Index: source/common/ucnv2022.cpp
===================================================================
--- source/common/ucnv2022.cpp (revision 259715)
+++ source/common/ucnv2022.cpp (working copy)
@@ -167,13 +167,19 @@
* all versions, not just JIS7 and JIS8.
* - ICU does not distinguish between different versions of JIS X 0208.
*/
+#if UCONFIG_NO_NON_HTML5_CONVERSION
+enum { MAX_JA_VERSION=0 };
+#else
enum { MAX_JA_VERSION=4 };
+#endif
static const uint16_t jpCharsetMasks[MAX_JA_VERSION+1]={
CSM(ASCII)|CSM(JISX201)|CSM(JISX208)|CSM(HWKANA_7BIT),
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
CSM(ASCII)|CSM(JISX201)|CSM(JISX208)|CSM(HWKANA_7BIT)|CSM(JISX212),
CSM(ASCII)|CSM(JISX201)|CSM(JISX208)|CSM(HWKANA_7BIT)|CSM(JISX212)|CSM(GB2312)|CSM(KSC5601)|CSM(ISO8859_1)|CSM(ISO8859_7),
CSM(ASCII)|CSM(JISX201)|CSM(JISX208)|CSM(HWKANA_7BIT)|CSM(JISX212)|CSM(GB2312)|CSM(KSC5601)|CSM(ISO8859_1)|CSM(ISO8859_7),
CSM(ASCII)|CSM(JISX201)|CSM(JISX208)|CSM(HWKANA_7BIT)|CSM(JISX212)|CSM(GB2312)|CSM(KSC5601)|CSM(ISO8859_1)|CSM(ISO8859_7)
+#endif
};
typedef enum {
@@ -361,14 +367,25 @@
};
+/* Enable ISO-2022-{KR,CN,CN-Ext} for now.
+ * TODO(jshin): Disable it when we know what to do about 'replacement'
+ * encodings. See http://crbug.com/277037 and
+ * https://codereview.chromium.org/145973021/
+ */
+#ifndef U_ENABLE_ISO_2022_KR_CN
+#define U_ENABLE_ISO_2022_KR_CN 1
+#endif
+
/* Type def for refactoring changeState_2022 code*/
typedef enum{
#ifdef U_ENABLE_GENERIC_ISO_2022
ISO_2022=0,
#endif
ISO_2022_JP=1,
+#ifdef U_ENABLE_ISO_2022_KR_CN
ISO_2022_KR=2,
ISO_2022_CN=3
+#endif
} Variant2022;
/*********** ISO 2022 Converter Protos ***********/
@@ -485,24 +502,28 @@
/* prevent indexing beyond jpCharsetMasks[] */
myConverterData->version = version = 0;
}
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
if(jpCharsetMasks[version]&CSM(ISO8859_7)) {
myConverterData->myConverterArray[ISO8859_7] =
ucnv_loadSharedData("ISO8859_7", &stackPieces, &stackArgs, errorCode);
}
+#endif
myConverterData->myConverterArray[JISX208] =
ucnv_loadSharedData("Shift-JIS", &stackPieces, &stackArgs, errorCode);
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
if(jpCharsetMasks[version]&CSM(JISX212)) {
myConverterData->myConverterArray[JISX212] =
ucnv_loadSharedData("jisx-212", &stackPieces, &stackArgs, errorCode);
}
if(jpCharsetMasks[version]&CSM(GB2312)) {
myConverterData->myConverterArray[GB2312] =
- ucnv_loadSharedData("ibm-5478", &stackPieces, &stackArgs, errorCode); /* gb_2312_80-1 */
+ ucnv_loadSharedData("noop-gb2312_gl", &stackPieces, &stackArgs, errorCode); /* gb_2312_80-1 */
}
if(jpCharsetMasks[version]&CSM(KSC5601)) {
myConverterData->myConverterArray[KSC5601] =
ucnv_loadSharedData("ksc_5601", &stackPieces, &stackArgs, errorCode);
}
+#endif
/* set the function pointers to appropriate funtions */
cnv->sharedData=(UConverterSharedData*)(&_ISO2022JPData);
@@ -513,6 +534,7 @@
myConverterData->name[len]=(char)(myConverterData->version+(int)'0');
myConverterData->name[len+1]='\0';
}
+#ifdef U_ENABLE_ISO_2022_KR_CN
else if(myLocale[0]=='k' && (myLocale[1]=='o'|| myLocale[1]=='r') &&
(myLocale[2]=='_' || myLocale[2]=='\0'))
{
@@ -558,13 +580,13 @@
/* open the required converters and cache them */
myConverterData->myConverterArray[GB2312_1] =
- ucnv_loadSharedData("ibm-5478", &stackPieces, &stackArgs, errorCode);
+ ucnv_loadSharedData("noop-gb2312_gl", &stackPieces, &stackArgs, errorCode);
if(version==1) {
myConverterData->myConverterArray[ISO_IR_165] =
- ucnv_loadSharedData("iso-ir-165", &stackPieces, &stackArgs, errorCode);
+ ucnv_loadSharedData("noop-iso-ir-165", &stackPieces, &stackArgs, errorCode);
}
myConverterData->myConverterArray[CNS_11643] =
- ucnv_loadSharedData("cns-11643-1992", &stackPieces, &stackArgs, errorCode);
+ ucnv_loadSharedData("noop-cns-11643", &stackPieces, &stackArgs, errorCode);
/* set the function pointers to appropriate funtions */
@@ -582,6 +604,7 @@
(void)uprv_strcpy(myConverterData->name,"ISO_2022,locale=zh,version=2");
}
}
+#endif // U_ENABLE_ISO_2022_KR_CN
else{
#ifdef U_ENABLE_GENERIC_ISO_2022
myConverterData->isFirstBuffer = TRUE;
Index: source/common/ucnvbocu.cpp
===================================================================
--- source/common/ucnvbocu.cpp (revision 259715)
+++ source/common/ucnvbocu.cpp (working copy)
@@ -19,7 +19,7 @@
#include "unicode/utypes.h"
-#if !UCONFIG_NO_CONVERSION
+#if !UCONFIG_NO_CONVERSION && !UCONFIG_NO_NON_HTML5_CONVERSION
#include "unicode/ucnv.h"
#include "unicode/ucnv_cb.h"
Index: source/common/ucnvisci.c
===================================================================
--- source/common/ucnvisci.c (revision 259715)
+++ source/common/ucnvisci.c (working copy)
@@ -17,7 +17,7 @@
#include "unicode/utypes.h"
-#if !UCONFIG_NO_CONVERSION && !UCONFIG_NO_LEGACY_CONVERSION
+#if !UCONFIG_NO_CONVERSION && !UCONFIG_NO_LEGACY_CONVERSION && !UCONFIG_NO_NON_HTML5_CONVERSION
#include "unicode/ucnv.h"
#include "unicode/ucnv_cb.h"
Index: source/common/ucnvscsu.c
===================================================================
--- source/common/ucnvscsu.c (revision 259715)
+++ source/common/ucnvscsu.c (working copy)
@@ -21,7 +21,7 @@
#include "unicode/utypes.h"
-#if !UCONFIG_NO_CONVERSION
+#if !UCONFIG_NO_CONVERSION && !UCONFIG_NO_NON_HTML5_CONVERSION
#include "unicode/ucnv.h"
#include "unicode/ucnv_cb.h"
Index: source/common/ucnv_u7.c
===================================================================
--- source/common/ucnv_u7.c (revision 259715)
+++ source/common/ucnv_u7.c (working copy)
@@ -16,7 +16,7 @@
#include "unicode/utypes.h"
-#if !UCONFIG_NO_CONVERSION
+#if !UCONFIG_NO_CONVERSION && !UCONFIG_NO_NON_HTML5_CONVERSION
#include "unicode/ucnv.h"
#include "ucnv_bld.h"
Index: source/common/unicode/uconfig.h
===================================================================
--- source/common/unicode/uconfig.h (revision 259715)
+++ source/common/unicode/uconfig.h (working copy)
@@ -265,6 +265,14 @@
#endif
/**
+ * This switch turns off all the converters NOT listed in
+ * the encoding standard : http://encoding.spec.whatwg.org
+ */
+#ifndef UCONFIG_NO_NON_HTML5_CONVERSION
+#define UCONFIG_NO_NON_HTML5_CONVERSION 0
+#endif
+
+/**
* \def UCONFIG_NO_LEGACY_CONVERSION
* This switch turns off all converters except for
* - Unicode charsets (UTF-7/8/16/32, CESU-8, SCSU, BOCU-1)
Index: source/common/ucnv_bld.cpp
===================================================================
--- source/common/ucnv_bld.cpp (revision 259715)
+++ source/common/ucnv_bld.cpp (working copy)
@@ -79,16 +79,25 @@
&_HZData,
#endif
+#if UCONFIG_NO_NON_HTML5_CONVERSION
+ NULL,
+#else
&_SCSUData,
+#endif
-#if UCONFIG_NO_LEGACY_CONVERSION
+
+#if UCONFIG_NO_LEGACY_CONVERSION || UCONFIG_NO_NON_HTML5_CONVERSION
NULL,
#else
&_ISCIIData,
#endif
&_ASCIIData,
+#if UCONFIG_NO_NON_HTML5_CONVERSION
+ NULL, NULL, &_UTF16Data, &_UTF32Data, NULL, NULL,
+#else
&_UTF7Data, &_Bocu1Data, &_UTF16Data, &_UTF32Data, &_CESU8Data, &_IMAPData,
+#endif
#if UCONFIG_NO_LEGACY_CONVERSION
NULL,
Index: source/common/ucnv_u8.c
===================================================================
--- source/common/ucnv_u8.c (revision 259715)
+++ source/common/ucnv_u8.c (working copy)
@@ -87,6 +87,15 @@
static const uint32_t
utf8_minChar32[7]={ 0, 0, 0x80, 0x800, 0x10000, 0xffffffff, 0xffffffff };
+static UBool hasCESU8Data(const UConverter *cnv)
+{
+#if UCONFIG_NO_NON_HTML5_CONVERSION
+ return FALSE;
+#else
+ return (UBool)(cnv->sharedData == &_CESU8Data);
+#endif
+}
+
static void ucnv_toUnicode_UTF8 (UConverterToUnicodeArgs * args,
UErrorCode * err)
{
@@ -96,10 +105,10 @@
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
const UChar *targetLimit = args->targetLimit;
unsigned char *toUBytes = cnv->toUBytes;
- UBool isCESU8 = (UBool)(cnv->sharedData == &_CESU8Data);
+ UBool isCESU8 = hasCESU8Data(cnv);
uint32_t ch, ch2 = 0;
int32_t i, inBytes;
-
+
/* Restore size of current sequence */
if (cnv->toUnicodeStatus && myTarget < targetLimit)
{
@@ -226,7 +235,7 @@
const unsigned char *sourceLimit = (unsigned char *) args->sourceLimit;
const UChar *targetLimit = args->targetLimit;
unsigned char *toUBytes = cnv->toUBytes;
- UBool isCESU8 = (UBool)(cnv->sharedData == &_CESU8Data);
+ UBool isCESU8 = hasCESU8Data(cnv);
uint32_t ch, ch2 = 0;
int32_t i, inBytes;
@@ -357,7 +366,7 @@
UChar32 ch;
uint8_t tempBuf[4];
int32_t indexToWrite;
- UBool isNotCESU8 = (UBool)(cnv->sharedData != &_CESU8Data);
+ UBool isNotCESU8 = !hasCESU8Data(cnv);
if (cnv->fromUChar32 && myTarget < targetLimit)
{
@@ -473,7 +482,7 @@
int32_t offsetNum, nextSourceIndex;
int32_t indexToWrite;
uint8_t tempBuf[4];
- UBool isNotCESU8 = (UBool)(cnv->sharedData != &_CESU8Data);
+ UBool isNotCESU8 = !hasCESU8Data(cnv);
if (cnv->fromUChar32 && myTarget < targetLimit)
{
Index: source/common/unicode/urename.h
===================================================================
--- source/common/unicode/urename.h (revision 259715)
+++ source/common/unicode/urename.h (working copy)
@@ -73,12 +73,16 @@
#define UDataMemory_setData U_ICU_ENTRY_POINT_RENAME(UDataMemory_setData)
#define UDatamemory_assign U_ICU_ENTRY_POINT_RENAME(UDatamemory_assign)
#define _ASCIIData U_ICU_ENTRY_POINT_RENAME(_ASCIIData)
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
#define _Bocu1Data U_ICU_ENTRY_POINT_RENAME(_Bocu1Data)
#define _CESU8Data U_ICU_ENTRY_POINT_RENAME(_CESU8Data)
+#endif
#define _CompoundTextData U_ICU_ENTRY_POINT_RENAME(_CompoundTextData)
#define _HZData U_ICU_ENTRY_POINT_RENAME(_HZData)
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
#define _IMAPData U_ICU_ENTRY_POINT_RENAME(_IMAPData)
#define _ISCIIData U_ICU_ENTRY_POINT_RENAME(_ISCIIData)
+#endif
#define _ISO2022Data U_ICU_ENTRY_POINT_RENAME(_ISO2022Data)
#define _LMBCSData1 U_ICU_ENTRY_POINT_RENAME(_LMBCSData1)
#define _LMBCSData11 U_ICU_ENTRY_POINT_RENAME(_LMBCSData11)
@@ -94,14 +98,18 @@
#define _LMBCSData8 U_ICU_ENTRY_POINT_RENAME(_LMBCSData8)
#define _Latin1Data U_ICU_ENTRY_POINT_RENAME(_Latin1Data)
#define _MBCSData U_ICU_ENTRY_POINT_RENAME(_MBCSData)
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
#define _SCSUData U_ICU_ENTRY_POINT_RENAME(_SCSUData)
+#endif
#define _UTF16BEData U_ICU_ENTRY_POINT_RENAME(_UTF16BEData)
#define _UTF16Data U_ICU_ENTRY_POINT_RENAME(_UTF16Data)
#define _UTF16LEData U_ICU_ENTRY_POINT_RENAME(_UTF16LEData)
#define _UTF32BEData U_ICU_ENTRY_POINT_RENAME(_UTF32BEData)
#define _UTF32Data U_ICU_ENTRY_POINT_RENAME(_UTF32Data)
#define _UTF32LEData U_ICU_ENTRY_POINT_RENAME(_UTF32LEData)
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
#define _UTF7Data U_ICU_ENTRY_POINT_RENAME(_UTF7Data)
+#endif
#define _UTF8Data U_ICU_ENTRY_POINT_RENAME(_UTF8Data)
#define cmemory_cleanup U_ICU_ENTRY_POINT_RENAME(cmemory_cleanup)
#define cmemory_inUse U_ICU_ENTRY_POINT_RENAME(cmemory_inUse)
Index: source/common/ucnv_cnv.h
===================================================================
--- source/common/ucnv_cnv.h (revision 259715)
+++ source/common/ucnv_cnv.h (working copy)
@@ -259,8 +259,13 @@
_ISO2022Data,
_LMBCSData1,_LMBCSData2, _LMBCSData3, _LMBCSData4, _LMBCSData5, _LMBCSData6,
_LMBCSData8,_LMBCSData11,_LMBCSData16,_LMBCSData17,_LMBCSData18,_LMBCSData19,
+#if !UCONFIG_NO_NON_HTML5_CONVERSION
_HZData,_ISCIIData, _SCSUData, _ASCIIData,
_UTF7Data, _Bocu1Data, _UTF16Data, _UTF32Data, _CESU8Data, _IMAPData, _CompoundTextData;
+#else
+ _HZData, _ASCIIData,
+ _UTF16Data, _UTF32Data, _CompoundTextData;
+#endif
U_CDECL_END