• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 // Copyright (C) 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 ******************************************************************************
5 *
6 *   Copyright (C) 1999-2012, International Business Machines
7 *   Corporation and others.  All Rights Reserved.
8 *
9 ******************************************************************************
10 *   file name:  utf_impl.c
11 *   encoding:   US-ASCII
12 *   tab size:   8 (not used)
13 *   indentation:4
14 *
15 *   created on: 1999sep13
16 *   created by: Markus W. Scherer
17 *
18 *   This file provides implementation functions for macros in the utfXX.h
19 *   that would otherwise be too long as macros.
20 */
21 
22 /* set import/export definitions */
23 #ifndef U_UTF8_IMPL
24 #   define U_UTF8_IMPL
25 #endif
26 
27 #include "unicode/utypes.h"
28 #include "unicode/utf.h"
29 #include "unicode/utf8.h"
30 #include "unicode/utf_old.h"
31 #include "uassert.h"
32 
33 /*
34  * This table could be replaced on many machines by
35  * a few lines of assembler code using an
36  * "index of first 0-bit from msb" instruction and
37  * one or two more integer instructions.
38  *
39  * For example, on an i386, do something like
40  * - MOV AL, leadByte
41  * - NOT AL         (8-bit, leave b15..b8==0..0, reverse only b7..b0)
42  * - MOV AH, 0
43  * - BSR BX, AX     (16-bit)
44  * - MOV AX, 6      (result)
45  * - JZ finish      (ZF==1 if leadByte==0xff)
46  * - SUB AX, BX (result)
47  * -finish:
48  * (BSR: Bit Scan Reverse, scans for a 1-bit, starting from the MSB)
49  *
50  * In Unicode, all UTF-8 byte sequences with more than 4 bytes are illegal;
51  * lead bytes above 0xf4 are illegal.
52  * We keep them in this table for skipping long ISO 10646-UTF-8 sequences.
53  */
54 U_EXPORT const uint8_t
55 utf8_countTrailBytes[256]={
56     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
57     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
58     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
59     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
60 
61     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
62     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
63     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
64     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
65 
66     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
67     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
68     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
69     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
70 
71     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
72     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
73 
74     2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
75     3, 3, 3, 3, 3,
76     3, 3, 3,    /* illegal in Unicode */
77     4, 4, 4, 4, /* illegal in Unicode */
78     5, 5,       /* illegal in Unicode */
79     0, 0        /* illegal bytes 0xfe and 0xff */
80 };
81 
82 static const UChar32
83 utf8_minLegal[4]={ 0, 0x80, 0x800, 0x10000 };
84 
85 static const UChar32
86 utf8_errorValue[6]={
87     UTF8_ERROR_VALUE_1, UTF8_ERROR_VALUE_2, UTF_ERROR_VALUE, 0x10ffff,
88     0x3ffffff, 0x7fffffff
89 };
90 
91 static UChar32
errorValue(int32_t count,int8_t strict)92 errorValue(int32_t count, int8_t strict) {
93     if(strict>=0) {
94         return utf8_errorValue[count];
95     } else if(strict==-3) {
96         return 0xfffd;
97     } else {
98         return U_SENTINEL;
99     }
100 }
101 
102 /*
103  * Handle the non-inline part of the U8_NEXT() and U8_NEXT_FFFD() macros
104  * and their obsolete sibling UTF8_NEXT_CHAR_SAFE().
105  *
106  * U8_NEXT() supports NUL-terminated strings indicated via length<0.
107  *
108  * The "strict" parameter controls the error behavior:
109  * <0  "Safe" behavior of U8_NEXT():
110  *     -1: All illegal byte sequences yield U_SENTINEL=-1.
111  *     -2: Same as -1, except for lenient treatment of surrogate code points as legal.
112  *         Some implementations use this for roundtripping of
113  *         Unicode 16-bit strings that are not well-formed UTF-16, that is, they
114  *         contain unpaired surrogates.
115  *     -3: All illegal byte sequences yield U+FFFD.
116  *  0  Obsolete "safe" behavior of UTF8_NEXT_CHAR_SAFE(..., FALSE):
117  *     All illegal byte sequences yield a positive code point such that this
118  *     result code point would be encoded with the same number of bytes as
119  *     the illegal sequence.
120  * >0  Obsolete "strict" behavior of UTF8_NEXT_CHAR_SAFE(..., TRUE):
121  *     Same as the obsolete "safe" behavior, but non-characters are also treated
122  *     like illegal sequences.
123  *
124  * Note that a UBool is the same as an int8_t.
125  */
126 U_CAPI UChar32 U_EXPORT2
utf8_nextCharSafeBody(const uint8_t * s,int32_t * pi,int32_t length,UChar32 c,UBool strict)127 utf8_nextCharSafeBody(const uint8_t *s, int32_t *pi, int32_t length, UChar32 c, UBool strict) {
128     int32_t i=*pi;
129     uint8_t count=U8_COUNT_TRAIL_BYTES(c);
130     U_ASSERT(count <= 5); /* U8_COUNT_TRAIL_BYTES returns value 0...5 */
131     if(i+count<=length || length<0) {
132         uint8_t trail;
133 
134         U8_MASK_LEAD_BYTE(c, count);
135         /* support NUL-terminated strings: do not read beyond the first non-trail byte */
136         switch(count) {
137         /* each branch falls through to the next one */
138         case 0:
139             /* count==0 for illegally leading trail bytes and the illegal bytes 0xfe and 0xff */
140         case 5:
141         case 4:
142             /* count>=4 is always illegal: no more than 3 trail bytes in Unicode's UTF-8 */
143             break;
144         case 3:
145             trail=s[i++]-0x80;
146             c=(c<<6)|trail;
147             /* c>=0x110 would result in code point>0x10ffff, outside Unicode */
148             if(c>=0x110 || trail>0x3f) { break; }
149         case 2:
150             trail=s[i++]-0x80;
151             c=(c<<6)|trail;
152             /*
153              * test for a surrogate d800..dfff unless we are lenient:
154              * before the last (c<<6), a surrogate is c=360..37f
155              */
156             if(((c&0xffe0)==0x360 && strict!=-2) || trail>0x3f) { break; }
157         case 1:
158             trail=s[i++]-0x80;
159             c=(c<<6)|trail;
160             if(trail>0x3f) { break; }
161             /* correct sequence - all trail bytes have (b7..b6)==(10) */
162             if(c>=utf8_minLegal[count] &&
163                     /* strict: forbid non-characters like U+fffe */
164                     (strict<=0 || !U_IS_UNICODE_NONCHAR(c))) {
165                 *pi=i;
166                 return c;
167             }
168         /* no default branch to optimize switch()  - all values are covered */
169         }
170     } else {
171         /* too few bytes left */
172         count=length-i;
173     }
174 
175     /* error handling */
176     i=*pi;
177     while(count>0 && U8_IS_TRAIL(s[i])) {
178         ++i;
179         --count;
180     }
181     c=errorValue(i-*pi, strict);
182     *pi=i;
183     return c;
184 }
185 
186 U_CAPI int32_t U_EXPORT2
utf8_appendCharSafeBody(uint8_t * s,int32_t i,int32_t length,UChar32 c,UBool * pIsError)187 utf8_appendCharSafeBody(uint8_t *s, int32_t i, int32_t length, UChar32 c, UBool *pIsError) {
188     if((uint32_t)(c)<=0x7ff) {
189         if((i)+1<(length)) {
190             (s)[(i)++]=(uint8_t)(((c)>>6)|0xc0);
191             (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80);
192             return i;
193         }
194     } else if((uint32_t)(c)<=0xffff) {
195         /* Starting with Unicode 3.2, surrogate code points must not be encoded in UTF-8. */
196         if((i)+2<(length) && !U_IS_SURROGATE(c)) {
197             (s)[(i)++]=(uint8_t)(((c)>>12)|0xe0);
198             (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80);
199             (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80);
200             return i;
201         }
202     } else if((uint32_t)(c)<=0x10ffff) {
203         if((i)+3<(length)) {
204             (s)[(i)++]=(uint8_t)(((c)>>18)|0xf0);
205             (s)[(i)++]=(uint8_t)((((c)>>12)&0x3f)|0x80);
206             (s)[(i)++]=(uint8_t)((((c)>>6)&0x3f)|0x80);
207             (s)[(i)++]=(uint8_t)(((c)&0x3f)|0x80);
208             return i;
209         }
210     }
211     /* c>0x10ffff or not enough space, write an error value */
212     if(pIsError!=NULL) {
213         *pIsError=TRUE;
214     } else {
215         length-=i;
216         if(length>0) {
217             int32_t offset;
218             if(length>3) {
219                 length=3;
220             }
221             s+=i;
222             offset=0;
223             c=utf8_errorValue[length-1];
224             UTF8_APPEND_CHAR_UNSAFE(s, offset, c);
225             i=i+offset;
226         }
227     }
228     return i;
229 }
230 
231 U_CAPI UChar32 U_EXPORT2
utf8_prevCharSafeBody(const uint8_t * s,int32_t start,int32_t * pi,UChar32 c,UBool strict)232 utf8_prevCharSafeBody(const uint8_t *s, int32_t start, int32_t *pi, UChar32 c, UBool strict) {
233     int32_t i=*pi;
234     uint8_t b, count=1, shift=6;
235 
236     if(!U8_IS_TRAIL(c)) { return errorValue(0, strict); }
237 
238     /* extract value bits from the last trail byte */
239     c&=0x3f;
240 
241     for(;;) {
242         if(i<=start) {
243             /* no lead byte at all */
244             return errorValue(0, strict);
245         }
246 
247         /* read another previous byte */
248         b=s[--i];
249         if((uint8_t)(b-0x80)<0x7e) { /* 0x80<=b<0xfe */
250             if(b&0x40) {
251                 /* lead byte, this will always end the loop */
252                 uint8_t shouldCount=U8_COUNT_TRAIL_BYTES(b);
253 
254                 if(count==shouldCount) {
255                     /* set the new position */
256                     *pi=i;
257                     U8_MASK_LEAD_BYTE(b, count);
258                     c|=(UChar32)b<<shift;
259                     if(count>=4 || c>0x10ffff || c<utf8_minLegal[count] || (U_IS_SURROGATE(c) && strict!=-2) || (strict>0 && U_IS_UNICODE_NONCHAR(c))) {
260                         /* illegal sequence or (strict and non-character) */
261                         if(count>=4) {
262                             count=3;
263                         }
264                         c=errorValue(count, strict);
265                     } else {
266                         /* exit with correct c */
267                     }
268                 } else {
269                     /* the lead byte does not match the number of trail bytes */
270                     /* only set the position to the lead byte if it would
271                        include the trail byte that we started with */
272                     if(count<shouldCount) {
273                         *pi=i;
274                         c=errorValue(count, strict);
275                     } else {
276                         c=errorValue(0, strict);
277                     }
278                 }
279                 break;
280             } else if(count<5) {
281                 /* trail byte */
282                 c|=(UChar32)(b&0x3f)<<shift;
283                 ++count;
284                 shift+=6;
285             } else {
286                 /* more than 5 trail bytes is illegal */
287                 c=errorValue(0, strict);
288                 break;
289             }
290         } else {
291             /* single-byte character precedes trailing bytes */
292             c=errorValue(0, strict);
293             break;
294         }
295     }
296     return c;
297 }
298 
299 U_CAPI int32_t U_EXPORT2
utf8_back1SafeBody(const uint8_t * s,int32_t start,int32_t i)300 utf8_back1SafeBody(const uint8_t *s, int32_t start, int32_t i) {
301     /* i had been decremented once before the function call */
302     int32_t I=i, Z;
303     uint8_t b;
304 
305     /* read at most the 6 bytes s[Z] to s[i], inclusively */
306     if(I-5>start) {
307         Z=I-5;
308     } else {
309         Z=start;
310     }
311 
312     /* return I if the sequence starting there is long enough to include i */
313     do {
314         b=s[I];
315         if((uint8_t)(b-0x80)>=0x7e) { /* not 0x80<=b<0xfe */
316             break;
317         } else if(b>=0xc0) {
318             if(U8_COUNT_TRAIL_BYTES(b)>=(i-I)) {
319                 return I;
320             } else {
321                 break;
322             }
323         }
324     } while(Z<=--I);
325 
326     /* return i itself to be consistent with the FWD_1 macro */
327     return i;
328 }
329