SQLite

Check-in [7695b88fe0]
Login

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

Overview
Comment:More bug fixes to the test_regexp.c module. Bring test_regexp.c into alignment with the regexp.c file in the Fossil sources.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 7695b88fe0d73fd0a36fb7d8f95350b80615a89e
User & Date: drh 2013-01-05 17:17:21.332
Context
2013-01-08
12:48
Do not raise an error if an unknown SQL function is found in a CHECK constraint while parsing the schema of an existing database. (check-in: cda790280a user: drh tags: trunk)
2013-01-07
13:26
Do not raise an error if an unknown SQL function is found in a CHECK constraint while parsing the schema of an existing database. This version causes a run-time error when processing STAT3 tables. (Closed-Leaf check-in: 0a1207c895 user: dan tags: mistake)
2013-01-05
17:17
More bug fixes to the test_regexp.c module. Bring test_regexp.c into alignment with the regexp.c file in the Fossil sources. (check-in: 7695b88fe0 user: drh tags: trunk)
2013-01-04
22:26
Remove superfluous text from a comment. (check-in: e9ac533960 user: mistachkin tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/test_regexp.c.
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
** be clear:  this routine converts utf8 to unicode.  This routine is 
** optimized for the common case where the next character is a single byte.
*/
static unsigned re_next_char(ReInput *p){
  unsigned c;
  if( p->i>=p->mx ) return 0;
  c = p->z[p->i++];
  if( c>0x80 ){
    if( (c&0xe0)==0xc0 && p->i<p->mx && (p->z[p->i]&0xc0)==0x80 ){
      c = (c&0x1f)<<6 | (p->z[p->i++]&0x3f);
      if( c<0x80 ) c = 0xfffd;
    }else if( (c&0xf0)==0xe0 && p->i+1<p->mx && (p->z[p->i]&0xc0)==0x80
           && (p->z[p->i+1]&0xc0)==0x80 ){
      c = (c&0x0f)<<12 | ((p->z[p->i]&0x3f)<<6) | (p->z[p->i+1]&0x3f);
      p->i += 2;
      if( c<0x3ff || (c>=0xd800 && c<=0xdfff) ) c = 0xfffd;
    }else if( (c&0xf8)==0xf0 && p->i+3<p->mx && (p->z[p->i]&0xc0)==0x80
           && (p->z[p->i+1]&0xc0)==0x80 && (p->z[p->i+2]&0xc0)==0x80 ){
      c = (c&0x07)<<18 | ((p->z[p->i]&0x3f)<<12) | ((p->z[p->i+1]&0x3f)<<6)
                       | (p->z[p->i+2]&0x3f);
      p->i += 3;
      if( c<0xffff ) c = 0xfffd;
    }else{
      c = 0xfffd;
    }
  }
  return c;
}
static unsigned re_next_char_nocase(ReInput *p){







|







|





|







125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
** be clear:  this routine converts utf8 to unicode.  This routine is 
** optimized for the common case where the next character is a single byte.
*/
static unsigned re_next_char(ReInput *p){
  unsigned c;
  if( p->i>=p->mx ) return 0;
  c = p->z[p->i++];
  if( c>=0x80 ){
    if( (c&0xe0)==0xc0 && p->i<p->mx && (p->z[p->i]&0xc0)==0x80 ){
      c = (c&0x1f)<<6 | (p->z[p->i++]&0x3f);
      if( c<0x80 ) c = 0xfffd;
    }else if( (c&0xf0)==0xe0 && p->i+1<p->mx && (p->z[p->i]&0xc0)==0x80
           && (p->z[p->i+1]&0xc0)==0x80 ){
      c = (c&0x0f)<<12 | ((p->z[p->i]&0x3f)<<6) | (p->z[p->i+1]&0x3f);
      p->i += 2;
      if( c<=0x3ff || (c>=0xd800 && c<=0xdfff) ) c = 0xfffd;
    }else if( (c&0xf8)==0xf0 && p->i+3<p->mx && (p->z[p->i]&0xc0)==0x80
           && (p->z[p->i+1]&0xc0)==0x80 && (p->z[p->i+2]&0xc0)==0x80 ){
      c = (c&0x07)<<18 | ((p->z[p->i]&0x3f)<<12) | ((p->z[p->i+1]&0x3f)<<6)
                       | (p->z[p->i+2]&0x3f);
      p->i += 3;
      if( c<=0xffff || c>0x10ffff ) c = 0xfffd;
    }else{
      c = 0xfffd;
    }
  }
  return c;
}
static unsigned re_next_char_nocase(ReInput *p){
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191


192
193
194
195
196
197
198
199
200

201
202
203
204
205
206
207
/* Return true if c is a "digit" character:  [0-9] */
static int re_digit_char(int c){
  return (c>='0' && c<='9');
}

/* Return true if c is a perl "space" character:  [ \t\r\n\v\f] */
static int re_space_char(int c){
  return c==' ' || c=='\t' || c=='\n' || c=='\v' || c=='\f';
}

/* Run a compiled regular expression on the zero-terminated input
** string zIn[].  Return true on a match and false if there is no match.
*/
int re_match(ReCompiled *pRe, const unsigned char *zIn, int nIn){
  ReStateSet aStateSet[2], *pThis, *pNext;
  ReStateNumber aSpace[100];
  ReStateNumber *pToFree;
  unsigned int i = 0;
  unsigned int iSwap = 0;
  int c = RE_EOF+1;
  int cPrev = 0;
  int rc = 0;
  ReInput in;

  in.z = zIn;
  in.i = 0;
  in.mx = nIn>=0 ? nIn : strlen((char*)zIn);


  if( pRe->nInit ){
    unsigned char x = pRe->zInit[0];
    while( in.i+pRe->nInit<=in.mx 
        && (zIn[in.i]!=x || memcmp(zIn+in.i, pRe->zInit, pRe->nInit)!=0)
    ){
      in.i++;
    }
    if( in.i+pRe->nInit>in.mx ) return 0;
  }

  if( pRe->nState<=(sizeof(aSpace)/(sizeof(aSpace[0])*2)) ){
    pToFree = 0;
    aStateSet[0].aState = aSpace;
  }else{
    pToFree = sqlite3_malloc( sizeof(ReStateNumber)*2*pRe->nState );
    if( pToFree==0 ) return -1;
    aStateSet[0].aState = pToFree;







|


















|
>
>









>







165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
/* Return true if c is a "digit" character:  [0-9] */
static int re_digit_char(int c){
  return (c>='0' && c<='9');
}

/* Return true if c is a perl "space" character:  [ \t\r\n\v\f] */
static int re_space_char(int c){
  return c==' ' || c=='\t' || c=='\n' || c=='\r' || c=='\v' || c=='\f';
}

/* Run a compiled regular expression on the zero-terminated input
** string zIn[].  Return true on a match and false if there is no match.
*/
int re_match(ReCompiled *pRe, const unsigned char *zIn, int nIn){
  ReStateSet aStateSet[2], *pThis, *pNext;
  ReStateNumber aSpace[100];
  ReStateNumber *pToFree;
  unsigned int i = 0;
  unsigned int iSwap = 0;
  int c = RE_EOF+1;
  int cPrev = 0;
  int rc = 0;
  ReInput in;

  in.z = zIn;
  in.i = 0;
  in.mx = nIn>=0 ? nIn : strlen((char const*)zIn);

  /* Look for the initial prefix match, if there is one. */
  if( pRe->nInit ){
    unsigned char x = pRe->zInit[0];
    while( in.i+pRe->nInit<=in.mx 
        && (zIn[in.i]!=x || memcmp(zIn+in.i, pRe->zInit, pRe->nInit)!=0)
    ){
      in.i++;
    }
    if( in.i+pRe->nInit>in.mx ) return 0;
  }

  if( pRe->nState<=(sizeof(aSpace)/(sizeof(aSpace[0])*2)) ){
    pToFree = 0;
    aStateSet[0].aState = aSpace;
  }else{
    pToFree = sqlite3_malloc( sizeof(ReStateNumber)*2*pRe->nState );
    if( pToFree==0 ) return -1;
    aStateSet[0].aState = pToFree;
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643









644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
  if( zIn[0]=='^' ){
    zIn++;
  }else{
    re_append(pRe, RE_OP_ANYSTAR, 0);
  }
  pRe->sIn.z = (unsigned char*)zIn;
  pRe->sIn.i = 0;
  pRe->sIn.mx = strlen((char*)pRe->sIn.z);
  zErr = re_subcompile_re(pRe);
  if( zErr ){
    re_free(pRe);
    return zErr;
  }
  if( rePeek(pRe)=='$' && pRe->sIn.i+1>=pRe->sIn.mx ){
    re_append(pRe, RE_OP_MATCH, RE_EOF);
    re_append(pRe, RE_OP_ACCEPT, 0);
    *ppRe = pRe;
  }else if( pRe->sIn.i>=pRe->sIn.mx ){
    re_append(pRe, RE_OP_ACCEPT, 0);
    *ppRe = pRe;
  }else{
    re_free(pRe);
    return "unrecognized character";
  }









  if( pRe->aOp[0]==RE_OP_ANYSTAR ){
    for(j=0, i=1; j<sizeof(pRe->zInit)-2 && pRe->aOp[i]==RE_OP_MATCH; i++){
      unsigned x = pRe->aArg[i];
      if( x<=127 ){
        pRe->zInit[j++] = x;
      }else if( x<=0xfff ){
        pRe->zInit[j++] = 0xc0 | (x>>6);
        pRe->zInit[j++] = 0x80 | (x&0x3f);
      }else if( x<=0xffff ){
        pRe->zInit[j++] = 0xd0 | (x>>12);
        pRe->zInit[j++] = 0x80 | ((x>>6)&0x3f);
        pRe->zInit[j++] = 0x80 | (0x3f);
      }else{
        break;
      }
    }
    if( j>0 && pRe->zInit[j-1]==0 ) j--;
    pRe->nInit = j;
  }







|
















>
>
>
>
>
>
>
>
>











|







623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
  if( zIn[0]=='^' ){
    zIn++;
  }else{
    re_append(pRe, RE_OP_ANYSTAR, 0);
  }
  pRe->sIn.z = (unsigned char*)zIn;
  pRe->sIn.i = 0;
  pRe->sIn.mx = strlen(zIn);
  zErr = re_subcompile_re(pRe);
  if( zErr ){
    re_free(pRe);
    return zErr;
  }
  if( rePeek(pRe)=='$' && pRe->sIn.i+1>=pRe->sIn.mx ){
    re_append(pRe, RE_OP_MATCH, RE_EOF);
    re_append(pRe, RE_OP_ACCEPT, 0);
    *ppRe = pRe;
  }else if( pRe->sIn.i>=pRe->sIn.mx ){
    re_append(pRe, RE_OP_ACCEPT, 0);
    *ppRe = pRe;
  }else{
    re_free(pRe);
    return "unrecognized character";
  }

  /* The following is a performance optimization.  If the regex begins with
  ** ".*" (if the input regex lacks an initial "^") and afterwards there are
  ** one or more matching characters, enter those matching characters into
  ** zInit[].  The re_match() routine can then search ahead in the input 
  ** string looking for the initial match without having to run the whole
  ** regex engine over the string.  Do not worry able trying to match
  ** unicode characters beyond plane 0 - those are very rare and this is
  ** just an optimization. */
  if( pRe->aOp[0]==RE_OP_ANYSTAR ){
    for(j=0, i=1; j<sizeof(pRe->zInit)-2 && pRe->aOp[i]==RE_OP_MATCH; i++){
      unsigned x = pRe->aArg[i];
      if( x<=127 ){
        pRe->zInit[j++] = x;
      }else if( x<=0xfff ){
        pRe->zInit[j++] = 0xc0 | (x>>6);
        pRe->zInit[j++] = 0x80 | (x&0x3f);
      }else if( x<=0xffff ){
        pRe->zInit[j++] = 0xd0 | (x>>12);
        pRe->zInit[j++] = 0x80 | ((x>>6)&0x3f);
        pRe->zInit[j++] = 0x80 | (x&0x3f);
      }else{
        break;
      }
    }
    if( j>0 && pRe->zInit[j-1]==0 ) j--;
    pRe->nInit = j;
  }