2 ********************************************************************************
3 * Copyright (C) 1996-2012, International Business Machines
4 * Corporation and others. All Rights Reserved.
5 ********************************************************************************
9 * Modification History:
11 * Date Name Description
12 * 04/02/97 aliu Creation.
13 * 4/15/99 Madhu Updated all the function definitions for C Implementation
14 * 5/20/99 Madhu Added the function u_getVersion()
15 * 8/19/1999 srl Upgraded scripts to Unicode3.0
16 * 11/11/1999 weiv added u_isalnum(), cleaned comments
17 * 01/11/2000 helena Renamed u_getVersion to u_getUnicodeVersion.
18 * 06/20/2000 helena OS/400 port changes; mostly typecast.
19 ******************************************************************************
22 #include "unicode/utypes.h"
23 #include "unicode/uchar.h"
24 #include "unicode/uscript.h"
25 #include "unicode/udata.h"
34 #define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
36 /* uchar_props_data.h is machine-generated by genprops --csource */
37 #define INCLUDED_FROM_UCHAR_C
38 #include "uchar_props_data.h"
40 /* constants and macros for access to the data ------------------------------ */
42 /* getting a uint32_t properties word from the data */
43 #define GET_PROPS(c, result) ((result)=UTRIE2_GET16(&propsTrie, c));
46 uprv_haveProperties(UErrorCode *pErrorCode) {
47 if(U_FAILURE(*pErrorCode)) {
53 /* API functions ------------------------------------------------------------ */
55 /* Gets the Unicode character's general category.*/
56 U_CAPI int8_t U_EXPORT2
57 u_charType(UChar32 c) {
60 return (int8_t)GET_CATEGORY(props);
63 /* Enumerate all code points with their general categories. */
64 struct _EnumTypeCallback {
65 UCharEnumTypeRange *enumRange;
69 static uint32_t U_CALLCONV
70 _enumTypeValue(const void *context, uint32_t value) {
71 return GET_CATEGORY(value);
74 static UBool U_CALLCONV
75 _enumTypeRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
76 /* just cast the value to UCharCategory */
77 return ((struct _EnumTypeCallback *)context)->
78 enumRange(((struct _EnumTypeCallback *)context)->context,
79 start, end+1, (UCharCategory)value);
83 u_enumCharTypes(UCharEnumTypeRange *enumRange, const void *context) {
84 struct _EnumTypeCallback callback;
90 callback.enumRange=enumRange;
91 callback.context=context;
92 utrie2_enum(&propsTrie, _enumTypeValue, _enumTypeRange, &callback);
95 /* Checks if ch is a lower case letter.*/
96 U_CAPI UBool U_EXPORT2
97 u_islower(UChar32 c) {
100 return (UBool)(GET_CATEGORY(props)==U_LOWERCASE_LETTER);
103 /* Checks if ch is an upper case letter.*/
104 U_CAPI UBool U_EXPORT2
105 u_isupper(UChar32 c) {
108 return (UBool)(GET_CATEGORY(props)==U_UPPERCASE_LETTER);
111 /* Checks if ch is a title case letter; usually upper case letters.*/
112 U_CAPI UBool U_EXPORT2
113 u_istitle(UChar32 c) {
116 return (UBool)(GET_CATEGORY(props)==U_TITLECASE_LETTER);
119 /* Checks if ch is a decimal digit. */
120 U_CAPI UBool U_EXPORT2
121 u_isdigit(UChar32 c) {
124 return (UBool)(GET_CATEGORY(props)==U_DECIMAL_DIGIT_NUMBER);
127 U_CAPI UBool U_EXPORT2
128 u_isxdigit(UChar32 c) {
131 /* check ASCII and Fullwidth ASCII a-fA-F */
133 (c<=0x66 && c>=0x41 && (c<=0x46 || c>=0x61)) ||
134 (c>=0xff21 && c<=0xff46 && (c<=0xff26 || c>=0xff41))
140 return (UBool)(GET_CATEGORY(props)==U_DECIMAL_DIGIT_NUMBER);
143 /* Checks if the Unicode character is a letter.*/
144 U_CAPI UBool U_EXPORT2
145 u_isalpha(UChar32 c) {
148 return (UBool)((CAT_MASK(props)&U_GC_L_MASK)!=0);
151 U_CAPI UBool U_EXPORT2
152 u_isUAlphabetic(UChar32 c) {
153 return (u_getUnicodeProperties(c, 1)&U_MASK(UPROPS_ALPHABETIC))!=0;
156 /* Checks if c is a letter or a decimal digit */
157 U_CAPI UBool U_EXPORT2
158 u_isalnum(UChar32 c) {
161 return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_ND_MASK))!=0);
165 * Checks if c is alphabetic, or a decimal digit; implements UCHAR_POSIX_ALNUM.
169 u_isalnumPOSIX(UChar32 c) {
170 return (UBool)(u_isUAlphabetic(c) || u_isdigit(c));
173 /* Checks if ch is a unicode character with assigned character type.*/
174 U_CAPI UBool U_EXPORT2
175 u_isdefined(UChar32 c) {
178 return (UBool)(GET_CATEGORY(props)!=0);
181 /* Checks if the Unicode character is a base form character that can take a diacritic.*/
182 U_CAPI UBool U_EXPORT2
183 u_isbase(UChar32 c) {
186 return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_N_MASK|U_GC_MC_MASK|U_GC_ME_MASK))!=0);
189 /* Checks if the Unicode character is a control character.*/
190 U_CAPI UBool U_EXPORT2
191 u_iscntrl(UChar32 c) {
194 return (UBool)((CAT_MASK(props)&(U_GC_CC_MASK|U_GC_CF_MASK|U_GC_ZL_MASK|U_GC_ZP_MASK))!=0);
197 U_CAPI UBool U_EXPORT2
198 u_isISOControl(UChar32 c) {
199 return (uint32_t)c<=0x9f && (c<=0x1f || c>=0x7f);
202 /* Some control characters that are used as space. */
203 #define IS_THAT_CONTROL_SPACE(c) \
204 (c<=0x9f && ((c>=TAB && c<=CR) || (c>=0x1c && c <=0x1f) || c==NL))
206 /* Java has decided that U+0085 New Line is not whitespace any more. */
207 #define IS_THAT_ASCII_CONTROL_SPACE(c) \
208 (c<=0x1f && c>=TAB && (c<=CR || c>=0x1c))
210 /* Checks if the Unicode character is a space character.*/
211 U_CAPI UBool U_EXPORT2
212 u_isspace(UChar32 c) {
215 return (UBool)((CAT_MASK(props)&U_GC_Z_MASK)!=0 || IS_THAT_CONTROL_SPACE(c));
218 U_CAPI UBool U_EXPORT2
219 u_isJavaSpaceChar(UChar32 c) {
222 return (UBool)((CAT_MASK(props)&U_GC_Z_MASK)!=0);
225 /* Checks if the Unicode character is a whitespace character.*/
226 U_CAPI UBool U_EXPORT2
227 u_isWhitespace(UChar32 c) {
231 ((CAT_MASK(props)&U_GC_Z_MASK)!=0 &&
232 c!=NBSP && c!=FIGURESP && c!=NNBSP) || /* exclude no-break spaces */
233 IS_THAT_ASCII_CONTROL_SPACE(c)
237 U_CAPI UBool U_EXPORT2
238 u_isblank(UChar32 c) {
239 if((uint32_t)c<=0x9f) {
240 return c==9 || c==0x20; /* TAB or SPACE */
245 return (UBool)(GET_CATEGORY(props)==U_SPACE_SEPARATOR);
249 U_CAPI UBool U_EXPORT2
250 u_isUWhiteSpace(UChar32 c) {
251 return (u_getUnicodeProperties(c, 1)&U_MASK(UPROPS_WHITE_SPACE))!=0;
254 /* Checks if the Unicode character is printable.*/
255 U_CAPI UBool U_EXPORT2
256 u_isprint(UChar32 c) {
259 /* comparing ==0 returns FALSE for the categories mentioned */
260 return (UBool)((CAT_MASK(props)&U_GC_C_MASK)==0);
264 * Checks if c is in \p{graph}\p{blank} - \p{cntrl}.
265 * Implements UCHAR_POSIX_PRINT.
269 u_isprintPOSIX(UChar32 c) {
273 * The only cntrl character in graph+blank is TAB (in blank).
274 * Here we implement (blank-TAB)=Zs instead of calling u_isblank().
276 return (UBool)((GET_CATEGORY(props)==U_SPACE_SEPARATOR) || u_isgraphPOSIX(c));
279 U_CAPI UBool U_EXPORT2
280 u_isgraph(UChar32 c) {
283 /* comparing ==0 returns FALSE for the categories mentioned */
284 return (UBool)((CAT_MASK(props)&
285 (U_GC_CC_MASK|U_GC_CF_MASK|U_GC_CS_MASK|U_GC_CN_MASK|U_GC_Z_MASK))
291 * [^\p{space}\p{gc=Control}\p{gc=Surrogate}\p{gc=Unassigned}]
292 * with space=\p{Whitespace} and Control=Cc.
293 * Implements UCHAR_POSIX_GRAPH.
297 u_isgraphPOSIX(UChar32 c) {
300 /* \p{space}\p{gc=Control} == \p{gc=Z}\p{Control} */
301 /* comparing ==0 returns FALSE for the categories mentioned */
302 return (UBool)((CAT_MASK(props)&
303 (U_GC_CC_MASK|U_GC_CS_MASK|U_GC_CN_MASK|U_GC_Z_MASK))
307 U_CAPI UBool U_EXPORT2
308 u_ispunct(UChar32 c) {
311 return (UBool)((CAT_MASK(props)&U_GC_P_MASK)!=0);
314 /* Checks if the Unicode character can start a Unicode identifier.*/
315 U_CAPI UBool U_EXPORT2
316 u_isIDStart(UChar32 c) {
317 /* same as u_isalpha() */
320 return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_NL_MASK))!=0);
323 /* Checks if the Unicode character can be a Unicode identifier part other than starting the
325 U_CAPI UBool U_EXPORT2
326 u_isIDPart(UChar32 c) {
331 (U_GC_ND_MASK|U_GC_NL_MASK|
333 U_GC_PC_MASK|U_GC_MC_MASK|U_GC_MN_MASK)
338 /*Checks if the Unicode character can be ignorable in a Java or Unicode identifier.*/
339 U_CAPI UBool U_EXPORT2
340 u_isIDIgnorable(UChar32 c) {
342 return u_isISOControl(c) && !IS_THAT_ASCII_CONTROL_SPACE(c);
346 return (UBool)(GET_CATEGORY(props)==U_FORMAT_CHAR);
350 /*Checks if the Unicode character can start a Java identifier.*/
351 U_CAPI UBool U_EXPORT2
352 u_isJavaIDStart(UChar32 c) {
355 return (UBool)((CAT_MASK(props)&(U_GC_L_MASK|U_GC_SC_MASK|U_GC_PC_MASK))!=0);
358 /*Checks if the Unicode character can be a Java identifier part other than starting the
361 U_CAPI UBool U_EXPORT2
362 u_isJavaIDPart(UChar32 c) {
367 (U_GC_ND_MASK|U_GC_NL_MASK|
369 U_GC_SC_MASK|U_GC_PC_MASK|
370 U_GC_MC_MASK|U_GC_MN_MASK)
375 U_CAPI int32_t U_EXPORT2
376 u_charDigitValue(UChar32 c) {
380 value=(int32_t)GET_NUMERIC_TYPE_VALUE(props)-UPROPS_NTV_DECIMAL_START;
388 U_CAPI double U_EXPORT2
389 u_getNumericValue(UChar32 c) {
393 ntv=(int32_t)GET_NUMERIC_TYPE_VALUE(props);
395 if(ntv==UPROPS_NTV_NONE) {
396 return U_NO_NUMERIC_VALUE;
397 } else if(ntv<UPROPS_NTV_DIGIT_START) {
399 return ntv-UPROPS_NTV_DECIMAL_START;
400 } else if(ntv<UPROPS_NTV_NUMERIC_START) {
402 return ntv-UPROPS_NTV_DIGIT_START;
403 } else if(ntv<UPROPS_NTV_FRACTION_START) {
405 return ntv-UPROPS_NTV_NUMERIC_START;
406 } else if(ntv<UPROPS_NTV_LARGE_START) {
408 int32_t numerator=(ntv>>4)-12;
409 int32_t denominator=(ntv&0xf)+1;
410 return (double)numerator/denominator;
411 } else if(ntv<UPROPS_NTV_BASE60_START) {
412 /* large, single-significant-digit integer */
414 int32_t mant=(ntv>>5)-14;
415 int32_t exp=(ntv&0x1f)+2;
418 /* multiply by 10^exp without math.h */
439 } else if(ntv<UPROPS_NTV_RESERVED_START) {
440 /* sexagesimal (base 60) integer */
441 int32_t numValue=(ntv>>2)-0xbf;
442 int32_t exp=(ntv&3)+1;
446 numValue*=60*60*60*60;
465 return U_NO_NUMERIC_VALUE;
469 U_CAPI int32_t U_EXPORT2
470 u_digit(UChar32 ch, int8_t radix) {
472 if((uint8_t)(radix-2)<=(36-2)) {
473 value=(int8_t)u_charDigitValue(ch);
475 /* ch is not a decimal digit, try latin letters */
476 if(ch>=0x61 && ch<=0x7A) {
477 value=(int8_t)(ch-0x57); /* ch - 'a' + 10 */
478 } else if(ch>=0x41 && ch<=0x5A) {
479 value=(int8_t)(ch-0x37); /* ch - 'A' + 10 */
480 } else if(ch>=0xFF41 && ch<=0xFF5A) {
481 value=(int8_t)(ch-0xFF37); /* fullwidth ASCII a-z */
482 } else if(ch>=0xFF21 && ch<=0xFF3A) {
483 value=(int8_t)(ch-0xFF17); /* fullwidth ASCII A-Z */
487 value=-1; /* invalid radix */
489 return (int8_t)((value<radix) ? value : -1);
492 U_CAPI UChar32 U_EXPORT2
493 u_forDigit(int32_t digit, int8_t radix) {
494 if((uint8_t)(radix-2)>(36-2) || (uint32_t)digit>=(uint32_t)radix) {
496 } else if(digit<10) {
497 return (UChar32)(0x30+digit);
499 return (UChar32)((0x61-10)+digit);
503 /* miscellaneous, and support for uprops.cpp -------------------------------- */
505 U_CAPI void U_EXPORT2
506 u_getUnicodeVersion(UVersionInfo versionArray) {
507 if(versionArray!=NULL) {
508 uprv_memcpy(versionArray, dataVersion, U_MAX_VERSION_LENGTH);
513 u_getMainProperties(UChar32 c) {
520 u_getUnicodeProperties(UChar32 c, int32_t column) {
522 if(column>=propsVectorsColumns) {
525 uint16_t vecIndex=UTRIE2_GET16(&propsVectorsTrie, c);
526 return propsVectors[vecIndex+column];
531 uprv_getMaxValues(int32_t column) {
534 return indexes[UPROPS_MAX_VALUES_INDEX];
536 return indexes[UPROPS_MAX_VALUES_2_INDEX];
542 U_CAPI void U_EXPORT2
543 u_charAge(UChar32 c, UVersionInfo versionArray) {
544 if(versionArray!=NULL) {
545 uint32_t version=u_getUnicodeProperties(c, 0)>>UPROPS_AGE_SHIFT;
546 versionArray[0]=(uint8_t)(version>>4);
547 versionArray[1]=(uint8_t)(version&0xf);
548 versionArray[2]=versionArray[3]=0;
552 U_CAPI UScriptCode U_EXPORT2
553 uscript_getScript(UChar32 c, UErrorCode *pErrorCode) {
555 if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
556 return USCRIPT_INVALID_CODE;
558 if((uint32_t)c>0x10ffff) {
559 *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
560 return USCRIPT_INVALID_CODE;
562 scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
563 if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
564 return (UScriptCode)scriptX;
565 } else if(scriptX<UPROPS_SCRIPT_X_WITH_INHERITED) {
566 return USCRIPT_COMMON;
567 } else if(scriptX<UPROPS_SCRIPT_X_WITH_OTHER) {
568 return USCRIPT_INHERITED;
570 return (UScriptCode)scriptExtensions[scriptX&UPROPS_SCRIPT_MASK];
574 U_CAPI UBool U_EXPORT2
575 uscript_hasScript(UChar32 c, UScriptCode sc) {
577 uint32_t scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
578 if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
579 return sc==(UScriptCode)scriptX;
582 scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
583 if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
584 scx=scriptExtensions+scx[1];
586 if(sc>=USCRIPT_CODE_LIMIT) {
587 /* Guard against bogus input that would make us go past the Script_Extensions terminator. */
593 return sc==(*scx&0x7fff);
596 U_CAPI int32_t U_EXPORT2
597 uscript_getScriptExtensions(UChar32 c,
598 UScriptCode *scripts, int32_t capacity,
599 UErrorCode *pErrorCode) {
604 if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
607 if(capacity<0 || (capacity>0 && scripts==NULL)) {
608 *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
611 scriptX=u_getUnicodeProperties(c, 0)&UPROPS_SCRIPT_X_MASK;
612 if(scriptX<UPROPS_SCRIPT_X_WITH_COMMON) {
614 *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
616 scripts[0]=(UScriptCode)scriptX;
621 scx=scriptExtensions+(scriptX&UPROPS_SCRIPT_MASK);
622 if(scriptX>=UPROPS_SCRIPT_X_WITH_OTHER) {
623 scx=scriptExtensions+scx[1];
628 if(length<capacity) {
629 scripts[length]=(UScriptCode)(sx&0x7fff);
633 if(length>capacity) {
634 *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
639 U_CAPI UBlockCode U_EXPORT2
640 ublock_getCode(UChar32 c) {
641 return (UBlockCode)((u_getUnicodeProperties(c, 0)&UPROPS_BLOCK_MASK)>>UPROPS_BLOCK_SHIFT);
644 /* property starts for UnicodeSet ------------------------------------------- */
646 static UBool U_CALLCONV
647 _enumPropertyStartsRange(const void *context, UChar32 start, UChar32 end, uint32_t value) {
648 /* add the start code point to the USet */
649 const USetAdder *sa=(const USetAdder *)context;
650 sa->add(sa->set, start);
654 #define USET_ADD_CP_AND_NEXT(sa, cp) sa->add(sa->set, cp); sa->add(sa->set, cp+1)
656 U_CFUNC void U_EXPORT2
657 uchar_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode) {
658 if(U_FAILURE(*pErrorCode)) {
662 /* add the start code point of each same-value range of the main trie */
663 utrie2_enum(&propsTrie, NULL, _enumPropertyStartsRange, sa);
665 /* add code points with hardcoded properties, plus the ones following them */
667 /* add for u_isblank() */
668 USET_ADD_CP_AND_NEXT(sa, TAB);
670 /* add for IS_THAT_CONTROL_SPACE() */
671 sa->add(sa->set, CR+1); /* range TAB..CR */
672 sa->add(sa->set, 0x1c);
673 sa->add(sa->set, 0x1f+1);
674 USET_ADD_CP_AND_NEXT(sa, NL);
676 /* add for u_isIDIgnorable() what was not added above */
677 sa->add(sa->set, DEL); /* range DEL..NBSP-1, NBSP added below */
678 sa->add(sa->set, HAIRSP);
679 sa->add(sa->set, RLM+1);
680 sa->add(sa->set, INHSWAP);
681 sa->add(sa->set, NOMDIG+1);
682 USET_ADD_CP_AND_NEXT(sa, ZWNBSP);
684 /* add no-break spaces for u_isWhitespace() what was not added above */
685 USET_ADD_CP_AND_NEXT(sa, NBSP);
686 USET_ADD_CP_AND_NEXT(sa, FIGURESP);
687 USET_ADD_CP_AND_NEXT(sa, NNBSP);
689 /* add for u_digit() */
690 sa->add(sa->set, U_a);
691 sa->add(sa->set, U_z+1);
692 sa->add(sa->set, U_A);
693 sa->add(sa->set, U_Z+1);
694 sa->add(sa->set, U_FW_a);
695 sa->add(sa->set, U_FW_z+1);
696 sa->add(sa->set, U_FW_A);
697 sa->add(sa->set, U_FW_Z+1);
699 /* add for u_isxdigit() */
700 sa->add(sa->set, U_f+1);
701 sa->add(sa->set, U_F+1);
702 sa->add(sa->set, U_FW_f+1);
703 sa->add(sa->set, U_FW_F+1);
705 /* add for UCHAR_DEFAULT_IGNORABLE_CODE_POINT what was not added above */
706 sa->add(sa->set, WJ); /* range WJ..NOMDIG */
707 sa->add(sa->set, 0xfff0);
708 sa->add(sa->set, 0xfffb+1);
709 sa->add(sa->set, 0xe0000);
710 sa->add(sa->set, 0xe0fff+1);
712 /* add for UCHAR_GRAPHEME_BASE and others */
713 USET_ADD_CP_AND_NEXT(sa, CGJ);
716 U_CFUNC void U_EXPORT2
717 upropsvec_addPropertyStarts(const USetAdder *sa, UErrorCode *pErrorCode) {
718 if(U_FAILURE(*pErrorCode)) {
722 /* add the start code point of each same-value range of the properties vectors trie */
723 if(propsVectorsColumns>0) {
724 /* if propsVectorsColumns==0 then the properties vectors trie may not be there at all */
725 utrie2_enum(&propsVectorsTrie, NULL, _enumPropertyStartsRange, sa);