AuroraMiddleware/scuffed-code
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
840affc200
scuffed-code/icu4c/source/common/ucnv2022.c
Markus Scherer d57871f769 ICU-3222 fix compiler warnings 
X-SVN-Rev: 13990
2003-12-04 01:09:57 +00:00

3000 lines
121 KiB
C
Raw Blame History

/*
**********************************************************************
* Copyright (C) 2000-2003, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* file name: ucnv2022.c
* encoding: US-ASCII
* tab size: 8 (not used)
* indentation:4
*
* created on: 2000feb03
* created by: Markus W. Scherer
*
* Change history:
*
* 06/29/2000 helena Major rewrite of the callback APIs.
* 08/08/2000 Ram Included support for ISO-2022-JP-2
* Changed implementation of toUnicode
* function
* 08/21/2000 Ram Added support for ISO-2022-KR
* 08/29/2000 Ram Seperated implementation of EBCDIC to
* ucnvebdc.c
* 09/20/2000 Ram Added support for ISO-2022-CN
* Added implementations for getNextUChar()
* for specific 2022 country variants.
* 10/31/2000 Ram Implemented offsets logic functions
*/
#include "unicode/utypes.h"
#if !UCONFIG_NO_LEGACY_CONVERSION
#include "unicode/ucnv.h"
#include "unicode/uset.h"
#include "unicode/ucnv_err.h"
#include "unicode/ucnv_cb.h"
#include "ucnv_bld.h"
#include "ucnv_cnv.h"
#include "ucnvmbcs.h"
#include "cstring.h"
#include "cmemory.h"
#ifdef U_ENABLE_GENERIC_ISO_2022
/*
* I am disabling the generic ISO-2022 converter after proposing to do so on
* the icu mailing list two days ago.
*
* Reasons:
* 1. It does not fully support the ISO-2022/ECMA-35 specification with all of
* its designation sequences, single shifts with return to the previous state,
* switch-with-no-return to UTF-16BE or similar, etc.
* This is unlike the language-specific variants like ISO-2022-JP which
* require a much smaller repertoire of ISO-2022 features.
* These variants continue to be supported.
* 2. I believe that no one is really using the generic ISO-2022 converter
* but rather always one of the language-specific variants.
* Note that ICU's generic ISO-2022 converter has always output one escape
* sequence followed by UTF-8 for the whole stream.
* 3. Switching between subcharsets is extremely slow, because each time
* the previous converter is closed and a new one opened,
* without any kind of caching, least-recently-used list, etc.
* 4. The code is currently buggy, and given the above it does not seem
* reasonable to spend the time on maintenance.
*
* Markus Scherer 2003-dec-03
*/
#endif
static const char UCNV_SS2[] = "\x1B\x4E";
static const char UCNV_SS3[] = "\x1B\x4F";
#define UCNV_SS2_LEN 2
#define UCNV_SS3_LEN 2
#define CR 0x0D
#define LF 0x0A
#define H_TAB 0x09
#define V_TAB 0x0B
#define SPACE 0x20
/* for ISO-2022-JP and -CN implementations */
typedef enum {
/* shared values */
INVALID_STATE=-1,
ASCII = 0,
SS2_STATE=0x10,
SS3_STATE,
/* JP */
ISO8859_1 = 1 ,
ISO8859_7 = 2 ,
JISX201 = 3,
JISX208 = 4,
JISX212 = 5,
GB2312 =6,
KSC5601 =7,
HWKANA_7BIT=8, /* Halfwidth Katakana 7 bit */
/* CN */
/* the first few enum constants must keep their values because they correspond to myConverterArray[] */
GB2312_1=1,
ISO_IR_165=2,
CNS_11643=3,
/*
* these are used in StateEnum and ISO2022State variables,
* but CNS_11643 must be used to index into myConverterArray[]
*/
CNS_11643_0=0x20,
CNS_11643_1,
CNS_11643_2,
CNS_11643_3,
CNS_11643_4,
CNS_11643_5,
CNS_11643_6,
CNS_11643_7
} StateEnum;
typedef enum {
ASCII1=0,
LATIN1,
SBCS,
DBCS,
MBCS
}Cnv2022Type;
typedef struct ISO2022State {
int8_t cs[4]; /* charset number for SI (G0)/SO (G1)/SS2 (G2)/SS3 (G3) */
int8_t g; /* 0..3 for G0..G3 (SI/SO/SS2/SS3) */
int8_t prevG; /* g before single shift (SS2 or SS3) */
} ISO2022State;
#define UCNV_OPTIONS_VERSION_MASK 0xf
#define UCNV_2022_MAX_CONVERTERS 10
typedef struct{
UConverter *currentConverter;
UConverter *fromUnicodeConverter;
UBool isFirstBuffer;
StateEnum toUnicodeCurrentState;
StateEnum fromUnicodeCurrentState;
StateEnum toUnicodeSaveState;
Cnv2022Type currentType;
ISO2022State toU2022State, fromU2022State;
UConverter* myConverterArray[UCNV_2022_MAX_CONVERTERS];
UBool isShiftAppended;
uint32_t key;
uint32_t version;
char locale[3];
char name[30];
}UConverterDataISO2022;
/* Protos */
/* ISO-2022 ----------------------------------------------------------------- */
/*Forward declaration */
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 void
_MBCSFromUnicodeWithOffsets(UConverterFromUnicodeArgs *pArgs,
UErrorCode *pErrorCode);
U_CFUNC void
_MBCSToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
UErrorCode *pErrorCode);
#define ESC_2022 0x1B /*ESC*/
typedef enum
{
INVALID_2022 = -1, /*Doesn't correspond to a valid iso 2022 escape sequence*/
VALID_NON_TERMINAL_2022 = 0, /*so far corresponds to a valid iso 2022 escape sequence*/
VALID_TERMINAL_2022 = 1, /*corresponds to a valid iso 2022 escape sequence*/
VALID_MAYBE_TERMINAL_2022 = 2 /*so far matches one iso 2022 escape sequence, but by adding more characters might match another escape sequence*/
} UCNV_TableStates_2022;
/*
* The way these state transition arrays work is:
* ex : ESC$B is the sequence for JISX208
* a) First Iteration: char is ESC
* i) Get the value of ESC from normalize_esq_chars_2022[] with int value of ESC as index
* int x = normalize_esq_chars_2022[27] which is equal to 1
* ii) Search for this value in escSeqStateTable_Key_2022[]
* value of x is stored at escSeqStateTable_Key_2022[0]
* iii) Save this index as offset
* iv) Get state of this sequence from escSeqStateTable_Value_2022[]
* escSeqStateTable_Value_2022[offset], which is VALID_NON_TERMINAL_2022
* b) Switch on this state and continue to next char
* i) Get the value of $ from normalize_esq_chars_2022[] with int value of $ as index
* which is normalize_esq_chars_2022[36] == 4
* ii) x is currently 1(from above)
* x<<=5 -- x is now 32
* x+=normalize_esq_chars_2022[36]
* now x is 36
* iii) Search for this value in escSeqStateTable_Key_2022[]
* value of x is stored at escSeqStateTable_Key_2022[2], so offset is 2
* iv) Get state of this sequence from escSeqStateTable_Value_2022[]
* escSeqStateTable_Value_2022[offset], which is VALID_NON_TERMINAL_2022
* c) Switch on this state and continue to next char
* i) Get the value of B from normalize_esq_chars_2022[] with int value of B as index
* ii) x is currently 36 (from above)
* x<<=5 -- x is now 1152
* x+=normalize_esq_chars_2022[66]
* now x is 1161
* iii) Search for this value in escSeqStateTable_Key_2022[]
* value of x is stored at escSeqStateTable_Key_2022[21], so offset is 21
* iv) Get state of this sequence from escSeqStateTable_Value_2022[21]
* escSeqStateTable_Value_2022[offset], which is VALID_TERMINAL_2022
* v) Get the converter name form escSeqStateTable_Result_2022[21] which is JISX208
*/
/*Below are the 3 arrays depicting a state transition table*/
static const int8_t normalize_esq_chars_2022[256] = {
/* 0 1 2 3 4 5 6 7 8 9 */
0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,1 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,4 ,7 ,29 ,0
,2 ,24 ,26 ,27 ,0 ,3 ,23 ,6 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,5 ,8 ,9 ,10 ,11 ,12
,13 ,14 ,15 ,16 ,17 ,18 ,19 ,20 ,25 ,28
,0 ,0 ,21 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,22 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0
,0 ,0 ,0 ,0 ,0 ,0
};
#ifdef U_ENABLE_GENERIC_ISO_2022
/*
* When the generic ISO-2022 converter is completely removed, not just disabled
* per #ifdef, then the following state table and the associated tables that are
* dimensioned with MAX_STATES_2022 should be trimmed.
*
* Especially, VALID_MAYBE_TERMINAL_2022 will not be used any more, and all of
* the associated escape sequences starting with ESC ( B should be removed.
* This includes the ones with key values 1097 and all of the ones above 1000000.
*
* For the latter, the tables can simply be truncated.
* For the former, since the tables must be kept parallel, it is probably best
* to simply duplicate an adjacent table cell, parallel in all tables.
*
* It may make sense to restructure the tables, especially by using small search
* tables for the variants instead of indexing them parallel to the table here.
*/
#endif
#define MAX_STATES_2022 74
static const int32_t escSeqStateTable_Key_2022[MAX_STATES_2022] = {
/* 0 1 2 3 4 5 6 7 8 9 */
1 ,34 ,36 ,39 ,55 ,57 ,60 ,61 ,1093 ,1096
,1097 ,1098 ,1099 ,1100 ,1101 ,1102 ,1103 ,1104 ,1105 ,1106
,1109 ,1154 ,1157 ,1160 ,1161 ,1176 ,1178 ,1179 ,1254 ,1257
,1768 ,1773 ,1957 ,35105 ,36933 ,36936 ,36937 ,36938 ,36939 ,36940
,36942 ,36943 ,36944 ,36945 ,36946 ,36947 ,36948 ,37640 ,37642 ,37644
,37646 ,37711 ,37744 ,37745 ,37746 ,37747 ,37748 ,40133 ,40136 ,40138
,40139 ,40140 ,40141 ,1123363 ,35947624 ,35947625 ,35947626 ,35947627 ,35947629 ,35947630
,35947631 ,35947635 ,35947636 ,35947638
};
#ifdef U_ENABLE_GENERIC_ISO_2022
static const char* const escSeqStateTable_Result_2022[MAX_STATES_2022] = {
/* 0 1 2 3 4 5 6 7 8 9 */
NULL ,NULL ,NULL ,NULL ,NULL ,NULL ,NULL ,NULL ,"latin1" ,"latin1"
,"latin1" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"JISX-201" ,"JISX-201" ,"latin1"
,"latin1" ,NULL ,"JISX-208" ,"ibm-5478" ,"JISX-208" ,NULL ,NULL ,NULL ,NULL ,"UTF8"
,"ISO-8859-1" ,"ISO-8859-7" ,"JIS-X-208" ,NULL ,"ibm-955" ,"ibm-367" ,"ibm-952" ,"ibm-949" ,"JISX-212" ,"ibm-1383"
,"ibm-952" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-5478" ,"ibm-949" ,"ISO-IR-165"
,"CNS-11643-1992,1" ,"CNS-11643-1992,2" ,"CNS-11643-1992,3" ,"CNS-11643-1992,4" ,"CNS-11643-1992,5" ,"CNS-11643-1992,6" ,"CNS-11643-1992,7" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian"
,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,NULL ,"latin1" ,"ibm-912" ,"ibm-913" ,"ibm-914" ,"ibm-813" ,"ibm-1089"
,"ibm-920" ,"ibm-915" ,"ibm-915" ,"latin1"
};
#endif
static const UCNV_TableStates_2022 escSeqStateTable_Value_2022[MAX_STATES_2022] = {
/* 0 1 2 3 4 5 6 7 8 9 */
VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_MAYBE_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_NON_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022 ,VALID_TERMINAL_2022
};
/* Type def for refactoring changeState_2022 code*/
typedef enum{
#ifdef U_ENABLE_GENERIC_ISO_2022
ISO_2022=0,
#endif
ISO_2022_JP=1,
ISO_2022_KR=2,
ISO_2022_CN=3
} Variant2022;
/*********** ISO 2022 Converter Protos ***********/
static void
_ISO2022Open(UConverter *cnv, const char *name, const char *locale,uint32_t options, UErrorCode *errorCode);
static void
_ISO2022Close(UConverter *converter);
static void
_ISO2022Reset(UConverter *converter, UConverterResetChoice choice);
static const char*
_ISO2022getName(const UConverter* cnv);
static void
_ISO_2022_WriteSub(UConverterFromUnicodeArgs *args, int32_t offsetIndex, UErrorCode *err);
static UConverter *
_ISO_2022_SafeClone(const UConverter *cnv, void *stackBuffer, int32_t *pBufferSize, UErrorCode *status);
static void
T_UConverter_toUnicode_ISO_2022_OFFSETS_LOGIC(UConverterToUnicodeArgs* args, UErrorCode* err);
/*const UConverterSharedData _ISO2022Data;*/
static const UConverterSharedData _ISO2022JPData;
static const UConverterSharedData _ISO2022KRData;
static const UConverterSharedData _ISO2022CNData;
/*************** Converter implementations ******************/
/**********/
static void
setInitialStateToUnicodeJPCN(UConverter* converter,UConverterDataISO2022 *myConverterData ){
myConverterData->toUnicodeCurrentState =ASCII;
myConverterData->currentConverter = NULL;
myConverterData->isFirstBuffer = TRUE;
myConverterData->toUnicodeSaveState = INVALID_STATE;
converter->mode = UCNV_SI;
}
static void
setInitialStateFromUnicodeJPCN(UConverter* converter,UConverterDataISO2022 *myConverterData){
myConverterData->fromUnicodeCurrentState= ASCII;
myConverterData->isShiftAppended=FALSE;
myConverterData->currentType = ASCII1;
converter->fromUnicodeStatus = FALSE;
}
static void
setInitialStateToUnicodeKR(UConverter* converter, UConverterDataISO2022 *myConverterData){
converter->mode = UCNV_SI;
myConverterData->currentConverter = myConverterData->fromUnicodeConverter;
}
static void
setInitialStateFromUnicodeKR(UConverter* converter,UConverterDataISO2022 *myConverterData){
/* in ISO-2022-KR the desginator sequence appears only once
* in a file so we append it only once
*/
if( converter->charErrorBufferLength==0){
converter->charErrorBufferLength = 4;
converter->charErrorBuffer[0] = 0x1b;
converter->charErrorBuffer[1] = 0x24;
converter->charErrorBuffer[2] = 0x29;
converter->charErrorBuffer[3] = 0x43;
}
myConverterData->isShiftAppended=FALSE;
}
static void
_ISO2022Open(UConverter *cnv, const char *name, const char *locale,uint32_t options, UErrorCode *errorCode){
char myLocale[6]={' ',' ',' ',' ',' ',' '};
cnv->extraInfo = uprv_malloc (sizeof (UConverterDataISO2022));
if(cnv->extraInfo != NULL) {
UConverterDataISO2022 *myConverterData=(UConverterDataISO2022 *) cnv->extraInfo;
uprv_memset(myConverterData, 0, sizeof(UConverterDataISO2022));
myConverterData->currentConverter = NULL;
myConverterData->fromUnicodeConverter = NULL;
myConverterData->currentType = ASCII1;
myConverterData->key =0;
myConverterData->isFirstBuffer = TRUE;
cnv->fromUnicodeStatus =FALSE;
if(locale){
uprv_strncpy(myLocale, locale, sizeof(myLocale));
}
myConverterData->version= 0;
myConverterData->myConverterArray[0] =NULL;
if(myLocale[0]=='j' && (myLocale[1]=='a'|| myLocale[1]=='p') &&
(myLocale[2]=='_' || myLocale[2]=='\0')){
int len=0;
/* open the required converters and cache them */
myConverterData->myConverterArray[0]= ucnv_open("ASCII", errorCode );
myConverterData->myConverterArray[1]= ucnv_open("ISO8859_1", errorCode);
myConverterData->myConverterArray[2]= ucnv_open("ISO8859_7", errorCode);
myConverterData->myConverterArray[3]= ucnv_open("jisx-201", errorCode);
myConverterData->myConverterArray[4]= ucnv_open("jisx-208", errorCode);
myConverterData->myConverterArray[5]= ucnv_open("jisx-212", errorCode);
myConverterData->myConverterArray[6]= ucnv_open("ibm-5478", errorCode); /* gb_2312_80-1 */
myConverterData->myConverterArray[7]= ucnv_open("ksc_5601", errorCode);
myConverterData->myConverterArray[8]= ucnv_open("jisx-201", errorCode);
myConverterData->myConverterArray[9]= NULL;
/* initialize the state variables */
setInitialStateToUnicodeJPCN(cnv, myConverterData);
setInitialStateFromUnicodeJPCN(cnv,myConverterData);
/* set the function pointers to appropriate funtions */
cnv->sharedData=(UConverterSharedData*)(&_ISO2022JPData);
uprv_strcpy(myConverterData->locale,"ja");
myConverterData->version =options & UCNV_OPTIONS_VERSION_MASK;
uprv_strcpy(myConverterData->name,"ISO_2022,locale=ja,version=");
len = uprv_strlen(myConverterData->name);
myConverterData->name[len]=(char)(myConverterData->version+(int)'0');
myConverterData->name[len+1]='\0';
}
else if(myLocale[0]=='k' && (myLocale[1]=='o'|| myLocale[1]=='r') &&
(myLocale[2]=='_' || myLocale[2]=='\0')){
/* initialize the state variables */
setInitialStateToUnicodeKR(cnv, myConverterData);
setInitialStateFromUnicodeKR(cnv,myConverterData);
if ((options & UCNV_OPTIONS_VERSION_MASK)==1){
myConverterData->version = 1;
myConverterData->currentConverter=myConverterData->fromUnicodeConverter=
ucnv_open("icu-internal-25546",errorCode);
uprv_strcpy(myConverterData->name,"ISO_2022,locale=ko,version=1");
}else{
myConverterData->currentConverter=myConverterData->fromUnicodeConverter = ucnv_open("ibm-949",errorCode);
myConverterData->version = 0;
uprv_strcpy(myConverterData->name,"ISO_2022,locale=ko,version=0");
}
/* set the function pointers to appropriate funtions */
cnv->sharedData=(UConverterSharedData*)&_ISO2022KRData;
cnv->mode=UCNV_SI;
uprv_strcpy(myConverterData->locale,"ko");
}
else if(((myLocale[0]=='z' && myLocale[1]=='h') || (myLocale[0]=='c'&& myLocale[1]=='n'))&&
(myLocale[2]=='_' || myLocale[2]=='\0')){
/* open the required converters and cache them */
myConverterData->myConverterArray[0] = NULL;
myConverterData->myConverterArray[GB2312_1] = ucnv_open("ibm-5478",errorCode);
myConverterData->myConverterArray[ISO_IR_165] = ucnv_open("iso-ir-165",errorCode);
myConverterData->myConverterArray[CNS_11643] = ucnv_open("cns-11643-1992",errorCode);
myConverterData->myConverterArray[4] = NULL;
/*initialize the state variables*/
setInitialStateToUnicodeJPCN(cnv, myConverterData);
setInitialStateFromUnicodeJPCN(cnv,myConverterData);
/* set the function pointers to appropriate funtions */
cnv->sharedData=(UConverterSharedData*)&_ISO2022CNData;
uprv_strcpy(myConverterData->locale,"cn");
if ((options & UCNV_OPTIONS_VERSION_MASK)==1){
myConverterData->version = 1;
uprv_strcpy(myConverterData->name,"ISO_2022,locale=zh,version=1");
}else{
uprv_strcpy(myConverterData->name,"ISO_2022,locale=zh,version=0");
myConverterData->version = 0;
}
}
else{
#ifdef U_ENABLE_GENERIC_ISO_2022
/* append the UTF-8 escape sequence */
cnv->charErrorBufferLength = 3;
cnv->charErrorBuffer[0] = 0x1b;
cnv->charErrorBuffer[1] = 0x25;
cnv->charErrorBuffer[2] = 0x42;
cnv->sharedData=(UConverterSharedData*)&_ISO2022Data;
/* initialize the state variables */
uprv_strcpy(myConverterData->name,"ISO_2022");
#else
*errorCode = U_UNSUPPORTED_ERROR;
return;
#endif
}
cnv->maxBytesPerUChar=cnv->sharedData->staticData->maxBytesPerChar;
} else {
*errorCode = U_MEMORY_ALLOCATION_ERROR;
}
}
static void
_ISO2022Close(UConverter *converter) {
UConverterDataISO2022* myData =(UConverterDataISO2022 *) (converter->extraInfo);
UConverter **array = myData->myConverterArray;
if (converter->extraInfo != NULL) {
/*close the array of converter pointers and free the memory*/
while(*array!=NULL){
if(*array==myData->currentConverter){
myData->currentConverter=NULL;
}
ucnv_close(*array++);
}
ucnv_close(myData->currentConverter); /* if not closed above */
if(!converter->isExtraLocal){
uprv_free (converter->extraInfo);
}
}
}
static void
_ISO2022Reset(UConverter *converter, UConverterResetChoice choice) {
UConverterDataISO2022 *myConverterData=(UConverterDataISO2022 *) (converter->extraInfo);
if(choice<=UCNV_RESET_TO_UNICODE) {
uprv_memset(&myConverterData->toU2022State, 0, sizeof(ISO2022State));
myConverterData->key = 0;
}
if(choice!=UCNV_RESET_TO_UNICODE) {
uprv_memset(&myConverterData->fromU2022State, 0, sizeof(ISO2022State));
}
#ifdef U_ENABLE_GENERIC_ISO_2022
if(myConverterData->locale[0] == 0){
if(choice<=UCNV_RESET_TO_UNICODE) {
myConverterData->isFirstBuffer = TRUE;
myConverterData->key = 0;
if (converter->mode == UCNV_SO){
ucnv_close (myConverterData->currentConverter);
myConverterData->currentConverter=NULL;
}
converter->mode = UCNV_SI;
}
if(choice!=UCNV_RESET_TO_UNICODE) {
/* re-append UTF-8 escape sequence */
converter->charErrorBufferLength = 3;
converter->charErrorBuffer[0] = 0x1b;
converter->charErrorBuffer[1] = 0x28;
converter->charErrorBuffer[2] = 0x42;
}
}
else
#endif
{
/* reset the state variables */
if(myConverterData->locale[0] == 'j' || myConverterData->locale[0] == 'c'){
if(choice<=UCNV_RESET_TO_UNICODE) {
setInitialStateToUnicodeJPCN(converter, myConverterData);
}
if(choice!=UCNV_RESET_TO_UNICODE) {
setInitialStateFromUnicodeJPCN(converter,myConverterData);
}
}
else if(myConverterData->locale[0] == 'k'){
if(choice<=UCNV_RESET_TO_UNICODE) {
setInitialStateToUnicodeKR(converter, myConverterData);
}
if(choice!=UCNV_RESET_TO_UNICODE) {
setInitialStateFromUnicodeKR(converter, myConverterData);
}
}
}
}
static const char*
_ISO2022getName(const UConverter* cnv){
if(cnv->extraInfo){
UConverterDataISO2022* myData= (UConverterDataISO2022*)cnv->extraInfo;
return myData->name;
}
return NULL;
}
/*************** to unicode *******************/
/****************************************************************************
* Recognized escape sequences are
* <ESC>(B ASCII
* <ESC>.A ISO-8859-1
* <ESC>.F ISO-8859-7
* <ESC>(J JISX-201
* <ESC>(I JISX-201
* <ESC>$B JISX-208
* <ESC>$@ JISX-208
* <ESC>$(D JISX-212
* <ESC>$A GB2312
* <ESC>$(C KSC5601
*/
static const StateEnum nextStateToUnicodeJP[5][MAX_STATES_2022]= {
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ASCII ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX201 ,HWKANA_7BIT ,JISX201 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,INVALID_STATE ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
},
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ASCII ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX201 ,HWKANA_7BIT ,JISX201 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,INVALID_STATE ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX212 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
},
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,SS2_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ASCII ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX201 ,HWKANA_7BIT ,JISX201 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,GB2312 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ISO8859_1 ,ISO8859_7 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,KSC5601 ,JISX212 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
},
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,SS2_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ASCII ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX201 ,HWKANA_7BIT ,JISX201 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,GB2312 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ISO8859_1 ,ISO8859_7 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,KSC5601 ,JISX212 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
},
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,SS2_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ASCII ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,JISX201 ,HWKANA_7BIT ,JISX201 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,JISX208 ,GB2312 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,ISO8859_1 ,ISO8859_7 ,JISX208 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,KSC5601 ,JISX212 ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
}
};
/*************** to unicode *******************/
static const StateEnum nextStateToUnicodeCN[2][MAX_STATES_2022]= {
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,SS2_STATE ,SS3_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,GB2312_1 ,INVALID_STATE ,INVALID_STATE
,CNS_11643_1 ,CNS_11643_2 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
},
{
/* 0 1 2 3 4 5 6 7 8 9 */
INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,SS2_STATE ,SS3_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,GB2312_1 ,INVALID_STATE ,ISO_IR_165
,CNS_11643_1 ,CNS_11643_2 ,CNS_11643_3 ,CNS_11643_4 ,CNS_11643_5 ,CNS_11643_6 ,CNS_11643_7 ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
,INVALID_STATE ,INVALID_STATE ,INVALID_STATE ,INVALID_STATE
}
};
static UCNV_TableStates_2022
getKey_2022(char c,int32_t* key,int32_t* offset){
int32_t togo;
int32_t low = 0;
int32_t hi = MAX_STATES_2022;
int32_t oldmid=0;
togo = normalize_esq_chars_2022[(uint8_t)c];
if(togo == 0) {
/* not a valid character anywhere in an escape sequence */
*key = 0;
*offset = 0;
return INVALID_2022;
}
togo = (*key << 5) + togo;
while (hi != low) /*binary search*/{
register int32_t mid = (hi+low) >> 1; /*Finds median*/
if (mid == oldmid)
break;
if (escSeqStateTable_Key_2022[mid] > togo){
hi = mid;
}
else if (escSeqStateTable_Key_2022[mid] < togo){
low = mid;
}
else /*we found it*/{
*key = togo;
*offset = mid;
return escSeqStateTable_Value_2022[mid];
}
oldmid = mid;
}
*key = 0;
*offset = 0;
return INVALID_2022;
}
/*runs through a state machine to determine the escape sequence - codepage correspondance
*/
static void
changeState_2022(UConverter* _this,
const char** source,
const char* sourceLimit,
Variant2022 var,
UErrorCode* err){
UConverter* myUConverter;
UCNV_TableStates_2022 value;
UConverterDataISO2022* myData2022 = ((UConverterDataISO2022*)_this->extraInfo);
uint32_t key = myData2022->key;
int32_t offset;
char c;
value = VALID_NON_TERMINAL_2022;
while (*source < sourceLimit) {
c = *(*source)++;
_this->toUBytes[_this->toULength++]=(uint8_t)c;
value = getKey_2022(c,(int32_t *) &key, &offset);
switch (value){
case VALID_NON_TERMINAL_2022 :
/* continue with the loop */
break;
case VALID_TERMINAL_2022:
key = 0;
goto DONE;
case INVALID_2022:
goto DONE;
case VALID_MAYBE_TERMINAL_2022:
#ifdef U_ENABLE_GENERIC_ISO_2022
/* ESC ( B is ambiguous only for ISO_2022 itself */
if(var == ISO_2022) {
/* discard toUBytes[] for ESC ( B because this sequence is correct and complete */
_this->toULength = 0;
/* TODO need to indicate that ESC ( B was seen; if failure, then need to replay from source or from MBCS-style replay */
/* continue with the loop */
value = VALID_NON_TERMINAL_2022;
break;
} else
#endif
{
/* not ISO_2022 itself, finish here */
value = VALID_TERMINAL_2022;
key = 0;
goto DONE;
}
}
}
DONE:
myData2022->key = key;
if (value == VALID_NON_TERMINAL_2022) {
/* indicate that the escape sequence is incomplete: key!=0 */
return;
} else if (value == INVALID_2022 ) {
*err = U_ILLEGAL_ESCAPE_SEQUENCE;
return;
} else /* value == VALID_TERMINAL_2022 */ {
switch(var){
#ifdef U_ENABLE_GENERIC_ISO_2022
case ISO_2022:
{
const char *chosenConverterName = escSeqStateTable_Result_2022[offset];
if(chosenConverterName == NULL) {
/* SS2 or SS3 */
*err = U_UNSUPPORTED_ESCAPE_SEQUENCE;
return;
}
_this->mode = UCNV_SI;
ucnv_close(myData2022->currentConverter);
myData2022->currentConverter = myUConverter = ucnv_open(chosenConverterName, err);
if(U_SUCCESS(*err)) {
myUConverter->fromCharErrorBehaviour = UCNV_TO_U_CALLBACK_STOP;
}
break;
}
#endif
case ISO_2022_JP:
{
StateEnum tempState=nextStateToUnicodeJP[myData2022->version][offset];
_this->mode = UCNV_SI;
if(tempState == INVALID_STATE) {
myData2022->toUnicodeCurrentState = INVALID_STATE;
myData2022->currentConverter = myUConverter = NULL;
*err = U_UNSUPPORTED_ESCAPE_SEQUENCE;
} else if(tempState == SS2_STATE) {
myUConverter = myData2022->currentConverter;
} else {
myData2022->toUnicodeCurrentState = tempState;
myData2022->currentConverter = myUConverter = myData2022->myConverterArray[tempState];
}
}
break;
case ISO_2022_CN:
{
StateEnum tempState=nextStateToUnicodeCN[myData2022->version][offset];
switch(tempState) {
case INVALID_STATE:
*err = U_UNSUPPORTED_ESCAPE_SEQUENCE;
break;
case SS2_STATE:
if(myData2022->toU2022State.cs[2]!=0) {
if(myData2022->toU2022State.g<2) {
myData2022->toU2022State.prevG=myData2022->toU2022State.g;
}
myData2022->toU2022State.g=2;
} else {
/* illegal to have SS2 before a matching designator */
*err = U_ILLEGAL_ESCAPE_SEQUENCE;
}
break;
case SS3_STATE:
if(myData2022->toU2022State.cs[3]!=0) {
if(myData2022->toU2022State.g<2) {
myData2022->toU2022State.prevG=myData2022->toU2022State.g;
}
myData2022->toU2022State.g=3;
} else {
/* illegal to have SS3 before a matching designator */
*err = U_ILLEGAL_ESCAPE_SEQUENCE;
}
break;
case GB2312_1:
case ISO_IR_165:
case CNS_11643_1:
myData2022->toU2022State.cs[1]=(int8_t)tempState;
break;
case CNS_11643_2:
myData2022->toU2022State.cs[2]=(int8_t)tempState;
break;
default:
/* other CNS 11643 planes */
myData2022->toU2022State.cs[3]=(int8_t)tempState;
break;
}
}
break;
case ISO_2022_KR:
if(offset==0x30){
_this->mode = UCNV_SI;
myUConverter = myData2022->currentConverter=myData2022->fromUnicodeConverter;
} else {
*err = U_UNSUPPORTED_ESCAPE_SEQUENCE;
}
break;
default:
*err = U_ILLEGAL_ESCAPE_SEQUENCE;
break;
}
}
if(U_SUCCESS(*err)) {
_this->mode = UCNV_SO;
_this->toULength = 0;
}
}
/*Checks the characters of the buffer against valid 2022 escape sequences
*if the match we return a pointer to the initial start of the sequence otherwise
*we return sourceLimit
*/
/*for 2022 looks ahead in the stream
*to determine the longest possible convertible
*data stream
*/
static const char*
getEndOfBuffer_2022(const char** source,
const char* sourceLimit,
UBool flush){
const char* mySource = *source;
if (*source >= sourceLimit)
return sourceLimit;
do{
if (*mySource == ESC_2022){
int8_t i;
int32_t key = 0;
int32_t offset;
UCNV_TableStates_2022 value = VALID_NON_TERMINAL_2022;
/* Kludge: I could not
* figure out the reason for validating an escape sequence
* twice - once here and once in changeState_2022().
* is it possible to have an ESC character in a ISO2022
* byte stream which is valid in a code page? Is it legal?
*/
for (i=0;
(mySource+i < sourceLimit)&&(value == VALID_NON_TERMINAL_2022);
i++) {
value = getKey_2022(*(mySource+i), &key, &offset);
}
if (value > 0 || *mySource==ESC_2022)
return mySource;
if ((value == VALID_NON_TERMINAL_2022)&&(!flush) )
return sourceLimit;
}
}while (++mySource < sourceLimit);
return sourceLimit;
}
static U_INLINE void
CONCAT_ESCAPE_EX(UConverterFromUnicodeArgs* args,
const UChar* source,
unsigned char** target,
const unsigned char* targetLimit,
int32_t** offsets,
const char* strToAppend,
int len,
UErrorCode* err)
{
unsigned char* myTarget = *target;
int32_t* myOffsets = *offsets;
while(len-->0){
if(myTarget < targetLimit){
*(myTarget++) = (unsigned char) *(strToAppend++);
if(myOffsets){
*(myOffsets++) = source - args->source -1;
}
}
else{
args->converter->charErrorBuffer[(int)args->converter->charErrorBufferLength++] = (unsigned char) *(strToAppend++);
*err =U_BUFFER_OVERFLOW_ERROR;
}
}
*target = myTarget;
*offsets = myOffsets;
}
/* This inline function replicates code in _MBCSFromUChar32() function in ucnvmbcs.c
* any future change in _MBCSFromUChar32() function should be reflected in
* this macro
*/
static U_INLINE void
MBCS_FROM_UCHAR32_ISO2022(UConverterSharedData* sharedData,
UChar32 c,
uint32_t* value,
UBool useFallback,
int32_t *length,
int outputType)
{
const uint16_t *table=sharedData->mbcs.fromUnicodeTable;
uint32_t stage2Entry;
uint32_t myValue=0;
const uint8_t *p;
/* BMP-only codepages are stored without stage 1 entries for supplementary code points */
if(c<0x10000 || (sharedData->mbcs.unicodeMask&UCNV_HAS_SUPPLEMENTARY)) {
stage2Entry=MBCS_STAGE_2_FROM_U(table, c);
/* get the bytes and the length for the output */
if(outputType==MBCS_OUTPUT_2){
myValue=MBCS_VALUE_2_FROM_STAGE_2(sharedData->mbcs.fromUnicodeBytes, stage2Entry, c);
if(myValue<=0xff) {
*length=1;
} else {
*length=2;
}
}else if(outputType==MBCS_OUTPUT_3){
p=MBCS_POINTER_3_FROM_STAGE_2(sharedData->mbcs.fromUnicodeBytes, stage2Entry, c);
myValue=((uint32_t)*p<<16)|((uint32_t)p[1]<<8)|p[2];
if(myValue<=0xff) {
*length=1;
} else if(myValue<=0xffff) {
*length=2;
} else {
*length=3;
}
}
/* is this code point assigned, or do we use fallbacks? */
if( (stage2Entry&(1<<(16+(c&0xf))))!=0 ||
(FROM_U_USE_FALLBACK(useFallback, c) && myValue!=0)
) {
/*
* We allow a 0 byte output if the "assigned" bit is set for this entry.
* There is no way with this data structure for fallback output
* to be a zero byte.
*/
/* assigned */
*value=myValue;
} else {
const int32_t *cx=sharedData->mbcs.extIndexes;
if(cx!=NULL) {
*length=ucnv_extSimpleMatchFromU(cx, c, value, useFallback);
} else {
/* unassigned */
*length=0;
}
}
}else{
*length=0;
}
}
/* This inline function replicates code in _MBCSSingleFromUChar32() function in ucnvmbcs.c
* any future change in _MBCSSingleFromUChar32() function should be reflected in
* this macro
*/
static U_INLINE void
MBCS_SINGLE_FROM_UCHAR32(UConverterSharedData* sharedData,
UChar32 c,
uint32_t* retval,
UBool useFallback)
{
const uint16_t *table;
int32_t value;
/* BMP-only codepages are stored without stage 1 entries for supplementary code points */
if(c>=0x10000 && !(sharedData->mbcs.unicodeMask&UCNV_HAS_SUPPLEMENTARY)) {
value= -1;
}
/* convert the Unicode code point in c into codepage bytes (same as in _MBCSFromUnicodeWithOffsets) */
table=sharedData->mbcs.fromUnicodeTable;
/* get the byte for the output */
value=MBCS_SINGLE_RESULT_FROM_U(table, (uint16_t *)sharedData->mbcs.fromUnicodeBytes, c);
/* is this code point assigned, or do we use fallbacks? */
if(useFallback ? value>=0x800 : value>=0xc00) {
value &=0xff;
} else {
value= -1;
}
*retval=(uint16_t) value;
}
#ifdef U_ENABLE_GENERIC_ISO_2022
/**********************************************************************************
* ISO-2022 Converter
*
*
*/
static void
T_UConverter_toUnicode_ISO_2022_OFFSETS_LOGIC(UConverterToUnicodeArgs* args,
UErrorCode* err){
const char* mySourceLimit, *realSourceLimit;
const char* sourceStart;
const UChar* myTargetStart;
UConverter* saveThis;
UConverterDataISO2022* myData;
int8_t length;
saveThis = args->converter;
myData=((UConverterDataISO2022*)(saveThis->extraInfo));
realSourceLimit = args->sourceLimit;
while (args->source < realSourceLimit) {
if(myData->key == 0) { /* are we in the middle of an escape sequence? */
/*Find the end of the buffer e.g : Next Escape Seq | end of Buffer*/
mySourceLimit = getEndOfBuffer_2022(&(args->source), realSourceLimit, args->flush);
if(args->source < mySourceLimit) {
if(myData->currentConverter==NULL) {
myData->currentConverter = ucnv_open("ASCII",err);
if(U_FAILURE(*err)){
return;
}
myData->currentConverter->fromCharErrorBehaviour = UCNV_TO_U_CALLBACK_STOP;
saveThis->mode = UCNV_SO;
}
/* convert to before the ESC or until the end of the buffer */
myData->isFirstBuffer=FALSE;
sourceStart = args->source;
myTargetStart = args->target;
args->converter = myData->currentConverter;
ucnv_toUnicode(args->converter,
&args->target,
args->targetLimit,
&args->source,
mySourceLimit,
args->offsets,
(UBool)(args->flush && mySourceLimit == realSourceLimit),
err);
args->converter = saveThis;
if (*err == U_BUFFER_OVERFLOW_ERROR) {
/* move the overflow buffer */
length = saveThis->UCharErrorBufferLength = myData->currentConverter->UCharErrorBufferLength;
myData->currentConverter->UCharErrorBufferLength = 0;
if(length > 0) {
uprv_memcpy(saveThis->UCharErrorBuffer,
myData->currentConverter->UCharErrorBuffer,
length*U_SIZEOF_UCHAR);
}
return;
}
/*
* At least one of:
* -Error while converting
* -Done with entire buffer
* -Need to write offsets or update the current offset
* (leave that up to the code in ucnv.c)
*
* or else we just stopped at an ESC byte and continue with changeState_2022()
*/
if (U_FAILURE(*err) ||
(args->source == realSourceLimit) ||
(args->offsets != NULL && (args->target != myTargetStart || args->source != sourceStart) ||
(mySourceLimit < realSourceLimit && myData->currentConverter->toULength > 0))
) {
/* copy partial or error input for truncated detection and error handling */
if(U_FAILURE(*err)) {
length = saveThis->invalidCharLength = myData->currentConverter->invalidCharLength;
if(length > 0) {
uprv_memcpy(saveThis->invalidCharBuffer, myData->currentConverter->invalidCharBuffer, length);
}
} else {
length = saveThis->toULength = myData->currentConverter->toULength;
if(length > 0) {
uprv_memcpy(saveThis->toUBytes, myData->currentConverter->toUBytes, length);
if(args->source < mySourceLimit) {
*err = U_TRUNCATED_CHAR_FOUND; /* truncated input before ESC */
}
}
}
return;
}
}
}
sourceStart = args->source;
changeState_2022(args->converter,
&(args->source),
realSourceLimit,
ISO_2022,
err);
if (U_FAILURE(*err) || (args->source != sourceStart && args->offsets != NULL)) {
/* let the ucnv.c code update its current offset */
return;
}
}
}
#endif
/*
* To Unicode Callback helper function
*/
static void
toUnicodeCallback(UConverter *cnv,
const uint32_t sourceChar, const uint32_t targetUniChar,
UErrorCode* err){
if(sourceChar>0xff){
cnv->toUBytes[0] = (uint8_t)(sourceChar>>8);
cnv->toUBytes[1] = (uint8_t)sourceChar;
cnv->toULength = 2;
}
else{
cnv->toUBytes[0] =(char) sourceChar;
cnv->toULength = 2;
}
if(targetUniChar == (missingCharMarker-1/*0xfffe*/)){
*err = U_INVALID_CHAR_FOUND;
}
else{
*err = U_ILLEGAL_CHAR_FOUND;
}
}
/**************************************ISO-2022-JP*************************************************/
/************************************** IMPORTANT **************************************************
* The UConverter_fromUnicode_ISO2022_JP converter does not use ucnv_fromUnicode() functions for SBCS,DBCS and
* MBCS; instead, the values are obtained directly by calling _MBCSFromUChar32().
* The converter iterates over each Unicode codepoint
* to obtain the equivalent codepoints from the codepages supported. Since the source buffer is
* processed one char at a time it would make sense to reduce the extra processing a canned converter
* would do as far as possible.
*
* If the implementation of these macros or structure of sharedData struct change in the future, make
* sure that ISO-2022 is also changed.
***************************************************************************************************
*/
/***************************************************************************************************
* Rules for ISO-2022-jp encoding
* (i) Escape sequences must be fully contained within a line they should not
* span new lines or CRs
* (ii) If the last character on a line is represented by two bytes then an ASCII or
* JIS-Roman character escape sequence should follow before the line terminates
* (iii) If the first character on the line is represented by two bytes then a two
* byte character escape sequence should precede it
* (iv) If no escape sequence is encountered then the characters are ASCII
* (v) Latin(ISO-8859-1) and Greek(ISO-8859-7) characters must be designated to G2,
* and invoked with SS2 (ESC N).
* (vi) If there is any G0 designation in text, there must be a switch to
* ASCII or to JIS X 0201-Roman before a space character (but not
* necessarily before "ESC 4/14 2/0" or "ESC N ' '") or control
* characters such as tab or CRLF.
* (vi) Supported encodings:
* ASCII, JISX201, JISX208, JISX212, GB2312, KSC5601, ISO-8859-1,ISO-8859-7
*
* source : RFC-1554
*
* JISX201, JISX208,JISX212 : new .cnv data files created
* KSC5601 : alias to ibm-949 mapping table
* GB2312 : alias to ibm-1386 mapping table
* ISO-8859-1 : Algorithmic implemented as LATIN1 case
* ISO-8859-7 : alisas to ibm-9409 mapping table
*/
#define MAX_VALID_CP_JP 9
static const Cnv2022Type myConverterType[MAX_VALID_CP_JP]={
ASCII1,
LATIN1,
SBCS,
SBCS,
DBCS,
DBCS,
DBCS,
DBCS,
SBCS,
};
static const StateEnum nextStateArray[5][MAX_VALID_CP_JP]= {
{JISX201 ,INVALID_STATE,INVALID_STATE,JISX208,ASCII,INVALID_STATE,INVALID_STATE,INVALID_STATE,INVALID_STATE},
{JISX201,INVALID_STATE,INVALID_STATE,JISX208,JISX212,ASCII,INVALID_STATE,INVALID_STATE,INVALID_STATE},
{ISO8859_1,ISO8859_7,JISX201,JISX208,JISX212,GB2312,KSC5601,ASCII,INVALID_STATE},
{JISX201,INVALID_STATE,INVALID_STATE,JISX208,JISX212,HWKANA_7BIT,INVALID_STATE,INVALID_STATE,ASCII},
{JISX201,INVALID_STATE,INVALID_STATE,JISX208,JISX212,ASCII,INVALID_STATE,INVALID_STATE,INVALID_STATE},
};
static const char escSeqChars[MAX_VALID_CP_JP][6] ={
"\x1B\x28\x42", /* <ESC>(B ASCII */
"\x1B\x2E\x41", /* <ESC>.A ISO-8859-1 */
"\x1B\x2E\x46", /* <ESC>.F ISO-8859-7 */
"\x1B\x28\x4A", /* <ESC>(J JISX-201 */
"\x1B\x24\x42", /* <ESC>$B JISX-208 */
"\x1B\x24\x28\x44", /* <ESC>$(D JISX-212 */
"\x1B\x24\x41", /* <ESC>$A GB2312 */
"\x1B\x24\x28\x43", /* <ESC>$(C KSC5601 */
"\x1B\x28\x49" /* <ESC>(I HWKANA_7BIT */
};
static const int32_t escSeqCharsLen[MAX_VALID_CP_JP] ={
3, /* length of <ESC>(B ASCII */
3, /* length of <ESC>.A ISO-8859-1 */
3, /* length of <ESC>.F ISO-8859-7 */
3, /* length of <ESC>(J JISX-201 */
3, /* length of <ESC>$B JISX-208 */
4, /* length of <ESC>$(D JISX-212 */
3, /* length of <ESC>$A GB2312 */
4, /* length of <ESC>$(C KSC5601 */
3 /* length of <ESC>(I HWKANA_7BIT */
};
/*
* The iteration over various code pages works this way:
* i) Get the currentState from myConverterData->currentState
* ii) Check if the character is mapped to a valid character in the currentState
* Yes -> a) set the initIterState to currentState
* b) remain in this state until an invalid character is found
* No -> a) go to the next code page and find the character
* iii) Before changing the state increment the current state check if the current state
* is equal to the intitIteration state
* Yes -> A character that cannot be represented in any of the supported encodings
* break and return a U_INVALID_CHARACTER error
* No -> Continue and find the character in next code page
*
*
* TODO: Implement a priority technique where the users are allowed to set the priority of code pages
*/
static void
UConverter_fromUnicode_ISO_2022_JP_OFFSETS_LOGIC(UConverterFromUnicodeArgs* args, UErrorCode* err){
UConverterDataISO2022 *converterData;
unsigned char* target = (unsigned char*) args->target;
const unsigned char* targetLimit = (const unsigned char*) args->targetLimit;
const UChar* source = args->source;
const UChar* sourceLimit = args->sourceLimit;
int32_t* offsets = args->offsets;
int32_t offset = 0;
uint32_t targetByteUnit = missingCharMarker;
UChar32 sourceChar =0x0000;
const char* escSeq = NULL;
int32_t len =0; /*length of escSeq chars*/
UConverterSharedData* sharedData=NULL;
UBool useFallback;
/* state variables*/
StateEnum* currentState;
StateEnum initIterState;
UConverter** currentConverter;
Cnv2022Type* currentType;
UConverter** convArray;
/* arguments check*/
if ((args->converter == NULL) || (targetLimit < target) || (sourceLimit < source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
/* Initialize */
converterData = (UConverterDataISO2022*)args->converter->extraInfo;
useFallback = args->converter->useFallback;
currentState = &converterData->fromUnicodeCurrentState;
initIterState = ASCII;
currentConverter = &converterData->fromUnicodeConverter;
convArray = converterData->myConverterArray;
initIterState = *currentState;
currentType = &converterData->currentType;
/* check if the last codepoint of previous buffer was a lead surrogate*/
if((sourceChar = args->converter->fromUChar32)!=0 && target< targetLimit) {
goto getTrail;
}
*currentConverter = convArray[(*currentConverter==NULL) ? 0 : (int)*currentState];
sharedData= (*currentConverter)->sharedData;
while( source < sourceLimit){
targetByteUnit = missingCharMarker;
if(target < targetLimit){
sourceChar = *(source++);
if(sourceChar > SPACE) {
do{
switch (*currentType){
/* most common case*/
case DBCS:
{
uint32_t value=0;
int32_t length=0;
/*if(2 == _MBCSFromUChar32(sharedData,sourceChar, &value, useFallback)) {
targetByteUnit = (uint16_t)value;
}*/
MBCS_FROM_UCHAR32_ISO2022(sharedData,sourceChar,&value,useFallback,&length,MBCS_OUTPUT_2);
if(length==2){
targetByteUnit = value;
}
}
break;
case ASCII1:
if(sourceChar < 0x7f){
targetByteUnit = sourceChar;
}
break;
case SBCS:
MBCS_SINGLE_FROM_UCHAR32(sharedData,sourceChar,&targetByteUnit,useFallback);
/*targetByteUnit=(uint16_t)_MBCSSingleFromUChar32(sharedData,sourceChar,useFallback);*/
/*
* If mySourceChar is unassigned, then _MBCSSingleFromUChar32() returns -1
* which becomes the same as missingCharMarker with the cast to uint16_t.
*/
/* Check if the sourceChar is in the HW Kana range*/
if(0xFF9F-sourceChar<=(0xFF9F-0xFF61)){
if( converterData->version==3){
/*we get a1-df from _MBCSSingleFromUChar32 so subtract 0x80*/
targetByteUnit-=0x80;
*currentState = HWKANA_7BIT;
}
else if( converterData->version==4){
*currentState = JISX201;
}
else{
targetByteUnit=missingCharMarker;
}
*currentConverter = convArray[(*currentConverter==NULL) ? 0 : (int)*currentState];
*currentType = (Cnv2022Type) myConverterType[*currentState];
}
break;
case LATIN1:
if(sourceChar <= 0x00FF){
targetByteUnit = sourceChar;
}
break;
default:
/*not expected */
break;
}
if(targetByteUnit==missingCharMarker){
*currentState = nextStateArray[converterData->version][*currentState];
*currentConverter = convArray[(*currentConverter==NULL) ? 0 : (int)*currentState];
*currentType = (Cnv2022Type) myConverterType[*currentState];
sharedData= (*currentConverter)->sharedData;
}
else
/*got the mapping so break from while loop*/
break;
}while(initIterState != *currentState);
}
else{
targetByteUnit = sourceChar;
*currentState = ASCII;
*currentType = (Cnv2022Type) myConverterType[*currentState];
}
if(targetByteUnit != missingCharMarker){
if( *currentState != initIterState){
escSeq = escSeqChars[(int)*currentState];
len = escSeqCharsLen[(int)*currentState];
CONCAT_ESCAPE_EX(args,source, &target,targetLimit, &offsets, escSeq,len,err);
/* Append SSN for shifting to G2 */
if(*currentState==ISO8859_1 || *currentState==ISO8859_7){
escSeq = UCNV_SS2;
len = UCNV_SS2_LEN;
CONCAT_ESCAPE_EX(args, source, &target, targetLimit,&offsets, escSeq,len,err);
}
}
initIterState = *currentState;
offset = source - args->source -1;
/* write the targetByteUnit to target */
if(targetByteUnit <= 0x00FF){
if( target <targetLimit){
*(target++) = (unsigned char) targetByteUnit;
if(offsets){
*(offsets++) = offset;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) targetByteUnit;
*err = U_BUFFER_OVERFLOW_ERROR;
}
}else{
if(target < targetLimit){
*(target++) =(unsigned char) (targetByteUnit>>8);
if(offsets){
*(offsets++) = offset;
}
if(target < targetLimit){
*(target++) =(unsigned char) (targetByteUnit);
if(offsets){
*(offsets++) = offset;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit);
*err = U_BUFFER_OVERFLOW_ERROR;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit>>8);
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit);
*err = U_BUFFER_OVERFLOW_ERROR;
}
}
}
else{
/* if we cannot find the character after checking all codepages
* then this is an error
*/
/*check if the char is a First surrogate*/
if(UTF_IS_SURROGATE(sourceChar)) {
if(UTF_IS_SURROGATE_FIRST(sourceChar)) {
getTrail:
/*look ahead to find the trail surrogate*/
if(source < sourceLimit) {
/* test the following code unit */
UChar trail=(UChar) *source;
if(UTF_IS_SECOND_SURROGATE(trail)) {
source++;
sourceChar=UTF16_GET_PAIR_VALUE(sourceChar, trail);
*err = U_INVALID_CHAR_FOUND;
/* convert this surrogate code point */
/* exit this condition tree */
} else {
/* this is an unmatched lead code unit (1st surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
}
} else {
/* no more input */
*err = U_ZERO_ERROR;
}
} else {
/* this is an unmatched trail code unit (2nd surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
}
} else {
/* callback(unassigned) for a BMP code point */
*err = U_INVALID_CHAR_FOUND;
}
args->converter->fromUChar32=sourceChar;
break;
}
} /* end if(myTargetIndex<myTargetLength) */
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}/* end while(mySourceIndex<mySourceLength) */
/*save the state and return */
args->source = source;
args->target = (char*)target;
}
/*************** to unicode *******************/
static void
UConverter_toUnicode_ISO_2022_JP_OFFSETS_LOGIC(UConverterToUnicodeArgs *args,
UErrorCode* err){
char tempBuf[2];
const char *mySource = ( char *) args->source;
UChar *myTarget = args->target;
const char *mySourceLimit = args->sourceLimit;
uint32_t targetUniChar = 0x0000;
uint32_t mySourceChar = 0x0000;
UConverterDataISO2022* myData;
StateEnum* currentState;
uint32_t* toUnicodeStatus;
if ((args->converter == NULL) || (myTarget < args->target) || (mySource < args->source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
myData=(UConverterDataISO2022*)(args->converter->extraInfo);
currentState = &myData->toUnicodeCurrentState;
toUnicodeStatus = &args->converter->toUnicodeStatus;
while(mySource< mySourceLimit){
targetUniChar = missingCharMarker;
if(myTarget < args->targetLimit){
mySourceChar= (unsigned char) *mySource++;
/* Consume the escape sequences and ascertain the state */
if(mySourceChar==UCNV_SI){
if(myData->version==3 && *toUnicodeStatus==0x00){
if(myData->toUnicodeSaveState!=INVALID_STATE){
*currentState = (StateEnum) myData->toUnicodeSaveState;
continue;
}
else{
*err =U_ILLEGAL_CHAR_FOUND;
goto CALLBACK;
}
}
else{
goto CALLBACK;
}
}else if(mySourceChar==UCNV_SO){
if(myData->version==3 && *toUnicodeStatus==0x00){
myData->toUnicodeSaveState= (int) *currentState;
*currentState = HWKANA_7BIT;
continue;
}
else{
goto CALLBACK;
}
}else if(mySourceChar==ESC_2022 || myData->key!=0){
if(*toUnicodeStatus== 0x00){
mySource--;
changeState_2022(args->converter,&(mySource),
mySourceLimit, ISO_2022_JP, err);
/*Invalid or illegal escape sequence */
if(U_SUCCESS(*err)){
continue;
}
else{
args->target = myTarget;
args->source = mySource;
return;
}
}
else{
goto CALLBACK;
}
}
switch(myConverterType[*currentState]){
case DBCS:
if(*toUnicodeStatus== 0x00){
*toUnicodeStatus= (UChar) mySourceChar;
continue;
}
else{
tempBuf[0] = (char) args->converter->toUnicodeStatus;
tempBuf[1] = (char) mySourceChar;
mySourceChar+= (args->converter->toUnicodeStatus)<<8;
*toUnicodeStatus= 0;
targetUniChar = _MBCSSimpleGetNextUChar(myData->currentConverter->sharedData, tempBuf, 2, args->converter->useFallback);
}
break;
case ASCII1:
if( mySourceChar < 0x7F){
targetUniChar = (UChar) mySourceChar;
}
else if((uint8_t)(mySourceChar - 0xa1) <= (0xdf - 0xa1) && myData->version==4) {
/* 8-bit halfwidth katakana in any single-byte mode for JIS8 */
targetUniChar = _MBCS_SINGLE_SIMPLE_GET_NEXT_BMP(myData->myConverterArray[JISX201]->sharedData, mySourceChar);
}
break;
case SBCS:
if((uint8_t)(mySourceChar - 0xa1) <= (0xdf - 0xa1) && myData->version==4) {
/* 8-bit halfwidth katakana in any single-byte mode for JIS8 */
targetUniChar = _MBCS_SINGLE_SIMPLE_GET_NEXT_BMP(myData->myConverterArray[JISX201]->sharedData, mySourceChar);
}
else if(*currentState==HWKANA_7BIT){
targetUniChar = _MBCS_SINGLE_SIMPLE_GET_NEXT_BMP(myData->myConverterArray[JISX201]->sharedData, mySourceChar+0x80);
}
else {
targetUniChar = _MBCS_SINGLE_SIMPLE_GET_NEXT_BMP(myData->currentConverter->sharedData, mySourceChar);
}
break;
case LATIN1:
targetUniChar = (UChar) mySourceChar;
break;
case INVALID_STATE:
*err = U_ILLEGAL_ESCAPE_SEQUENCE;
args->target = myTarget;
args->source = mySource;
return;
default:
/* For non-valid state MBCS and others */
break;
}
if(targetUniChar < (missingCharMarker-1/*0xfffe*/)){
if(args->offsets){
args->offsets[myTarget - args->target]= mySource - args->source - 2
+(myConverterType[*currentState] <= SBCS);
}
*(myTarget++)=(UChar)targetUniChar;
targetUniChar=missingCharMarker;
}
else{
CALLBACK:
/* Call the callback function*/
toUnicodeCallback(args->converter,mySourceChar,targetUniChar,err);
break;
}
}
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}
args->target = myTarget;
args->source = mySource;
}
/***************************************************************
* Rules for ISO-2022-KR encoding
* i) The KSC5601 designator sequence should appear only once in a file,
* at the begining of a line before any KSC5601 characters. This usually
* means that it appears by itself on the first line of the file
* ii) There are only 2 shifting sequences SO to shift into double byte mode
* and SI to shift into single byte mode
*/
static void
UConverter_fromUnicode_ISO_2022_KR_OFFSETS_LOGIC_IBM(UConverterFromUnicodeArgs* args, UErrorCode* err){
UConverter* saveConv = args->converter;
UConverterDataISO2022 *myConverterData=(UConverterDataISO2022*)args->converter->extraInfo;
args->converter=myConverterData->currentConverter;
_MBCSFromUnicodeWithOffsets(args,err);
if(U_FAILURE(*err)){
if(args->converter->charErrorBufferLength!=0){
uprv_memcpy(saveConv->charErrorBuffer, args->converter->charErrorBuffer,
args->converter->charErrorBufferLength);
saveConv->charErrorBufferLength=args->converter->charErrorBufferLength;
args->converter->charErrorBufferLength=0;
}
if(args->converter->toULength!=0){
uprv_memcpy(saveConv->toUBytes, args->converter->toUBytes,
args->converter->toULength);
saveConv->toULength=args->converter->toULength;
args->converter->toULength=0;
}
}
args->converter=saveConv;
}
static void
UConverter_fromUnicode_ISO_2022_KR_OFFSETS_LOGIC(UConverterFromUnicodeArgs* args, UErrorCode* err){
const UChar *source = args->source;
const UChar *sourceLimit = args->sourceLimit;
unsigned char *target = (unsigned char *) args->target;
unsigned char *targetLimit = (unsigned char *) args->targetLimit;
int32_t* offsets = args->offsets;
uint32_t targetByteUnit = 0x0000;
UChar32 sourceChar = 0x0000;
UBool isTargetByteDBCS;
UBool oldIsTargetByteDBCS;
UConverterDataISO2022 *converterData;
UConverterSharedData* sharedData;
UBool useFallback;
int32_t length =0;
if ((args->converter == NULL) || (args->targetLimit < args->target) || (sourceLimit < args->source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
/* initialize data */
converterData=(UConverterDataISO2022*)args->converter->extraInfo;
sharedData = converterData->fromUnicodeConverter->sharedData;
useFallback = args->converter->useFallback;
isTargetByteDBCS=(UBool)args->converter->fromUnicodeStatus;
oldIsTargetByteDBCS = isTargetByteDBCS;
/* if the version is 1 then the user is requesting
* conversion with ibm-25546 pass the arguments to
* MBCS converter and return
*/
if(converterData->version==1){
UConverter_fromUnicode_ISO_2022_KR_OFFSETS_LOGIC_IBM(args,err);
return;
}
isTargetByteDBCS = (UBool) args->converter->fromUnicodeStatus;
if((sourceChar = args->converter->fromUChar32)!=0 && target <targetLimit) {
goto getTrail;
}
while(source < sourceLimit){
targetByteUnit = missingCharMarker;
if(target < (unsigned char*) args->targetLimit){
sourceChar = *source++;
/* length= _MBCSFromUChar32(converterData->fromUnicodeConverter->sharedData,
sourceChar,&targetByteUnit,args->converter->useFallback);*/
MBCS_FROM_UCHAR32_ISO2022(sharedData,sourceChar,&targetByteUnit,useFallback,&length,MBCS_OUTPUT_2);
/* only DBCS or SBCS characters are expected*/
/* DB haracters with high bit set to 1 are expected */
if(length > 2 || length==0 ||(((targetByteUnit & 0x8080) != 0x8080)&& length==2)){
targetByteUnit=missingCharMarker;
}
if (targetByteUnit != missingCharMarker){
oldIsTargetByteDBCS = isTargetByteDBCS;
isTargetByteDBCS = (UBool)(targetByteUnit>0x00FF);
/* append the shift sequence */
if (oldIsTargetByteDBCS != isTargetByteDBCS ){
if (isTargetByteDBCS)
*target++ = UCNV_SO;
else
*target++ = UCNV_SI;
if(offsets)
*(offsets++)= source - args->source-1;
}
/* write the targetUniChar to target */
if(targetByteUnit <= 0x00FF){
if( target < targetLimit){
*(target++) = (unsigned char) targetByteUnit;
if(offsets){
*(offsets++) = source - args->source-1;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit);
*err = U_BUFFER_OVERFLOW_ERROR;
}
}else{
if(target < targetLimit){
*(target++) =(unsigned char) ((targetByteUnit>>8) -0x80);
if(offsets){
*(offsets++) = source - args->source-1;
}
if(target < targetLimit){
*(target++) =(unsigned char) (targetByteUnit -0x80);
if(offsets){
*(offsets++) = source - args->source-1;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit -0x80);
*err = U_BUFFER_OVERFLOW_ERROR;
}
}else{
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) ((targetByteUnit>>8) -0x80);
args->converter->charErrorBuffer[args->converter->charErrorBufferLength++] = (unsigned char) (targetByteUnit-0x80);
*err = U_BUFFER_OVERFLOW_ERROR;
}
}
}
else{
/* oops.. the code point is unassingned
* set the error and reason
*/
/*check if the char is a First surrogate*/
if(UTF_IS_SURROGATE(sourceChar)) {
if(UTF_IS_SURROGATE_FIRST(sourceChar)) {
getTrail:
/*look ahead to find the trail surrogate*/
if(source < sourceLimit) {
/* test the following code unit */
UChar trail=(UChar) *source;
if(UTF_IS_SECOND_SURROGATE(trail)) {
source++;
sourceChar=UTF16_GET_PAIR_VALUE(sourceChar, trail);
*err = U_INVALID_CHAR_FOUND;
/* convert this surrogate code point */
/* exit this condition tree */
} else {
/* this is an unmatched lead code unit (1st surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
}
} else {
/* no more input */
*err = U_ZERO_ERROR;
}
} else {
/* this is an unmatched trail code unit (2nd surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
}
} else {
/* callback(unassigned) for a BMP code point */
*err = U_INVALID_CHAR_FOUND;
}
args->converter->fromUChar32=sourceChar;
args->converter->fromUnicodeStatus = (int32_t)isTargetByteDBCS;
break;
}
} /* end if(myTargetIndex<myTargetLength) */
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}/* end while(mySourceIndex<mySourceLength) */
/*save the state and return */
args->source = source;
args->target = (char*)target;
args->converter->fromUnicodeStatus = (uint32_t)isTargetByteDBCS;
}
/************************ To Unicode ***************************************/
static void
UConverter_toUnicode_ISO_2022_KR_OFFSETS_LOGIC_IBM(UConverterToUnicodeArgs *args,
UErrorCode* err){
const char* mySourceLimit;
char const* sourceStart;
UConverter* saveThis;
UConverterDataISO2022* myData=(UConverterDataISO2022*)(args->converter->extraInfo);
do{
/*Find the end of the buffer e.g : Next Escape Seq | end of Buffer*/
mySourceLimit = getEndOfBuffer_2022(&(args->source), args->sourceLimit, args->flush);
if (args->converter->mode == UCNV_SO) /*Already doing some conversion*/{
saveThis = args->converter;
args->offsets = NULL;
args->converter = myData->currentConverter;
_MBCSToUnicodeWithOffsets(args,err);
if(U_FAILURE(*err)){
uprv_memcpy(saveThis->invalidUCharBuffer, args->converter->invalidUCharBuffer,
args->converter->invalidUCharLength);
saveThis->invalidUCharLength=args->converter->invalidUCharLength;
}
args->converter = saveThis;
}
/*-Done with buffer with entire buffer
-Error while converting
*/
if (U_FAILURE(*err) || (args->source == args->sourceLimit))
return;
sourceStart = args->source;
changeState_2022(args->converter,
&(args->source),
args->sourceLimit,
ISO_2022_KR,
err);
/* args->source = sourceStart; */
}while(args->source < args->sourceLimit);
/* return*/
}
static void
UConverter_toUnicode_ISO_2022_KR_OFFSETS_LOGIC(UConverterToUnicodeArgs *args,
UErrorCode* err){
char tempBuf[2];
const char *mySource = ( char *) args->source;
UChar *myTarget = args->target;
const char *mySourceLimit = args->sourceLimit;
UChar32 targetUniChar = 0x0000;
UChar mySourceChar = 0x0000;
UConverterDataISO2022* myData;
UConverterSharedData* sharedData ;
UBool useFallback;
if ((args->converter == NULL) || (args->targetLimit < args->target) || (mySourceLimit < args->source)){
*err = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
/* initialize state */
myData=(UConverterDataISO2022*)(args->converter->extraInfo);
sharedData = myData->fromUnicodeConverter->sharedData;
useFallback = args->converter->useFallback;
if(myData->version==1){
UConverter_toUnicode_ISO_2022_KR_OFFSETS_LOGIC_IBM(args,err);
return;
}
while(mySource< mySourceLimit){
targetUniChar = missingCharMarker;
if(myTarget < args->targetLimit){
mySourceChar= (unsigned char) *mySource++;
if(mySourceChar==UCNV_SI){
myData->currentType = SBCS;
/*consume the source */
continue;
}else if(mySourceChar==UCNV_SO){
myData->currentType = DBCS;
/*consume the source */
continue;
}else if(mySourceChar==ESC_2022 || myData->key!=0){
/*
* Commented out this part to be lenient and allow for
* more escape sequences in ISO-2022-KR byte stream
*
* Already doing some conversion and found escape Sequence
* if(args->converter->mode == UCNV_SO){
* *err = U_ILLEGAL_ESCAPE_SEQUENCE;
* }
* else{
*
*/
mySource--;
changeState_2022(args->converter,&(mySource),
mySourceLimit, ISO_2022_KR, err);
/*}*/
if(U_FAILURE(*err)){
args->target = myTarget;
args->source = mySource;
return;
}
continue;
}
if(myData->currentType==DBCS){
if(args->converter->toUnicodeStatus == 0x00){
args->converter->toUnicodeStatus = (UChar) mySourceChar;
continue;
}
else{
tempBuf[0] = (char) (args->converter->toUnicodeStatus+0x80);
tempBuf[1] = (char) (mySourceChar+0x80);
mySourceChar = (UChar)(mySourceChar + (args->converter->toUnicodeStatus<<8));
args->converter->toUnicodeStatus =0x00;
targetUniChar = _MBCSSimpleGetNextUChar(sharedData, tempBuf, 2, useFallback);
}
}
else{
if(args->converter->fromUnicodeStatus == 0x00){
targetUniChar = _MBCSSimpleGetNextUChar(sharedData, mySource - 1, 1, useFallback);
}
}
if(targetUniChar < 0xfffe){
if(args->offsets)
args->offsets[myTarget - args->target]= mySource - args->source - 1-(myData->currentType==DBCS);
*(myTarget++)=(UChar)targetUniChar;
}
else {
/* Call the callback function*/
toUnicodeCallback(args->converter,mySourceChar,targetUniChar,err);
break;
}
}
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}
args->target = myTarget;
args->source = mySource;
}
/*************************** END ISO2022-KR *********************************/
/*************************** ISO-2022-CN *********************************
*
* Rules for ISO-2022-CN Encoding:
* i) The designator sequence must appear once on a line before any instance
* of character set it designates.
* ii) If two lines contain characters from the same character set, both lines
* must include the designator sequence.
* iii) Once the designator sequence is known, a shifting sequence has to be found
* to invoke the shifting
* iv) All lines start in ASCII and end in ASCII.
* v) Four shifting sequences are employed for this purpose:
*
* Sequcence ASCII Eq Charsets
* ---------- ------- ---------
* SI <SI> US-ASCII
* SO <SO> CNS-11643-1992 Plane 1, GB2312, ISO-IR-165
* SS2 <ESC>N CNS-11643-1992 Plane 2
* SS3 <ESC>O CNS-11643-1992 Planes 3-7
*
* vi)
* SOdesignator : ESC "$" ")" finalchar_for_SO
* SS2designator : ESC "$" "*" finalchar_for_SS2
* SS3designator : ESC "$" "+" finalchar_for_SS3
*
* ESC $ ) A Indicates the bytes following SO are Chinese
* characters as defined in GB 2312-80, until
* another SOdesignation appears
*
*
* ESC $ ) E Indicates the bytes following SO are as defined
* in ISO-IR-165 (for details, see section 2.1),
* until another SOdesignation appears
*
* ESC $ ) G Indicates the bytes following SO are as defined
* in CNS 11643-plane-1, until another
* SOdesignation appears
*
* ESC $ * H Indicates the two bytes immediately following
* SS2 is a Chinese character as defined in CNS
* 11643-plane-2, until another SS2designation
* appears
* (Meaning <ESC>N must preceed every 2 byte
* sequence.)
*
* ESC $ + I Indicates the immediate two bytes following SS3
* is a Chinese character as defined in CNS
* 11643-plane-3, until another SS3designation
* appears
* (Meaning <ESC>O must preceed every 2 byte
* sequence.)
*
* ESC $ + J Indicates the immediate two bytes following SS3
* is a Chinese character as defined in CNS
* 11643-plane-4, until another SS3designation
* appears
* (In English: <ESC>O must preceed every 2 byte
* sequence.)
*
* ESC $ + K Indicates the immediate two bytes following SS3
* is a Chinese character as defined in CNS
* 11643-plane-5, until another SS3designation
* appears
*
* ESC $ + L Indicates the immediate two bytes following SS3
* is a Chinese character as defined in CNS
* 11643-plane-6, until another SS3designation
* appears
*
* ESC $ + M Indicates the immediate two bytes following SS3
* is a Chinese character as defined in CNS
* 11643-plane-7, until another SS3designation
* appears
*
* As in ISO-2022-CN, each line starts in ASCII, and ends in ASCII, and
* has its own designation information before any Chinese characters
* appear
*
*/
/* The following are defined this way to make the strings truely readonly */
static const char SHIFT_IN_STR[] = "\x0F";
static const char SHIFT_OUT_STR[] = "\x0E";
static const char GB_2312_80_STR[] = "\x1B\x24\x29\x41";
static const char ISO_IR_165_STR[] = "\x1B\x24\x29\x45";
static const char CNS_11643_1992_Plane_1_STR[] = "\x1B\x24\x29\x47";
static const char CNS_11643_1992_Plane_2_STR[] = "\x1B\x24\x2A\x48";
static const char CNS_11643_1992_Plane_3_STR[] = "\x1B\x24\x2B\x49";
static const char CNS_11643_1992_Plane_4_STR[] = "\x1B\x24\x2B\x4A";
static const char CNS_11643_1992_Plane_5_STR[] = "\x1B\x24\x2B\x4B";
static const char CNS_11643_1992_Plane_6_STR[] = "\x1B\x24\x2B\x4C";
static const char CNS_11643_1992_Plane_7_STR[] = "\x1B\x24\x2B\x4D";
/********************** ISO2022-CN Data **************************/
static const char* const escSeqCharsCN[10] ={
SHIFT_IN_STR, /* ASCII */
GB_2312_80_STR,
ISO_IR_165_STR,
CNS_11643_1992_Plane_1_STR,
CNS_11643_1992_Plane_2_STR,
CNS_11643_1992_Plane_3_STR,
CNS_11643_1992_Plane_4_STR,
CNS_11643_1992_Plane_5_STR,
CNS_11643_1992_Plane_6_STR,
CNS_11643_1992_Plane_7_STR
};
static void
UConverter_fromUnicode_ISO_2022_CN_OFFSETS_LOGIC(UConverterFromUnicodeArgs* args, UErrorCode* err){
UConverterDataISO2022 *converterData;
uint8_t *target = (uint8_t *) args->target;
const uint8_t *targetLimit = (const uint8_t *) args->targetLimit;
const UChar* source = args->source;
const UChar* sourceLimit = args->sourceLimit;
int32_t* offsets = args->offsets;
UChar32 sourceChar;
char buffer[8];
int32_t len;
int8_t choices[3];
int32_t choiceCount;
uint32_t targetValue;
UBool useFallback;
/* set up the state */
converterData = (UConverterDataISO2022*)args->converter->extraInfo;
useFallback = args->converter->useFallback;
choiceCount = 0;
/* check if the last codepoint of previous buffer was a lead surrogate*/
if((sourceChar = args->converter->fromUChar32)!=0 && target< targetLimit) {
goto getTrail;
}
while( source < sourceLimit){
if(target < targetLimit){
sourceChar = *(source++);
/*check if the char is a First surrogate*/
if(UTF_IS_SURROGATE(sourceChar)) {
if(UTF_IS_SURROGATE_FIRST(sourceChar)) {
getTrail:
/*look ahead to find the trail surrogate*/
if(source < sourceLimit) {
/* test the following code unit */
UChar trail=(UChar) *source;
if(UTF_IS_SECOND_SURROGATE(trail)) {
source++;
sourceChar=UTF16_GET_PAIR_VALUE(sourceChar, trail);
args->converter->fromUChar32=0x00;
/* convert this supplementary code point */
/* exit this condition tree */
} else {
/* this is an unmatched lead code unit (1st surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
args->converter->fromUChar32=sourceChar;
break;
}
} else {
/* no more input */
args->converter->fromUChar32=sourceChar;
break;
}
} else {
/* this is an unmatched trail code unit (2nd surrogate) */
/* callback(illegal) */
*err=U_ILLEGAL_CHAR_FOUND;
args->converter->fromUChar32=sourceChar;
break;
}
}
/* do the conversion */
if(sourceChar <= 0x007f ){
/* US-ASCII */
if(converterData->fromU2022State.g == 0) {
buffer[0] = (char)sourceChar;
len = 1;
} else {
buffer[0] = UCNV_SI;
buffer[1] = (char)sourceChar;
len = 2;
converterData->fromU2022State.g = 0;
}
if(sourceChar == CR || sourceChar == LF) {
/* reset the state at the end of a line */
uprv_memset(&converterData->fromU2022State, 0, sizeof(ISO2022State));
}
}
else{
/* convert U+0080..U+10ffff */
UConverter *cnv;
int32_t i;
int8_t cs, g;
if(choiceCount == 0) {
/* try the current SO/G1 converter first */
choices[0] = converterData->fromU2022State.cs[1];
/* default to GB2312_1 if none is designated yet */
if(choices[0] == 0) {
choices[0] = GB2312_1;
}
if(converterData->version == 0) {
/* ISO-2022-CN */
/* try the other SO/G1 converter; a CNS_11643_1 lookup may result in any plane */
if(choices[0] == GB2312_1) {
choices[1] = (int8_t)CNS_11643_1;
} else {
choices[1] = (int8_t)GB2312_1;
}
choiceCount = 2;
} else {
/* ISO-2022-CN-EXT */
/* try one of the other converters */
switch(choices[0]) {
case GB2312_1:
choices[1] = (int8_t)CNS_11643_1;
choices[2] = (int8_t)ISO_IR_165;
break;
case ISO_IR_165:
choices[1] = (int8_t)GB2312_1;
choices[2] = (int8_t)CNS_11643_1;
break;
default: /* CNS_11643_x */
choices[1] = (int8_t)GB2312_1;
choices[2] = (int8_t)ISO_IR_165;
break;
}
choiceCount = 3;
}
}
cs = g = 0;
len = 0;
for(i = 0; i < choiceCount && len == 0; ++i) {
cs = choices[i];
if(cs > 0) {
if(cs > CNS_11643_0) {
cnv = converterData->myConverterArray[CNS_11643];
MBCS_FROM_UCHAR32_ISO2022(cnv->sharedData,sourceChar,&targetValue,useFallback,&len,MBCS_OUTPUT_3);
if(len==3) {
cs = (int8_t)(CNS_11643_0 + (targetValue >> 16) - 0x80);
len = 2;
if(cs == CNS_11643_1) {
g = 1;
} else if(cs == CNS_11643_2) {
g = 2;
} else /* plane 3..7 */ if(converterData->version == 1) {
g = 3;
} else {
/* ISO-2022-CN (without -EXT) does not support plane 3..7 */
len = 0;
}
}
} else {
/* GB2312_1 or ISO-IR-165 */
cnv = converterData->myConverterArray[cs];
MBCS_FROM_UCHAR32_ISO2022(cnv->sharedData,sourceChar,&targetValue,useFallback,&len,MBCS_OUTPUT_2);
g = 1; /* used if len == 2 */
}
}
}
if(len > 0) {
len = 0; /* count output bytes; it must have been len == 2 */
/* write the designation sequence if necessary */
if(cs != converterData->fromU2022State.cs[g]) {
if(cs < CNS_11643) {
uprv_memcpy(buffer, escSeqCharsCN[cs], 4);
} else {
uprv_memcpy(buffer, escSeqCharsCN[CNS_11643 + (cs - CNS_11643_1)], 4);
}
len = 4;
converterData->fromU2022State.cs[g] = cs;
if(g == 1) {
/* changing the SO/G1 charset invalidates the choices[] */
choiceCount = 0;
}
}
/* write the shift sequence if necessary */
if(g != converterData->fromU2022State.g) {
switch(g) {
case 1:
buffer[len++] = UCNV_SO;
/* set the new state only if it is the locking shift SO/G1, not for SS2 or SS3 */
converterData->fromU2022State.g = 1;
break;
case 2:
buffer[len++] = 0x1b;
buffer[len++] = 0x4e;
break;
default: /* case 3 */
buffer[len++] = 0x1b;
buffer[len++] = 0x4f;
break;
}
}
/* write the two output bytes */
buffer[len++] = (char)(targetValue >> 8);
buffer[len++] = (char)targetValue;
} else {
/* if we cannot find the character after checking all codepages
* then this is an error
*/
*err = U_INVALID_CHAR_FOUND;
args->converter->fromUChar32=sourceChar;
break;
}
}
/* output len>0 bytes in buffer[] */
if(len == 1) {
*target++ = buffer[0];
if(offsets) {
*offsets++ = source - args->source - 1; /* -1: known to be ASCII */
}
} else if(len == 2 && (target + 2) <= targetLimit) {
*target++ = buffer[0];
*target++ = buffer[1];
if(offsets) {
int32_t sourceIndex = (int32_t)(source - args->source - U16_LENGTH(sourceChar));
*offsets++ = sourceIndex;
*offsets++ = sourceIndex;
}
} else {
ucnv_fromUWriteBytes(
args->converter,
buffer, len,
(char **)&target, (const char *)targetLimit,
&offsets, (int32_t)(source - args->source - U16_LENGTH(sourceChar)),
err);
if(U_FAILURE(*err)) {
break;
}
}
} /* end if(myTargetIndex<myTargetLength) */
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}/* end while(mySourceIndex<mySourceLength) */
/*save the state and return */
args->source = source;
args->target = (char*)target;
}
static void
UConverter_toUnicode_ISO_2022_CN_OFFSETS_LOGIC(UConverterToUnicodeArgs *args,
UErrorCode* err){
char tempBuf[3];
const char *mySource = ( char *) args->source;
UChar *myTarget = args->target;
const char *mySourceLimit = args->sourceLimit;
uint32_t targetUniChar = 0x0000;
uint32_t mySourceChar = 0x0000;
UConverterDataISO2022* myData;
myData=(UConverterDataISO2022*)(args->converter->extraInfo);
if(myData->key != 0) {
/* continue with a partial escape sequence */
goto escape;
} else if(args->converter->toULength == 1 && mySource < mySourceLimit && myTarget < args->targetLimit) {
/* continue with a partial double-byte character */
mySourceChar = args->converter->toUBytes[0];
args->converter->toULength = 0;
goto getTrailByte;
}
while(mySource < mySourceLimit){
targetUniChar =missingCharMarker;
if(myTarget < args->targetLimit){
mySourceChar= (unsigned char) *mySource++;
switch(mySourceChar){
case UCNV_SI:
myData->toU2022State.g=0;
continue;
case UCNV_SO:
if(myData->toU2022State.cs[1] != 0) {
myData->toU2022State.g=1;
continue;
} else {
/* illegal to have SO before a matching designator */
break;
}
case ESC_2022:
mySource--;
escape:
changeState_2022(args->converter,&(mySource),
mySourceLimit, ISO_2022_CN,err);
/* invalid or illegal escape sequence */
if(U_FAILURE(*err)){
args->target = myTarget;
args->source = mySource;
return;
}
continue;
/* ISO-2022-CN does not use single-byte (C1) SS2 and SS3 */
case CR:
/*falls through*/
case LF:
uprv_memset(&myData->toU2022State, 0, sizeof(ISO2022State));
/* falls through */
default:
/* convert one or two bytes */
if(myData->toU2022State.g != 0) {
getTrailByte:
if(mySource < mySourceLimit) {
UConverter *cnv;
StateEnum tempState;
int32_t tempBufLen;
char trailByte = *mySource++;
tempState = (StateEnum)myData->toU2022State.cs[myData->toU2022State.g];
if(tempState > CNS_11643_0) {
cnv = myData->myConverterArray[CNS_11643];
tempBuf[0] = (char) (0x80+(tempState-CNS_11643_0));
tempBuf[1] = (char) (mySourceChar);
tempBuf[2] = trailByte;
tempBufLen = 3;
}else{
cnv = myData->myConverterArray[tempState];
tempBuf[0] = (char) (mySourceChar);
tempBuf[1] = trailByte;
tempBufLen = 2;
}
mySourceChar = (mySourceChar << 8) | (uint8_t)(trailByte);
if(myData->toU2022State.g>=2) {
/* return from a single-shift state to the previous one */
myData->toU2022State.g=myData->toU2022State.prevG;
}
targetUniChar = _MBCSSimpleGetNextUChar(cnv->sharedData, tempBuf, tempBufLen, FALSE);
} else {
args->converter->toUBytes[0] = (uint8_t)mySourceChar;
args->converter->toULength = 1;
goto endloop;
}
}
else{
if(mySourceChar <= 0x7f) {
targetUniChar = (UChar) mySourceChar;
}
}
break;
}
if(targetUniChar < (missingCharMarker-1/*0xfffe*/)){
if(args->offsets){
args->offsets[myTarget - args->target]= mySource - args->source - (mySourceChar <= 0xff ? 1 : 2);
}
*(myTarget++)=(UChar)targetUniChar;
}
else if(targetUniChar > missingCharMarker){
/* disassemble the surrogate pair and write to output*/
targetUniChar-=0x0010000;
*myTarget = (UChar)(0xd800+(UChar)(targetUniChar>>10));
if(args->offsets){
args->offsets[myTarget - args->target]= mySource - args->source - (mySourceChar <= 0xff ? 1 : 2);
}
++myTarget;
if(myTarget< args->targetLimit){
*myTarget = (UChar)(0xdc00+(UChar)(targetUniChar&0x3ff));
if(args->offsets){
args->offsets[myTarget - args->target]= mySource - args->source - (mySourceChar <= 0xff ? 1 : 2);
}
++myTarget;
}else{
args->converter->UCharErrorBuffer[args->converter->UCharErrorBufferLength++]=
(UChar)(0xdc00+(UChar)(targetUniChar&0x3ff));
}
}
else{
/* Call the callback function*/
toUnicodeCallback(args->converter,mySourceChar,targetUniChar,err);
break;
}
}
else{
*err =U_BUFFER_OVERFLOW_ERROR;
break;
}
}
endloop:
args->target = myTarget;
args->source = mySource;
}
static void
_ISO_2022_WriteSub(UConverterFromUnicodeArgs *args, int32_t offsetIndex, UErrorCode *err) {
UConverter *cnv = args->converter;
UConverterDataISO2022 *myConverterData=(UConverterDataISO2022 *) cnv->extraInfo;
char *p;
char buffer[4];
p = buffer;
switch(myConverterData->locale[0]){
case 'j':
if(myConverterData->fromUnicodeCurrentState!= ASCII){
myConverterData->fromUnicodeCurrentState= ASCII;
myConverterData->currentType = (Cnv2022Type) myConverterType[myConverterData->fromUnicodeCurrentState];
*p++ = '\x1b';
*p++ = '\x28';
*p++ = '\x42';
}
*p++ = cnv->subChar[0];
break;
case 'c':
if(myConverterData->fromU2022State.g != 0) {
/* not in ASCII mode: switch to ASCII */
myConverterData->fromU2022State.g = 0;
*p++ = UCNV_SI;
}
*p++ = cnv->subChar[0];
break;
case 'k':
if(args->converter->fromUnicodeStatus){
args->converter->fromUnicodeStatus=0x00;
*p++= UCNV_SI;
}
*p++ = cnv->subChar[0];
default:
/* not expected */
break;
}
ucnv_cbFromUWriteBytes(args,
buffer, (int32_t)(p - buffer),
offsetIndex, err);
}
/* structure for SafeClone calculations */
struct cloneStruct
{
UConverter cnv;
UConverterDataISO2022 mydata;
UConverter currentCnv; /**< for ISO_2022 converter if the current converter is open */
UConverter clonedConverters[1]; /* Actually a variable sized array for all of the sub converters to be cloned. */
};
static UConverter *
_ISO_2022_SafeClone(
const UConverter *cnv,
void *stackBuffer,
int32_t *pBufferSize,
UErrorCode *status)
{
struct cloneStruct * localClone;
int32_t bufferSizeNeeded = sizeof(struct cloneStruct);
UConverterDataISO2022* cnvData = (UConverterDataISO2022*)cnv->extraInfo;
int32_t i;
int32_t sizes[UCNV_2022_MAX_CONVERTERS];
int32_t numConverters = 0;
int32_t currentConverterIndex = -1;
int32_t fromUnicodeConverterIndex = -1;
int32_t currentConverterSize = 0;
char *ptr; /* buffer pointer */
if (U_FAILURE(*status)) {
return 0;
}
for(i=0;(i<UCNV_2022_MAX_CONVERTERS)&&cnvData->myConverterArray[i];i++) {
int32_t size;
size = 0;
ucnv_safeClone(cnvData->myConverterArray[i], NULL, &size, status);
bufferSizeNeeded += size;
sizes[i] = size;
numConverters++;
if(cnvData->currentConverter == cnvData->myConverterArray[i]) {
currentConverterIndex = i;
}
if(cnvData->fromUnicodeConverter == cnvData->myConverterArray[i]) {
fromUnicodeConverterIndex = i;
}
}
if(currentConverterIndex == -1) { /* -1 means - not found in array. Clone separately */
currentConverterSize = 0;
if(cnvData->currentConverter) {
ucnv_safeClone(cnvData->currentConverter, NULL, &currentConverterSize, status);
bufferSizeNeeded += currentConverterSize;
}
}
for(;i<UCNV_2022_MAX_CONVERTERS;i++) { /* zero the other sizes */
sizes[i]=0;
}
if (*pBufferSize == 0) { /* 'preflighting' request - set needed size into *pBufferSize */
*pBufferSize = bufferSizeNeeded;
return 0;
}
if(*pBufferSize < bufferSizeNeeded) {
*status = U_BUFFER_OVERFLOW_ERROR;
return 0;
}
localClone = (struct cloneStruct *)stackBuffer;
uprv_memcpy(&localClone->cnv, cnv, sizeof(UConverter));
uprv_memcpy(&localClone->mydata, cnv->extraInfo, sizeof(UConverterDataISO2022));
/* clone back sub cnvs */
ptr = (char*)&localClone->clonedConverters;
for(i=0;i<numConverters;i++) {
int32_t size;
size = sizes[i];
localClone->mydata.myConverterArray[i] = ucnv_safeClone(cnvData->myConverterArray[i], (UConverter*)ptr, &size, status);
ptr += size;
}
for(;i<UCNV_2022_MAX_CONVERTERS;i++) {
localClone->mydata.myConverterArray[i] = NULL;
}
if(currentConverterIndex == -1) { /* -1 = not found in list */
/* KR version 1 also uses the state in currentConverter for preserving state
* so we need to clone it too!
*/
if(cnvData->currentConverter) {
localClone->mydata.currentConverter = ucnv_safeClone(cnvData->currentConverter, ptr, &currentConverterSize, status);
ptr += currentConverterSize;
} else {
localClone->mydata.currentConverter = NULL;
}
} else {
localClone->mydata.currentConverter = localClone->mydata.myConverterArray[currentConverterIndex];
}
if(fromUnicodeConverterIndex != -1) {
/* fromUnicodeConverter is in the list */
localClone->mydata.fromUnicodeConverter = localClone->mydata.myConverterArray[fromUnicodeConverterIndex];
} else if(cnvData->currentConverter == cnvData->fromUnicodeConverter) {
/* fromUnicodeConverter is the same as currentConverter */
localClone->mydata.fromUnicodeConverter = localClone->mydata.currentConverter;
} else {
/* fromUnicodeConverter is NULL */
localClone->mydata.fromUnicodeConverter = NULL;
}
localClone->cnv.extraInfo = &localClone->mydata; /* set pointer to extra data */
return &localClone->cnv;
}
static void
_ISO_2022_GetUnicodeSet(const UConverter *cnv,
USet *set,
UConverterUnicodeSet which,
UErrorCode *pErrorCode)
{
int32_t i;
USet *cnvSet;
UConverterDataISO2022* cnvData;
if (U_FAILURE(*pErrorCode)) {
return;
}
if (cnv->sharedData == &_ISO2022Data) {
/* We use UTF-8 in this case */
uset_addRange(set, 0, 0xd7FF);
uset_addRange(set, 0xE000, 0x10FFFF);
return;
}
cnvData = (UConverterDataISO2022*)cnv->extraInfo;
if (cnv->sharedData == &_ISO2022KRData && cnvData->currentConverter != NULL) {
ucnv_getUnicodeSet(cnvData->currentConverter, set, which, pErrorCode);
return;
}
cnvSet = uset_open(0, 0);
if (!cnvSet) {
*pErrorCode =U_MEMORY_ALLOCATION_ERROR;
return;
}
for (i=0; i<UCNV_2022_MAX_CONVERTERS; i++) {
if(cnvData->myConverterArray[i]!=NULL) {
ucnv_getUnicodeSet(cnvData->myConverterArray[i], cnvSet, which, pErrorCode);
uset_addAll(set, cnvSet /* pErrorCode */);
}
}
uset_close(cnvSet);
}
static const UConverterImpl _ISO2022Impl={
UCNV_ISO_2022,
NULL,
NULL,
_ISO2022Open,
_ISO2022Close,
_ISO2022Reset,
#ifdef U_ENABLE_GENERIC_ISO_2022
T_UConverter_toUnicode_ISO_2022_OFFSETS_LOGIC,
T_UConverter_toUnicode_ISO_2022_OFFSETS_LOGIC,
T_UConverter_fromUnicode_UTF8,
T_UConverter_fromUnicode_UTF8_OFFSETS_LOGIC,
#else
NULL,
NULL,
NULL,
NULL,
#endif
NULL,
NULL,
_ISO2022getName,
_ISO_2022_WriteSub,
_ISO_2022_SafeClone,
_ISO_2022_GetUnicodeSet
};
static const UConverterStaticData _ISO2022StaticData={
sizeof(UConverterStaticData),
"ISO_2022",
2022,
UCNV_IBM,
UCNV_ISO_2022,
1,
3, /* max 3 bytes per UChar from UTF-8 (4 bytes from surrogate _pair_) */
{ 0x1a, 0, 0, 0 },
1,
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 _ISO2022Data={
sizeof(UConverterSharedData),
~((uint32_t) 0),
NULL,
NULL,
&_ISO2022StaticData,
FALSE,
&_ISO2022Impl,
0
};
/*************JP****************/
static const UConverterImpl _ISO2022JPImpl={
UCNV_ISO_2022,
NULL,
NULL,
_ISO2022Open,
_ISO2022Close,
_ISO2022Reset,
UConverter_toUnicode_ISO_2022_JP_OFFSETS_LOGIC,
UConverter_toUnicode_ISO_2022_JP_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_JP_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_JP_OFFSETS_LOGIC,
NULL,
NULL,
_ISO2022getName,
_ISO_2022_WriteSub,
_ISO_2022_SafeClone,
_ISO_2022_GetUnicodeSet
};
static const UConverterStaticData _ISO2022JPStaticData={
sizeof(UConverterStaticData),
"ISO_2022_JP",
0,
UCNV_IBM,
UCNV_ISO_2022,
1,
6, /* max 6 bytes per UChar: 4-byte escape sequence + DBCS */
{ 0x1a, 0, 0, 0 },
1,
FALSE,
FALSE,
0,
0,
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
};
static const UConverterSharedData _ISO2022JPData={
sizeof(UConverterSharedData),
~((uint32_t) 0),
NULL,
NULL,
&_ISO2022JPStaticData,
FALSE,
&_ISO2022JPImpl,
0
};
/************* KR ***************/
static const UConverterImpl _ISO2022KRImpl={
UCNV_ISO_2022,
NULL,
NULL,
_ISO2022Open,
_ISO2022Close,
_ISO2022Reset,
UConverter_toUnicode_ISO_2022_KR_OFFSETS_LOGIC,
UConverter_toUnicode_ISO_2022_KR_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_KR_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_KR_OFFSETS_LOGIC,
NULL,
NULL,
_ISO2022getName,
_ISO_2022_WriteSub,
_ISO_2022_SafeClone,
_ISO_2022_GetUnicodeSet
};
static const UConverterStaticData _ISO2022KRStaticData={
sizeof(UConverterStaticData),
"ISO_2022_KR",
0,
UCNV_IBM,
UCNV_ISO_2022,
1,
3, /* max 3 bytes per UChar: SO+DBCS */
{ 0x1a, 0, 0, 0 },
1,
FALSE,
FALSE,
0,
0,
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
};
static const UConverterSharedData _ISO2022KRData={
sizeof(UConverterSharedData),
~((uint32_t) 0),
NULL,
NULL,
&_ISO2022KRStaticData,
FALSE,
&_ISO2022KRImpl,
0
};
/*************** CN ***************/
static const UConverterImpl _ISO2022CNImpl={
UCNV_ISO_2022,
NULL,
NULL,
_ISO2022Open,
_ISO2022Close,
_ISO2022Reset,
UConverter_toUnicode_ISO_2022_CN_OFFSETS_LOGIC,
UConverter_toUnicode_ISO_2022_CN_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_CN_OFFSETS_LOGIC,
UConverter_fromUnicode_ISO_2022_CN_OFFSETS_LOGIC,
NULL,
NULL,
_ISO2022getName,
_ISO_2022_WriteSub,
_ISO_2022_SafeClone,
_ISO_2022_GetUnicodeSet
};
static const UConverterStaticData _ISO2022CNStaticData={
sizeof(UConverterStaticData),
"ISO_2022_CN",
0,
UCNV_IBM,
UCNV_ISO_2022,
2,
8, /* max 8 bytes per UChar: 4-byte CNS designator + 2 bytes for SS2/SS3 + DBCS */
{ 0x1a, 0, 0, 0 },
1,
FALSE,
FALSE,
0,
0,
{ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } /* reserved */
};
static const UConverterSharedData _ISO2022CNData={
sizeof(UConverterSharedData),
~((uint32_t) 0),
NULL,
NULL,
&_ISO2022CNStaticData,
FALSE,
&_ISO2022CNImpl,
0
};
#endif /* #if !UCONFIG_NO_LEGACY_CONVERSION */
Reference in New Issue View Git Blame Copy Permalink