SQLite

Check-in [0c6736df9c]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:When converting UTF8 or UTF16 strings, change overlong strings and other illegal codes to 0xFFFD. Ticket #2029. (CVS 3479)
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 0c6736df9cb4c3c8f6224e30df939cead9cd5369
User & Date: drh 2006-10-19 01:58:44.000
Context
2006-10-19
20:27
fts2 support for testing. These are a prelude to adding some test scripts. (CVS 3480) (check-in: 004ad1943f user: shess tags: trunk)
01:58
When converting UTF8 or UTF16 strings, change overlong strings and other illegal codes to 0xFFFD. Ticket #2029. (CVS 3479) (check-in: 0c6736df9c user: drh tags: trunk)
2006-10-18
23:26
Fix a problems that arise if malloc() fails while compiling SELECT statements within a TRIGGER. (CVS 3478) (check-in: ee4894b499 user: drh tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/utf.c.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** This file contains routines used to translate between UTF-8, 
** UTF-16, UTF-16BE, and UTF-16LE.
**
** $Id: utf.c,v 1.42 2006/10/05 11:43:53 drh Exp $
**
** Notes on UTF-8:
**
**   Byte-0    Byte-1    Byte-2    Byte-3    Value
**  0xxxxxxx                                 00000000 00000000 0xxxxxxx
**  110yyyyy  10xxxxxx                       00000000 00000yyy yyxxxxxx
**  1110zzzz  10yyyyyy  10xxxxxx             00000000 zzzzyyyy yyxxxxxx







|







8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** This file contains routines used to translate between UTF-8, 
** UTF-16, UTF-16BE, and UTF-16LE.
**
** $Id: utf.c,v 1.43 2006/10/19 01:58:44 drh Exp $
**
** Notes on UTF-8:
**
**   Byte-0    Byte-1    Byte-2    Byte-3    Value
**  0xxxxxxx                                 00000000 00000000 0xxxxxxx
**  110yyyyy  10xxxxxx                       00000000 00000yyy yyxxxxxx
**  1110zzzz  10yyyyyy  10xxxxxx             00000000 zzzzyyyy yyxxxxxx
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108













109
110
111
112
113
114
115
116
117
118
119
120



121
122
123
124
125
126
127
*/
#include "sqliteInt.h"
#include <assert.h>
#include "vdbeInt.h"

/*
** This table maps from the first byte of a UTF-8 character to the number
** of trailing bytes expected. A value '255' indicates that the table key
** is not a legal first byte for a UTF-8 character.
*/
static const u8 xtra_utf8_bytes[256]  = {
/* 0xxxxxxx */
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,

/* 10wwwwww */
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,

/* 110yyyyy */
1, 1, 1, 1, 1, 1, 1, 1,     1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,     1, 1, 1, 1, 1, 1, 1, 1,

/* 1110zzzz */
2, 2, 2, 2, 2, 2, 2, 2,     2, 2, 2, 2, 2, 2, 2, 2,

/* 11110yyy */
3, 3, 3, 3, 3, 3, 3, 3,     255, 255, 255, 255, 255, 255, 255, 255,
};

/*
** This table maps from the number of trailing bytes in a UTF-8 character
** to an integer constant that is effectively calculated for each character
** read by a naive implementation of a UTF-8 character reader. The code
** in the READ_UTF8 macro explains things best.
*/
static const int xtra_utf8_bits[4] =  {
0,
12416,          /* (0xC0 << 6) + (0x80) */
925824,         /* (0xE0 << 12) + (0x80 << 6) + (0x80) */
63447168        /* (0xF0 << 18) + (0x80 << 12) + (0x80 << 6) + 0x80 */













};

#define READ_UTF8(zIn, c) { \
  int xtra;                                            \
  c = *(zIn)++;                                        \
  xtra = xtra_utf8_bytes[c];                           \
  switch( xtra ){                                      \
    case 255: c = (int)0xFFFD; break;                  \
    case 3: c = (c<<6) + *(zIn)++;                     \
    case 2: c = (c<<6) + *(zIn)++;                     \
    case 1: c = (c<<6) + *(zIn)++;                     \
    c -= xtra_utf8_bits[xtra];                         \



  }                                                    \
}
int sqlite3ReadUtf8(const unsigned char *z){
  int c;
  READ_UTF8(z, c);
  return c;
}







|














|
|
|
|









|








|
|
|
|
|
>
>
>
>
>
>
>
>
>
>
>
>
>







|




>
>
>







60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
*/
#include "sqliteInt.h"
#include <assert.h>
#include "vdbeInt.h"

/*
** This table maps from the first byte of a UTF-8 character to the number
** of trailing bytes expected. A value '4' indicates that the table key
** is not a legal first byte for a UTF-8 character.
*/
static const u8 xtra_utf8_bytes[256]  = {
/* 0xxxxxxx */
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,     0, 0, 0, 0, 0, 0, 0, 0,

/* 10wwwwww */
4, 4, 4, 4, 4, 4, 4, 4,     4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4,     4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4,     4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4,     4, 4, 4, 4, 4, 4, 4, 4,

/* 110yyyyy */
1, 1, 1, 1, 1, 1, 1, 1,     1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,     1, 1, 1, 1, 1, 1, 1, 1,

/* 1110zzzz */
2, 2, 2, 2, 2, 2, 2, 2,     2, 2, 2, 2, 2, 2, 2, 2,

/* 11110yyy */
3, 3, 3, 3, 3, 3, 3, 3,     4, 4, 4, 4, 4, 4, 4, 4,
};

/*
** This table maps from the number of trailing bytes in a UTF-8 character
** to an integer constant that is effectively calculated for each character
** read by a naive implementation of a UTF-8 character reader. The code
** in the READ_UTF8 macro explains things best.
*/
static const int xtra_utf8_bits[] =  {
  0,
  12416,          /* (0xC0 << 6) + (0x80) */
  925824,         /* (0xE0 << 12) + (0x80 << 6) + (0x80) */
  63447168        /* (0xF0 << 18) + (0x80 << 12) + (0x80 << 6) + 0x80 */
};

/*
** If a UTF-8 character contains N bytes extra bytes (N bytes follow
** the initial byte so that the total character length is N+1) then
** masking the character with utf8_mask[N] must produce a non-zero
** result.  Otherwise, we have an (illegal) overlong encoding.
*/
static const int utf_mask[] = {
  0x00000000,
  0xffffff80,
  0xfffff800,
  0xffff0000,
};

#define READ_UTF8(zIn, c) { \
  int xtra;                                            \
  c = *(zIn)++;                                        \
  xtra = xtra_utf8_bytes[c];                           \
  switch( xtra ){                                      \
    case 4: c = (int)0xFFFD; break;                    \
    case 3: c = (c<<6) + *(zIn)++;                     \
    case 2: c = (c<<6) + *(zIn)++;                     \
    case 1: c = (c<<6) + *(zIn)++;                     \
    c -= xtra_utf8_bits[xtra];                         \
    if( (utf_mask[xtra]&c)==0                          \
        || (c&0xFFFFF800)==0xD800                      \
        || (c&0xFFFFFFFE)==0xFFFE ){  c = 0xFFFD; }    \
  }                                                    \
}
int sqlite3ReadUtf8(const unsigned char *z){
  int c;
  READ_UTF8(z, c);
  return c;
}
177
178
179
180
181
182
183

184
185
186
187
188
189
190
191
192
193

194
195
196
197
198
199
200
#define READ_UTF16LE(zIn, c){                                         \
  c = (*zIn++);                                                       \
  c += ((*zIn++)<<8);                                                 \
  if( c>=0xD800 && c<=0xE000 ){                                       \
    int c2 = (*zIn++);                                                \
    c2 += ((*zIn++)<<8);                                              \
    c = (c2&0x03FF) + ((c&0x003F)<<10) + (((c&0x03C0)+0x0040)<<10);   \

  }                                                                   \
}

#define READ_UTF16BE(zIn, c){                                         \
  c = ((*zIn++)<<8);                                                  \
  c += (*zIn++);                                                      \
  if( c>=0xD800 && c<=0xE000 ){                                       \
    int c2 = ((*zIn++)<<8);                                           \
    c2 += (*zIn++);                                                   \
    c = (c2&0x03FF) + ((c&0x003F)<<10) + (((c&0x03C0)+0x0040)<<10);   \

  }                                                                   \
}

#define SKIP_UTF16BE(zIn){                                            \
  if( *zIn>=0xD8 && (*zIn<0xE0 || (*zIn==0xE0 && *(zIn+1)==0x00)) ){  \
    zIn += 4;                                                         \
  }else{                                                              \







>










>







193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
#define READ_UTF16LE(zIn, c){                                         \
  c = (*zIn++);                                                       \
  c += ((*zIn++)<<8);                                                 \
  if( c>=0xD800 && c<=0xE000 ){                                       \
    int c2 = (*zIn++);                                                \
    c2 += ((*zIn++)<<8);                                              \
    c = (c2&0x03FF) + ((c&0x003F)<<10) + (((c&0x03C0)+0x0040)<<10);   \
    if( (c & 0xFFFF0000)==0 ) c = 0xFFFD;                             \
  }                                                                   \
}

#define READ_UTF16BE(zIn, c){                                         \
  c = ((*zIn++)<<8);                                                  \
  c += (*zIn++);                                                      \
  if( c>=0xD800 && c<=0xE000 ){                                       \
    int c2 = ((*zIn++)<<8);                                           \
    c2 += (*zIn++);                                                   \
    c = (c2&0x03FF) + ((c&0x003F)<<10) + (((c&0x03C0)+0x0040)<<10);   \
    if( (c & 0xFFFF0000)==0 ) c = 0xFFFD;                             \
  }                                                                   \
}

#define SKIP_UTF16BE(zIn){                                            \
  if( *zIn>=0xD8 && (*zIn<0xE0 || (*zIn==0xE0 && *(zIn+1)==0x00)) ){  \
    zIn += 4;                                                         \
  }else{                                                              \
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570



571
572
573
574
575
576
577
578
#if defined(SQLITE_TEST)
/*
** This routine is called from the TCL test function "translate_selftest".
** It checks that the primitives for serializing and deserializing
** characters in each encoding are inverses of each other.
*/
void sqlite3utfSelfTest(){
  unsigned int i;
  unsigned char zBuf[20];
  unsigned char *z;
  int n;
  unsigned int c;

  for(i=0; i<0x00110000; i++){
    z = zBuf;
    WRITE_UTF8(z, i);
    n = z-zBuf;
    z = zBuf;
    READ_UTF8(z, c);



    assert( c==i );
    assert( (z-zBuf)==n );
  }
  for(i=0; i<0x00110000; i++){
    if( i>=0xD800 && i<=0xE000 ) continue;
    z = zBuf;
    WRITE_UTF16LE(z, i);
    n = z-zBuf;







|











>
>
>
|







570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
#if defined(SQLITE_TEST)
/*
** This routine is called from the TCL test function "translate_selftest".
** It checks that the primitives for serializing and deserializing
** characters in each encoding are inverses of each other.
*/
void sqlite3utfSelfTest(){
  unsigned int i, t;
  unsigned char zBuf[20];
  unsigned char *z;
  int n;
  unsigned int c;

  for(i=0; i<0x00110000; i++){
    z = zBuf;
    WRITE_UTF8(z, i);
    n = z-zBuf;
    z = zBuf;
    READ_UTF8(z, c);
    t = i;
    if( i>=0xD800 && i<=0xDFFF ) t = 0xFFFD;
    if( (i&0xFFFFFFFE)==0xFFFE ) t = 0xFFFD;
    assert( c==t );
    assert( (z-zBuf)==n );
  }
  for(i=0; i<0x00110000; i++){
    if( i>=0xD800 && i<=0xE000 ) continue;
    z = zBuf;
    WRITE_UTF16LE(z, i);
    n = z-zBuf;