michael@0: /*
michael@0: ********************************************************************************
michael@0: *   Copyright (C) 1996-2012, International Business Machines
michael@0: *   Corporation and others.  All Rights Reserved.
michael@0: ********************************************************************************
michael@0: *
michael@0: * File UCHAR.C
michael@0: *
michael@0: * Modification History:
michael@0: *
michael@0: *   Date        Name        Description
michael@0: *   04/02/97    aliu        Creation.
michael@0: *   4/15/99     Madhu       Updated all the function definitions for C Implementation
michael@0: *   5/20/99     Madhu       Added the function u_getVersion()
michael@0: *   8/19/1999   srl         Upgraded scripts to Unicode3.0 
michael@0: *   11/11/1999  weiv        added u_isalnum(), cleaned comments
michael@0: *   01/11/2000  helena      Renamed u_getVersion to u_getUnicodeVersion.
michael@0: *   06/20/2000  helena      OS/400 port changes; mostly typecast.
michael@0: ******************************************************************************
michael@0: */
michael@0: 
michael@0: #include "unicode/utypes.h"
michael@0: #include "unicode/uchar.h"
michael@0: #include "unicode/uscript.h"
michael@0: #include "unicode/udata.h"
michael@0: #include "uassert.h"
michael@0: #include "cmemory.h"
michael@0: #include "ucln_cmn.h"
michael@0: #include "utrie2.h"
michael@0: #include "udataswp.h"
michael@0: #include "uprops.h"
michael@0: #include "ustr_imp.h"
michael@0: 
michael@0: #define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
michael@0: 
michael@0: /* uchar_props_data.h is machine-generated by genprops --csource */
michael@0: #define INCLUDED_FROM_UCHAR_C
michael@0: #include "uchar_props_data.h"
michael@0: 
michael@0: /* constants and macros for access to the data ------------------------------ */
michael@0: 
michael@0: /* getting a uint32_t properties word from the data */
michael@0: #define GET_PROPS(c, result) ((result)=UTRIE2_GET16(&propsTrie, c));
michael@0: 
michael@0: U_CFUNC UBool
michael@0: uprv_haveProperties(UErrorCode *pErrorCode) {
michael@0:     if(U_FAILURE(*pErrorCode)) {
michael@0:         return FALSE;
michael@0:     }
michael@0:     return TRUE;
michael@0: }
michael@0: 
michael@0: /* API functions ------------------------------------------------------------ */
michael@0: 
michael@0: /* Gets the Unicode character's general category.*/
michael@0: U_CAPI int8_t U_EXPORT2
michael@0: u_charType(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (int8_t)GET_CATEGORY(props);
michael@0: }
michael@0: 
michael@0: /* Enumerate all code points with their general categories. */
michael@0: struct _EnumTypeCallback {
michael@0:     UCharEnumTypeRange *enumRange;
michael@0:     const void *context;
michael@0: };
michael@0: 
michael@0: static uint32_t U_CALLCONV
michael@0: _enumTypeValue(const void *context, uint32_t value) {
michael@0:     return GET_CATEGORY(value);
michael@0: }
michael@0: 
michael@0: static UBool U_CALLCONV
michael@0: _enumTypeRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
michael@0:     /* just cast the value to UCharCategory */
michael@0:     return ((struct _EnumTypeCallback *)context)->
michael@0:         enumRange(((struct _EnumTypeCallback *)context)->context,
michael@0:                   start, end+1, (UCharCategory)value);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: u_enumCharTypes(UCharEnumTypeRange *enumRange, const void *context) {
michael@0:     struct _EnumTypeCallback callback;
michael@0: 
michael@0:     if(enumRange==NULL) {
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     callback.enumRange=enumRange;
michael@0:     callback.context=context;
michael@0:     utrie2_enum(&propsTrie, _enumTypeValue, _enumTypeRange, &callback);
michael@0: }
michael@0: 
michael@0: /* Checks if ch is a lower case letter.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_islower(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)==U_LOWERCASE_LETTER);
michael@0: }
michael@0: 
michael@0: /* Checks if ch is an upper case letter.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isupper(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)==U_UPPERCASE_LETTER);
michael@0: }
michael@0: 
michael@0: /* Checks if ch is a title case letter; usually upper case letters.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_istitle(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)==U_TITLECASE_LETTER);
michael@0: }
michael@0: 
michael@0: /* Checks if ch is a decimal digit. */
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isdigit(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)==U_DECIMAL_DIGIT_NUMBER);
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isxdigit(UChar32 c) {
michael@0:     uint32_t props;
michael@0: 
michael@0:     /* check ASCII and Fullwidth ASCII a-fA-F */
michael@0:     if(
michael@0:         (c<=0x66 && c>=0x41 && (c<=0x46 || c>=0x61)) ||
michael@0:         (c>=0xff21 && c<=0xff46 && (c<=0xff26 || c>=0xff41))
michael@0:     ) {
michael@0:         return TRUE;
michael@0:     }
michael@0: 
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)==U_DECIMAL_DIGIT_NUMBER);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character is a letter.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isalpha(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&U_GC_L_MASK)!=0);
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isUAlphabetic(UChar32 c) {
michael@0:     return (u_getUnicodeProperties(c, 1)&U_MASK(UPROPS_ALPHABETIC))!=0;
michael@0: }
michael@0: 
michael@0: /* Checks if c is a letter or a decimal digit */
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isalnum(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_ND_MASK))!=0);
michael@0: }
michael@0: 
michael@0: /**
michael@0:  * Checks if c is alphabetic, or a decimal digit; implements UCHAR_POSIX_ALNUM.
michael@0:  * @internal
michael@0:  */
michael@0: U_CFUNC UBool
michael@0: u_isalnumPOSIX(UChar32 c) {
michael@0:     return (UBool)(u_isUAlphabetic(c) || u_isdigit(c));
michael@0: }
michael@0: 
michael@0: /* Checks if ch is a unicode character with assigned character type.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isdefined(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(GET_CATEGORY(props)!=0);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character is a base form character that can take a diacritic.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isbase(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_N_MASK|U_GC_MC_MASK|U_GC_ME_MASK))!=0);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character is a control character.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_iscntrl(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&(U_GC_CC_MASK|U_GC_CF_MASK|U_GC_ZL_MASK|U_GC_ZP_MASK))!=0);
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isISOControl(UChar32 c) {
michael@0:     return (uint32_t)c<=0x9f && (c<=0x1f || c>=0x7f);
michael@0: }
michael@0: 
michael@0: /* Some control characters that are used as space. */
michael@0: #define IS_THAT_CONTROL_SPACE(c) \
michael@0:     (c<=0x9f && ((c>=TAB && c<=CR) || (c>=0x1c && c <=0x1f) || c==NL))
michael@0: 
michael@0: /* Java has decided that U+0085 New Line is not whitespace any more. */
michael@0: #define IS_THAT_ASCII_CONTROL_SPACE(c) \
michael@0:     (c<=0x1f && c>=TAB && (c<=CR || c>=0x1c))
michael@0: 
michael@0: /* Checks if the Unicode character is a space character.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isspace(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&U_GC_Z_MASK)!=0 || IS_THAT_CONTROL_SPACE(c));
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isJavaSpaceChar(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&U_GC_Z_MASK)!=0);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character is a whitespace character.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isWhitespace(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(
michael@0:                 ((CAT_MASK(props)&U_GC_Z_MASK)!=0 &&
michael@0:                     c!=NBSP && c!=FIGURESP && c!=NNBSP) || /* exclude no-break spaces */
michael@0:                 IS_THAT_ASCII_CONTROL_SPACE(c)
michael@0:            );
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isblank(UChar32 c) {
michael@0:     if((uint32_t)c<=0x9f) {
michael@0:         return c==9 || c==0x20; /* TAB or SPACE */
michael@0:     } else {
michael@0:         /* Zs */
michael@0:         uint32_t props;
michael@0:         GET_PROPS(c, props);
michael@0:         return (UBool)(GET_CATEGORY(props)==U_SPACE_SEPARATOR);
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isUWhiteSpace(UChar32 c) {
michael@0:     return (u_getUnicodeProperties(c, 1)&U_MASK(UPROPS_WHITE_SPACE))!=0;
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character is printable.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isprint(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     /* comparing ==0 returns FALSE for the categories mentioned */
michael@0:     return (UBool)((CAT_MASK(props)&U_GC_C_MASK)==0);
michael@0: }
michael@0: 
michael@0: /**
michael@0:  * Checks if c is in \p{graph}\p{blank} - \p{cntrl}.
michael@0:  * Implements UCHAR_POSIX_PRINT.
michael@0:  * @internal
michael@0:  */
michael@0: U_CFUNC UBool
michael@0: u_isprintPOSIX(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     /*
michael@0:      * The only cntrl character in graph+blank is TAB (in blank).
michael@0:      * Here we implement (blank-TAB)=Zs instead of calling u_isblank().
michael@0:      */
michael@0:     return (UBool)((GET_CATEGORY(props)==U_SPACE_SEPARATOR) || u_isgraphPOSIX(c));
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isgraph(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     /* comparing ==0 returns FALSE for the categories mentioned */
michael@0:     return (UBool)((CAT_MASK(props)&
michael@0:                     (U_GC_CC_MASK|U_GC_CF_MASK|U_GC_CS_MASK|U_GC_CN_MASK|U_GC_Z_MASK))
michael@0:                    ==0);
michael@0: }
michael@0: 
michael@0: /**
michael@0:  * Checks if c is in
michael@0:  * [^\p{space}\p{gc=Control}\p{gc=Surrogate}\p{gc=Unassigned}]
michael@0:  * with space=\p{Whitespace} and Control=Cc.
michael@0:  * Implements UCHAR_POSIX_GRAPH.
michael@0:  * @internal
michael@0:  */
michael@0: U_CFUNC UBool
michael@0: u_isgraphPOSIX(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     /* \p{space}\p{gc=Control} == \p{gc=Z}\p{Control} */
michael@0:     /* comparing ==0 returns FALSE for the categories mentioned */
michael@0:     return (UBool)((CAT_MASK(props)&
michael@0:                     (U_GC_CC_MASK|U_GC_CS_MASK|U_GC_CN_MASK|U_GC_Z_MASK))
michael@0:                    ==0);
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_ispunct(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&U_GC_P_MASK)!=0);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character can start a Unicode identifier.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isIDStart(UChar32 c) {
michael@0:     /* same as u_isalpha() */
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_NL_MASK))!=0);
michael@0: }
michael@0: 
michael@0: /* Checks if the Unicode character can be a Unicode identifier part other than starting the
michael@0:  identifier.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isIDPart(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(
michael@0:            (CAT_MASK(props)&
michael@0:             (U_GC_ND_MASK|U_GC_NL_MASK|
michael@0:              U_GC_L_MASK|
michael@0:              U_GC_PC_MASK|U_GC_MC_MASK|U_GC_MN_MASK)
michael@0:            )!=0 ||
michael@0:            u_isIDIgnorable(c));
michael@0: }
michael@0: 
michael@0: /*Checks if the Unicode character can be ignorable in a Java or Unicode identifier.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isIDIgnorable(UChar32 c) {
michael@0:     if(c<=0x9f) {
michael@0:         return u_isISOControl(c) && !IS_THAT_ASCII_CONTROL_SPACE(c);
michael@0:     } else {
michael@0:         uint32_t props;
michael@0:         GET_PROPS(c, props);
michael@0:         return (UBool)(GET_CATEGORY(props)==U_FORMAT_CHAR);
michael@0:     }
michael@0: }
michael@0: 
michael@0: /*Checks if the Unicode character can start a Java identifier.*/
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isJavaIDStart(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_SC_MASK|U_GC_PC_MASK))!=0);
michael@0: }
michael@0: 
michael@0: /*Checks if the Unicode character can be a Java identifier part other than starting the
michael@0:  * identifier.
michael@0:  */
michael@0: U_CAPI UBool U_EXPORT2
michael@0: u_isJavaIDPart(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return (UBool)(
michael@0:            (CAT_MASK(props)&
michael@0:             (U_GC_ND_MASK|U_GC_NL_MASK|
michael@0:              U_GC_L_MASK|
michael@0:              U_GC_SC_MASK|U_GC_PC_MASK|
michael@0:              U_GC_MC_MASK|U_GC_MN_MASK)
michael@0:            )!=0 ||
michael@0:            u_isIDIgnorable(c));
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: u_charDigitValue(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     int32_t value;
michael@0:     GET_PROPS(c, props);
michael@0:     value=(int32_t)GET_NUMERIC_TYPE_VALUE(props)-UPROPS_NTV_DECIMAL_START;
michael@0:     if(value<=9) {
michael@0:         return value;
michael@0:     } else {
michael@0:         return -1;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI double U_EXPORT2
michael@0: u_getNumericValue(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     int32_t ntv;
michael@0:     GET_PROPS(c, props);
michael@0:     ntv=(int32_t)GET_NUMERIC_TYPE_VALUE(props);
michael@0: 
michael@0:     if(ntv==UPROPS_NTV_NONE) {
michael@0:         return U_NO_NUMERIC_VALUE;
michael@0:     } else if(ntv<UPROPS_NTV_DIGIT_START) {
michael@0:         /* decimal digit */
michael@0:         return ntv-UPROPS_NTV_DECIMAL_START;
michael@0:     } else if(ntv<UPROPS_NTV_NUMERIC_START) {
michael@0:         /* other digit */
michael@0:         return ntv-UPROPS_NTV_DIGIT_START;
michael@0:     } else if(ntv<UPROPS_NTV_FRACTION_START) {
michael@0:         /* small integer */
michael@0:         return ntv-UPROPS_NTV_NUMERIC_START;
michael@0:     } else if(ntv<UPROPS_NTV_LARGE_START) {
michael@0:         /* fraction */
michael@0:         int32_t numerator=(ntv>>4)-12;
michael@0:         int32_t denominator=(ntv&0xf)+1;
michael@0:         return (double)numerator/denominator;
michael@0:     } else if(ntv<UPROPS_NTV_BASE60_START) {
michael@0:         /* large, single-significant-digit integer */
michael@0:         double numValue;
michael@0:         int32_t mant=(ntv>>5)-14;
michael@0:         int32_t exp=(ntv&0x1f)+2;
michael@0:         numValue=mant;
michael@0: 
michael@0:         /* multiply by 10^exp without math.h */
michael@0:         while(exp>=4) {
michael@0:             numValue*=10000.;
michael@0:             exp-=4;
michael@0:         }
michael@0:         switch(exp) {
michael@0:         case 3:
michael@0:             numValue*=1000.;
michael@0:             break;
michael@0:         case 2:
michael@0:             numValue*=100.;
michael@0:             break;
michael@0:         case 1:
michael@0:             numValue*=10.;
michael@0:             break;
michael@0:         case 0:
michael@0:         default:
michael@0:             break;
michael@0:         }
michael@0: 
michael@0:         return numValue;
michael@0:     } else if(ntv<UPROPS_NTV_RESERVED_START) {
michael@0:         /* sexagesimal (base 60) integer */
michael@0:         int32_t numValue=(ntv>>2)-0xbf;
michael@0:         int32_t exp=(ntv&3)+1;
michael@0: 
michael@0:         switch(exp) {
michael@0:         case 4:
michael@0:             numValue*=60*60*60*60;
michael@0:             break;
michael@0:         case 3:
michael@0:             numValue*=60*60*60;
michael@0:             break;
michael@0:         case 2:
michael@0:             numValue*=60*60;
michael@0:             break;
michael@0:         case 1:
michael@0:             numValue*=60;
michael@0:             break;
michael@0:         case 0:
michael@0:         default:
michael@0:             break;
michael@0:         }
michael@0: 
michael@0:         return numValue;
michael@0:     } else {
michael@0:         /* reserved */
michael@0:         return U_NO_NUMERIC_VALUE;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: u_digit(UChar32 ch, int8_t radix) {
michael@0:     int8_t value;
michael@0:     if((uint8_t)(radix-2)<=(36-2)) {
michael@0:         value=(int8_t)u_charDigitValue(ch);
michael@0:         if(value<0) {
michael@0:             /* ch is not a decimal digit, try latin letters */
michael@0:             if(ch>=0x61 && ch<=0x7A) {
michael@0:                 value=(int8_t)(ch-0x57);  /* ch - 'a' + 10 */
michael@0:             } else if(ch>=0x41 && ch<=0x5A) {
michael@0:                 value=(int8_t)(ch-0x37);  /* ch - 'A' + 10 */
michael@0:             } else if(ch>=0xFF41 && ch<=0xFF5A) {
michael@0:                 value=(int8_t)(ch-0xFF37);  /* fullwidth ASCII a-z */
michael@0:             } else if(ch>=0xFF21 && ch<=0xFF3A) {
michael@0:                 value=(int8_t)(ch-0xFF17);  /* fullwidth ASCII A-Z */
michael@0:             }
michael@0:         }
michael@0:     } else {
michael@0:         value=-1;   /* invalid radix */
michael@0:     }
michael@0:     return (int8_t)((value<radix) ? value : -1);
michael@0: }
michael@0: 
michael@0: U_CAPI UChar32 U_EXPORT2
michael@0: u_forDigit(int32_t digit, int8_t radix) {
michael@0:     if((uint8_t)(radix-2)>(36-2) || (uint32_t)digit>=(uint32_t)radix) {
michael@0:         return 0;
michael@0:     } else if(digit<10) {
michael@0:         return (UChar32)(0x30+digit);
michael@0:     } else {
michael@0:         return (UChar32)((0x61-10)+digit);
michael@0:     }
michael@0: }
michael@0: 
michael@0: /* miscellaneous, and support for uprops.cpp -------------------------------- */
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: u_getUnicodeVersion(UVersionInfo versionArray) {
michael@0:     if(versionArray!=NULL) {
michael@0:         uprv_memcpy(versionArray, dataVersion, U_MAX_VERSION_LENGTH);
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CFUNC uint32_t
michael@0: u_getMainProperties(UChar32 c) {
michael@0:     uint32_t props;
michael@0:     GET_PROPS(c, props);
michael@0:     return props;
michael@0: }
michael@0: 
michael@0: U_CFUNC uint32_t
michael@0: u_getUnicodeProperties(UChar32 c, int32_t column) {
michael@0:     U_ASSERT(column>=0);
michael@0:     if(column>=propsVectorsColumns) {
michael@0:         return 0;
michael@0:     } else {
michael@0:         uint16_t vecIndex=UTRIE2_GET16(&propsVectorsTrie, c);
michael@0:         return propsVectors[vecIndex+column];
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CFUNC int32_t
michael@0: uprv_getMaxValues(int32_t column) {
michael@0:     switch(column) {
michael@0:     case 0:
michael@0:         return indexes[UPROPS_MAX_VALUES_INDEX];
michael@0:     case 2:
michael@0:         return indexes[UPROPS_MAX_VALUES_2_INDEX];
michael@0:     default:
michael@0:         return 0;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: u_charAge(UChar32 c, UVersionInfo versionArray) {
michael@0:     if(versionArray!=NULL) {
michael@0:         uint32_t version=u_getUnicodeProperties(c, 0)>>UPROPS_AGE_SHIFT;
michael@0:         versionArray[0]=(uint8_t)(version>>4);
michael@0:         versionArray[1]=(uint8_t)(version&0xf);
michael@0:         versionArray[2]=versionArray[3]=0;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI UScriptCode U_EXPORT2
michael@0: uscript_getScript(UChar32 c, UErrorCode *pErrorCode) {
michael@0:     uint32_t scriptX;
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return USCRIPT_INVALID_CODE;
michael@0:     }
michael@0:     if((uint32_t)c>0x10ffff) {
michael@0:         *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return USCRIPT_INVALID_CODE;
michael@0:     }
michael@0:     scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
michael@0:     if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
michael@0:         return (UScriptCode)scriptX;
michael@0:     } else if(scriptX<UPROPS_SCRIPT_X_WITH_INHERITED) {
michael@0:         return USCRIPT_COMMON;
michael@0:     } else if(scriptX<UPROPS_SCRIPT_X_WITH_OTHER) {
michael@0:         return USCRIPT_INHERITED;
michael@0:     } else {
michael@0:         return (UScriptCode)scriptExtensions[scriptX&UPROPS_SCRIPT_MASK];
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI UBool U_EXPORT2
michael@0: uscript_hasScript(UChar32 c, UScriptCode sc) {
michael@0:     const uint16_t *scx;
michael@0:     uint32_t scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
michael@0:     if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
michael@0:         return sc==(UScriptCode)scriptX;
michael@0:     }
michael@0: 
michael@0:     scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
michael@0:     if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
michael@0:         scx=scriptExtensions+scx[1];
michael@0:     }
michael@0:     if(sc>=USCRIPT_CODE_LIMIT) {
michael@0:         /* Guard against bogus input that would make us go past the Script_Extensions terminator. */
michael@0:         return FALSE;
michael@0:     }
michael@0:     while(sc>*scx) {
michael@0:         ++scx;
michael@0:     }
michael@0:     return sc==(*scx&0x7fff);
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: uscript_getScriptExtensions(UChar32 c,
michael@0:                             UScriptCode *scripts, int32_t capacity,
michael@0:                             UErrorCode *pErrorCode) {
michael@0:     uint32_t scriptX;
michael@0:     int32_t length;
michael@0:     const uint16_t *scx;
michael@0:     uint16_t sx;
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return 0;
michael@0:     }
michael@0:     if(capacity<0 || (capacity>0 && scripts==NULL)) {
michael@0:         *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return 0;
michael@0:     }
michael@0:     scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
michael@0:     if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
michael@0:         if(capacity==0) {
michael@0:             *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
michael@0:         } else {
michael@0:             scripts[0]=(UScriptCode)scriptX;
michael@0:         }
michael@0:         return 1;
michael@0:     }
michael@0: 
michael@0:     scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
michael@0:     if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
michael@0:         scx=scriptExtensions+scx[1];
michael@0:     }
michael@0:     length=0;
michael@0:     do {
michael@0:         sx=*scx++;
michael@0:         if(length<capacity) {
michael@0:             scripts[length]=(UScriptCode)(sx&0x7fff);
michael@0:         }
michael@0:         ++length;
michael@0:     } while(sx<0x8000);
michael@0:     if(length>capacity) {
michael@0:         *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
michael@0:     }
michael@0:     return length;
michael@0: }
michael@0: 
michael@0: U_CAPI UBlockCode U_EXPORT2
michael@0: ublock_getCode(UChar32 c) {
michael@0:     return (UBlockCode)((u_getUnicodeProperties(c, 0)&UPROPS_BLOCK_MASK)>>UPROPS_BLOCK_SHIFT);
michael@0: }
michael@0: 
michael@0: /* property starts for UnicodeSet ------------------------------------------- */
michael@0: 
michael@0: static UBool U_CALLCONV
michael@0: _enumPropertyStartsRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
michael@0:     /* add the start code point to the USet */
michael@0:     const USetAdder *sa=(const USetAdder *)context;
michael@0:     sa->add(sa->set, start);
michael@0:     return TRUE;
michael@0: }
michael@0: 
michael@0: #define USET_ADD_CP_AND_NEXT(sa, cp) sa->add(sa->set, cp); sa->add(sa->set, cp+1)
michael@0: 
michael@0: U_CFUNC void U_EXPORT2
michael@0: uchar_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode) {
michael@0:     if(U_FAILURE(*pErrorCode)) {
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     /* add the start code point of each same-value range of the main trie */
michael@0:     utrie2_enum(&propsTrie, NULL, _enumPropertyStartsRange, sa);
michael@0: 
michael@0:     /* add code points with hardcoded properties, plus the ones following them */
michael@0: 
michael@0:     /* add for u_isblank() */
michael@0:     USET_ADD_CP_AND_NEXT(sa, TAB);
michael@0: 
michael@0:     /* add for IS_THAT_CONTROL_SPACE() */
michael@0:     sa->add(sa->set, CR+1); /* range TAB..CR */
michael@0:     sa->add(sa->set, 0x1c);
michael@0:     sa->add(sa->set, 0x1f+1);
michael@0:     USET_ADD_CP_AND_NEXT(sa, NL);
michael@0: 
michael@0:     /* add for u_isIDIgnorable() what was not added above */
michael@0:     sa->add(sa->set, DEL); /* range DEL..NBSP-1, NBSP added below */
michael@0:     sa->add(sa->set, HAIRSP);
michael@0:     sa->add(sa->set, RLM+1);
michael@0:     sa->add(sa->set, INHSWAP);
michael@0:     sa->add(sa->set, NOMDIG+1);
michael@0:     USET_ADD_CP_AND_NEXT(sa, ZWNBSP);
michael@0: 
michael@0:     /* add no-break spaces for u_isWhitespace() what was not added above */
michael@0:     USET_ADD_CP_AND_NEXT(sa, NBSP);
michael@0:     USET_ADD_CP_AND_NEXT(sa, FIGURESP);
michael@0:     USET_ADD_CP_AND_NEXT(sa, NNBSP);
michael@0: 
michael@0:     /* add for u_digit() */
michael@0:     sa->add(sa->set, U_a);
michael@0:     sa->add(sa->set, U_z+1);
michael@0:     sa->add(sa->set, U_A);
michael@0:     sa->add(sa->set, U_Z+1);
michael@0:     sa->add(sa->set, U_FW_a);
michael@0:     sa->add(sa->set, U_FW_z+1);
michael@0:     sa->add(sa->set, U_FW_A);
michael@0:     sa->add(sa->set, U_FW_Z+1);
michael@0: 
michael@0:     /* add for u_isxdigit() */
michael@0:     sa->add(sa->set, U_f+1);
michael@0:     sa->add(sa->set, U_F+1);
michael@0:     sa->add(sa->set, U_FW_f+1);
michael@0:     sa->add(sa->set, U_FW_F+1);
michael@0: 
michael@0:     /* add for UCHAR_DEFAULT_IGNORABLE_CODE_POINT what was not added above */
michael@0:     sa->add(sa->set, WJ); /* range WJ..NOMDIG */
michael@0:     sa->add(sa->set, 0xfff0);
michael@0:     sa->add(sa->set, 0xfffb+1);
michael@0:     sa->add(sa->set, 0xe0000);
michael@0:     sa->add(sa->set, 0xe0fff+1);
michael@0: 
michael@0:     /* add for UCHAR_GRAPHEME_BASE and others */
michael@0:     USET_ADD_CP_AND_NEXT(sa, CGJ);
michael@0: }
michael@0: 
michael@0: U_CFUNC void U_EXPORT2
michael@0: upropsvec_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode) {
michael@0:     if(U_FAILURE(*pErrorCode)) {
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     /* add the start code point of each same-value range of the properties vectors trie */
michael@0:     if(propsVectorsColumns>0) {
michael@0:         /* if propsVectorsColumns==0 then the properties vectors trie may not be there at all */
michael@0:         utrie2_enum(&propsVectorsTrie, NULL, _enumPropertyStartsRange, sa);
michael@0:     }
michael@0: }