Make GB18030/GBK compliant to HTML5

source/common/ucnvmbcs.cpp

 /*
 ******************************************************************************
 *
 *   Copyright (C) 2000-2014, International Business Machines
 *   Corporation and others.  All Rights Reserved.
 *
 ******************************************************************************
 *   file name:  ucnvmbcs.cpp
 *   encoding:   US-ASCII
 *   tab size:   8 (not used)
@@ skipping 534 matching lines @@
     {0x3CE1, 0x4055, LINEAR(0x8231D438), LINEAR(0x8232AF32)},
     {0x361B, 0x3917, LINEAR(0x8230A633), LINEAR(0x8230F237)},
     {0x49B8, 0x4C76, LINEAR(0x8234A131), LINEAR(0x8234E733)},
     {0x4160, 0x4336, LINEAR(0x8232C937), LINEAR(0x8232F837)},
     {0x478E, 0x4946, LINEAR(0x8233E838), LINEAR(0x82349638)},
     {0x44D7, 0x464B, LINEAR(0x8233A339), LINEAR(0x8233C931)},
     {0xFFE6, 0xFFFF, LINEAR(0x8431A234), LINEAR(0x8431A439)}
 };
 
 /* bit flag for UConverter.options indicating GB 18030 special handling */
-#define _MBCS_OPTION_GB18030 0x8000
+#define _MBCS_OPTION_GB18030_TOU 0x08000
+#define _MBCS_OPTION_GB18030_FROMU 0x10000
 
 /* bit flag for UConverter.options indicating KEIS,JEF,JIF special handling */
 #define _MBCS_OPTION_KEIS 0x01000
 #define _MBCS_OPTION_JEF  0x02000
 #define _MBCS_OPTION_JIPS 0x04000
 
 #define KEIS_SO_CHAR_1 0x0A
 #define KEIS_SO_CHAR_2 0x42
 #define KEIS_SI_CHAR_1 0x0A
 #define KEIS_SI_CHAR_2 0x41
@@ skipping 496 matching lines @@
             UCNV_SET_FILTER_DBCS_ONLY :
             UCNV_SET_FILTER_NONE,
         pErrorCode);
 }
 
 static void
 ucnv_MBCSGetUnicodeSet(const UConverter *cnv,
                    const USetAdder *sa,
                    UConverterUnicodeSet which,
                    UErrorCode *pErrorCode) {
-    if(cnv->options&_MBCS_OPTION_GB18030) {
+    if(cnv->options&_MBCS_OPTION_GB18030_FROMU) {
         sa->addRange(sa->set, 0, 0xd7ff);
         sa->addRange(sa->set, 0xe000, 0x10ffff);
     } else {
         ucnv_MBCSGetUnicodeSetForUnicode(cnv->sharedData, sa, which, pErrorCode);
     }
 }
 
 /* conversion extensions for input not in the main table -------------------- */
 
 /*
@@ skipping 29 matching lines @@
             cp, source, sourceLimit,
             (char **)target, (char *)targetLimit,
             offsets, sourceIndex,
             flush,
             pErrorCode)
     ) {
         return 0; /* an extension mapping handled the input */
     }
 
     /* GB 18030 */
-    if((cnv->options&_MBCS_OPTION_GB18030)!=0) {
+    if((cnv->options&_MBCS_OPTION_GB18030_FROMU)!=0) {
         const uint32_t *range;
         int32_t i;
 
         range=gb18030Ranges[0];
         for(i=0; i<UPRV_LENGTHOF(gb18030Ranges); range+=4, ++i) {
             if(range[0]<=(uint32_t)cp && (uint32_t)cp<=range[1]) {
                 /* found the Unicode code point, output the four-byte sequence for it */
                 uint32_t linear;
                 char bytes[4];
 
@@ skipping 44 matching lines @@
             length, (const char **)source, (const char *)sourceLimit,
             target, targetLimit,
             offsets, sourceIndex,
             flush,
             pErrorCode)
     ) {
         return 0; /* an extension mapping handled the input */
     }
 
     /* GB 18030 */
-    if(length==4 && (cnv->options&_MBCS_OPTION_GB18030)!=0) {
+    if(length==4 && (cnv->options&_MBCS_OPTION_GB18030_TOU)!=0) {
         const uint32_t *range;
         uint32_t linear;
         int32_t i;
 
         linear=LINEAR_18030(cnv->toUBytes[0], cnv->toUBytes[1], cnv->toUBytes[2], cnv->toUBytes[3]);
         range=gb18030Ranges[0];
         for(i=0; i<UPRV_LENGTHOF(gb18030Ranges); range+=4, ++i) {
             if(range[2]<=linear && linear<=range[3]) {
                 /* found the sequence, output the Unicode code point for it */
                 *pErrorCode=U_ZERO_ERROR;
@@ skipping 726 matching lines @@
 
                 /* the option does not apply, remove it */
                 cnv->options=pArgs->options&=~UCNV_OPTION_SWAP_LFNL;
             }
         }
     }
 
     if(uprv_strstr(pArgs->name, "18030")!=NULL) {
         if(uprv_strstr(pArgs->name, "gb18030")!=NULL || uprv_strstr(pArgs->name, "GB18030")!=NULL) {
             /* set a flag for GB 18030 mode, which changes the callback behavior */
-            cnv->options|=_MBCS_OPTION_GB18030;
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
+            cnv->options|=_MBCS_OPTION_GB18030_FROMU;
+#endif
+            cnv->options|=_MBCS_OPTION_GB18030_TOU;
         }
     } else if((uprv_strstr(pArgs->name, "KEIS")!=NULL) || (uprv_strstr(pArgs->name, "keis")!=NULL)) {
         /* set a flag for KEIS converter, which changes the SI/SO character sequence */
         cnv->options|=_MBCS_OPTION_KEIS;
     } else if((uprv_strstr(pArgs->name, "JEF")!=NULL) || (uprv_strstr(pArgs->name, "jef")!=NULL)) {
         /* set a flag for JEF converter, which changes the SI/SO character sequence */
         cnv->options|=_MBCS_OPTION_JEF;
     } else if((uprv_strstr(pArgs->name, "JIPS")!=NULL) || (uprv_strstr(pArgs->name, "jips")!=NULL)) {
         /* set a flag for JIPS converter, which changes the SI/SO character sequence */
         cnv->options|=_MBCS_OPTION_JIPS;
@@ skipping 3758 matching lines @@
         return (UConverterType)UCNV_SBCS;
     } else if((converter->sharedData->mbcs.outputType&0xff)==MBCS_OUTPUT_2_SISO) {
         return (UConverterType)UCNV_EBCDIC_STATEFUL;
     } else if(converter->sharedData->staticData->minBytesPerChar==2 && converter->sharedData->staticData->maxBytesPerChar==2) {
         return (UConverterType)UCNV_DBCS;
     }
     return (UConverterType)UCNV_MBCS;
 }
 
 #endif /* #if !UCONFIG_NO_LEGACY_CONVERSION */