003-07-24 00:28:47 +00:00
|
|
|
args->converter->fromUChar32 = ch;
|
2003-10-02 17:02:14 +00:00
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[1] = (uint8_t) (ch >> 16 & 0x1F);
|
|
|
|
temp[2] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[3] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++) {
|
|
|
|
if (myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
*(myOffsets++) = offsetNum;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
}
|
2004-12-23 21:03:30 +00:00
|
|
|
offsetNum = offsetNum + 1 + (temp[1] != 0);
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
args->offsets = myOffsets;
|
|
|
|
}
|
|
|
|
|
|
|
|
static UChar32
|
|
|
|
T_UConverter_getNextUChar_UTF32_BE(UConverterToUnicodeArgs* args,
|
|
|
|
UErrorCode* err)
|
|
|
|
{
|
2003-08-01 14:58:43 +00:00
|
|
|
const uint8_t *mySource;
|
2002-07-01 19:32:52 +00:00
|
|
|
UChar32 myUChar;
|
|
|
|
int32_t length;
|
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
mySource = (const uint8_t *)args->source;
|
|
|
|
if (mySource >= (const uint8_t *)args->sourceLimit)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-08-01 14:58:43 +00:00
|
|
|
/* no input */
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
length = (int32_t)((const uint8_t *)args->sourceLimit - mySource);
|
|
|
|
if (length < 4)
|
|
|
|
{
|
|
|
|
/* got a partial character */
|
|
|
|
uprv_memcpy(args->converter->toUBytes, mySource, length);
|
|
|
|
args->converter->toULength = (int8_t)length;
|
|
|
|
args->source = (const char *)(mySource + length);
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
/* Don't even try to do a direct cast because the value may be on an odd address. */
|
|
|
|
myUChar = ((UChar32)mySource[0] << 24)
|
|
|
|
| ((UChar32)mySource[1] << 16)
|
|
|
|
| ((UChar32)mySource[2] << 8)
|
|
|
|
| ((UChar32)mySource[3]);
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
args->source = (const char *)(mySource + 4);
|
2003-10-02 17:02:14 +00:00
|
|
|
if ((uint32_t)myUChar <= MAXIMUM_UTF && !U_IS_SURROGATE(myUChar)) {
|
2003-08-01 14:58:43 +00:00
|
|
|
return myUChar;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
uprv_memcpy(args->converter->toUBytes, mySource, 4);
|
|
|
|
args->converter->toULength = 4;
|
|
|
|
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
return 0xffff;
|
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF32BEImpl = {
|
|
|
|
UCNV_UTF32_BigEndian,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
T_UConverter_toUnicode_UTF32_BE,
|
|
|
|
T_UConverter_toUnicode_UTF32_BE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE,
|
|
|
|
T_UConverter_fromUnicode_UTF32_BE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_getNextUChar_UTF32_BE,
|
|
|
|
|
|
|
|
NULL,
|
2003-04-05 01:33:02 +00:00
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
2005-06-17 07:04:03 +00:00
|
|
|
ucnv_getNonSurrogateUnicodeSet
|
2002-07-01 19:32:52 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/* The 1232 CCSID refers to any version of Unicode with any endianess of UTF-32 */
|
|
|
|
static const UConverterStaticData _UTF32BEStaticData = {
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-32BE",
|
|
|
|
1232,
|
|
|
|
UCNV_IBM, UCNV_UTF32_BigEndian, 4, 4,
|
|
|
|
{ 0, 0, 0xff, 0xfd }, 4, FALSE, FALSE,
|
|
|
|
0,
|
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
|
|
|
};
|
|
|
|
|
2015-05-29 19:33:24 +00:00
|
|
|
const UConverterSharedData _UTF32BEData =
|
|
|
|
UCNV_IMMUTABLE_SHARED_DATA_INITIALIZER(&_UTF32BEStaticData, &_UTF32BEImpl);
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
/* UTF-32LE ---------------------------------------------------------- */
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_toUnicode_UTF32_LE(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
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;
|
|
|
|
|
2006-05-05 07:08:37 +00:00
|
|
|
/* Restore state of current sequence */
|
2002-07-01 19:32:52 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
2006-05-05 07:08:37 +00:00
|
|
|
args->converter->toULength = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
/* 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
|
|
|
|
{
|
2003-07-22 04:22:57 +00:00
|
|
|
/* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UTF && !U_IS_SURROGATE(ch)) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Normal valid byte when the loop has not prematurely terminated (i < inBytes) */
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UCS2) {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* fits in 16 bits */
|
|
|
|
*(myTarget++) = (UChar) ch;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* write out the surrogates */
|
2003-10-02 17:02:14 +00:00
|
|
|
*(myTarget++) = U16_LEAD(ch);
|
|
|
|
ch = U16_TRAIL(ch);
|
|
|
|
if (myTarget < targetLimit) {
|
2002-07-01 19:32:52 +00:00
|
|
|
*(myTarget++) = (UChar)ch;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* Put in overflow buffer (not handled here) */
|
|
|
|
args->converter->UCharErrorBuffer[0] = (UChar) ch;
|
|
|
|
args->converter->UCharErrorBufferLength = 1;
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2003-08-01 14:58:43 +00:00
|
|
|
args->converter->toULength = (int8_t)i;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
2002-07-01 19:32:52 +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;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_toUnicode_UTF32_LE_OFFSET_LOGIC(UConverterToUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
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;
|
|
|
|
|
2006-05-05 07:08:37 +00:00
|
|
|
/* Restore state of current sequence */
|
2002-07-01 19:32:52 +00:00
|
|
|
if (args->converter->toUnicodeStatus && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
i = args->converter->toULength; /* restore # of bytes consumed */
|
2006-05-05 07:08:37 +00:00
|
|
|
args->converter->toULength = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
/* 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
|
|
|
|
{
|
2003-07-22 04:22:57 +00:00
|
|
|
/* stores a partially calculated target*/
|
|
|
|
/* + 1 to make 0 a valid character */
|
|
|
|
args->converter->toUnicodeStatus = ch + 1;
|
|
|
|
args->converter->toULength = (int8_t) i;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto donefornow;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-10-02 17:02:14 +00:00
|
|
|
if (ch <= MAXIMUM_UTF && !U_IS_SURROGATE(ch))
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
|
|
|
/* 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;
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
2002-07-01 19:32:52 +00:00
|
|
|
/* write out the surrogates */
|
2003-10-02 17:02:14 +00:00
|
|
|
*(myTarget++) = U16_LEAD(ch);
|
2002-07-01 19:32:52 +00:00
|
|
|
*(myOffsets++) = offsetNum;
|
2003-10-02 17:02:14 +00:00
|
|
|
ch = U16_TRAIL(ch);
|
2002-07-01 19:32:52 +00:00
|
|
|
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
|
|
|
|
{
|
2003-08-01 14:58:43 +00:00
|
|
|
args->converter->toULength = (int8_t)i;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
offsetNum += i;
|
|
|
|
}
|
|
|
|
|
|
|
|
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;
|
|
|
|
args->offsets = myOffsets;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
2005-06-12 21:15:18 +00:00
|
|
|
unsigned char *myTarget;
|
2002-07-01 19:32:52 +00:00
|
|
|
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)];
|
|
|
|
|
2005-06-12 21:15:18 +00:00
|
|
|
if(mySource >= sourceLimit) {
|
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the BOM if necessary */
|
|
|
|
if(args->converter->fromUnicodeStatus==UCNV_NEED_TO_WRITE_BOM) {
|
|
|
|
static const char bom[]={ (char)0xff, (char)0xfe, 0, 0 };
|
|
|
|
ucnv_fromUWriteBytes(args->converter,
|
|
|
|
bom, 4,
|
|
|
|
&args->target, args->targetLimit,
|
|
|
|
&args->offsets, -1,
|
|
|
|
err);
|
|
|
|
args->converter->fromUnicodeStatus=0;
|
|
|
|
}
|
|
|
|
|
|
|
|
myTarget = (unsigned char *) args->target;
|
2002-07-01 19:32:52 +00:00
|
|
|
temp[3] = 0;
|
|
|
|
|
2003-07-24 00:28:47 +00:00
|
|
|
if (args->converter->fromUChar32)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-07-24 00:28:47 +00:00
|
|
|
ch = args->converter->fromUChar32;
|
|
|
|
args->converter->fromUChar32 = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U16_IS_SURROGATE(ch)) {
|
|
|
|
if (U16_IS_LEAD(ch))
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-10-02 17:02:14 +00:00
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-10-02 17:02:14 +00:00
|
|
|
ch2 = *mySource;
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U16_IS_TRAIL(ch2)) {
|
2003-10-02 17:02:14 +00:00
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
if (args->flush) {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
}
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
2003-07-24 00:28:47 +00:00
|
|
|
args->converter->fromUChar32 = ch;
|
2003-10-02 17:02:14 +00:00
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[2] = (uint8_t) (ch >> 16 & 0x1F);
|
|
|
|
temp[1] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[0] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
|
|
|
|
for (indexToWrite = 0; indexToWrite <= sizeof(uint32_t) - 1; indexToWrite++)
|
|
|
|
{
|
|
|
|
if (myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
*(myTarget++) = temp[indexToWrite];
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = temp[indexToWrite];
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE_OFFSET_LOGIC(UConverterFromUnicodeArgs * args,
|
|
|
|
UErrorCode * err)
|
|
|
|
{
|
|
|
|
const UChar *mySource = args->source;
|
2005-06-12 21:15:18 +00:00
|
|
|
unsigned char *myTarget;
|
|
|
|
int32_t *myOffsets;
|
2002-07-01 19:32:52 +00:00
|
|
|
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;
|
|
|
|
|
2005-06-12 21:15:18 +00:00
|
|
|
if(mySource >= sourceLimit) {
|
|
|
|
/* no input, nothing to do */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the BOM if necessary */
|
|
|
|
if(args->converter->fromUnicodeStatus==UCNV_NEED_TO_WRITE_BOM) {
|
|
|
|
static const char bom[]={ (char)0xff, (char)0xfe, 0, 0 };
|
|
|
|
ucnv_fromUWriteBytes(args->converter,
|
|
|
|
bom, 4,
|
|
|
|
&args->target, args->targetLimit,
|
|
|
|
&args->offsets, -1,
|
|
|
|
err);
|
|
|
|
args->converter->fromUnicodeStatus=0;
|
|
|
|
}
|
|
|
|
|
|
|
|
myTarget = (unsigned char *) args->target;
|
|
|
|
myOffsets = args->offsets;
|
2002-07-01 19:32:52 +00:00
|
|
|
temp[3] = 0;
|
|
|
|
|
2003-07-24 00:28:47 +00:00
|
|
|
if (args->converter->fromUChar32)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-07-24 00:28:47 +00:00
|
|
|
ch = args->converter->fromUChar32;
|
|
|
|
args->converter->fromUChar32 = 0;
|
2002-07-01 19:32:52 +00:00
|
|
|
goto lowsurogate;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (mySource < sourceLimit && myTarget < targetLimit)
|
|
|
|
{
|
|
|
|
ch = *(mySource++);
|
|
|
|
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U16_IS_SURROGATE(ch)) {
|
|
|
|
if (U16_IS_LEAD(ch))
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-10-02 17:02:14 +00:00
|
|
|
lowsurogate:
|
|
|
|
if (mySource < sourceLimit)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-10-02 17:02:14 +00:00
|
|
|
ch2 = *mySource;
|
2011-07-27 05:53:56 +00:00
|
|
|
if (U16_IS_TRAIL(ch2))
|
2003-10-02 17:02:14 +00:00
|
|
|
{
|
|
|
|
ch = ((ch - SURROGATE_HIGH_START) << HALF_SHIFT) + ch2 + SURROGATE_LOW_BASE;
|
|
|
|
mySource++;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* ran out of source */
|
|
|
|
args->converter->fromUChar32 = ch;
|
|
|
|
if (args->flush) {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
|
|
|
}
|
|
|
|
break;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
2003-10-02 17:02:14 +00:00
|
|
|
else {
|
|
|
|
/* this is an unmatched trail code unit (2nd surrogate) */
|
|
|
|
/* callback(illegal) */
|
2003-07-24 00:28:47 +00:00
|
|
|
args->converter->fromUChar32 = ch;
|
2003-10-02 17:02:14 +00:00
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We cannot get any larger than 10FFFF because we are coming from UTF-16 */
|
|
|
|
temp[2] = (uint8_t) (ch >> 16 & 0x1F);
|
|
|
|
temp[1] = (uint8_t) (ch >> 8); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
temp[0] = (uint8_t) (ch); /* unsigned cast implicitly does (ch & FF) */
|
|
|
|
|
|
|
|
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];
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
}
|
2004-12-23 21:03:30 +00:00
|
|
|
offsetNum = offsetNum + 1 + (temp[2] != 0);
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (mySource < sourceLimit && myTarget >= targetLimit && U_SUCCESS(*err))
|
|
|
|
{
|
|
|
|
*err = U_BUFFER_OVERFLOW_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
args->target = (char *) myTarget;
|
|
|
|
args->source = mySource;
|
|
|
|
args->offsets = myOffsets;
|
|
|
|
}
|
|
|
|
|
|
|
|
static UChar32
|
|
|
|
T_UConverter_getNextUChar_UTF32_LE(UConverterToUnicodeArgs* args,
|
|
|
|
UErrorCode* err)
|
|
|
|
{
|
2003-08-01 14:58:43 +00:00
|
|
|
const uint8_t *mySource;
|
2002-07-01 19:32:52 +00:00
|
|
|
UChar32 myUChar;
|
|
|
|
int32_t length;
|
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
mySource = (const uint8_t *)args->source;
|
|
|
|
if (mySource >= (const uint8_t *)args->sourceLimit)
|
2002-07-01 19:32:52 +00:00
|
|
|
{
|
2003-08-01 14:58:43 +00:00
|
|
|
/* no input */
|
|
|
|
*err = U_INDEX_OUTOFBOUNDS_ERROR;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
length = (int32_t)((const uint8_t *)args->sourceLimit - mySource);
|
|
|
|
if (length < 4)
|
|
|
|
{
|
|
|
|
/* got a partial character */
|
|
|
|
uprv_memcpy(args->converter->toUBytes, mySource, length);
|
|
|
|
args->converter->toULength = (int8_t)length;
|
|
|
|
args->source = (const char *)(mySource + length);
|
|
|
|
*err = U_TRUNCATED_CHAR_FOUND;
|
|
|
|
return 0xffff;
|
|
|
|
}
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
/* Don't even try to do a direct cast because the value may be on an odd address. */
|
|
|
|
myUChar = ((UChar32)mySource[3] << 24)
|
|
|
|
| ((UChar32)mySource[2] << 16)
|
|
|
|
| ((UChar32)mySource[1] << 8)
|
|
|
|
| ((UChar32)mySource[0]);
|
2002-07-01 19:32:52 +00:00
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
args->source = (const char *)(mySource + 4);
|
2003-10-02 17:02:14 +00:00
|
|
|
if ((uint32_t)myUChar <= MAXIMUM_UTF && !U_IS_SURROGATE(myUChar)) {
|
2003-08-01 14:58:43 +00:00
|
|
|
return myUChar;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
2003-08-01 14:58:43 +00:00
|
|
|
uprv_memcpy(args->converter->toUBytes, mySource, 4);
|
|
|
|
args->converter->toULength = 4;
|
|
|
|
|
|
|
|
*err = U_ILLEGAL_CHAR_FOUND;
|
2002-07-01 19:32:52 +00:00
|
|
|
return 0xffff;
|
|
|
|
}
|
|
|
|
|
|
|
|
static const UConverterImpl _UTF32LEImpl = {
|
|
|
|
UCNV_UTF32_LittleEndian,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
|
|
|
|
T_UConverter_toUnicode_UTF32_LE,
|
|
|
|
T_UConverter_toUnicode_UTF32_LE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE,
|
|
|
|
T_UConverter_fromUnicode_UTF32_LE_OFFSET_LOGIC,
|
|
|
|
T_UConverter_getNextUChar_UTF32_LE,
|
|
|
|
|
|
|
|
NULL,
|
2003-04-05 01:33:02 +00:00
|
|
|
NULL,
|
|
|
|
NULL,
|
|
|
|
NULL,
|
2005-06-17 07:04:03 +00:00
|
|
|
ucnv_getNonSurrogateUnicodeSet
|
2002-07-01 19:32:52 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/* The 1232 CCSID refers to any version of Unicode with any endianess of UTF-32 */
|
|
|
|
static const UConverterStaticData _UTF32LEStaticData = {
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-32LE",
|
2002-07-24 19:15:27 +00:00
|
|
|
1234,
|
2002-07-01 19:32:52 +00:00
|
|
|
UCNV_IBM, UCNV_UTF32_LittleEndian, 4, 4,
|
|
|
|
{ 0xfd, 0xff, 0, 0 }, 4, FALSE, FALSE,
|
|
|
|
0,
|
|
|
|
0,
|
|
|
|
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
|
|
|
|
};
|
|
|
|
|
|
|
|
|
2015-05-29 19:33:24 +00:00
|
|
|
const UConverterSharedData _UTF32LEData =
|
|
|
|
UCNV_IMMUTABLE_SHARED_DATA_INITIALIZER(&_UTF32LEStaticData, &_UTF32LEImpl);
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
/* 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 */
|
2005-06-12 21:15:18 +00:00
|
|
|
cnv->fromUnicodeStatus=UCNV_NEED_TO_WRITE_BOM;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
_UTF32Open(UConverter *cnv,
|
2009-03-13 22:05:01 +00:00
|
|
|
UConverterLoadArgs *pArgs,
|
2002-07-01 19:32:52 +00:00
|
|
|
UErrorCode *pErrorCode) {
|
2009-04-07 18:31:32 +00:00
|
|
|
_UTF32Reset(cnv, UCNV_RESET_BOTH);
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
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 */
|
2004-12-08 01:05:40 +00:00
|
|
|
offsetDelta=(int32_t)(source-pArgs->source);
|
2002-07-01 19:32:52 +00:00
|
|
|
} else if(state==8) {
|
|
|
|
state=9; /* detect UTF-32LE */
|
2004-12-08 01:05:40 +00:00
|
|
|
offsetDelta=(int32_t)(source-pArgs->source);
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* switch to UTF-32BE and pass the previous bytes */
|
2004-12-08 01:05:40 +00:00
|
|
|
int32_t count=(int32_t)(source-pArgs->source); /* number of bytes from this buffer */
|
2002-07-01 19:32:52 +00:00
|
|
|
|
|
|
|
/* 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;
|
|
|
|
}
|
|
|
|
state=8;
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2003-04-21 02:06:03 +00:00
|
|
|
pArgs->source=source;
|
|
|
|
|
2002-07-01 19:32:52 +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 */
|
|
|
|
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);
|
2003-04-21 02:06:03 +00:00
|
|
|
pArgs->source=source;
|
2002-07-01 19:32:52 +00:00
|
|
|
pArgs->sourceLimit=sourceLimit;
|
2003-07-22 04:22:57 +00:00
|
|
|
state=8;
|
2002-07-01 19:32:52 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2003-07-22 04:22:57 +00:00
|
|
|
|
|
|
|
cnv->mode=state;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
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:
|
2003-07-22 04:22:57 +00:00
|
|
|
return UCNV_GET_NEXT_UCHAR_USE_TO_U;
|
2002-07-01 19:32:52 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
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,
|
2003-04-05 01:33:02 +00:00
|
|
|
NULL,
|
2005-06-17 07:04:03 +00:00
|
|
|
ucnv_getNonSurrogateUnicodeSet
|
2002-07-01 19:32:52 +00:00
|
|
|
};
|
|
|
|
|
2005-05-31 22:04:26 +00:00
|
|
|
/* The 1236 CCSID refers to any version of Unicode with a BOM sensitive endianess of UTF-32 */
|
2002-07-01 19:32:52 +00:00
|
|
|
static const UConverterStaticData _UTF32StaticData = {
|
|
|
|
sizeof(UConverterStaticData),
|
|
|
|
"UTF-32",
|
2005-05-31 22:04:26 +00:00
|
|
|
1236,
|
2002-07-01 19:32:52 +00:00
|
|
|
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 */
|
|
|
|
};
|
|
|
|
|
2015-05-29 19:33:24 +00:00
|
|
|
const UConverterSharedData _UTF32Data =
|
|
|
|
UCNV_IMMUTABLE_SHARED_DATA_INITIALIZER(&_UTF32StaticData, &_UTF32Impl);
|
2004-08-26 22:51:40 +00:00
|
|
|
|
|
|
|
#endif
|