ICU 56.1  56.1
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
utf8.h
Go to the documentation of this file.
1 /*
2 *******************************************************************************
3 *
4 * Copyright (C) 1999-2015, International Business Machines
5 * Corporation and others. All Rights Reserved.
6 *
7 *******************************************************************************
8 * file name: utf8.h
9 * encoding: US-ASCII
10 * tab size: 8 (not used)
11 * indentation:4
12 *
13 * created on: 1999sep13
14 * created by: Markus W. Scherer
15 */
16 
32 #ifndef __UTF8_H__
33 #define __UTF8_H__
34 
35 #include "unicode/umachine.h"
36 #ifndef __UTF_H__
37 # include "unicode/utf.h"
38 #endif
39 
40 /* internal definitions ----------------------------------------------------- */
41 
53 #ifdef U_UTF8_IMPL
54 U_EXPORT const uint8_t
55 #elif defined(U_STATIC_IMPLEMENTATION) || defined(U_COMMON_IMPLEMENTATION)
56 U_CFUNC const uint8_t
57 #else
58 U_CFUNC U_IMPORT const uint8_t /* U_IMPORT2? */ /*U_IMPORT*/
59 #endif
61 
80 #define U8_COUNT_TRAIL_BYTES(leadByte) \
81  ((uint8_t)(leadByte)<0xf0 ? \
82  ((uint8_t)(leadByte)>=0xc0)+((uint8_t)(leadByte)>=0xe0) : \
83  (uint8_t)(leadByte)<0xfe ? 3+((uint8_t)(leadByte)>=0xf8)+((uint8_t)(leadByte)>=0xfc) : 0)
84 
96 #define U8_COUNT_TRAIL_BYTES_UNSAFE(leadByte) \
97  (((leadByte)>=0xc0)+((leadByte)>=0xe0)+((leadByte)>=0xf0))
98 
106 #define U8_MASK_LEAD_BYTE(leadByte, countTrailBytes) ((leadByte)&=(1<<(6-(countTrailBytes)))-1)
107 
117 U_STABLE UChar32 U_EXPORT2
118 utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict);
119 
129 U_STABLE int32_t U_EXPORT2
130 utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool *pIsError);
131 
141 U_STABLE UChar32 U_EXPORT2
142 utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict);
143 
153 U_STABLE int32_t U_EXPORT2
154 utf8_back1SafeBody(const uint8_t *s, int32_t start, int32_t i);
155 
156 /* single-code point definitions -------------------------------------------- */
157 
164 #define U8_IS_SINGLE(c) (((c)&0x80)==0)
165 
172 #define U8_IS_LEAD(c) ((uint8_t)((c)-0xc0)<0x3e)
173 
180 #define U8_IS_TRAIL(c) (((c)&0xc0)==0x80)
181 
189 #define U8_LENGTH(c) \
190  ((uint32_t)(c)<=0x7f ? 1 : \
191  ((uint32_t)(c)<=0x7ff ? 2 : \
192  ((uint32_t)(c)<=0xd7ff ? 3 : \
193  ((uint32_t)(c)<=0xdfff || (uint32_t)(c)>0x10ffff ? 0 : \
194  ((uint32_t)(c)<=0xffff ? 3 : 4)\
195  ) \
196  ) \
197  ) \
198  )
199 
205 #define U8_MAX_LENGTH 4
206 
223 #define U8_GET_UNSAFE(s, i, c) { \
224  int32_t _u8_get_unsafe_index=(int32_t)(i); \
225  U8_SET_CP_START_UNSAFE(s, _u8_get_unsafe_index); \
226  U8_NEXT_UNSAFE(s, _u8_get_unsafe_index, c); \
227 }
228 
250 #define U8_GET(s, start, i, length, c) { \
251  int32_t _u8_get_index=(i); \
252  U8_SET_CP_START(s, start, _u8_get_index); \
253  U8_NEXT(s, _u8_get_index, length, c); \
254 }
255 
281 #define U8_GET_OR_FFFD(s, start, i, length, c) { \
282  int32_t _u8_get_index=(i); \
283  U8_SET_CP_START(s, start, _u8_get_index); \
284  U8_NEXT_OR_FFFD(s, _u8_get_index, length, c); \
285 }
286 
287 /* definitions with forward iteration --------------------------------------- */
288 
306 #define U8_NEXT_UNSAFE(s, i, c) { \
307  (c)=(uint8_t)(s)[(i)++]; \
308  if((c)>=0x80) { \
309  if((c)<0xe0) { \
310  (c)=(((c)&0x1f)<<6)|((s)[(i)++]&0x3f); \
311  } else if((c)<0xf0) { \
312  /* no need for (c&0xf) because the upper bits are truncated after <<12 in the cast to (UChar) */ \
313  (c)=(UChar)(((c)<<12)|(((s)[i]&0x3f)<<6)|((s)[(i)+1]&0x3f)); \
314  (i)+=2; \
315  } else { \
316  (c)=(((c)&7)<<18)|(((s)[i]&0x3f)<<12)|(((s)[(i)+1]&0x3f)<<6)|((s)[(i)+2]&0x3f); \
317  (i)+=3; \
318  } \
319  } \
320 }
321 
342 #define U8_NEXT(s, i, length, c) { \
343  (c)=(uint8_t)(s)[(i)++]; \
344  if((c)>=0x80) { \
345  uint8_t __t1, __t2; \
346  if( /* handle U+1000..U+CFFF inline */ \
347  (0xe0<(c) && (c)<=0xec) && \
348  (((i)+1)<(length) || (length)<0) && \
349  (__t1=(uint8_t)((s)[i]-0x80))<=0x3f && \
350  (__t2=(uint8_t)((s)[(i)+1]-0x80))<= 0x3f \
351  ) { \
352  /* no need for (c&0xf) because the upper bits are truncated after <<12 in the cast to (UChar) */ \
353  (c)=(UChar)(((c)<<12)|(__t1<<6)|__t2); \
354  (i)+=2; \
355  } else if( /* handle U+0080..U+07FF inline */ \
356  ((c)<0xe0 && (c)>=0xc2) && \
357  ((i)!=(length)) && \
358  (__t1=(uint8_t)((s)[i]-0x80))<=0x3f \
359  ) { \
360  (c)=(((c)&0x1f)<<6)|__t1; \
361  ++(i); \
362  } else { \
363  /* function call for "complicated" and error cases */ \
364  (c)=utf8_nextCharSafeBody((const uint8_t *)s, &(i), (length), c, -1); \
365  } \
366  } \
367 }
368 
393 #define U8_NEXT_OR_FFFD(s, i, length, c) { \
394  (c)=(uint8_t)(s)[(i)++]; \
395  if((c)>=0x80) { \
396  uint8_t __t1, __t2; \
397  if( /* handle U+1000..U+CFFF inline */ \
398  (0xe0<(c) && (c)<=0xec) && \
399  (((i)+1)<(length) || (length)<0) && \
400  (__t1=(uint8_t)((s)[i]-0x80))<=0x3f && \
401  (__t2=(uint8_t)((s)[(i)+1]-0x80))<= 0x3f \
402  ) { \
403  /* no need for (c&0xf) because the upper bits are truncated after <<12 in the cast to (UChar) */ \
404  (c)=(UChar)(((c)<<12)|(__t1<<6)|__t2); \
405  (i)+=2; \
406  } else if( /* handle U+0080..U+07FF inline */ \
407  ((c)<0xe0 && (c)>=0xc2) && \
408  ((i)!=(length)) && \
409  (__t1=(uint8_t)((s)[i]-0x80))<=0x3f \
410  ) { \
411  (c)=(((c)&0x1f)<<6)|__t1; \
412  ++(i); \
413  } else { \
414  /* function call for "complicated" and error cases */ \
415  (c)=utf8_nextCharSafeBody((const uint8_t *)s, &(i), (length), c, -3); \
416  } \
417  } \
418 }
419 
433 #define U8_APPEND_UNSAFE(s, i, c) { \
434  if((uint32_t)(c)<=0x7f) { \
435  (s)[(i)++]=(uint8_t)(c); \
436  } else { \
437  if((uint32_t)(c)<=0x7ff) { \
438  (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \
439  } else { \
440  if((uint32_t)(c)<=0xffff) { \
441  (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \
442  } else { \
443  (s)[(i)++]=(uint8_t)(((c)>>18)|0xf0); \
444  (s)[(i)++]=(uint8_t)((((c)>>12)&0x3f)|0x80); \
445  } \
446  (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \
447  } \
448  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
449  } \
450 }
451 
469 #define U8_APPEND(s, i, capacity, c, isError) { \
470  if((uint32_t)(c)<=0x7f) { \
471  (s)[(i)++]=(uint8_t)(c); \
472  } else if((uint32_t)(c)<=0x7ff && (i)+1<(capacity)) { \
473  (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \
474  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
475  } else if((uint32_t)(c)<=0xd7ff && (i)+2<(capacity)) { \
476  (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \
477  (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \
478  (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \
479  } else { \
480  (i)=utf8_appendCharSafeBody(s, (i), (capacity), c, &(isError)); \
481  } \
482 }
483 
494 #define U8_FWD_1_UNSAFE(s, i) { \
495  (i)+=1+U8_COUNT_TRAIL_BYTES_UNSAFE((uint8_t)(s)[i]); \
496 }
497 
511 #define U8_FWD_1(s, i, length) { \
512  uint8_t __b=(uint8_t)(s)[(i)++]; \
513  if(U8_IS_LEAD(__b)) { \
514  uint8_t __count=U8_COUNT_TRAIL_BYTES(__b); \
515  if((i)+__count>(length) && (length)>=0) { \
516  __count=(uint8_t)((length)-(i)); \
517  } \
518  while(__count>0 && U8_IS_TRAIL((s)[i])) { \
519  ++(i); \
520  --__count; \
521  } \
522  } \
523 }
524 
537 #define U8_FWD_N_UNSAFE(s, i, n) { \
538  int32_t __N=(n); \
539  while(__N>0) { \
540  U8_FWD_1_UNSAFE(s, i); \
541  --__N; \
542  } \
543 }
544 
560 #define U8_FWD_N(s, i, length, n) { \
561  int32_t __N=(n); \
562  while(__N>0 && ((i)<(length) || ((length)<0 && (s)[i]!=0))) { \
563  U8_FWD_1(s, i, length); \
564  --__N; \
565  } \
566 }
567 
581 #define U8_SET_CP_START_UNSAFE(s, i) { \
582  while(U8_IS_TRAIL((s)[i])) { --(i); } \
583 }
584 
599 #define U8_SET_CP_START(s, start, i) { \
600  if(U8_IS_TRAIL((s)[(i)])) { \
601  (i)=utf8_back1SafeBody(s, start, (i)); \
602  } \
603 }
604 
605 /* definitions with backward iteration -------------------------------------- */
606 
626 #define U8_PREV_UNSAFE(s, i, c) { \
627  (c)=(uint8_t)(s)[--(i)]; \
628  if(U8_IS_TRAIL(c)) { \
629  uint8_t __b, __count=1, __shift=6; \
630 \
631  /* c is a trail byte */ \
632  (c)&=0x3f; \
633  for(;;) { \
634  __b=(uint8_t)(s)[--(i)]; \
635  if(__b>=0xc0) { \
636  U8_MASK_LEAD_BYTE(__b, __count); \
637  (c)|=(UChar32)__b<<__shift; \
638  break; \
639  } else { \
640  (c)|=(UChar32)(__b&0x3f)<<__shift; \
641  ++__count; \
642  __shift+=6; \
643  } \
644  } \
645  } \
646 }
647 
668 #define U8_PREV(s, start, i, c) { \
669  (c)=(uint8_t)(s)[--(i)]; \
670  if((c)>=0x80) { \
671  (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -1); \
672  } \
673 }
674 
699 #define U8_PREV_OR_FFFD(s, start, i, c) { \
700  (c)=(uint8_t)(s)[--(i)]; \
701  if((c)>=0x80) { \
702  (c)=utf8_prevCharSafeBody((const uint8_t *)s, start, &(i), c, -3); \
703  } \
704 }
705 
717 #define U8_BACK_1_UNSAFE(s, i) { \
718  while(U8_IS_TRAIL((s)[--(i)])) {} \
719 }
720 
733 #define U8_BACK_1(s, start, i) { \
734  if(U8_IS_TRAIL((s)[--(i)])) { \
735  (i)=utf8_back1SafeBody(s, start, (i)); \
736  } \
737 }
738 
752 #define U8_BACK_N_UNSAFE(s, i, n) { \
753  int32_t __N=(n); \
754  while(__N>0) { \
755  U8_BACK_1_UNSAFE(s, i); \
756  --__N; \
757  } \
758 }
759 
774 #define U8_BACK_N(s, start, i, n) { \
775  int32_t __N=(n); \
776  while(__N>0 && (i)>(start)) { \
777  U8_BACK_1(s, start, i); \
778  --__N; \
779  } \
780 }
781 
795 #define U8_SET_CP_LIMIT_UNSAFE(s, i) { \
796  U8_BACK_1_UNSAFE(s, i); \
797  U8_FWD_1_UNSAFE(s, i); \
798 }
799 
817 #define U8_SET_CP_LIMIT(s, start, i, length) { \
818  if((start)<(i) && ((i)<(length) || (length)<0)) { \
819  U8_BACK_1(s, start, i); \
820  U8_FWD_1(s, i, length); \
821  } \
822 }
823 
824 #endif