Fossil

Check-in [a13e0a20a8]
Login

Check-in [a13e0a20a8]

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

Overview
Comment:regexp.c (re_next_char): 3 off-by-one errors, and disallow characters > 0x10ffff unicode.c/diff.c: fix some gcc warnings
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: a13e0a20a8752b5c62a6c77c4b43a95c1162af19
User & Date: jan.nijtmans 2013-01-03 13:38:01.387
Context
2013-02-11
19:30
Merging from trunk. I still have merge conflicts since my previous merge. ... (check-in: bdeb633a6c user: viriketo tags: annotate_links)
2013-01-03
20:02
Rename re_execute() to re_match(). Other fixes to regex matcher bugs that were introduced when porting the code from SQLite. ... (check-in: c84051f38d user: drh tags: trunk)
13:38
regexp.c (re_next_char): 3 off-by-one errors, and disallow characters > 0x10ffff unicode.c/diff.c: fix some gcc warnings ... (check-in: a13e0a20a8 user: jan.nijtmans tags: trunk)
09:27
Patched re_space_char() to conform to its docs (now recognizes \r as a space). ... (check-in: c77995e650 user: stephan tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/diff.c.
︙︙
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435

*/
static int re_dline_match(
  ReCompiled *pRe,    /* The regular expression to be matched */
  DLine *aDLine,      /* First of N DLines to compare against */
  int N               /* Number of DLines to check */
){
  while( N-- ){
    if( re_execute(pRe, aDLine->z, LENGTH(aDLine)) ){
      return 1;
    }
    aDLine++;
  }
  return 0;
}









|








421
422
423
424
425
426
427
428
429
430
431
432
433
434
435

*/
static int re_dline_match(
  ReCompiled *pRe,    /* The regular expression to be matched */
  DLine *aDLine,      /* First of N DLines to compare against */
  int N               /* Number of DLines to check */
){
  while( N-- ){
    if( re_execute(pRe, (const unsigned char *)aDLine->z, LENGTH(aDLine)) ){
      return 1;
    }
    aDLine++;
  }
  return 0;
}
︙︙
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937

  Blob *pA_Blob,   /* FROM file */
  Blob *pB_Blob,   /* TO file */
  Blob *pOut,      /* Write diff here if not NULL */
  ReCompiled *pRe, /* Only output changes where this Regexp matches */
  u64 diffFlags    /* DIFF_* flags defined above */
){
  int ignoreEolWs; /* Ignore whitespace at the end of lines */
  int nContext;    /* Amount of context to display */
  DContext c;

  if( diffFlags & DIFF_INVERT ){
    Blob *pTemp = pA_Blob;
    pA_Blob = pB_Blob;
    pB_Blob = pTemp;
  }








<








1923
1924
1925
1926
1927
1928
1929

1930
1931
1932
1933
1934
1935
1936

  Blob *pA_Blob,   /* FROM file */
  Blob *pB_Blob,   /* TO file */
  Blob *pOut,      /* Write diff here if not NULL */
  ReCompiled *pRe, /* Only output changes where this Regexp matches */
  u64 diffFlags    /* DIFF_* flags defined above */
){
  int ignoreEolWs; /* Ignore whitespace at the end of lines */

  DContext c;

  if( diffFlags & DIFF_INVERT ){
    Blob *pTemp = pA_Blob;
    pA_Blob = pB_Blob;
    pB_Blob = pTemp;
  }
︙︙
Changes to src/regexp.c.
︙︙
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160

** 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){








|







|





|








132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160

** 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){
︙︙
Changes to src/unicode.c.
︙︙
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143

    0xFFFFFFFF, 0xFC00FFFF, 0xF8000001, 0xF8000001,
  };

  if( c<128 ){
    return ( (aAscii[c >> 5] & (1 << (c & 0x001F)))==0 );
  }else if( c<(1<<22) ){
    unsigned int key = (((unsigned int)c)<<10) | 0x000003FF;
    int iRes;
    int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
    int iLo = 0;
    while( iHi>=iLo ){
      int iTest = (iHi + iLo) / 2;
      if( key >= aEntry[iTest] ){
        iRes = iTest;
        iLo = iTest+1;








|








129
130
131
132
133
134
135
136
137
138
139
140
141
142
143

    0xFFFFFFFF, 0xFC00FFFF, 0xF8000001, 0xF8000001,
  };

  if( c<128 ){
    return ( (aAscii[c >> 5] & (1 << (c & 0x001F)))==0 );
  }else if( c<(1<<22) ){
    unsigned int key = (((unsigned int)c)<<10) | 0x000003FF;
    int iRes = 0;
    int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
    int iLo = 0;
    while( iHi>=iLo ){
      int iTest = (iHi + iLo) / 2;
      if( key >= aEntry[iTest] ){
        iRes = iTest;
        iLo = iTest+1;
︙︙