ICU 74.1 74.1
utf8.h
Go to the documentation of this file.
1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4*******************************************************************************
5*
6* Copyright (C) 1999-2015, International Business Machines
7* Corporation and others. All Rights Reserved.
8*
9*******************************************************************************
10* file name: utf8.h
11* encoding: UTF-8
12* tab size: 8 (not used)
13* indentation:4
14*
15* created on: 1999sep13
16* created by: Markus W. Scherer
17*/
18
34#ifndef __UTF8_H__
35#define __UTF8_H__
36
37#include <stdbool.h>
38#include "unicode/umachine.h"
39#ifndef __UTF_H__
40# include "unicode/utf.h"
41#endif
42
43/* internal definitions ----------------------------------------------------- */
44
56#define U8_COUNT_TRAIL_BYTES(leadByte) \
57 (U8_IS_LEAD(leadByte) ? \
58 ((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0)+1 : 0)
59
71#define U8_COUNT_TRAIL_BYTES_UNSAFE(leadByte) \
72 (((uint8_t)(leadByte)>=0xc2)+((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0))
73
81#define U8_MASK_LEAD_BYTE(leadByte, countTrailBytes) ((leadByte)&=(1<<(6-(countTrailBytes)))-1)
82
91#define U8_LEAD3_T1_BITS "\x20\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x10\x30\x30"
92
98#define U8_IS_VALID_LEAD3_AND_T1(lead, t1) (U8_LEAD3_T1_BITS[(lead)&0xf]&(1<<((uint8_t)(t1)>>5)))
99
108#define U8_LEAD4_T1_BITS "\x00\x00\x00\x00\x00\x00\x00\x00\x1E\x0F\x0F\x0F\x00\x00\x00\x00"
109
115#define U8_IS_VALID_LEAD4_AND_T1(lead, t1) (U8_LEAD4_T1_BITS[(uint8_t)(t1)>>4]&(1<<((lead)&7)))
116
126U_CAPI UChar32 U_EXPORT2
127utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict);
128
138U_CAPI int32_t U_EXPORT2
139utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool *pIsError);
140
150U_CAPI UChar32 U_EXPORT2
151utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict);
152
162U_CAPI int32_t U_EXPORT2
163utf8_back1SafeBody(const uint8_t *s, int32_t start, int32_t i);
164
165/* single-code point definitions -------------------------------------------- */
166
173#define U8_IS_SINGLE(c) (((c)&0x80)==0)
174
181#define U8_IS_LEAD(c) ((uint8_t)((c)-0xc2)<=0x32)
182// 0x32=0xf4-0xc2
183
190#define U8_IS_TRAIL(c) ((int8_t)(c)<-0x40)
191
199#define U8_LENGTH(c) \
200 ((uint32_t)(c)<=0x7f ? 1 : \
201 ((uint32_t)(c)<=0x7ff ? 2 : \
202 ((uint32_t)(c)<=0xd7ff ? 3 : \
203 ((uint32_t)(c)<=0xdfff || (uint32_t)(c)>0x10ffff ? 0 : \
204 ((uint32_t)(c)<=0xffff ? 3 : 4)\
205 ) \
206 ) \
207 ) \
208 )
209
215#define U8_MAX_LENGTH 4
216
233#define U8_GET_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
234 int32_t _u8_get_unsafe_index=(int32_t)(i); \
235 U8_SET_CP_START_UNSAFE(s, _u8_get_unsafe_index); \
236 U8_NEXT_UNSAFE(s, _u8_get_unsafe_index, c); \
237} UPRV_BLOCK_MACRO_END
238
260#define U8_GET(s, start, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
261 int32_t _u8_get_index=(i); \
262 U8_SET_CP_START(s, start, _u8_get_index); \
263 U8_NEXT(s, _u8_get_index, length, c); \
264} UPRV_BLOCK_MACRO_END
265
291#define U8_GET_OR_FFFD(s, start, i, length, c) UPRV_BLOCK_MACRO_BEGIN { \
292 int32_t _u8_get_index=(i); \
293 U8_SET_CP_START(s, start, _u8_get_index); \
294 U8_NEXT_OR_FFFD(s, _u8_get_index, length, c); \
295} UPRV_BLOCK_MACRO_END
296
297/* definitions with forward iteration --------------------------------------- */
298
316#define U8_NEXT_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
317 (c)=(uint8_t)(s)[(i)++]; \
318 if(!U8_IS_SINGLE(c)) { \
319 if((c)<0xe0) { \
320 (c)=(((c)&0x1f)<<6)|((s)[(i)++]&0x3f); \
321 } else if((c)<0xf0) { \
322 /* no need for (c&0xf) because the upper bits are truncated after <<12 in the cast to (UChar) */ \
323 (c)=(UChar)(((c)<<12)|(((s)[i]&0x3f)<<6)|((s)[(i)+1]&0x3f)); \
324 (i)+=2; \
325 } else { \
326 (c)=(((c)&7)<<18)|(((s)[i]&0x3f)<<12)|(((s)[(i)+1]&0x3f)<<6)|((s)[(i)+2]&0x3f); \
327 (i)+=3; \
328 } \
329 } \
330} UPRV_BLOCK_MACRO_END
331
352#define U8_NEXT(s, i, length, c) U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, U_SENTINEL)
353
378#define U8_NEXT_OR_FFFD(s, i, length, c) U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, 0xfffd)
379
381#define U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, sub) UPRV_BLOCK_MACRO_BEGIN { \
382 (c)=(uint8_t)(s)[(i)++]; \
383 if(!U8_IS_SINGLE(c)) { \
384 uint8_t __t = 0; \
385 if((i)!=(length) && \
386 /* fetch/validate/assemble all but last trail byte */ \
387 ((c)>=0xe0 ? \
388 ((c)<0xf0 ? /* U+0800..U+FFFF except surrogates */ \
389 U8_LEAD3_T1_BITS[(c)&=0xf]&(1<<((__t=(s)[i])>>5)) && \
390 (__t&=0x3f, 1) \
391 : /* U+10000..U+10FFFF */ \
392 ((c)-=0xf0)<=4 && \
393 U8_LEAD4_T1_BITS[(__t=(s)[i])>>4]&(1<<(c)) && \
394 ((c)=((c)<<6)|(__t&0x3f), ++(i)!=(length)) && \
395 (__t=(s)[i]-0x80)<=0x3f) && \
396 /* valid second-to-last trail byte */ \
397 ((c)=((c)<<6)|__t, ++(i)!=(length)) \
398 : /* U+0080..U+07FF */ \
399 (c)>=0xc2 && ((c)&=0x1f, 1)) && \
400 /* last trail byte */ \
401 (__t=(s)[i]-0x80)<=0x3f && \
402 ((c)=((c)<<6)|__t, ++(i), 1)) { \
403 } else { \
404 (c)=(sub); /* ill-formed*/ \
405 } \
406 } \
407} UPRV_BLOCK_MACRO_END
408
422#define U8_APPEND_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
423 uint32_t __uc=(c); \
424 if(__uc<=0x7f) { \
425 (s)[(i)++]=(uint8_t)__uc; \
426 } else { \
427 if(__uc<=0x7ff) { \
428 (s)[(i)++]=(uint8_t)((__uc>>6)|0xc0); \
429 } else { \
430 if(__uc<=0xffff) { \
431 (s)[(i)++]=(uint8_t)((__uc>>12)|0xe0); \
432 } else { \
433 (s)[(i)++]=(uint8_t)((__uc>>18)|0xf0); \
434 (s)[(i)++]=(uint8_t)(((__uc>>12)&0x3f)|0x80); \
435 } \
436 (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
437 } \
438 (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
439 } \
440} UPRV_BLOCK_MACRO_END
441
459#define U8_APPEND(s, i, capacity, c, isError) UPRV_BLOCK_MACRO_BEGIN { \
460 uint32_t __uc=(c); \
461 if(__uc<=0x7f) { \
462 (s)[(i)++]=(uint8_t)__uc; \
463 } else if(__uc<=0x7ff && (i)+1<(capacity)) { \
464 (s)[(i)++]=(uint8_t)((__uc>>6)|0xc0); \
465 (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
466 } else if((__uc<=0xd7ff || (0xe000<=__uc && __uc<=0xffff)) && (i)+2<(capacity)) { \
467 (s)[(i)++]=(uint8_t)((__uc>>12)|0xe0); \
468 (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
469 (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
470 } else if(0xffff<__uc && __uc<=0x10ffff && (i)+3<(capacity)) { \
471 (s)[(i)++]=(uint8_t)((__uc>>18)|0xf0); \
472 (s)[(i)++]=(uint8_t)(((__uc>>12)&0x3f)|0x80); \
473 (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
474 (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
475 } else { \
476 (isError)=true; \
477 } \
478} UPRV_BLOCK_MACRO_END
479
490#define U8_FWD_1_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
491 (i)+=1+U8_COUNT_TRAIL_BYTES_UNSAFE((s)[i]); \
492} UPRV_BLOCK_MACRO_END
493
507#define U8_FWD_1(s, i, length) UPRV_BLOCK_MACRO_BEGIN { \
508 uint8_t __b=(s)[(i)++]; \
509 if(U8_IS_LEAD(__b) && (i)!=(length)) { \
510 uint8_t __t1=(s)[i]; \
511 if((0xe0<=__b && __b<0xf0)) { \
512 if(U8_IS_VALID_LEAD3_AND_T1(__b, __t1) && \
513 ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
514 ++(i); \
515 } \
516 } else if(__b<0xe0) { \
517 if(U8_IS_TRAIL(__t1)) { \
518 ++(i); \
519 } \
520 } else /* c>=0xf0 */ { \
521 if(U8_IS_VALID_LEAD4_AND_T1(__b, __t1) && \
522 ++(i)!=(length) && U8_IS_TRAIL((s)[i]) && \
523 ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
524 ++(i); \
525 } \
526 } \
527 } \
528} UPRV_BLOCK_MACRO_END
529
542#define U8_FWD_N_UNSAFE(s, i, n) UPRV_BLOCK_MACRO_BEGIN { \
543 int32_t __N=(n); \
544 while(__N>0) { \
545 U8_FWD_1_UNSAFE(s, i); \
546 --__N; \
547 } \
548} UPRV_BLOCK_MACRO_END
549
565#define U8_FWD_N(s, i, length, n) UPRV_BLOCK_MACRO_BEGIN { \
566 int32_t __N=(n); \
567 while(__N>0 && ((i)<(length) || ((length)<0 && (s)[i]!=0))) { \
568 U8_FWD_1(s, i, length); \
569 --__N; \
570 } \
571} UPRV_BLOCK_MACRO_END
572
586#define U8_SET_CP_START_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
587 while(U8_IS_TRAIL((s)[i])) { --(i); } \
588} UPRV_BLOCK_MACRO_END
589
607#define U8_SET_CP_START(s, start, i) UPRV_BLOCK_MACRO_BEGIN { \
608 if(U8_IS_TRAIL((s)[(i)])) { \
609 (i)=utf8_back1SafeBody(s, start, (i)); \
610 } \
611} UPRV_BLOCK_MACRO_END
612
639#define U8_TRUNCATE_IF_INCOMPLETE(s, start, length) UPRV_BLOCK_MACRO_BEGIN { \
640 if((length)>(start)) { \
641 uint8_t __b1=s[(length)-1]; \
642 if(U8_IS_SINGLE(__b1)) { \
643 /* common ASCII character */ \
644 } else if(U8_IS_LEAD(__b1)) { \
645 --(length); \
646 } else if(U8_IS_TRAIL(__b1) && ((length)-2)>=(start)) { \
647 uint8_t __b2=s[(length)-2]; \
648 if(0xe0<=__b2 && __b2<=0xf4) { \
649 if(__b2<0xf0 ? U8_IS_VALID_LEAD3_AND_T1(__b2, __b1) : \
650 U8_IS_VALID_LEAD4_AND_T1(__b2, __b1)) { \
651 (length)-=2; \
652 } \
653 } else if(U8_IS_TRAIL(__b2) && ((length)-3)>=(start)) { \
654 uint8_t __b3=s[(length)-3]; \
655 if(0xf0<=__b3 && __b3<=0xf4 && U8_IS_VALID_LEAD4_AND_T1(__b3, __b2)) { \
656 (length)-=3; \
657 } \
658 } \
659 } \
660 } \
661} UPRV_BLOCK_MACRO_END
662
663/* definitions with backward iteration -------------------------------------- */
664
684#define U8_PREV_UNSAFE(s, i, c) UPRV_BLOCK_MACRO_BEGIN { \
685 (c)=(uint8_t)(s)[--(i)]; \
686 if(U8_IS_TRAIL(c)) { \
687 uint8_t __b, __count=1, __shift=6; \
688\
689 /* c is a trail byte */ \
690 (c)&=0x3f; \
691 for(;;) { \
692 __b=(s)[--(i)]; \
693 if(__b>=0xc0) { \
694 U8_MASK_LEAD_BYTE(__b, __count); \
695 (c)|=(UChar32)__b<<__shift; \
696 break; \
697 } else { \
698 (c)|=(UChar32)(__b&0x3f)<<__shift; \
699 ++__count; \
700 __shift+=6; \
701 } \
702 } \
703 } \
704} UPRV_BLOCK_MACRO_END
705
726#define U8_PREV(s, start, i, c) UPRV_BLOCK_MACRO_BEGIN { \
727 (c)=(uint8_t)(s)[--(i)]; \
728 if(!U8_IS_SINGLE(c)) { \
729 (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -1); \
730 } \
731} UPRV_BLOCK_MACRO_END
732
757#define U8_PREV_OR_FFFD(s, start, i, c) UPRV_BLOCK_MACRO_BEGIN { \
758 (c)=(uint8_t)(s)[--(i)]; \
759 if(!U8_IS_SINGLE(c)) { \
760 (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -3); \
761 } \
762} UPRV_BLOCK_MACRO_END
763
775#define U8_BACK_1_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
776 while(U8_IS_TRAIL((s)[--(i)])) {} \
777} UPRV_BLOCK_MACRO_END
778
791#define U8_BACK_1(s, start, i) UPRV_BLOCK_MACRO_BEGIN { \
792 if(U8_IS_TRAIL((s)[--(i)])) { \
793 (i)=utf8_back1SafeBody(s, start, (i)); \
794 } \
795} UPRV_BLOCK_MACRO_END
796
810#define U8_BACK_N_UNSAFE(s, i, n) UPRV_BLOCK_MACRO_BEGIN { \
811 int32_t __N=(n); \
812 while(__N>0) { \
813 U8_BACK_1_UNSAFE(s, i); \
814 --__N; \
815 } \
816} UPRV_BLOCK_MACRO_END
817
832#define U8_BACK_N(s, start, i, n) UPRV_BLOCK_MACRO_BEGIN { \
833 int32_t __N=(n); \
834 while(__N>0 && (i)>(start)) { \
835 U8_BACK_1(s, start, i); \
836 --__N; \
837 } \
838} UPRV_BLOCK_MACRO_END
839
853#define U8_SET_CP_LIMIT_UNSAFE(s, i) UPRV_BLOCK_MACRO_BEGIN { \
854 U8_BACK_1_UNSAFE(s, i); \
855 U8_FWD_1_UNSAFE(s, i); \
856} UPRV_BLOCK_MACRO_END
857
875#define U8_SET_CP_LIMIT(s, start, i, length) UPRV_BLOCK_MACRO_BEGIN { \
876 if((start)<(i) && ((i)<(length) || (length)<0)) { \
877 U8_BACK_1(s, start, i); \
878 U8_FWD_1(s, i, length); \
879 } \
880} UPRV_BLOCK_MACRO_END
881
882#endif
Basic types and constants for UTF.
int32_t UChar32
Define UChar32 as a type for single Unicode code points.
Definition: umachine.h:435
int8_t UBool
The ICU boolean type, a signed-byte integer.
Definition: umachine.h:247
#define U_CAPI
This is used to declare a function as a public ICU C API.
Definition: umachine.h:110
U_CAPI int32_t utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool *pIsError)
Function for handling "append code point" with error-checking.
U_CAPI int32_t utf8_back1SafeBody(const uint8_t *s, int32_t start, int32_t i)
Function for handling "skip backward one code point" with error-checking.
U_CAPI UChar32 utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict)
Function for handling "previous code point" with error-checking.
U_CAPI UChar32 utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict)
Function for handling "next code point" with error-checking.
C API: Code point macros.