1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
4 ********************************************************************************
5 * Copyright (C) 1996-2016, International Business Machines
6 * Corporation and others. All Rights Reserved.
7 ********************************************************************************
11 * Modification History:
13 * Date Name Description
14 * 04/02/97 aliu Creation.
15 * 4/15/99 Madhu Updated all the function definitions for C Implementation
16 * 5/20/99 Madhu Added the function u_getVersion()
17 * 8/19/1999 srl Upgraded scripts to Unicode3.0
18 * 11/11/1999 weiv added u_isalnum(), cleaned comments
19 * 01/11/2000 helena Renamed u_getVersion to u_getUnicodeVersion.
20 * 06/20/2000 helena OS/400 port changes; mostly typecast.
21 ******************************************************************************
24 #include "unicode/utypes.h"
25 #include "unicode/uchar.h"
26 #include "unicode/uscript.h"
27 #include "unicode/udata.h"
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));
45 /* API functions ------------------------------------------------------------ */
47 /* Gets the Unicode character's general category.*/
48 U_CAPI
int8_t U_EXPORT2
49 u_charType(UChar32 c
) {
52 return (int8_t)GET_CATEGORY(props
);
55 /* Enumerate all code points with their general categories. */
56 struct _EnumTypeCallback
{
57 UCharEnumTypeRange
*enumRange
;
61 static uint32_t U_CALLCONV
62 _enumTypeValue(const void *context
, uint32_t value
) {
64 return GET_CATEGORY(value
);
67 static UBool U_CALLCONV
68 _enumTypeRange(const void *context
, UChar32 start
, UChar32 end
, uint32_t value
) {
69 /* just cast the value to UCharCategory */
70 return ((struct _EnumTypeCallback
*)context
)->
71 enumRange(((struct _EnumTypeCallback
*)context
)->context
,
72 start
, end
+1, (UCharCategory
)value
);
76 u_enumCharTypes(UCharEnumTypeRange
*enumRange
, const void *context
) {
77 struct _EnumTypeCallback callback
;
83 callback
.enumRange
=enumRange
;
84 callback
.context
=context
;
85 utrie2_enum(&propsTrie
, _enumTypeValue
, _enumTypeRange
, &callback
);
88 /* Checks if ch is a lower case letter.*/
89 U_CAPI UBool U_EXPORT2
90 u_islower(UChar32 c
) {
93 return (UBool
)(GET_CATEGORY(props
)==U_LOWERCASE_LETTER
);
96 /* Checks if ch is an upper case letter.*/
97 U_CAPI UBool U_EXPORT2
98 u_isupper(UChar32 c
) {
101 return (UBool
)(GET_CATEGORY(props
)==U_UPPERCASE_LETTER
);
104 /* Checks if ch is a title case letter; usually upper case letters.*/
105 U_CAPI UBool U_EXPORT2
106 u_istitle(UChar32 c
) {
109 return (UBool
)(GET_CATEGORY(props
)==U_TITLECASE_LETTER
);
112 /* Checks if ch is a decimal digit. */
113 U_CAPI UBool U_EXPORT2
114 u_isdigit(UChar32 c
) {
117 return (UBool
)(GET_CATEGORY(props
)==U_DECIMAL_DIGIT_NUMBER
);
120 U_CAPI UBool U_EXPORT2
121 u_isxdigit(UChar32 c
) {
124 /* check ASCII and Fullwidth ASCII a-fA-F */
126 (c
<=0x66 && c
>=0x41 && (c
<=0x46 || c
>=0x61)) ||
127 (c
>=0xff21 && c
<=0xff46 && (c
<=0xff26 || c
>=0xff41))
133 return (UBool
)(GET_CATEGORY(props
)==U_DECIMAL_DIGIT_NUMBER
);
136 /* Checks if the Unicode character is a letter.*/
137 U_CAPI UBool U_EXPORT2
138 u_isalpha(UChar32 c
) {
141 return (UBool
)((CAT_MASK(props
)&U_GC_L_MASK
)!=0);
144 U_CAPI UBool U_EXPORT2
145 u_isUAlphabetic(UChar32 c
) {
146 return (u_getUnicodeProperties(c
, 1)&U_MASK(UPROPS_ALPHABETIC
))!=0;
149 /* Checks if c is a letter or a decimal digit */
150 U_CAPI UBool U_EXPORT2
151 u_isalnum(UChar32 c
) {
154 return (UBool
)((CAT_MASK(props
)&(U_GC_L_MASK
|U_GC_ND_MASK
))!=0);
158 * Checks if c is alphabetic, or a decimal digit; implements UCHAR_POSIX_ALNUM.
162 u_isalnumPOSIX(UChar32 c
) {
163 return (UBool
)(u_isUAlphabetic(c
) || u_isdigit(c
));
166 /* Checks if ch is a unicode character with assigned character type.*/
167 U_CAPI UBool U_EXPORT2
168 u_isdefined(UChar32 c
) {
171 return (UBool
)(GET_CATEGORY(props
)!=0);
174 /* Checks if the Unicode character is a base form character that can take a diacritic.*/
175 U_CAPI UBool U_EXPORT2
176 u_isbase(UChar32 c
) {
179 return (UBool
)((CAT_MASK(props
)&(U_GC_L_MASK
|U_GC_N_MASK
|U_GC_MC_MASK
|U_GC_ME_MASK
))!=0);
182 /* Checks if the Unicode character is a control character.*/
183 U_CAPI UBool U_EXPORT2
184 u_iscntrl(UChar32 c
) {
187 return (UBool
)((CAT_MASK(props
)&(U_GC_CC_MASK
|U_GC_CF_MASK
|U_GC_ZL_MASK
|U_GC_ZP_MASK
))!=0);
190 U_CAPI UBool U_EXPORT2
191 u_isISOControl(UChar32 c
) {
192 return (uint32_t)c
<=0x9f && (c
<=0x1f || c
>=0x7f);
195 /* Some control characters that are used as space. */
196 #define IS_THAT_CONTROL_SPACE(c) \
197 (c<=0x9f && ((c>=TAB && c<=CR) || (c>=0x1c && c <=0x1f) || c==NL))
199 /* Java has decided that U+0085 New Line is not whitespace any more. */
200 #define IS_THAT_ASCII_CONTROL_SPACE(c) \
201 (c<=0x1f && c>=TAB && (c<=CR || c>=0x1c))
203 /* Checks if the Unicode character is a space character.*/
204 U_CAPI UBool U_EXPORT2
205 u_isspace(UChar32 c
) {
208 return (UBool
)((CAT_MASK(props
)&U_GC_Z_MASK
)!=0 || IS_THAT_CONTROL_SPACE(c
));
211 U_CAPI UBool U_EXPORT2
212 u_isJavaSpaceChar(UChar32 c
) {
215 return (UBool
)((CAT_MASK(props
)&U_GC_Z_MASK
)!=0);
218 /* Checks if the Unicode character is a whitespace character.*/
219 U_CAPI UBool U_EXPORT2
220 u_isWhitespace(UChar32 c
) {
224 ((CAT_MASK(props
)&U_GC_Z_MASK
)!=0 &&
225 c
!=NBSP
&& c
!=FIGURESP
&& c
!=NNBSP
) || /* exclude no-break spaces */
226 IS_THAT_ASCII_CONTROL_SPACE(c
)
230 U_CAPI UBool U_EXPORT2
231 u_isblank(UChar32 c
) {
232 if((uint32_t)c
<=0x9f) {
233 return c
==9 || c
==0x20; /* TAB or SPACE */
238 return (UBool
)(GET_CATEGORY(props
)==U_SPACE_SEPARATOR
);
242 U_CAPI UBool U_EXPORT2
243 u_isUWhiteSpace(UChar32 c
) {
244 return (u_getUnicodeProperties(c
, 1)&U_MASK(UPROPS_WHITE_SPACE
))!=0;
247 /* Checks if the Unicode character is printable.*/
248 U_CAPI UBool U_EXPORT2
249 u_isprint(UChar32 c
) {
252 /* comparing ==0 returns FALSE for the categories mentioned */
253 return (UBool
)((CAT_MASK(props
)&U_GC_C_MASK
)==0);
257 * Checks if c is in \p{graph}\p{blank} - \p{cntrl}.
258 * Implements UCHAR_POSIX_PRINT.
262 u_isprintPOSIX(UChar32 c
) {
266 * The only cntrl character in graph+blank is TAB (in blank).
267 * Here we implement (blank-TAB)=Zs instead of calling u_isblank().
269 return (UBool
)((GET_CATEGORY(props
)==U_SPACE_SEPARATOR
) || u_isgraphPOSIX(c
));
272 U_CAPI UBool U_EXPORT2
273 u_isgraph(UChar32 c
) {
276 /* comparing ==0 returns FALSE for the categories mentioned */
277 return (UBool
)((CAT_MASK(props
)&
278 (U_GC_CC_MASK
|U_GC_CF_MASK
|U_GC_CS_MASK
|U_GC_CN_MASK
|U_GC_Z_MASK
))
284 * [^\p{space}\p{gc=Control}\p{gc=Surrogate}\p{gc=Unassigned}]
285 * with space=\p{Whitespace} and Control=Cc.
286 * Implements UCHAR_POSIX_GRAPH.
290 u_isgraphPOSIX(UChar32 c
) {
293 /* \p{space}\p{gc=Control} == \p{gc=Z}\p{Control} */
294 /* comparing ==0 returns FALSE for the categories mentioned */
295 return (UBool
)((CAT_MASK(props
)&
296 (U_GC_CC_MASK
|U_GC_CS_MASK
|U_GC_CN_MASK
|U_GC_Z_MASK
))
300 U_CAPI UBool U_EXPORT2
301 u_ispunct(UChar32 c
) {
304 return (UBool
)((CAT_MASK(props
)&U_GC_P_MASK
)!=0);
307 /* Checks if the Unicode character can start a Unicode identifier.*/
308 U_CAPI UBool U_EXPORT2
309 u_isIDStart(UChar32 c
) {
310 /* same as u_isalpha() */
313 return (UBool
)((CAT_MASK(props
)&(U_GC_L_MASK
|U_GC_NL_MASK
))!=0);
316 /* Checks if the Unicode character can be a Unicode identifier part other than starting the
318 U_CAPI UBool U_EXPORT2
319 u_isIDPart(UChar32 c
) {
324 (U_GC_ND_MASK
|U_GC_NL_MASK
|
326 U_GC_PC_MASK
|U_GC_MC_MASK
|U_GC_MN_MASK
)
331 /*Checks if the Unicode character can be ignorable in a Java or Unicode identifier.*/
332 U_CAPI UBool U_EXPORT2
333 u_isIDIgnorable(UChar32 c
) {
335 return u_isISOControl(c
) && !IS_THAT_ASCII_CONTROL_SPACE(c
);
339 return (UBool
)(GET_CATEGORY(props
)==U_FORMAT_CHAR
);
343 /*Checks if the Unicode character can start a Java identifier.*/
344 U_CAPI UBool U_EXPORT2
345 u_isJavaIDStart(UChar32 c
) {
348 return (UBool
)((CAT_MASK(props
)&(U_GC_L_MASK
|U_GC_SC_MASK
|U_GC_PC_MASK
))!=0);
351 /*Checks if the Unicode character can be a Java identifier part other than starting the
354 U_CAPI UBool U_EXPORT2
355 u_isJavaIDPart(UChar32 c
) {
360 (U_GC_ND_MASK
|U_GC_NL_MASK
|
362 U_GC_SC_MASK
|U_GC_PC_MASK
|
363 U_GC_MC_MASK
|U_GC_MN_MASK
)
368 U_CAPI
int32_t U_EXPORT2
369 u_charDigitValue(UChar32 c
) {
373 value
=(int32_t)GET_NUMERIC_TYPE_VALUE(props
)-UPROPS_NTV_DECIMAL_START
;
381 U_CAPI
double U_EXPORT2
382 u_getNumericValue(UChar32 c
) {
386 ntv
=(int32_t)GET_NUMERIC_TYPE_VALUE(props
);
388 if(ntv
==UPROPS_NTV_NONE
) {
389 return U_NO_NUMERIC_VALUE
;
390 } else if(ntv
<UPROPS_NTV_DIGIT_START
) {
392 return ntv
-UPROPS_NTV_DECIMAL_START
;
393 } else if(ntv
<UPROPS_NTV_NUMERIC_START
) {
395 return ntv
-UPROPS_NTV_DIGIT_START
;
396 } else if(ntv
<UPROPS_NTV_FRACTION_START
) {
398 return ntv
-UPROPS_NTV_NUMERIC_START
;
399 } else if(ntv
<UPROPS_NTV_LARGE_START
) {
401 int32_t numerator
=(ntv
>>4)-12;
402 int32_t denominator
=(ntv
&0xf)+1;
403 return (double)numerator
/denominator
;
404 } else if(ntv
<UPROPS_NTV_BASE60_START
) {
405 /* large, single-significant-digit integer */
407 int32_t mant
=(ntv
>>5)-14;
408 int32_t exp
=(ntv
&0x1f)+2;
411 /* multiply by 10^exp without math.h */
432 } else if(ntv
<UPROPS_NTV_FRACTION20_START
) {
433 /* sexagesimal (base 60) integer */
434 int32_t numValue
=(ntv
>>2)-0xbf;
435 int32_t exp
=(ntv
&3)+1;
439 numValue
*=60*60*60*60;
456 } else if(ntv
<UPROPS_NTV_RESERVED_START
) {
457 // fraction-20 e.g. 3/80
458 int32_t frac20
=ntv
-UPROPS_NTV_FRACTION20_START
; // 0..0x17
459 int32_t numerator
=2*(frac20
&3)+1;
460 int32_t denominator
=20<<(frac20
>>2);
461 return (double)numerator
/denominator
;
464 return U_NO_NUMERIC_VALUE
;
468 U_CAPI
int32_t U_EXPORT2
469 u_digit(UChar32 ch
, int8_t radix
) {
471 if((uint8_t)(radix
-2)<=(36-2)) {
472 value
=(int8_t)u_charDigitValue(ch
);
474 /* ch is not a decimal digit, try latin letters */
475 if(ch
>=0x61 && ch
<=0x7A) {
476 value
=(int8_t)(ch
-0x57); /* ch - 'a' + 10 */
477 } else if(ch
>=0x41 && ch
<=0x5A) {
478 value
=(int8_t)(ch
-0x37); /* ch - 'A' + 10 */
479 } else if(ch
>=0xFF41 && ch
<=0xFF5A) {
480 value
=(int8_t)(ch
-0xFF37); /* fullwidth ASCII a-z */
481 } else if(ch
>=0xFF21 && ch
<=0xFF3A) {
482 value
=(int8_t)(ch
-0xFF17); /* fullwidth ASCII A-Z */
486 value
=-1; /* invalid radix */
488 return (int8_t)((value
<radix
) ? value
: -1);
491 U_CAPI UChar32 U_EXPORT2
492 u_forDigit(int32_t digit
, int8_t radix
) {
493 if((uint8_t)(radix
-2)>(36-2) || (uint32_t)digit
>=(uint32_t)radix
) {
495 } else if(digit
<10) {
496 return (UChar32
)(0x30+digit
);
498 return (UChar32
)((0x61-10)+digit
);
502 /* miscellaneous, and support for uprops.cpp -------------------------------- */
504 U_CAPI
void U_EXPORT2
505 u_getUnicodeVersion(UVersionInfo versionArray
) {
506 if(versionArray
!=NULL
) {
507 uprv_memcpy(versionArray
, dataVersion
, U_MAX_VERSION_LENGTH
);
512 u_getMainProperties(UChar32 c
) {
519 u_getUnicodeProperties(UChar32 c
, int32_t column
) {
521 if(column
>=propsVectorsColumns
) {
524 uint16_t vecIndex
=UTRIE2_GET16(&propsVectorsTrie
, c
);
525 return propsVectors
[vecIndex
+column
];
530 uprv_getMaxValues(int32_t column
) {
533 return indexes
[UPROPS_MAX_VALUES_INDEX
];
535 return indexes
[UPROPS_MAX_VALUES_2_INDEX
];
541 U_CAPI
void U_EXPORT2
542 u_charAge(UChar32 c
, UVersionInfo versionArray
) {
543 if(versionArray
!=NULL
) {
544 uint32_t version
=u_getUnicodeProperties(c
, 0)>>UPROPS_AGE_SHIFT
;
545 versionArray
[0]=(uint8_t)(version
>>4);
546 versionArray
[1]=(uint8_t)(version
&0xf);
547 versionArray
[2]=versionArray
[3]=0;
551 U_CAPI UScriptCode U_EXPORT2
552 uscript_getScript(UChar32 c
, UErrorCode
*pErrorCode
) {
554 if(pErrorCode
==NULL
|| U_FAILURE(*pErrorCode
)) {
555 return USCRIPT_INVALID_CODE
;
557 if((uint32_t)c
>0x10ffff) {
558 *pErrorCode
=U_ILLEGAL_ARGUMENT_ERROR
;
559 return USCRIPT_INVALID_CODE
;
561 scriptX
=u_getUnicodeProperties(c
, 0)&UPROPS_SCRIPT_X_MASK
;
562 if(scriptX
<UPROPS_SCRIPT_X_WITH_COMMON
) {
563 return (UScriptCode
)scriptX
;
564 } else if(scriptX
<UPROPS_SCRIPT_X_WITH_INHERITED
) {
565 return USCRIPT_COMMON
;
566 } else if(scriptX
<UPROPS_SCRIPT_X_WITH_OTHER
) {
567 return USCRIPT_INHERITED
;
569 return (UScriptCode
)scriptExtensions
[scriptX
&UPROPS_SCRIPT_MASK
];
573 U_CAPI UBool U_EXPORT2
574 uscript_hasScript(UChar32 c
, UScriptCode sc
) {
576 uint32_t scriptX
=u_getUnicodeProperties(c
, 0)&UPROPS_SCRIPT_X_MASK
;
577 if(scriptX
<UPROPS_SCRIPT_X_WITH_COMMON
) {
578 return sc
==(UScriptCode
)scriptX
;
581 scx
=scriptExtensions
+(scriptX
&UPROPS_SCRIPT_MASK
);
582 if(scriptX
>=UPROPS_SCRIPT_X_WITH_OTHER
) {
583 scx
=scriptExtensions
+scx
[1];
585 if(sc
>=USCRIPT_CODE_LIMIT
) {
586 /* Guard against bogus input that would make us go past the Script_Extensions terminator. */
592 return sc
==(*scx
&0x7fff);
595 U_CAPI
int32_t U_EXPORT2
596 uscript_getScriptExtensions(UChar32 c
,
597 UScriptCode
*scripts
, int32_t capacity
,
598 UErrorCode
*pErrorCode
) {
603 if(pErrorCode
==NULL
|| U_FAILURE(*pErrorCode
)) {
606 if(capacity
<0 || (capacity
>0 && scripts
==NULL
)) {
607 *pErrorCode
=U_ILLEGAL_ARGUMENT_ERROR
;
610 scriptX
=u_getUnicodeProperties(c
, 0)&UPROPS_SCRIPT_X_MASK
;
611 if(scriptX
<UPROPS_SCRIPT_X_WITH_COMMON
) {
613 *pErrorCode
=U_BUFFER_OVERFLOW_ERROR
;
615 scripts
[0]=(UScriptCode
)scriptX
;
620 scx
=scriptExtensions
+(scriptX
&UPROPS_SCRIPT_MASK
);
621 if(scriptX
>=UPROPS_SCRIPT_X_WITH_OTHER
) {
622 scx
=scriptExtensions
+scx
[1];
627 if(length
<capacity
) {
628 scripts
[length
]=(UScriptCode
)(sx
&0x7fff);
632 if(length
>capacity
) {
633 *pErrorCode
=U_BUFFER_OVERFLOW_ERROR
;
638 U_CAPI UBlockCode U_EXPORT2
639 ublock_getCode(UChar32 c
) {
640 return (UBlockCode
)((u_getUnicodeProperties(c
, 0)&UPROPS_BLOCK_MASK
)>>UPROPS_BLOCK_SHIFT
);
643 /* property starts for UnicodeSet ------------------------------------------- */
645 static UBool U_CALLCONV
646 _enumPropertyStartsRange(const void *context
, UChar32 start
, UChar32 end
, uint32_t value
) {
647 /* add the start code point to the USet */
648 const USetAdder
*sa
=(const USetAdder
*)context
;
649 sa
->add(sa
->set
, start
);
655 #define USET_ADD_CP_AND_NEXT(sa, cp) sa->add(sa->set, cp); sa->add(sa->set, cp+1)
657 U_CFUNC
void U_EXPORT2
658 uchar_addPropertyStarts(const USetAdder
*sa
, UErrorCode
*pErrorCode
) {
659 if(U_FAILURE(*pErrorCode
)) {
663 /* add the start code point of each same-value range of the main trie */
664 utrie2_enum(&propsTrie
, NULL
, _enumPropertyStartsRange
, sa
);
666 /* add code points with hardcoded properties, plus the ones following them */
668 /* add for u_isblank() */
669 USET_ADD_CP_AND_NEXT(sa
, TAB
);
671 /* add for IS_THAT_CONTROL_SPACE() */
672 sa
->add(sa
->set
, CR
+1); /* range TAB..CR */
673 sa
->add(sa
->set
, 0x1c);
674 sa
->add(sa
->set
, 0x1f+1);
675 USET_ADD_CP_AND_NEXT(sa
, NL
);
677 /* add for u_isIDIgnorable() what was not added above */
678 sa
->add(sa
->set
, DEL
); /* range DEL..NBSP-1, NBSP added below */
679 sa
->add(sa
->set
, HAIRSP
);
680 sa
->add(sa
->set
, RLM
+1);
681 sa
->add(sa
->set
, INHSWAP
);
682 sa
->add(sa
->set
, NOMDIG
+1);
683 USET_ADD_CP_AND_NEXT(sa
, ZWNBSP
);
685 /* add no-break spaces for u_isWhitespace() what was not added above */
686 USET_ADD_CP_AND_NEXT(sa
, NBSP
);
687 USET_ADD_CP_AND_NEXT(sa
, FIGURESP
);
688 USET_ADD_CP_AND_NEXT(sa
, NNBSP
);
690 /* add for u_digit() */
691 sa
->add(sa
->set
, U_a
);
692 sa
->add(sa
->set
, U_z
+1);
693 sa
->add(sa
->set
, U_A
);
694 sa
->add(sa
->set
, U_Z
+1);
695 sa
->add(sa
->set
, U_FW_a
);
696 sa
->add(sa
->set
, U_FW_z
+1);
697 sa
->add(sa
->set
, U_FW_A
);
698 sa
->add(sa
->set
, U_FW_Z
+1);
700 /* add for u_isxdigit() */
701 sa
->add(sa
->set
, U_f
+1);
702 sa
->add(sa
->set
, U_F
+1);
703 sa
->add(sa
->set
, U_FW_f
+1);
704 sa
->add(sa
->set
, U_FW_F
+1);
706 /* add for UCHAR_DEFAULT_IGNORABLE_CODE_POINT what was not added above */
707 sa
->add(sa
->set
, WJ
); /* range WJ..NOMDIG */
708 sa
->add(sa
->set
, 0xfff0);
709 sa
->add(sa
->set
, 0xfffb+1);
710 sa
->add(sa
->set
, 0xe0000);
711 sa
->add(sa
->set
, 0xe0fff+1);
713 /* add for UCHAR_GRAPHEME_BASE and others */
714 USET_ADD_CP_AND_NEXT(sa
, CGJ
);
717 U_CFUNC
void U_EXPORT2
718 upropsvec_addPropertyStarts(const USetAdder
*sa
, UErrorCode
*pErrorCode
) {
719 if(U_FAILURE(*pErrorCode
)) {
723 /* add the start code point of each same-value range of the properties vectors trie */
724 utrie2_enum(&propsVectorsTrie
, NULL
, _enumPropertyStartsRange
, sa
);