ICU 60.2  60.2
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) { \
352  (c)=(uint8_t)(s)[(i)++]; \
353  if(!U8_IS_SINGLE(c)) { \
354  uint8_t __t1, __t2; \
355  if( /* handle U+0800..U+FFFF inline */ \
356  (0xe0<=(c) && (c)<0xf0) && \
357  (((i)+1)<(length) || (length)<0) && \
358  U8_IS_VALID_LEAD3_AND_T1((c), __t1=(s)[i]) && \
359  (__t2=(s)[(i)+1]-0x80)<=0x3f) { \
360  (c)=(((c)&0xf)<<12)|((__t1&0x3f)<<6)|__t2; \
361  (i)+=2; \
362  } else if( /* handle U+0080..U+07FF inline */ \
363  ((c)<0xe0 && (c)>=0xc2) && \
364  ((i)!=(length)) && \
365  (__t1=(s)[i]-0x80)<=0x3f) { \
366  (c)=(((c)&0x1f)<<6)|__t1; \
367  ++(i); \
368  } else { \
369  /* function call for "complicated" and error cases */ \
370  (c)=utf8_nextCharSafeBody((const uint8_t *)s, &(i), (length), c, -1); \
371  } \
372  } \
373 }
374 
399 #define U8_NEXT_OR_FFFD(s, i, length, c) { \
400  (c)=(uint8_t)(s)[(i)++]; \
401  if(!U8_IS_SINGLE(c)) { \
402  uint8_t __t1, __t2; \
403  if( /* handle U+0800..U+FFFF inline */ \
404  (0xe0<=(c) && (c)<0xf0) && \
405  (((i)+1)<(length) || (length)<0) && \
406  U8_IS_VALID_LEAD3_AND_T1((c), __t1=(s)[i]) && \
407  (__t2=(s)[(i)+1]-0x80)<=0x3f) { \
408  (c)=(((c)&0xf)<<12)|((__t1&0x3f)<<6)|__t2; \
409  (i)+=2; \
410  } else if( /* handle U+0080..U+07FF inline */ \
411  ((c)<0xe0 && (c)>=0xc2) && \
412  ((i)!=(length)) && \
413  (__t1=(s)[i]-0x80)<=0x3f) { \
414  (c)=(((c)&0x1f)<<6)|__t1; \
415  ++(i); \
416  } else { \
417  /* function call for "complicated" and error cases */ \
418  (c)=utf8_nextCharSafeBody((const uint8_t *)s, &(i), (length), c, -3); \
419  } \
420  } \
421 }
422 
436 #define U8_APPEND_UNSAFE(s, i, c) { \
437  if((uint32_t)(c)<=0x7f) { \
438  (s)[(i)++]=(uint8_t)(c); \
439  } else { \
440  if((uint32_t)(c)<=0x7ff) { \
441  (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \
442  } else { \
443  if((uint32_t)(c)<=0xffff) { \
444  (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \
445  } else { \
446  (s)[(i)++]=(uint8_t)(((c)>>18)|0xf0); \
447  (s)[(i)++]=(uint8_t)((((c)>>12)&0x3f)|0x80); \
448  } \
449  (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \
450  } \
451  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
452  } \
453 }
454 
472 #define U8_APPEND(s, i, capacity, c, isError) { \
473  if((uint32_t)(c)<=0x7f) { \
474  (s)[(i)++]=(uint8_t)(c); \
475  } else if((uint32_t)(c)<=0x7ff && (i)+1<(capacity)) { \
476  (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \
477  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
478  } else if((uint32_t)(c)<=0xd7ff && (i)+2<(capacity)) { \
479  (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \
480  (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \
481  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
482  } else { \
483  (i)=utf8_appendCharSafeBody(s, (i), (capacity), c, &(isError)); \
484  } \
485 }
486 
497 #define U8_FWD_1_UNSAFE(s, i) { \
498  (i)+=1+U8_COUNT_TRAIL_BYTES_UNSAFE((s)[i]); \
499 }
500 
514 #define U8_FWD_1(s, i, length) { \
515  uint8_t __b=(s)[(i)++]; \
516  if(U8_IS_LEAD(__b) && (i)!=(length)) { \
517  uint8_t __t1=(s)[i]; \
518  if((0xe0<=__b && __b<0xf0)) { \
519  if(U8_IS_VALID_LEAD3_AND_T1(__b, __t1) && \
520  ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
521  ++(i); \
522  } \
523  } else if(__b<0xe0) { \
524  if(U8_IS_TRAIL(__t1)) { \
525  ++(i); \
526  } \
527  } else /* c>=0xf0 */ { \
528  if(U8_IS_VALID_LEAD4_AND_T1(__b, __t1) && \
529  ++(i)!=(length) && U8_IS_TRAIL((s)[i]) && \
530  ++(i)!=(length) && U8_IS_TRAIL((s)[i])) { \
531  ++(i); \
532  } \
533  } \
534  } \
535 }
536 
549 #define U8_FWD_N_UNSAFE(s, i, n) { \
550  int32_t __N=(n); \
551  while(__N>0) { \
552  U8_FWD_1_UNSAFE(s, i); \
553  --__N; \
554  } \
555 }
556 
572 #define U8_FWD_N(s, i, length, n) { \
573  int32_t __N=(n); \
574  while(__N>0 && ((i)<(length) || ((length)<0 && (s)[i]!=0))) { \
575  U8_FWD_1(s, i, length); \
576  --__N; \
577  } \
578 }
579 
593 #define U8_SET_CP_START_UNSAFE(s, i) { \
594  while(U8_IS_TRAIL((s)[i])) { --(i); } \
595 }
596 
611 #define U8_SET_CP_START(s, start, i) { \
612  if(U8_IS_TRAIL((s)[(i)])) { \
613  (i)=utf8_back1SafeBody(s, start, (i)); \
614  } \
615 }
616 
617 /* definitions with backward iteration -------------------------------------- */
618 
638 #define U8_PREV_UNSAFE(s, i, c) { \
639  (c)=(uint8_t)(s)[--(i)]; \
640  if(U8_IS_TRAIL(c)) { \
641  uint8_t __b, __count=1, __shift=6; \
642 \
643  /* c is a trail byte */ \
644  (c)&=0x3f; \
645  for(;;) { \
646  __b=(s)[--(i)]; \
647  if(__b>=0xc0) { \
648  U8_MASK_LEAD_BYTE(__b, __count); \
649  (c)|=(UChar32)__b<<__shift; \
650  break; \
651  } else { \
652  (c)|=(UChar32)(__b&0x3f)<<__shift; \
653  ++__count; \
654  __shift+=6; \
655  } \
656  } \
657  } \
658 }
659 
680 #define U8_PREV(s, start, i, c) { \
681  (c)=(uint8_t)(s)[--(i)]; \
682  if(!U8_IS_SINGLE(c)) { \
683  (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -1); \
684  } \
685 }
686 
711 #define U8_PREV_OR_FFFD(s, start, i, c) { \
712  (c)=(uint8_t)(s)[--(i)]; \
713  if(!U8_IS_SINGLE(c)) { \
714  (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -3); \
715  } \
716 }
717 
729 #define U8_BACK_1_UNSAFE(s, i) { \
730  while(U8_IS_TRAIL((s)[--(i)])) {} \
731 }
732 
745 #define U8_BACK_1(s, start, i) { \
746  if(U8_IS_TRAIL((s)[--(i)])) { \
747  (i)=utf8_back1SafeBody(s, start, (i)); \
748  } \
749 }
750 
764 #define U8_BACK_N_UNSAFE(s, i, n) { \
765  int32_t __N=(n); \
766  while(__N>0) { \
767  U8_BACK_1_UNSAFE(s, i); \
768  --__N; \
769  } \
770 }
771 
786 #define U8_BACK_N(s, start, i, n) { \
787  int32_t __N=(n); \
788  while(__N>0 && (i)>(start)) { \
789  U8_BACK_1(s, start, i); \
790  --__N; \
791  } \
792 }
793 
807 #define U8_SET_CP_LIMIT_UNSAFE(s, i) { \
808  U8_BACK_1_UNSAFE(s, i); \
809  U8_FWD_1_UNSAFE(s, i); \
810 }
811 
829 #define U8_SET_CP_LIMIT(s, start, i, length) { \
830  if((start)<(i) && ((i)<(length) || (length)<0)) { \
831  U8_BACK_1(s, start, i); \
832  U8_FWD_1(s, i, length); \
833  } \
834 }
835 
836 #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:396
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