/*
********************************************************************************
-* Copyright (C) 1996-2008, International Business Machines
+* Copyright (C) 1996-2014, International Business Machines
* Corporation and others. All Rights Reserved.
********************************************************************************
*
#include "unicode/uchar.h"
#include "unicode/uscript.h"
#include "unicode/udata.h"
-#include "umutex.h"
+#include "uassert.h"
#include "cmemory.h"
#include "ucln_cmn.h"
-#include "utrie.h"
+#include "utrie2.h"
#include "udataswp.h"
-#include "unormimp.h" /* JAMO_L_BASE etc. */
#include "uprops.h"
+#include "ustr_imp.h"
-#define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
-
-/* dynamically loaded Unicode character properties -------------------------- */
-
-#define UCHAR_HARDCODE_DATA 1
-
-#if UCHAR_HARDCODE_DATA
-
-/* uchar_props_data.c is machine-generated by genprops --csource */
-#include "uchar_props_data.c"
-
-#else
-
-/*
- * loaded uprops.dat -
- * for a description of the file format, see icu/source/tools/genprops/store.c
- */
-static const char DATA_NAME[] = "uprops";
-static const char DATA_TYPE[] = "icu";
-
-static UDataMemory *propsData=NULL;
-static UErrorCode dataErrorCode=U_ZERO_ERROR;
-
-static uint8_t formatVersion[4]={ 0, 0, 0, 0 };
-static UVersionInfo dataVersion={ 0, 0, 0, 0 };
-
-static UTrie propsTrie={ 0 }, propsVectorsTrie={ 0 };
-static const uint32_t *pData32=NULL, *propsVectors=NULL;
-static int32_t countPropsVectors=0, propsVectorsColumns=0;
-
-static int8_t havePropsData=0; /* == 0 -> Data has not been loaded.
- * < 0 -> Error occured attempting to load data.
- * > 0 -> Data has been successfully loaded.
- */
-
-/* index values loaded from uprops.dat */
-static int32_t indexes[UPROPS_INDEX_COUNT];
-
-static UBool U_CALLCONV
-isAcceptable(void *context,
- const char *type, const char *name,
- const UDataInfo *pInfo) {
- if(
- pInfo->size>=20 &&
- pInfo->isBigEndian==U_IS_BIG_ENDIAN &&
- pInfo->charsetFamily==U_CHARSET_FAMILY &&
- pInfo->dataFormat[0]==0x55 && /* dataFormat="UPro" */
- pInfo->dataFormat[1]==0x50 &&
- pInfo->dataFormat[2]==0x72 &&
- pInfo->dataFormat[3]==0x6f &&
- pInfo->formatVersion[0]==4 &&
- pInfo->formatVersion[2]==UTRIE_SHIFT &&
- pInfo->formatVersion[3]==UTRIE_INDEX_SHIFT
- ) {
- uprv_memcpy(formatVersion, pInfo->formatVersion, 4);
- uprv_memcpy(dataVersion, pInfo->dataVersion, 4);
- return TRUE;
- } else {
- return FALSE;
- }
-}
-
-static UBool U_CALLCONV uchar_cleanup(void)
-{
- if (propsData) {
- udata_close(propsData);
- propsData=NULL;
- }
- pData32=NULL;
- propsVectors=NULL;
- countPropsVectors=0;
- uprv_memset(dataVersion, 0, U_MAX_VERSION_LENGTH);
- dataErrorCode=U_ZERO_ERROR;
- havePropsData=0;
-
- return TRUE;
-}
-
-struct UCharProps {
- UDataMemory *propsData;
- UTrie propsTrie, propsVectorsTrie;
- const uint32_t *pData32;
-};
-typedef struct UCharProps UCharProps;
-
-/* open uprops.icu */
-static void
-_openProps(UCharProps *ucp, UErrorCode *pErrorCode) {
- const uint32_t *p;
- int32_t length;
-
- ucp->propsData=udata_openChoice(NULL, DATA_TYPE, DATA_NAME, isAcceptable, NULL, pErrorCode);
- if(U_FAILURE(*pErrorCode)) {
- return;
- }
-
- ucp->pData32=p=(const uint32_t *)udata_getMemory(ucp->propsData);
-
- /* unserialize the trie; it is directly after the int32_t indexes[UPROPS_INDEX_COUNT] */
- length=(int32_t)p[UPROPS_PROPS32_INDEX]*4;
- length=utrie_unserialize(&ucp->propsTrie, (const uint8_t *)(p+UPROPS_INDEX_COUNT), length-64, pErrorCode);
- if(U_FAILURE(*pErrorCode)) {
- return;
- }
-
- /* unserialize the properties vectors trie */
- length=(int32_t)(p[UPROPS_ADDITIONAL_VECTORS_INDEX]-p[UPROPS_ADDITIONAL_TRIE_INDEX])*4;
- if(length>0) {
- length=utrie_unserialize(&ucp->propsVectorsTrie, (const uint8_t *)(p+p[UPROPS_ADDITIONAL_TRIE_INDEX]), length, pErrorCode);
- }
- if(length<=0 || U_FAILURE(*pErrorCode)) {
- /*
- * length==0:
- * Allow the properties vectors trie to be missing -
- * also requires propsVectorsColumns=indexes[UPROPS_ADDITIONAL_VECTORS_COLUMNS_INDEX]
- * to be zero so that this trie is never accessed.
- */
- uprv_memset(&ucp->propsVectorsTrie, 0, sizeof(ucp->propsVectorsTrie));
- }
-}
-
-#endif
-
-#if !UCHAR_HARDCODE_DATA
-static int8_t
-uprv_loadPropsData(UErrorCode *pErrorCode) {
- /* load Unicode character properties data from file if necessary */
-
- /*
- * This lazy intialization with double-checked locking (without mutex protection for
- * haveNormData==0) is transiently unsafe under certain circumstances.
- * Check the readme and use u_init() if necessary.
- */
- if(havePropsData==0) {
- UCharProps ucp={ NULL };
-
- if(U_FAILURE(*pErrorCode)) {
- return havePropsData;
- }
-
- /* open the data outside the mutex block */
- _openProps(&ucp, pErrorCode);
-
- if(U_SUCCESS(*pErrorCode)) {
- /* in the mutex block, set the data for this process */
- umtx_lock(NULL);
- if(propsData==NULL) {
- propsData=ucp.propsData;
- ucp.propsData=NULL;
- pData32=ucp.pData32;
- ucp.pData32=NULL;
- uprv_memcpy(&propsTrie, &ucp.propsTrie, sizeof(propsTrie));
- uprv_memcpy(&propsVectorsTrie, &ucp.propsVectorsTrie, sizeof(propsVectorsTrie));
- }
-
- /* initialize some variables */
- uprv_memcpy(indexes, pData32, sizeof(indexes));
-
- /* additional properties */
- if(indexes[UPROPS_ADDITIONAL_VECTORS_INDEX]!=0) {
- propsVectors=pData32+indexes[UPROPS_ADDITIONAL_VECTORS_INDEX];
- countPropsVectors=indexes[UPROPS_RESERVED_INDEX]-indexes[UPROPS_ADDITIONAL_VECTORS_INDEX];
- propsVectorsColumns=indexes[UPROPS_ADDITIONAL_VECTORS_COLUMNS_INDEX];
- }
-
- havePropsData=1;
- umtx_unlock(NULL);
- } else {
- dataErrorCode=*pErrorCode;
- havePropsData=-1;
- }
- ucln_common_registerCleanup(UCLN_COMMON_UCHAR, uchar_cleanup);
-
- /* if a different thread set it first, then close the extra data */
- udata_close(ucp.propsData); /* NULL if it was set correctly */
- }
-
- return havePropsData;
-}
-
-static int8_t
-loadPropsData(void) {
- UErrorCode errorCode = U_ZERO_ERROR;
- int8_t retVal = uprv_loadPropsData(&errorCode);
- return retVal;
-}
-
-#endif
+/* uchar_props_data.h is machine-generated by genprops --csource */
+#define INCLUDED_FROM_UCHAR_C
+#include "uchar_props_data.h"
/* constants and macros for access to the data ------------------------------ */
/* getting a uint32_t properties word from the data */
-#if UCHAR_HARDCODE_DATA
-
-#define GET_PROPS(c, result) UTRIE_GET16(&propsTrie, c, result);
-
-#else
-
-#define HAVE_DATA (havePropsData>0 || loadPropsData()>0)
-#define GET_PROPS_UNSAFE(c, result) \
- UTRIE_GET16(&propsTrie, c, result);
-#define GET_PROPS(c, result) \
- if(HAVE_DATA) { \
- GET_PROPS_UNSAFE(c, result); \
- } else { \
- (result)=0; \
- }
-
-#endif
+#define GET_PROPS(c, result) ((result)=UTRIE2_GET16(&propsTrie, c));
U_CFUNC UBool
uprv_haveProperties(UErrorCode *pErrorCode) {
if(U_FAILURE(*pErrorCode)) {
return FALSE;
}
-#if !UCHAR_HARDCODE_DATA
- if(havePropsData==0) {
- uprv_loadPropsData(pErrorCode);
- }
- if(havePropsData<0) {
- *pErrorCode=dataErrorCode;
- return FALSE;
- }
-#endif
return TRUE;
}
}
static UBool U_CALLCONV
-_enumTypeRange(const void *context, UChar32 start, UChar32 limit, uint32_t value) {
+_enumTypeRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
/* just cast the value to UCharCategory */
return ((struct _EnumTypeCallback *)context)->
enumRange(((struct _EnumTypeCallback *)context)->context,
- start, limit, (UCharCategory)value);
+ start, end+1, (UCharCategory)value);
}
U_CAPI void U_EXPORT2
u_enumCharTypes(UCharEnumTypeRange *enumRange, const void *context) {
struct _EnumTypeCallback callback;
- if(enumRange==NULL
-#if !UCHAR_HARDCODE_DATA
- || !HAVE_DATA
-#endif
- ) {
+ if(enumRange==NULL) {
return;
}
callback.enumRange=enumRange;
callback.context=context;
- utrie_enum(&propsTrie, _enumTypeValue, _enumTypeRange, &callback);
+ utrie2_enum(&propsTrie, _enumTypeValue, _enumTypeRange, &callback);
}
/* Checks if ch is a lower case letter.*/
#define IS_THAT_CONTROL_SPACE(c) \
(c<=0x9f && ((c>=TAB && c<=CR) || (c>=0x1c && c <=0x1f) || c==NL))
+/* Java has decided that U+0085 New Line is not whitespace any more. */
+#define IS_THAT_ASCII_CONTROL_SPACE(c) \
+ (c<=0x1f && c>=TAB && (c<=CR || c>=0x1c))
+
/* Checks if the Unicode character is a space character.*/
U_CAPI UBool U_EXPORT2
u_isspace(UChar32 c) {
return (UBool)(
((CAT_MASK(props)&U_GC_Z_MASK)!=0 &&
c!=NBSP && c!=FIGURESP && c!=NNBSP) || /* exclude no-break spaces */
- IS_THAT_CONTROL_SPACE(c)
+ IS_THAT_ASCII_CONTROL_SPACE(c)
);
}
U_CAPI UBool U_EXPORT2
u_isIDIgnorable(UChar32 c) {
if(c<=0x9f) {
- return u_isISOControl(c) && !IS_THAT_CONTROL_SPACE(c);
+ return u_isISOControl(c) && !IS_THAT_ASCII_CONTROL_SPACE(c);
} else {
uint32_t props;
GET_PROPS(c, props);
U_CAPI int32_t U_EXPORT2
u_charDigitValue(UChar32 c) {
uint32_t props;
+ int32_t value;
GET_PROPS(c, props);
-
- if(GET_NUMERIC_TYPE(props)==1) {
- return GET_NUMERIC_VALUE(props);
+ value=(int32_t)GET_NUMERIC_TYPE_VALUE(props)-UPROPS_NTV_DECIMAL_START;
+ if(value<=9) {
+ return value;
} else {
return -1;
}
U_CAPI double U_EXPORT2
u_getNumericValue(UChar32 c) {
- uint32_t props, numericType, numericValue;
+ uint32_t props;
+ int32_t ntv;
GET_PROPS(c, props);
- numericType=GET_NUMERIC_TYPE(props);
+ ntv=(int32_t)GET_NUMERIC_TYPE_VALUE(props);
- if(numericType==0 || numericType>=UPROPS_NT_COUNT) {
+ if(ntv==UPROPS_NTV_NONE) {
return U_NO_NUMERIC_VALUE;
- }
-
- numericValue=GET_NUMERIC_VALUE(props);
-
- if(numericType<U_NT_COUNT) {
- /* normal type, the value is stored directly */
- return numericValue;
- } else if(numericType==UPROPS_NT_FRACTION) {
- /* fraction value */
- int32_t numerator;
- uint32_t denominator;
-
- numerator=(int32_t)numericValue>>UPROPS_FRACTION_NUM_SHIFT;
- denominator=(numericValue&UPROPS_FRACTION_DEN_MASK)+UPROPS_FRACTION_DEN_OFFSET;
-
- if(numerator==0) {
- numerator=-1;
- }
- return (double)numerator/(double)denominator;
- } else /* numericType==UPROPS_NT_LARGE */ {
- /* large value with exponent */
+ } else if(ntv<UPROPS_NTV_DIGIT_START) {
+ /* decimal digit */
+ return ntv-UPROPS_NTV_DECIMAL_START;
+ } else if(ntv<UPROPS_NTV_NUMERIC_START) {
+ /* other digit */
+ return ntv-UPROPS_NTV_DIGIT_START;
+ } else if(ntv<UPROPS_NTV_FRACTION_START) {
+ /* small integer */
+ return ntv-UPROPS_NTV_NUMERIC_START;
+ } else if(ntv<UPROPS_NTV_LARGE_START) {
+ /* fraction */
+ int32_t numerator=(ntv>>4)-12;
+ int32_t denominator=(ntv&0xf)+1;
+ return (double)numerator/denominator;
+ } else if(ntv<UPROPS_NTV_BASE60_START) {
+ /* large, single-significant-digit integer */
double numValue;
- int32_t mant, exp;
-
- mant=(int32_t)numericValue>>UPROPS_LARGE_MANT_SHIFT;
- exp=(int32_t)numericValue&UPROPS_LARGE_EXP_MASK;
- if(mant==0) {
- mant=1;
- exp+=UPROPS_LARGE_EXP_OFFSET_EXTRA;
- } else if(mant>9) {
- return U_NO_NUMERIC_VALUE; /* reserved mantissa value */
- } else {
- exp+=UPROPS_LARGE_EXP_OFFSET;
- }
-
+ int32_t mant=(ntv>>5)-14;
+ int32_t exp=(ntv&0x1f)+2;
numValue=mant;
/* multiply by 10^exp without math.h */
}
return numValue;
- }
-}
+ } else if(ntv<UPROPS_NTV_RESERVED_START) {
+ /* sexagesimal (base 60) integer */
+ int32_t numValue=(ntv>>2)-0xbf;
+ int32_t exp=(ntv&3)+1;
-/* ICU 3.4: bidi/shaping properties moved to ubidi_props.c */
+ switch(exp) {
+ case 4:
+ numValue*=60*60*60*60;
+ break;
+ case 3:
+ numValue*=60*60*60;
+ break;
+ case 2:
+ numValue*=60*60;
+ break;
+ case 1:
+ numValue*=60;
+ break;
+ case 0:
+ default:
+ break;
+ }
-/* ICU 2.1: u_getCombiningClass() moved to unorm.cpp */
+ return numValue;
+ } else {
+ /* reserved */
+ return U_NO_NUMERIC_VALUE;
+ }
+}
U_CAPI int32_t U_EXPORT2
u_digit(UChar32 ch, int8_t radix) {
}
}
-/* miscellaneous, and support for uprops.c ---------------------------------- */
+/* miscellaneous, and support for uprops.cpp -------------------------------- */
U_CAPI void U_EXPORT2
u_getUnicodeVersion(UVersionInfo versionArray) {
}
U_CFUNC uint32_t
-u_getUnicodeProperties(UChar32 c, int32_t column) {
- uint16_t vecIndex;
+u_getMainProperties(UChar32 c) {
+ uint32_t props;
+ GET_PROPS(c, props);
+ return props;
+}
- if(column==-1) {
- uint32_t props;
- GET_PROPS(c, props);
- return props;
- } else if(
-#if !UCHAR_HARDCODE_DATA
- !HAVE_DATA || countPropsVectors==0 ||
-#endif
- column<0 || column>=propsVectorsColumns
- ) {
+U_CFUNC uint32_t
+u_getUnicodeProperties(UChar32 c, int32_t column) {
+ U_ASSERT(column>=0);
+ if(column>=propsVectorsColumns) {
return 0;
} else {
- UTRIE_GET16(&propsVectorsTrie, c, vecIndex);
+ uint16_t vecIndex=UTRIE2_GET16(&propsVectorsTrie, c);
return propsVectors[vecIndex+column];
}
}
U_CFUNC int32_t
uprv_getMaxValues(int32_t column) {
-#if !UCHAR_HARDCODE_DATA
- if(HAVE_DATA) {
-#endif
- switch(column) {
- case 0:
- return indexes[UPROPS_MAX_VALUES_INDEX];
- case 2:
- return indexes[UPROPS_MAX_VALUES_2_INDEX];
- default:
- return 0;
- }
-#if !UCHAR_HARDCODE_DATA
- } else {
+ switch(column) {
+ case 0:
+ return indexes[UPROPS_MAX_VALUES_INDEX];
+ case 2:
+ return indexes[UPROPS_MAX_VALUES_2_INDEX];
+ default:
return 0;
}
-#endif
-}
-
-/*
- * get Hangul Syllable Type
- * implemented here so that uchar.c (uhst_addPropertyStarts())
- * does not depend on uprops.c (u_getIntPropertyValue(c, UCHAR_HANGUL_SYLLABLE_TYPE))
- */
-U_CFUNC UHangulSyllableType
-uchar_getHST(UChar32 c) {
- /* purely algorithmic; hardcode known characters, check for assigned new ones */
- if(c<JAMO_L_BASE) {
- /* U_HST_NOT_APPLICABLE */
- } else if(c<=0x11ff) {
- /* Jamo range */
- if(c<=0x115f) {
- /* Jamo L range, HANGUL CHOSEONG ... */
- if(c==0x115f || c<=0x1159 || u_charType(c)==U_OTHER_LETTER) {
- return U_HST_LEADING_JAMO;
- }
- } else if(c<=0x11a7) {
- /* Jamo V range, HANGUL JUNGSEONG ... */
- if(c<=0x11a2 || u_charType(c)==U_OTHER_LETTER) {
- return U_HST_VOWEL_JAMO;
- }
- } else {
- /* Jamo T range */
- if(c<=0x11f9 || u_charType(c)==U_OTHER_LETTER) {
- return U_HST_TRAILING_JAMO;
- }
- }
- } else if((c-=HANGUL_BASE)<0) {
- /* U_HST_NOT_APPLICABLE */
- } else if(c<HANGUL_COUNT) {
- /* Hangul syllable */
- return c%JAMO_T_COUNT==0 ? U_HST_LV_SYLLABLE : U_HST_LVT_SYLLABLE;
- }
- return U_HST_NOT_APPLICABLE;
}
U_CAPI void U_EXPORT2
U_CAPI UScriptCode U_EXPORT2
uscript_getScript(UChar32 c, UErrorCode *pErrorCode) {
+ uint32_t scriptX;
if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
return USCRIPT_INVALID_CODE;
}
*pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
return USCRIPT_INVALID_CODE;
}
-
- return (UScriptCode)(u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_MASK);
-}
-
-U_CAPI UBlockCode U_EXPORT2
-ublock_getCode(UChar32 c) {
- return (UBlockCode)((u_getUnicodeProperties(c, 0)&UPROPS_BLOCK_MASK)>>UPROPS_BLOCK_SHIFT);
+ scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
+ if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
+ return (UScriptCode)scriptX;
+ } else if(scriptX<UPROPS_SCRIPT_X_WITH_INHERITED) {
+ return USCRIPT_COMMON;
+ } else if(scriptX<UPROPS_SCRIPT_X_WITH_OTHER) {
+ return USCRIPT_INHERITED;
+ } else {
+ return (UScriptCode)scriptExtensions[scriptX&UPROPS_SCRIPT_MASK];
+ }
}
-/* property starts for UnicodeSet ------------------------------------------- */
-
-/* for Hangul_Syllable_Type */
-U_CFUNC void U_EXPORT2
-uhst_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode) {
- UChar32 c;
- int32_t value, value2;
-
- if(U_FAILURE(*pErrorCode)) {
- return;
+U_CAPI UBool U_EXPORT2
+uscript_hasScript(UChar32 c, UScriptCode sc) {
+ const uint16_t *scx;
+ uint32_t scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
+ if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
+ return sc==(UScriptCode)scriptX;
}
-#if !UCHAR_HARDCODE_DATA
- if(!HAVE_DATA) {
- *pErrorCode=dataErrorCode;
- return;
+ scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
+ if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
+ scx=scriptExtensions+scx[1];
}
-#endif
-
- /* add code points with hardcoded properties, plus the ones following them */
-
- /*
- * Add Jamo type boundaries for UCHAR_HANGUL_SYLLABLE_TYPE.
- * First, we add fixed boundaries for the blocks of Jamos.
- * Then we check in loops to see where the current Unicode version
- * actually stops assigning such Jamos. We start each loop
- * at the end of the per-Jamo-block assignments in Unicode 4 or earlier.
- * (These have not changed since Unicode 2.)
- */
- sa->add(sa->set, 0x1100);
- value=U_HST_LEADING_JAMO;
- for(c=0x115a; c<=0x115f; ++c) {
- value2=uchar_getHST(c);
- if(value!=value2) {
- value=value2;
- sa->add(sa->set, c);
- }
+ if(sc>=USCRIPT_CODE_LIMIT) {
+ /* Guard against bogus input that would make us go past the Script_Extensions terminator. */
+ return FALSE;
+ }
+ while(sc>*scx) {
+ ++scx;
}
+ return sc==(*scx&0x7fff);
+}
- sa->add(sa->set, 0x1160);
- value=U_HST_VOWEL_JAMO;
- for(c=0x11a3; c<=0x11a7; ++c) {
- value2=uchar_getHST(c);
- if(value!=value2) {
- value=value2;
- sa->add(sa->set, c);
+U_CAPI int32_t U_EXPORT2
+uscript_getScriptExtensions(UChar32 c,
+ UScriptCode *scripts, int32_t capacity,
+ UErrorCode *pErrorCode) {
+ uint32_t scriptX;
+ int32_t length;
+ const uint16_t *scx;
+ uint16_t sx;
+ if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
+ return 0;
+ }
+ if(capacity<0 || (capacity>0 && scripts==NULL)) {
+ *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
+ return 0;
+ }
+ scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
+ if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
+ if(capacity==0) {
+ *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
+ } else {
+ scripts[0]=(UScriptCode)scriptX;
}
+ return 1;
}
- sa->add(sa->set, 0x11a8);
- value=U_HST_TRAILING_JAMO;
- for(c=0x11fa; c<=0x11ff; ++c) {
- value2=uchar_getHST(c);
- if(value!=value2) {
- value=value2;
- sa->add(sa->set, c);
+ scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
+ if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
+ scx=scriptExtensions+scx[1];
+ }
+ length=0;
+ do {
+ sx=*scx++;
+ if(length<capacity) {
+ scripts[length]=(UScriptCode)(sx&0x7fff);
}
+ ++length;
+ } while(sx<0x8000);
+ if(length>capacity) {
+ *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
}
+ return length;
+}
- /* Add Hangul type boundaries for UCHAR_HANGUL_SYLLABLE_TYPE. */
- for(c=HANGUL_BASE; c<(HANGUL_BASE+HANGUL_COUNT); c+=JAMO_T_COUNT) {
- sa->add(sa->set, c);
- sa->add(sa->set, c+1);
- }
- sa->add(sa->set, c);
+U_CAPI UBlockCode U_EXPORT2
+ublock_getCode(UChar32 c) {
+ return (UBlockCode)((u_getUnicodeProperties(c, 0)&UPROPS_BLOCK_MASK)>>UPROPS_BLOCK_SHIFT);
}
+/* property starts for UnicodeSet ------------------------------------------- */
+
static UBool U_CALLCONV
-_enumPropertyStartsRange(const void *context, UChar32 start, UChar32 limit, uint32_t value) {
+_enumPropertyStartsRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
/* add the start code point to the USet */
const USetAdder *sa=(const USetAdder *)context;
sa->add(sa->set, start);
return;
}
-#if !UCHAR_HARDCODE_DATA
- if(!HAVE_DATA) {
- *pErrorCode=dataErrorCode;
- return;
- }
-#endif
-
/* add the start code point of each same-value range of the main trie */
- utrie_enum(&propsTrie, NULL, _enumPropertyStartsRange, sa);
+ utrie2_enum(&propsTrie, NULL, _enumPropertyStartsRange, sa);
/* add code points with hardcoded properties, plus the ones following them */
return;
}
-#if !UCHAR_HARDCODE_DATA
- if(!HAVE_DATA) {
- *pErrorCode=dataErrorCode;
- return;
- }
-#endif
-
/* add the start code point of each same-value range of the properties vectors trie */
if(propsVectorsColumns>0) {
/* if propsVectorsColumns==0 then the properties vectors trie may not be there at all */
- utrie_enum(&propsVectorsTrie, NULL, _enumPropertyStartsRange, sa);
+ utrie2_enum(&propsVectorsTrie, NULL, _enumPropertyStartsRange, sa);
}
}
projects / apple / icu.git / blobdiff