/*
**********************************************************************
-* Copyright (C) 2000-2006, International Business Machines
+* Copyright (C) 2000-2011, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* file name: ucnvhz.c
#include "unicode/ucnv.h"
#include "unicode/ucnv_cb.h"
#include "unicode/uset.h"
+#include "unicode/utf16.h"
#include "ucnv_bld.h"
#include "ucnv_cnv.h"
+#include "ucnv_imp.h"
#define UCNV_TILDE 0x7E /* ~ */
#define UCNV_OPEN_BRACE 0x7B /* { */
UBool isEscapeAppended;
UBool isStateDBCS;
UBool isTargetUCharDBCS;
+ UBool isEmptySegment;
}UConverterDataHZ;
static void
-_HZOpen(UConverter *cnv, const char *name,const char *locale,uint32_t options, UErrorCode *errorCode){
+_HZOpen(UConverter *cnv, UConverterLoadArgs *pArgs, UErrorCode *errorCode){
+ UConverter *gbConverter;
+ if(pArgs->onlyTestIsLoadable) {
+ ucnv_canCreateConverter("GBK", errorCode); /* errorCode carries result */
+ return;
+ }
+ gbConverter = ucnv_open("GBK", errorCode);
+ if(U_FAILURE(*errorCode)) {
+ return;
+ }
cnv->toUnicodeStatus = 0;
cnv->fromUnicodeStatus= 0;
cnv->mode=0;
cnv->fromUChar32=0x0000;
- cnv->extraInfo = uprv_malloc(sizeof(UConverterDataHZ));
+ cnv->extraInfo = uprv_calloc(1, sizeof(UConverterDataHZ));
if(cnv->extraInfo != NULL){
- uprv_memset(cnv->extraInfo, 0, sizeof(UConverterDataHZ));
- ((UConverterDataHZ*)cnv->extraInfo)->gbConverter = ucnv_open("ibm-1386",errorCode);
+ ((UConverterDataHZ*)cnv->extraInfo)->gbConverter = gbConverter;
}
else {
+ ucnv_close(gbConverter);
*errorCode = U_MEMORY_ALLOCATION_ERROR;
return;
}
cnv->mode=0;
if(cnv->extraInfo != NULL){
((UConverterDataHZ*)cnv->extraInfo)->isStateDBCS = FALSE;
+ ((UConverterDataHZ*)cnv->extraInfo)->isEmptySegment = FALSE;
}
}
if(choice!=UCNV_RESET_TO_UNICODE) {
* from-GB code '~}' ($7E7D) is outside the defined GB range.)
*
* Source: RFC 1842
+*
+* Note that the formal syntax in RFC 1842 is invalid. I assume that the
+* intended definition of single-byte-segment is as follows (pedberg):
+* single-byte-segment = single-byte-seq 1*single-byte-char
*/
UChar *myTarget = args->target;
const char *mySourceLimit = args->sourceLimit;
UChar32 targetUniChar = 0x0000;
- UChar mySourceChar = 0x0000;
+ int32_t mySourceChar = 0x0000;
UConverterDataHZ* myData=(UConverterDataHZ*)(args->converter->extraInfo);
tempBuf[0]=0;
tempBuf[1]=0;
- if ((args->converter == NULL) || (args->targetLimit < args->target) || (mySourceLimit < args->source)){
+
+ /* Calling code already handles this situation. */
+ /*if ((args->converter == NULL) || (args->targetLimit < args->target) || (mySourceLimit < args->source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
- }
+ }*/
while(mySource< mySourceLimit){
mySourceChar= (unsigned char) *mySource++;
- switch(mySourceChar){
+ if(args->converter->mode == UCNV_TILDE) {
+ /* second byte after ~ */
+ args->converter->mode=0;
+ switch(mySourceChar) {
case 0x0A:
- if(args->converter->mode ==UCNV_TILDE){
- args->converter->mode=0;
-
- }
- *(myTarget++)=(UChar)mySourceChar;
+ /* no output for ~\n (line-continuation marker) */
continue;
-
case UCNV_TILDE:
- if(args->converter->mode ==UCNV_TILDE){
- *(myTarget++)=(UChar)mySourceChar;
- args->converter->mode=0;
- continue;
-
+ if(args->offsets) {
+ args->offsets[myTarget - args->target]=(int32_t)(mySource - args->source - 2);
}
- else if(args->converter->toUnicodeStatus !=0){
- args->converter->mode=0;
- break;
- }
- else{
- args->converter->mode = UCNV_TILDE;
- continue;
- }
-
-
+ *(myTarget++)=(UChar)mySourceChar;
+ myData->isEmptySegment = FALSE;
+ continue;
case UCNV_OPEN_BRACE:
- if(args->converter->mode == UCNV_TILDE){
- args->converter->mode=0;
- myData->isStateDBCS = TRUE;
- continue;
- }
- else{
- break;
- }
-
-
case UCNV_CLOSE_BRACE:
- if(args->converter->mode == UCNV_TILDE){
- args->converter->mode=0;
- myData->isStateDBCS = FALSE;
- continue;
- }
- else{
- break;
+ myData->isStateDBCS = (mySourceChar == UCNV_OPEN_BRACE);
+ if (myData->isEmptySegment) {
+ myData->isEmptySegment = FALSE; /* we are handling it, reset to avoid future spurious errors */
+ *err = U_ILLEGAL_ESCAPE_SEQUENCE;
+ args->converter->toUCallbackReason = UCNV_IRREGULAR;
+ args->converter->toUBytes[0] = UCNV_TILDE;
+ args->converter->toUBytes[1] = mySourceChar;
+ args->converter->toULength = 2;
+ args->target = myTarget;
+ args->source = mySource;
+ return;
}
-
+ myData->isEmptySegment = TRUE;
+ continue;
default:
/* if the first byte is equal to TILDE and the trail byte
* is not a valid byte then it is an error condition
*/
- if(args->converter->mode == UCNV_TILDE){
- args->converter->mode=0;
- mySourceChar= (UChar)(((UCNV_TILDE+0x80) << 8) | ((mySourceChar & 0x00ff)+0x80));
- goto SAVE_STATE;
+ /*
+ * Ticket 5691: consistent illegal sequences:
+ * - We include at least the first byte in the illegal sequence.
+ * - If any of the non-initial bytes could be the start of a character,
+ * we stop the illegal sequence before the first one of those.
+ */
+ myData->isEmptySegment = FALSE; /* different error here, reset this to avoid spurious future error */
+ *err = U_ILLEGAL_ESCAPE_SEQUENCE;
+ args->converter->toUBytes[0] = UCNV_TILDE;
+ if( myData->isStateDBCS ?
+ (0x21 <= mySourceChar && mySourceChar <= 0x7e) :
+ mySourceChar <= 0x7f
+ ) {
+ /* The current byte could be the start of a character: Back it out. */
+ args->converter->toULength = 1;
+ --mySource;
+ } else {
+ /* Include the current byte in the illegal sequence. */
+ args->converter->toUBytes[1] = mySourceChar;
+ args->converter->toULength = 2;
}
-
- break;
-
- }
-
- if(myData->isStateDBCS){
+ args->target = myTarget;
+ args->source = mySource;
+ return;
+ }
+ } else if(myData->isStateDBCS) {
if(args->converter->toUnicodeStatus == 0x00){
- args->converter->toUnicodeStatus = (UChar) mySourceChar;
+ /* lead byte */
+ if(mySourceChar == UCNV_TILDE) {
+ args->converter->mode = UCNV_TILDE;
+ } else {
+ /* add another bit to distinguish a 0 byte from not having seen a lead byte */
+ args->converter->toUnicodeStatus = (uint32_t) (mySourceChar | 0x100);
+ myData->isEmptySegment = FALSE; /* the segment has something, either valid or will produce a different error, so reset this */
+ }
continue;
}
else{
- tempBuf[0] = (char) (args->converter->toUnicodeStatus+0x80) ;
- tempBuf[1] = (char) (mySourceChar+0x80);
- mySourceChar= (UChar)(((args->converter->toUnicodeStatus+0x80) << 8) | ((mySourceChar & 0x00ff)+0x80));
+ /* trail byte */
+ int leadIsOk, trailIsOk;
+ uint32_t leadByte = args->converter->toUnicodeStatus & 0xff;
+ targetUniChar = 0xffff;
+ /*
+ * Ticket 5691: consistent illegal sequences:
+ * - We include at least the first byte in the illegal sequence.
+ * - If any of the non-initial bytes could be the start of a character,
+ * we stop the illegal sequence before the first one of those.
+ *
+ * In HZ DBCS, if the second byte is in the 21..7e range,
+ * we report only the first byte as the illegal sequence.
+ * Otherwise we convert or report the pair of bytes.
+ */
+ leadIsOk = (uint8_t)(leadByte - 0x21) <= (0x7d - 0x21);
+ trailIsOk = (uint8_t)(mySourceChar - 0x21) <= (0x7e - 0x21);
+ if (leadIsOk && trailIsOk) {
+ tempBuf[0] = (char) (leadByte+0x80) ;
+ tempBuf[1] = (char) (mySourceChar+0x80);
+ targetUniChar = ucnv_MBCSSimpleGetNextUChar(myData->gbConverter->sharedData,
+ tempBuf, 2, args->converter->useFallback);
+ mySourceChar= (leadByte << 8) | mySourceChar;
+ } else if (trailIsOk) {
+ /* report a single illegal byte and continue with the following DBCS starter byte */
+ --mySource;
+ mySourceChar = (int32_t)leadByte;
+ } else {
+ /* report a pair of illegal bytes if the second byte is not a DBCS starter */
+ /* add another bit so that the code below writes 2 bytes in case of error */
+ mySourceChar= 0x10000 | (leadByte << 8) | mySourceChar;
+ }
args->converter->toUnicodeStatus =0x00;
- targetUniChar = ucnv_MBCSSimpleGetNextUChar(myData->gbConverter->sharedData,
- tempBuf, 2, args->converter->useFallback);
}
}
else{
- if(args->converter->fromUnicodeStatus == 0x00){
- targetUniChar = ucnv_MBCSSimpleGetNextUChar(myData->gbConverter->sharedData,
- mySource - 1, 1, args->converter->useFallback);
- }
- else{
- goto SAVE_STATE;
+ if(mySourceChar == UCNV_TILDE) {
+ args->converter->mode = UCNV_TILDE;
+ continue;
+ } else if(mySourceChar <= 0x7f) {
+ targetUniChar = (UChar)mySourceChar; /* ASCII */
+ myData->isEmptySegment = FALSE; /* the segment has something valid */
+ } else {
+ targetUniChar = 0xffff;
+ myData->isEmptySegment = FALSE; /* different error here, reset this to avoid spurious future error */
}
-
}
if(targetUniChar < 0xfffe){
if(args->offsets) {
*(myTarget++)=(UChar)targetUniChar;
}
- else if(targetUniChar>=0xfffe){
-SAVE_STATE:
+ else /* targetUniChar>=0xfffe */ {
if(targetUniChar == 0xfffe){
*err = U_INVALID_CHAR_FOUND;
}
else{
*err = U_ILLEGAL_CHAR_FOUND;
}
- if(myData->isStateDBCS){
- /* this should never occur since isStateDBCS is set to true
- * only after tempBuf[0] and tempBuf[1]
- * are set to the input .. just to please BEAM
- */
- if(tempBuf[0]==0 || tempBuf[1]==0){
- *err = U_INTERNAL_PROGRAM_ERROR;
- }else{
- args->converter->toUBytes[0] = (uint8_t)(tempBuf[0]-0x80);
- args->converter->toUBytes[1] = (uint8_t)(tempBuf[1]-0x80);
- args->converter->toULength=2;
- }
+ if(mySourceChar > 0xff){
+ args->converter->toUBytes[0] = (uint8_t)(mySourceChar >> 8);
+ args->converter->toUBytes[1] = (uint8_t)mySourceChar;
+ args->converter->toULength=2;
}
else{
args->converter->toUBytes[0] = (uint8_t)mySourceChar;
int len =0;
const char* escSeq=NULL;
- if ((args->converter == NULL) || (args->targetLimit < myTarget) || (args->sourceLimit < args->source)){
+ /* Calling code already handles this situation. */
+ /*if ((args->converter == NULL) || (args->targetLimit < myTarget) || (args->sourceLimit < args->source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
- }
+ }*/
if(args->converter->fromUChar32!=0 && myTargetIndex < targetLength) {
goto getTrail;
}
escSeq = TILDE_ESCAPE;
CONCAT_ESCAPE_MACRO(args, myTargetIndex, targetLength, escSeq,err,len,mySourceIndex);
continue;
- }
- else{
+ } else if(mySourceChar <= 0x7f) {
+ length = 1;
+ targetUniChar = mySourceChar;
+ } else {
length= ucnv_MBCSFromUChar32(myConverterData->gbConverter->sharedData,
mySourceChar,&targetUniChar,args->converter->useFallback);
-
- }
- /* only DBCS or SBCS characters are expected*/
- /* DB haracters with high bit set to 1 are expected */
- if(length > 2 || length==0 ||(((targetUniChar & 0x8080) != 0x8080)&& length==2)){
- targetUniChar= missingCharMarker;
+ /* we can only use lead bytes 21..7D and trail bytes 21..7E */
+ if( length == 2 &&
+ (uint16_t)(targetUniChar - 0xa1a1) <= (0xfdfe - 0xa1a1) &&
+ (uint8_t)(targetUniChar - 0xa1) <= (0xfe - 0xa1)
+ ) {
+ targetUniChar -= 0x8080;
+ } else {
+ targetUniChar = missingCharMarker;
+ }
}
if (targetUniChar != missingCharMarker){
myConverterData->isTargetUCharDBCS = isTargetUCharDBCS = (UBool)(targetUniChar>0x00FF);
if(isTargetUCharDBCS){
if( myTargetIndex <targetLength){
- myTarget[myTargetIndex++] =(char) ((targetUniChar >> 8) -0x80);
+ myTarget[myTargetIndex++] =(char) (targetUniChar >> 8);
if(offsets){
*(offsets++) = mySourceIndex-1;
}
if(myTargetIndex < targetLength){
- myTarget[myTargetIndex++] =(char) ((targetUniChar & 0x00FF) -0x80);
+ myTarget[myTargetIndex++] =(char) targetUniChar;
if(offsets){
*(offsets++) = mySourceIndex-1;
}
}else{
- args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (char) ((targetUniChar & 0x00FF) -0x80);
+ args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (char) targetUniChar;
*err = U_BUFFER_OVERFLOW_ERROR;
}
}else{
- args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] =(char) ((targetUniChar >> 8) -0x80);
- args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (char) ((targetUniChar & 0x00FF) -0x80);
+ args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] =(char) (targetUniChar >> 8);
+ args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (char) targetUniChar;
*err = U_BUFFER_OVERFLOW_ERROR;
}
/* oops.. the code point is unassigned */
/*Handle surrogates */
/*check if the char is a First surrogate*/
- if(UTF_IS_SURROGATE(mySourceChar)) {
- if(UTF_IS_SURROGATE_FIRST(mySourceChar)) {
+ if(U16_IS_SURROGATE(mySourceChar)) {
+ if(U16_IS_SURROGATE_LEAD(mySourceChar)) {
args->converter->fromUChar32=mySourceChar;
getTrail:
/*look ahead to find the trail surrogate*/
if(mySourceIndex < mySourceLength) {
/* test the following code unit */
UChar trail=(UChar) args->source[mySourceIndex];
- if(UTF_IS_SECOND_SURROGATE(trail)) {
+ if(U16_IS_TRAIL(trail)) {
++mySourceIndex;
- mySourceChar=UTF16_GET_PAIR_VALUE(args->converter->fromUChar32, trail);
+ mySourceChar=U16_GET_SUPPLEMENTARY(args->converter->fromUChar32, trail);
args->converter->fromUChar32=0x00;
/* there are no surrogates in GB2312*/
*err = U_INVALID_CHAR_FOUND;
const USetAdder *sa,
UConverterUnicodeSet which,
UErrorCode *pErrorCode) {
- /* the tilde '~' is hardcoded in the converter */
- sa->add(sa->set, 0x7e);
+ /* HZ converts all of ASCII */
+ sa->addRange(sa->set, 0, 0x7f);
/* add all of the code points that the sub-converter handles */
- ((UConverterDataHZ*)cnv->extraInfo)->
- gbConverter->sharedData->impl->
- getUnicodeSet(((UConverterDataHZ*)cnv->extraInfo)->gbConverter,
- sa, which, pErrorCode);
+ ucnv_MBCSGetFilteredUnicodeSetForUnicode(
+ ((UConverterDataHZ*)cnv->extraInfo)->gbConverter->sharedData,
+ sa, which, UCNV_SET_FILTER_HZ,
+ pErrorCode);
}
static const UConverterImpl _HZImpl={
projects / apple / icu.git / blobdiff