ICU 62.1  62.1
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 "unicode/umachine.h"
38 #ifndef __UTF_H__
39 # include "unicode/utf.h"
40 #endif
41 
42 /* internal definitions ----------------------------------------------------- */
43 
55 #define U8_COUNT_TRAIL_BYTES(leadByte) \
56  (U8_IS_LEAD(leadByte) ? \
57  ((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0)+1 : 0)
58 
70 #define U8_COUNT_TRAIL_BYTES_UNSAFE(leadByte) \
71  (((uint8_t)(leadByte)>=0xc2)+((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0))
72 
80 #define U8_MASK_LEAD_BYTE(leadByte, countTrailBytes) ((leadByte)&=(1<<(6-(countTrailBytes)))-1)
81 
90 #define U8_LEAD3_T1_BITS "\x20\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x30\x10\x30\x30"
91 
97 #define U8_IS_VALID_LEAD3_AND_T1(lead, t1) (U8_LEAD3_T1_BITS[(lead)&0xf]&(1<<((uint8_t)(t1)>>5)))
98 
107 #define U8_LEAD4_T1_BITS "\x00\x00\x00\x00\x00\x00\x00\x00\x1E\x0F\x0F\x0F\x00\x00\x00\x00"
108 
114 #define U8_IS_VALID_LEAD4_AND_T1(lead, t1) (U8_LEAD4_T1_BITS[(uint8_t)(t1)>>4]&(1<<((lead)&7)))
115 
125 U_STABLE UChar32 U_EXPORT2
126 utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict);
127 
137 U_STABLE int32_t U_EXPORT2
138 utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool *pIsError);
139 
149 U_STABLE UChar32 U_EXPORT2
150 utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict);
151 
161 U_STABLE int32_t U_EXPORT2
162 utf8_back1SafeBody(const uint8_t *s, int32_t start, int32_t i);
163 
164 /* single-code point definitions -------------------------------------------- */
165 
172 #define U8_IS_SINGLE(c) (((c)&0x80)==0)
173 
180 #define U8_IS_LEAD(c) ((uint8_t)((c)-0xc2)<=0x32)
181 // 0x32=0xf4-0xc2
182 
189 #define U8_IS_TRAIL(c) ((int8_t)(c)<-0x40)
190 
198 #define U8_LENGTH(c) \
199  ((uint32_t)(c)<=0x7f ? 1 : \
200  ((uint32_t)(c)<=0x7ff ? 2 : \
201  ((uint32_t)(c)<=0xd7ff ? 3 : \
202  ((uint32_t)(c)<=0xdfff || (uint32_t)(c)>0x10ffff ? 0 : \
203  ((uint32_t)(c)<=0xffff ? 3 : 4)\
204  ) \
205  ) \
206  ) \
207  )
208 
214 #define U8_MAX_LENGTH 4
215 
232 #define U8_GET_UNSAFE(s, i, c) { \
233  int32_t _u8_get_unsafe_index=(int32_t)(i); \
234  U8_SET_CP_START_UNSAFE(s, _u8_get_unsafe_index); \
235  U8_NEXT_UNSAFE(s, _u8_get_unsafe_index, c); \
236 }
237 
259 #define U8_GET(s, start, i, length, c) { \
260  int32_t _u8_get_index=(i); \
261  U8_SET_CP_START(s, start, _u8_get_index); \
262  U8_NEXT(s, _u8_get_index, length, c); \
263 }
264 
290 #define U8_GET_OR_FFFD(s, start, i, length, c) { \
291  int32_t _u8_get_index=(i); \
292  U8_SET_CP_START(s, start, _u8_get_index); \
293  U8_NEXT_OR_FFFD(s, _u8_get_index, length, c); \
294 }
295 
296 /* definitions with forward iteration --------------------------------------- */
297 
315 #define U8_NEXT_UNSAFE(s, i, c) { \
316  (c)=(uint8_t)(s)[(i)++]; \
317  if(!U8_IS_SINGLE(c)) { \
318  if((c)<0xe0) { \
319  (c)=(((c)&0x1f)<<6)|((s)[(i)++]&0x3f); \
320  } else if((c)<0xf0) { \
321  /* no need for (c&0xf) because the upper bits are truncated after <<12 in the cast to (UChar) */ \
322  (c)=(UChar)(((c)<<12)|(((s)[i]&0x3f)<<6)|((s)[(i)+1]&0x3f)); \
323  (i)+=2; \
324  } else { \
325  (c)=(((c)&7)<<18)|(((s)[i]&0x3f)<<12)|(((s)[(i)+1]&0x3f)<<6)|((s)[(i)+2]&0x3f); \
326  (i)+=3; \
327  } \
328  } \
329 }
330 
351 #define U8_NEXT(s, i, length, c) U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, U_SENTINEL)
352 
377 #define U8_NEXT_OR_FFFD(s, i, length, c) U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, 0xfffd)
378 
380 #define U8_INTERNAL_NEXT_OR_SUB(s, i, length, c, sub) { \
381  (c)=(uint8_t)(s)[(i)++]; \
382  if(!U8_IS_SINGLE(c)) { \
383  uint8_t __t = 0; \
384  if((i)!=(length) && \
385  /* fetch/validate/assemble all but last trail byte */ \
386  ((c)>=0xe0 ? \
387  ((c)<0xf0 ? /* U+0800..U+FFFF except surrogates */ \
388  U8_LEAD3_T1_BITS[(c)&=0xf]&(1<<((__t=(s)[i])>>5)) && \
389  (__t&=0x3f, 1) \
390  : /* U+10000..U+10FFFF */ \
391  ((c)-=0xf0)<=4 && \
392  U8_LEAD4_T1_BITS[(__t=(s)[i])>>4]&(1<<(c)) && \
393  ((c)=((c)<<6)|(__t&0x3f), ++(i)!=(length)) && \
394  (__t=(s)[i]-0x80)<=0x3f) && \
395  /* valid second-to-last trail byte */ \
396  ((c)=((c)<<6)|__t, ++(i)!=(length)) \
397  : /* U+0080..U+07FF */ \
398  (c)>=0xc2 && ((c)&=0x1f, 1)) && \
399  /* last trail byte */ \
400  (__t=(s)[i]-0x80)<=0x3f && \
401  ((c)=((c)<<6)|__t, ++(i), 1)) { \
402  } else { \
403  (c)=(sub); /* ill-formed*/ \
404  } \
405  } \
406 }
407 
421 #define U8_APPEND_UNSAFE(s, i, c) { \
422  uint32_t __uc=(c); \
423  if(__uc<=0x7f) { \
424  (s)[(i)++]=(uint8_t)__uc; \
425  } else { \
426  if(__uc<=0x7ff) { \
427  (s)[(i)++]=(uint8_t)((__uc>>6)|0xc0); \
428  } else { \
429  if(__uc<=0xffff) { \
430  (s)[(i)++]=(uint8_t)((__uc>>12)|0xe0); \
431  } else { \
432  (s)[(i)++]=(uint8_t)((__uc>>18)|0xf0); \
433  (s)[(i)++]=(uint8_t)(((__uc>>12)&0x3f)|0x80); \
434  } \
435  (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
436  } \
437  (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
438  } \
439 }
440 
458 #define U8_APPEND(s, i, capacity, c, isError) { \
459  uint32_t __uc=(c); \
460  if(__uc<=0x7f) { \
461  (s)[(i)++]=(uint8_t)__uc; \
462  } else if(__uc<=0x7ff && (i)+1<(capacity)) { \
463  (s)[(i)++]=(uint8_t)((__uc>>6)|0xc0); \
464  (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
465  } else if((__uc<=0xd7ff || (0xe000<=__uc && __uc<=0xffff)) && (i)+2<(capacity)) { \
466  (s)[(i)++]=(uint8_t)((__uc>>12)|0xe0); \
467  (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
468  (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
469  } else if(0xffff<__uc && __uc<=0x10ffff && (i)+3<(capacity)) { \
470  (s)[(i)++]=(uint8_t)((__uc>>18)|0xf0); \
471  (s)[(i)++]=(uint8_t)(((__uc>>12)&0x3f)|0x80); \
472  (s)[(i)++]=(uint8_t)(((__uc>>6)&0x3f)|0x80); \
473  (s)[(i)++]=(uint8_t)((__uc&0x3f)|0x80); \
474  } else { \
475  (isError)=TRUE; \
476  } \
477 }
478 
489 #define U8_FWD_1_UNSAFE(s, i) { \
490  (i)+=1+U8_COUNT_TRAIL_BYTES_UNSAFE((s)[i]); \
491 }
492 
506 #define U8_FWD_1(s, i, length) { \
507  uint8_t __b=(s)[(i)++]; \
508  if(U8_IS_LEAD(__b) && (i)!=(length)) { \
509  uint8_t __t1=(s)[i]; \
510  if((0xe0<=__b && __b<0xf0)) { \
511  if(U8_IS_VALID_LEAD3_AND_T1(__b, __t1) && \
512  ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
513  ++(i); \
514  } \
515  } else if(__b<0xe0) { \
516  if(U8_IS_TRAIL(__t1)) { \
517  ++(i); \
518  } \
519  } else /* c>=0xf0 */ { \
520  if(U8_IS_VALID_LEAD4_AND_T1(__b, __t1) && \
521  ++(i)!=(length) && U8_IS_TRAIL((s)[i]) && \
522  ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
523  ++(i); \
524  } \
525  } \
526  } \
527 }
528 
541 #define U8_FWD_N_UNSAFE(s, i, n) { \
542  int32_t __N=(n); \
543  while(__N>0) { \
544  U8_FWD_1_UNSAFE(s, i); \
545  --__N; \
546  } \
547 }
548 
564 #define U8_FWD_N(s, i, length, n) { \
565  int32_t __N=(n); \
566  while(__N>0 && ((i)<(length) || ((length)<0 && (s)[i]!=0))) { \
567  U8_FWD_1(s, i, length); \
568  --__N; \
569  } \
570 }
571 
585 #define U8_SET_CP_START_UNSAFE(s, i) { \
586  while(U8_IS_TRAIL((s)[i])) { --(i); } \
587 }
588 
606 #define U8_SET_CP_START(s, start, i) { \
607  if(U8_IS_TRAIL((s)[(i)])) { \
608  (i)=utf8_back1SafeBody(s, start, (i)); \
609  } \
610 }
611 
612 #ifndef U_HIDE_DRAFT_API
613 
639 #define U8_TRUNCATE_IF_INCOMPLETE(s, start, length) \
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 #endif // U_HIDE_DRAFT_API
662 
663 /* definitions with backward iteration -------------------------------------- */
664 
684 #define U8_PREV_UNSAFE(s, i, c) { \
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 }
705 
726 #define U8_PREV(s, start, i, c) { \
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 }
732 
757 #define U8_PREV_OR_FFFD(s, start, i, c) { \
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 }
763 
775 #define U8_BACK_1_UNSAFE(s, i) { \
776  while(U8_IS_TRAIL((s)[--(i)])) {} \
777 }
778 
791 #define U8_BACK_1(s, start, i) { \
792  if(U8_IS_TRAIL((s)[--(i)])) { \
793  (i)=utf8_back1SafeBody(s, start, (i)); \
794  } \
795 }
796 
810 #define U8_BACK_N_UNSAFE(s, i, n) { \
811  int32_t __N=(n); \
812  while(__N>0) { \
813  U8_BACK_1_UNSAFE(s, i); \
814  --__N; \
815  } \
816 }
817 
832 #define U8_BACK_N(s, start, i, n) { \
833  int32_t __N=(n); \
834  while(__N>0 && (i)>(start)) { \
835  U8_BACK_1(s, start, i); \
836  --__N; \
837  } \
838 }
839 
853 #define U8_SET_CP_LIMIT_UNSAFE(s, i) { \
854  U8_BACK_1_UNSAFE(s, i); \
855  U8_FWD_1_UNSAFE(s, i); \
856 }
857 
875 #define U8_SET_CP_LIMIT(s, start, i, length) { \
876  if((start)<(i) && ((i)<(length) || (length)<0)) { \
877  U8_BACK_1(s, start, i); \
878  U8_FWD_1(s, i, length); \
879  } \
880 }
881 
882 #endif
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.
Basic types and constants for UTF.
C API: Code point macros.
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.
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.
int32_t UChar32
Define UChar32 as a type for single Unicode code points.
Definition: umachine.h:400
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.
#define U_STABLE
This is used to declare a function as a stable public ICU C API.
Definition: umachine.h:111
int8_t UBool
The ICU boolean type.
Definition: umachine.h:236