#include "unicode/utf.h"
Go to the source code of this file.
Defines | |
#define | UTF8_COUNT_TRAIL_BYTES(leadByte) (utf8_countTrailBytes[(uint8_t)leadByte]) |
Count the trail bytes for a lead byte - this macro should be used so that the assembler code that is mentioned in utf_impl.c could be used here. More... | |
#define | UTF8_MASK_LEAD_BYTE(leadByte, countTrailBytes) ((leadByte)&=(1<<(6-(countTrailBytes)))-1) |
#define | UTF8_IS_SINGLE(uchar) (((uchar)&0x80)==0) |
Is this this code point a single code unit (byte)? More... | |
#define | UTF8_IS_LEAD(uchar) ((uint8_t)((uchar)-0xc0)<0x3e) |
Is this this code unit the lead code unit (byte) of a code point? More... | |
#define | UTF8_IS_TRAIL(uchar) (((uchar)&0xc0)==0x80) |
Is this this code unit a trailing code unit (byte) of a code point? More... | |
#define | UTF8_NEED_MULTIPLE_UCHAR(c) ((uint32_t)(c)>0x7f) |
Does this scalar Unicode value need multiple code units for storage? More... | |
#define | UTF8_CHAR_LENGTH(c) |
Given the lead character, how many bytes are taken by this code point. More... | |
#define | UTF8_MAX_CHAR_LENGTH 4 |
The maximum number of bytes per code point. More... | |
#define | UTF8_ARRAY_SIZE(size) ((5*(size))/2) |
Average number of code units compared to UTF-16. More... | |
#define | UTF8_GET_CHAR_UNSAFE(s, i, c) |
#define | UTF8_GET_CHAR_SAFE(s, start, i, length, c, strict) |
#define | UTF8_NEXT_CHAR_UNSAFE(s, i, c) |
Read a Unicode scalar value from an array of UTF-8 bytes. More... | |
#define | UTF8_APPEND_CHAR_UNSAFE(s, i, c) |
#define | UTF8_FWD_1_UNSAFE(s, i) |
#define | UTF8_FWD_N_UNSAFE(s, i, n) |
#define | UTF8_SET_CHAR_START_UNSAFE(s, i) |
#define | UTF8_NEXT_CHAR_SAFE(s, i, length, c, strict) |
#define | UTF8_APPEND_CHAR_SAFE(s, i, length, c) |
#define | UTF8_FWD_1_SAFE(s, i, length) |
#define | UTF8_FWD_N_SAFE(s, i, length, n) |
#define | UTF8_SET_CHAR_START_SAFE(s, start, i) |
#define | UTF8_PREV_CHAR_UNSAFE(s, i, c) |
#define | UTF8_BACK_1_UNSAFE(s, i) |
#define | UTF8_BACK_N_UNSAFE(s, i, n) |
#define | UTF8_SET_CHAR_LIMIT_UNSAFE(s, i) |
#define | UTF8_PREV_CHAR_SAFE(s, start, i, c, strict) |
#define | UTF8_BACK_1_SAFE(s, start, i) |
#define | UTF8_BACK_N_SAFE(s, start, i, n) |
#define | UTF8_SET_CHAR_LIMIT_SAFE(s, start, i, length) |
Functions | |
U_CAPI UChar32 U_EXPORT2 | utf8_nextCharSafeBody (const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict, UBool *pIsError) |
U_CAPI int32_t U_EXPORT2 | utf8_appendCharSafeBody (uint8_t *s, int32_t i, int32_t length, UChar32 c) |
U_CAPI UChar32 U_EXPORT2 | utf8_prevCharSafeBody (const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict) |
U_CAPI int32_t U_EXPORT2 | utf8_back1SafeBody (const uint8_t *s, int32_t start, int32_t i) |
Variables | |
U_CFUNC U_IMPORT const uint8_t | utf8_countTrailBytes [256] |
This file defines macros to deal with UTF-8 code units and code points. Signatures and semantics are the same as for the similarly named macros in utf16.h. utf8.h is included by utf.h after unicode/umachine.h and some common definitions.
Usage: ICU coding guidelines for if() statements should be followed when using these macros. Compound statements (curly braces {}) must be used for if-else-while... bodies and all macro statements should be terminated with semicolon.
Definition in file utf8.h.
|
Value: { \ if((uint32_t)(c)<=0x7f) { \ (s)[(i)++]=(uint8_t)(c); \ } else { \ (i)=utf8_appendCharSafeBody(s, (int32_t)(i), (int32_t)(length), c); \ } \ } |
|
Value: { \ if((uint32_t)(c)<=0x7f) { \ (s)[(i)++]=(uint8_t)(c); \ } else { \ if((uint32_t)(c)<=0x7ff) { \ (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0); \ } else { \ if((uint32_t)(c)<=0xffff) { \ (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0); \ } else { \ (s)[(i)++]=(uint8_t)(((c)>>18)|0xf0); \ (s)[(i)++]=(uint8_t)((((c)>>12)&0x3f)|0x80); \ } \ (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80); \ } \ (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80); \ } \ } |
|
Average number of code units compared to UTF-16.
|
|
Value: { \ if(UTF8_IS_TRAIL((s)[--(i)])) { \ (i)=utf8_back1SafeBody(s, start, (int32_t)(i)); \ } \ } |
|
Value: { \ while(UTF8_IS_TRAIL((s)[--(i)])) {} \ } |
|
Value: { \ int32_t __N=(n); \ while(__N>0 && (i)>(start)) { \ UTF8_BACK_1_SAFE(s, start, i); \ --__N; \ } \ } |
|
Value: { \ int32_t __N=(n); \ while(__N>0) { \ UTF8_BACK_1_UNSAFE(s, i); \ --__N; \ } \ } |
|
Value: ((uint32_t)(c)<=0x7f ? 1 : \ ((uint32_t)(c)<=0x7ff ? 2 : \ ((uint32_t)((c)-0x10000)>0xfffff ? 3 : 4) \ ) \ ) ICU does not deal with code points >0x10ffff unless necessary for advancing in the byte stream. These length macros take into account that for values >0x10ffff the "safe" append macros would write the error code point 0xffff with 3 bytes. Code point comparisons need to be in uint32_t because UChar32 may be a signed type, and negative values must be recognized. |
|
Count the trail bytes for a lead byte - this macro should be used so that the assembler code that is mentioned in utf_impl.c could be used here.
|
|
Value: { \ uint8_t __b=(s)[(i)++]; \ if(UTF8_IS_LEAD(__b)) { \ uint8_t __count=UTF8_COUNT_TRAIL_BYTES(__b); \ if((i)+__count>(length)) { \ __count=(uint8_t)((length)-(i)); \ } \ while(__count>0 && UTF8_IS_TRAIL((s)[i])) { \ ++(i); \ --__count; \ } \ } \ } |
|
Value: { \ (i)+=1+UTF8_COUNT_TRAIL_BYTES((s)[i]); \ } |
|
Value: { \ int32_t __N=(n); \ while(__N>0 && (i)<(length)) { \ UTF8_FWD_1_SAFE(s, i, length); \ --__N; \ } \ } |
|
Value: { \ int32_t __N=(n); \ while(__N>0) { \ UTF8_FWD_1_UNSAFE(s, i); \ --__N; \ } \ } |
|
Value: { \ int32_t __I=(int32_t)(i); \ UTF8_SET_CHAR_START_SAFE(s, start, __I); \ UTF8_NEXT_CHAR_SAFE(s, __I, length, c, strict); \ } |
|
Value: { \ int32_t __I=(int32_t)(i); \ UTF8_SET_CHAR_START_UNSAFE(s, __I); \ UTF8_NEXT_CHAR_UNSAFE(s, __I, c); \ } |
|
Is this this code unit the lead code unit (byte) of a code point?
|
|
Is this this code point a single code unit (byte)?
|
|
Is this this code unit a trailing code unit (byte) of a code point?
|
|
|
|
The maximum number of bytes per code point.
|
|
Does this scalar Unicode value need multiple code units for storage?
|
|
Value: { \ (c)=(s)[(i)++]; \ if((c)>=0x80) { \ if(UTF8_IS_LEAD(c)) { \ (c)=utf8_nextCharSafeBody(s, &(i), (int32_t)(length), c, strict, NULL); \ } else { \ (c)=UTF8_ERROR_VALUE_1; \ } \ } \ } |
|
Value: { \ (c)=(s)[(i)++]; \ if((uint8_t)((c)-0xc0)<0x35) { \ uint8_t __count=UTF8_COUNT_TRAIL_BYTES(c); \ UTF8_MASK_LEAD_BYTE(c, __count); \ switch(__count) { \ \ case 3: \ (c)=((c)<<6)|((s)[(i)++]&0x3f); \ case 2: \ (c)=((c)<<6)|((s)[(i)++]&0x3f); \ case 1: \ (c)=((c)<<6)|((s)[(i)++]&0x3f); \ \ break; \ } \ } \ } Only values <=0x10ffff are accepted, and if an error occurs, then c will be set such that UTF_IS_ERROR(c). The _UNSAFE macro is fast and does not check for errors. The _SAFE macro checks for errors and optionally for irregular sequences, too, i.e., for sequences that are longer than necessary, such as <c0 80> instead of <0>. The strict checks also check for non-characters. |
|
Value: { \ (c)=(s)[--(i)]; \ if((c)>=0x80) { \ if((c)<=0xbf) { \ (c)=utf8_prevCharSafeBody(s, start, &(i), c, strict); \ } else { \ (c)=UTF8_ERROR_VALUE_1; \ } \ } \ } |
|
Value: { \ (c)=(s)[--(i)]; \ if(UTF8_IS_TRAIL(c)) { \ uint8_t __b, __count=1, __shift=6; \ \ \ (c)&=0x3f; \ for(;;) { \ __b=(s)[--(i)]; \ if(__b>=0xc0) { \ UTF8_MASK_LEAD_BYTE(__b, __count); \ (c)|=(UChar32)__b<<__shift; \ break; \ } else { \ (c)|=(UChar32)(__b&0x3f)<<__shift; \ ++__count; \ __shift+=6; \ } \ } \ } \ } |
|
Value: { \ if((start)<(i) && (i)<(length)) { \ UTF8_BACK_1_SAFE(s, start, i); \ UTF8_FWD_1_SAFE(s, i, length); \ } \ } |
|
Value: { \ UTF8_BACK_1_UNSAFE(s, i); \ UTF8_FWD_1_UNSAFE(s, i); \ } |
|
Value: { \ if(UTF8_IS_TRAIL((s)[(i)])) { \ (i)=utf8_back1SafeBody(s, start, (int32_t)(i)); \ } \ } |
|
Value: { \ while(UTF8_IS_TRAIL((s)[i])) { --(i); } \ } |
|
|
|
|
|
|
|
|
|
|