/[pcre]/code/tags/pcre-8.33/pcre_compile.c
ViewVC logotype

Diff of /code/tags/pcre-8.33/pcre_compile.c

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1152 by chpe, Sun Oct 21 16:53:51 2012 UTC revision 1305 by ph10, Mon Apr 1 14:50:45 2013 UTC
# Line 80  to check them every time. */ Line 80  to check them every time. */
80  /* Definitions to allow mutual recursion */  /* Definitions to allow mutual recursion */
81    
82  static int  static int
83    add_list_to_class(pcre_uint8 *, pcre_uchar **, int, compile_data *,    add_list_to_class(pcre_uint8 *, pcre_uchar **, int, compile_data *,
84      const pcre_uint32 *, unsigned int);      const pcre_uint32 *, unsigned int);
85    
86  static BOOL  static BOOL
87    compile_regex(int, pcre_uchar **, const pcre_uchar **, int *, BOOL, BOOL, int, int,    compile_regex(int, pcre_uchar **, const pcre_uchar **, int *, BOOL, BOOL, int, int,
88      pcre_uint32 *, pcre_int32 *, pcre_uint32 *, pcre_int32 *, branch_chain *,      pcre_uint32 *, pcre_int32 *, pcre_uint32 *, pcre_int32 *, branch_chain *,
89      compile_data *, int *);      compile_data *, int *);
90    
91    
# Line 667  find_error_text(int n) Line 667  find_error_text(int n)
667  const char *s = error_texts;  const char *s = error_texts;
668  for (; n > 0; n--)  for (; n > 0; n--)
669    {    {
670    while (*s++ != 0) {};    while (*s++ != CHAR_NULL) {};
671    if (*s == 0) return "Error text not found (please report)";    if (*s == CHAR_NULL) return "Error text not found (please report)";
672    }    }
673  return s;  return s;
674  }  }
# Line 773  Returns:         zero => a data characte Line 773  Returns:         zero => a data characte
773  */  */
774    
775  static int  static int
776  check_escape(const pcre_uchar **ptrptr, pcre_uint32 *chptr, int *errorcodeptr,  check_escape(const pcre_uchar **ptrptr, pcre_uint32 *chptr, int *errorcodeptr,
777    int bracount, int options, BOOL isclass)    int bracount, int options, BOOL isclass)
778  {  {
779  /* PCRE_UTF16 has the same value as PCRE_UTF8. */  /* PCRE_UTF16 has the same value as PCRE_UTF8. */
# Line 788  ptr--;                            /* Set Line 788  ptr--;                            /* Set
788    
789  /* If backslash is at the end of the pattern, it's an error. */  /* If backslash is at the end of the pattern, it's an error. */
790    
791  if (c == 0) *errorcodeptr = ERR1;  if (c == CHAR_NULL) *errorcodeptr = ERR1;
792    
793  /* Non-alphanumerics are literals. For digits or letters, do an initial lookup  /* Non-alphanumerics are literals. For digits or letters, do an initial lookup
794  in a table. A non-zero result is something that can be returned immediately.  in a table. A non-zero result is something that can be returned immediately.
# Line 797  Otherwise further processing may be requ Line 797  Otherwise further processing may be requ
797  #ifndef EBCDIC  /* ASCII/UTF-8 coding */  #ifndef EBCDIC  /* ASCII/UTF-8 coding */
798  /* Not alphanumeric */  /* Not alphanumeric */
799  else if (c < CHAR_0 || c > CHAR_z) {}  else if (c < CHAR_0 || c > CHAR_z) {}
800  else if ((i = escapes[c - CHAR_0]) != 0) { if (i > 0) c = (pcre_uint32)i; else escape = -i; }  else if ((i = escapes[c - CHAR_0]) != 0)
801      { if (i > 0) c = (pcre_uint32)i; else escape = -i; }
802    
803  #else           /* EBCDIC coding */  #else           /* EBCDIC coding */
804  /* Not alphanumeric */  /* Not alphanumeric */
# Line 847  else Line 848  else
848            }            }
849    
850  #if defined COMPILE_PCRE8  #if defined COMPILE_PCRE8
851          if (c > (utf ? 0x10ffff : 0xff))          if (c > (utf ? 0x10ffffU : 0xffU))
852  #elif defined COMPILE_PCRE16  #elif defined COMPILE_PCRE16
853          if (c > (utf ? 0x10ffff : 0xffff))          if (c > (utf ? 0x10ffffU : 0xffffU))
854  #elif defined COMPILE_PCRE32  #elif defined COMPILE_PCRE32
855          if (utf && c > 0x10ffff)          if (utf && c > 0x10ffffU)
856  #endif  #endif
857            {            {
858            *errorcodeptr = ERR76;            *errorcodeptr = ERR76;
# Line 898  else Line 899  else
899      if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)      if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
900        {        {
901        const pcre_uchar *p;        const pcre_uchar *p;
902        for (p = ptr+2; *p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET; p++)        for (p = ptr+2; *p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET; p++)
903          if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break;          if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break;
904        if (*p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET)        if (*p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET)
905          {          {
906          escape = ESC_k;          escape = ESC_k;
907          break;          break;
# Line 1085  else Line 1086  else
1086  #endif  #endif
1087    
1088  #if defined COMPILE_PCRE8  #if defined COMPILE_PCRE8
1089          if (c > (utf ? 0x10ffff : 0xff)) { overflow = TRUE; break; }          if (c > (utf ? 0x10ffffU : 0xffU)) { overflow = TRUE; break; }
1090  #elif defined COMPILE_PCRE16  #elif defined COMPILE_PCRE16
1091          if (c > (utf ? 0x10ffff : 0xffff)) { overflow = TRUE; break; }          if (c > (utf ? 0x10ffffU : 0xffffU)) { overflow = TRUE; break; }
1092  #elif defined COMPILE_PCRE32  #elif defined COMPILE_PCRE32
1093          if (utf && c > 0x10ffff) { overflow = TRUE; break; }          if (utf && c > 0x10ffffU) { overflow = TRUE; break; }
1094  #endif  #endif
1095          }          }
1096    
# Line 1134  else Line 1135  else
1135    
1136      case CHAR_c:      case CHAR_c:
1137      c = *(++ptr);      c = *(++ptr);
1138      if (c == 0)      if (c == CHAR_NULL)
1139        {        {
1140        *errorcodeptr = ERR2;        *errorcodeptr = ERR2;
1141        break;        break;
# Line 1211  Returns:         TRUE if the type value Line 1212  Returns:         TRUE if the type value
1212  */  */
1213    
1214  static BOOL  static BOOL
1215  get_ucp(const pcre_uchar **ptrptr, BOOL *negptr, unsigned int *ptypeptr,  get_ucp(const pcre_uchar **ptrptr, BOOL *negptr, unsigned int *ptypeptr,
1216    unsigned int *pdataptr, int *errorcodeptr)    unsigned int *pdataptr, int *errorcodeptr)
1217  {  {
1218  pcre_uchar c;  pcre_uchar c;
# Line 1220  const pcre_uchar *ptr = *ptrptr; Line 1221  const pcre_uchar *ptr = *ptrptr;
1221  pcre_uchar name[32];  pcre_uchar name[32];
1222    
1223  c = *(++ptr);  c = *(++ptr);
1224  if (c == 0) goto ERROR_RETURN;  if (c == CHAR_NULL) goto ERROR_RETURN;
1225    
1226  *negptr = FALSE;  *negptr = FALSE;
1227    
# Line 1237  if (c == CHAR_LEFT_CURLY_BRACKET) Line 1238  if (c == CHAR_LEFT_CURLY_BRACKET)
1238    for (i = 0; i < (int)(sizeof(name) / sizeof(pcre_uchar)) - 1; i++)    for (i = 0; i < (int)(sizeof(name) / sizeof(pcre_uchar)) - 1; i++)
1239      {      {
1240      c = *(++ptr);      c = *(++ptr);
1241      if (c == 0) goto ERROR_RETURN;      if (c == CHAR_NULL) goto ERROR_RETURN;
1242      if (c == CHAR_RIGHT_CURLY_BRACKET) break;      if (c == CHAR_RIGHT_CURLY_BRACKET) break;
1243      name[i] = c;      name[i] = c;
1244      }      }
# Line 1408  if (ptr[0] == CHAR_LEFT_PARENTHESIS) Line 1409  if (ptr[0] == CHAR_LEFT_PARENTHESIS)
1409    {    {
1410    /* Handle specials such as (*SKIP) or (*UTF8) etc. */    /* Handle specials such as (*SKIP) or (*UTF8) etc. */
1411    
1412    if (ptr[1] == CHAR_ASTERISK) ptr += 2;    if (ptr[1] == CHAR_ASTERISK)
1413        {
1414        ptr += 2;
1415        while (ptr < cd->end_pattern && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
1416        }
1417    
1418    /* Handle a normal, unnamed capturing parenthesis. */    /* Handle a normal, unnamed capturing parenthesis. */
1419    
# Line 1432  if (ptr[0] == CHAR_LEFT_PARENTHESIS) Line 1437  if (ptr[0] == CHAR_LEFT_PARENTHESIS)
1437    
1438    else if (ptr[2] == CHAR_NUMBER_SIGN)    else if (ptr[2] == CHAR_NUMBER_SIGN)
1439      {      {
1440      for (ptr += 3; *ptr != 0; ptr++) if (*ptr == CHAR_RIGHT_PARENTHESIS) break;      for (ptr += 3; *ptr != CHAR_NULL; ptr++)
1441          if (*ptr == CHAR_RIGHT_PARENTHESIS) break;
1442      goto FAIL_EXIT;      goto FAIL_EXIT;
1443      }      }
1444    
# Line 1445  if (ptr[0] == CHAR_LEFT_PARENTHESIS) Line 1451  if (ptr[0] == CHAR_LEFT_PARENTHESIS)
1451      ptr += 2;      ptr += 2;
1452      if (ptr[1] != CHAR_QUESTION_MARK)      if (ptr[1] != CHAR_QUESTION_MARK)
1453        {        {
1454        while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;        while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
1455        if (*ptr != 0) ptr++;        if (*ptr != CHAR_NULL) ptr++;
1456        }        }
1457      }      }
1458    
# Line 1489  for (; ptr < cd->end_pattern; ptr++) Line 1495  for (; ptr < cd->end_pattern; ptr++)
1495    
1496    if (*ptr == CHAR_BACKSLASH)    if (*ptr == CHAR_BACKSLASH)
1497      {      {
1498      if (*(++ptr) == 0) goto FAIL_EXIT;      if (*(++ptr) == CHAR_NULL) goto FAIL_EXIT;
1499      if (*ptr == CHAR_Q) for (;;)      if (*ptr == CHAR_Q) for (;;)
1500        {        {
1501        while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};        while (*(++ptr) != CHAR_NULL && *ptr != CHAR_BACKSLASH) {};
1502        if (*ptr == 0) goto FAIL_EXIT;        if (*ptr == CHAR_NULL) goto FAIL_EXIT;
1503        if (*(++ptr) == CHAR_E) break;        if (*(++ptr) == CHAR_E) break;
1504        }        }
1505      continue;      continue;
# Line 1537  for (; ptr < cd->end_pattern; ptr++) Line 1543  for (; ptr < cd->end_pattern; ptr++)
1543    
1544      while (*(++ptr) != CHAR_RIGHT_SQUARE_BRACKET)      while (*(++ptr) != CHAR_RIGHT_SQUARE_BRACKET)
1545        {        {
1546        if (*ptr == 0) return -1;        if (*ptr == CHAR_NULL) return -1;
1547        if (*ptr == CHAR_BACKSLASH)        if (*ptr == CHAR_BACKSLASH)
1548          {          {
1549          if (*(++ptr) == 0) goto FAIL_EXIT;          if (*(++ptr) == CHAR_NULL) goto FAIL_EXIT;
1550          if (*ptr == CHAR_Q) for (;;)          if (*ptr == CHAR_Q) for (;;)
1551            {            {
1552            while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};            while (*(++ptr) != CHAR_NULL && *ptr != CHAR_BACKSLASH) {};
1553            if (*ptr == 0) goto FAIL_EXIT;            if (*ptr == CHAR_NULL) goto FAIL_EXIT;
1554            if (*(++ptr) == CHAR_E) break;            if (*(++ptr) == CHAR_E) break;
1555            }            }
1556          continue;          continue;
# Line 1558  for (; ptr < cd->end_pattern; ptr++) Line 1564  for (; ptr < cd->end_pattern; ptr++)
1564    if (xmode && *ptr == CHAR_NUMBER_SIGN)    if (xmode && *ptr == CHAR_NUMBER_SIGN)
1565      {      {
1566      ptr++;      ptr++;
1567      while (*ptr != 0)      while (*ptr != CHAR_NULL)
1568        {        {
1569        if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }        if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
1570        ptr++;        ptr++;
# Line 1566  for (; ptr < cd->end_pattern; ptr++) Line 1572  for (; ptr < cd->end_pattern; ptr++)
1572        if (utf) FORWARDCHAR(ptr);        if (utf) FORWARDCHAR(ptr);
1573  #endif  #endif
1574        }        }
1575      if (*ptr == 0) goto FAIL_EXIT;      if (*ptr == CHAR_NULL) goto FAIL_EXIT;
1576      continue;      continue;
1577      }      }
1578    
# Line 1576  for (; ptr < cd->end_pattern; ptr++) Line 1582  for (; ptr < cd->end_pattern; ptr++)
1582      {      {
1583      int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, count);      int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, count);
1584      if (rc > 0) return rc;      if (rc > 0) return rc;
1585      if (*ptr == 0) goto FAIL_EXIT;      if (*ptr == CHAR_NULL) goto FAIL_EXIT;
1586      }      }
1587    
1588    else if (*ptr == CHAR_RIGHT_PARENTHESIS)    else if (*ptr == CHAR_RIGHT_PARENTHESIS)
# Line 1641  matching closing parens. That is why we Line 1647  matching closing parens. That is why we
1647  for (;;)  for (;;)
1648    {    {
1649    rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, &count);    rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, &count);
1650    if (rc > 0 || *ptr++ == 0) break;    if (rc > 0 || *ptr++ == CHAR_NULL) break;
1651    }    }
1652    
1653  return rc;  return rc;
# Line 1878  for (;;) Line 1884  for (;;)
1884    
1885      case OP_TYPEEXACT:      case OP_TYPEEXACT:
1886      branchlength += GET2(cc,1);      branchlength += GET2(cc,1);
1887      if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP)      if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP)
1888        cc += 2;        cc += 2;
1889      cc += 1 + IMM2_SIZE + 1;      cc += 1 + IMM2_SIZE + 1;
1890      break;      break;
# Line 2121  for (;;) Line 2127  for (;;)
2127        case OP_TYPEMINUPTO:        case OP_TYPEMINUPTO:
2128        case OP_TYPEEXACT:        case OP_TYPEEXACT:
2129        case OP_TYPEPOSUPTO:        case OP_TYPEPOSUPTO:
2130        if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)        if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2131          code += 2;          code += 2;
2132        break;        break;
2133    
2134        case OP_MARK:        case OP_MARK:
2135        case OP_PRUNE_ARG:        case OP_PRUNE_ARG:
2136        case OP_SKIP_ARG:        case OP_SKIP_ARG:
       code += code[1];  
       break;  
   
2137        case OP_THEN_ARG:        case OP_THEN_ARG:
2138        code += code[1];        code += code[1];
2139        break;        break;
# Line 2241  for (;;) Line 2244  for (;;)
2244        case OP_TYPEUPTO:        case OP_TYPEUPTO:
2245        case OP_TYPEMINUPTO:        case OP_TYPEMINUPTO:
2246        case OP_TYPEEXACT:        case OP_TYPEEXACT:
2247        if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)        if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2248          code += 2;          code += 2;
2249        break;        break;
2250    
2251        case OP_MARK:        case OP_MARK:
2252        case OP_PRUNE_ARG:        case OP_PRUNE_ARG:
2253        case OP_SKIP_ARG:        case OP_SKIP_ARG:
       code += code[1];  
       break;  
   
2254        case OP_THEN_ARG:        case OP_THEN_ARG:
2255        code += code[1];        code += code[1];
2256        break;        break;
# Line 2567  for (code = first_significant_code(code Line 2567  for (code = first_significant_code(code
2567      case OP_TYPEUPTO:      case OP_TYPEUPTO:
2568      case OP_TYPEMINUPTO:      case OP_TYPEMINUPTO:
2569      case OP_TYPEPOSUPTO:      case OP_TYPEPOSUPTO:
2570      if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)      if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2571        code += 2;        code += 2;
2572      break;      break;
2573    
# Line 2615  for (code = first_significant_code(code Line 2615  for (code = first_significant_code(code
2615      case OP_MARK:      case OP_MARK:
2616      case OP_PRUNE_ARG:      case OP_PRUNE_ARG:
2617      case OP_SKIP_ARG:      case OP_SKIP_ARG:
     code += code[1];  
     break;  
   
2618      case OP_THEN_ARG:      case OP_THEN_ARG:
2619      code += code[1];      code += code[1];
2620      break;      break;
# Line 2717  check_posix_syntax(const pcre_uchar *ptr Line 2714  check_posix_syntax(const pcre_uchar *ptr
2714  {  {
2715  pcre_uchar terminator;          /* Don't combine these lines; the Solaris cc */  pcre_uchar terminator;          /* Don't combine these lines; the Solaris cc */
2716  terminator = *(++ptr);   /* compiler warns about "non-constant" initializer. */  terminator = *(++ptr);   /* compiler warns about "non-constant" initializer. */
2717  for (++ptr; *ptr != 0; ptr++)  for (++ptr; *ptr != CHAR_NULL; ptr++)
2718    {    {
2719    if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)    if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
2720      ptr++;      ptr++;
# Line 2902  PUT(previous_callout, 2 + LINK_SIZE, len Line 2899  PUT(previous_callout, 2 + LINK_SIZE, len
2899  /* This function is passed the start and end of a class range, in UTF-8 mode  /* This function is passed the start and end of a class range, in UTF-8 mode
2900  with UCP support. It searches up the characters, looking for ranges of  with UCP support. It searches up the characters, looking for ranges of
2901  characters in the "other" case. Each call returns the next one, updating the  characters in the "other" case. Each call returns the next one, updating the
2902  start address. A character with multiple other cases is returned on its own  start address. A character with multiple other cases is returned on its own
2903  with a special return value.  with a special return value.
2904    
2905  Arguments:  Arguments:
# Line 2914  Arguments: Line 2911  Arguments:
2911  Yield:        -1 when no more  Yield:        -1 when no more
2912                 0 when a range is returned                 0 when a range is returned
2913                >0 the CASESET offset for char with multiple other cases                >0 the CASESET offset for char with multiple other cases
2914                  in this case, ocptr contains the original                  in this case, ocptr contains the original
2915  */  */
2916    
2917  static int  static int
# Line 2924  get_othercase_range(pcre_uint32 *cptr, p Line 2921  get_othercase_range(pcre_uint32 *cptr, p
2921  pcre_uint32 c, othercase, next;  pcre_uint32 c, othercase, next;
2922  unsigned int co;  unsigned int co;
2923    
2924  /* Find the first character that has an other case. If it has multiple other  /* Find the first character that has an other case. If it has multiple other
2925  cases, return its case offset value. */  cases, return its case offset value. */
2926    
2927  for (c = *cptr; c <= d; c++)  for (c = *cptr; c <= d; c++)
2928    {    {
2929    if ((co = UCD_CASESET(c)) != 0)    if ((co = UCD_CASESET(c)) != 0)
2930      {      {
2931      *ocptr = c++;   /* Character that has the set */      *ocptr = c++;   /* Character that has the set */
2932      *cptr = c;      /* Rest of input range */      *cptr = c;      /* Rest of input range */
2933      return (int)co;      return (int)co;
2934      }      }
2935    if ((othercase = UCD_OTHERCASE(c)) != c) break;    if ((othercase = UCD_OTHERCASE(c)) != c) break;
2936    }    }
2937    
2938  if (c > d) return -1;  /* Reached end of range */  if (c > d) return -1;  /* Reached end of range */
# Line 3018  switch(ptype) Line 3015  switch(ptype)
3015    return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||    return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
3016            PRIV(ucp_gentype)[prop->chartype] == ucp_N ||            PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
3017            c == CHAR_UNDERSCORE) == negated;            c == CHAR_UNDERSCORE) == negated;
3018    
3019  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3020    case PT_CLIST:    case PT_CLIST:
3021    p = PRIV(ucd_caseless_sets) + prop->caseset;    p = PRIV(ucd_caseless_sets) + prop->caseset;
3022    for (;;)    for (;;)
3023      {      {
3024      if ((unsigned int)c < *p) return !negated;      if (c < *p) return !negated;
3025      if ((unsigned int)c == *p++) return negated;      if (c == *p++) return negated;
3026      }      }
3027    break;  /* Control never reaches here */    break;  /* Control never reaches here */
3028  #endif  #endif
# Line 3074  if ((options & PCRE_EXTENDED) != 0) Line 3071  if ((options & PCRE_EXTENDED) != 0)
3071      if (*ptr == CHAR_NUMBER_SIGN)      if (*ptr == CHAR_NUMBER_SIGN)
3072        {        {
3073        ptr++;        ptr++;
3074        while (*ptr != 0)        while (*ptr != CHAR_NULL)
3075          {          {
3076          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
3077          ptr++;          ptr++;
# Line 3093  value is a character, a negative value i Line 3090  value is a character, a negative value i
3090  if (*ptr == CHAR_BACKSLASH)  if (*ptr == CHAR_BACKSLASH)
3091    {    {
3092    int temperrorcode = 0;    int temperrorcode = 0;
3093    escape = check_escape(&ptr, &next, &temperrorcode, cd->bracount, options, FALSE);    escape = check_escape(&ptr, &next, &temperrorcode, cd->bracount, options,
3094        FALSE);
3095    if (temperrorcode != 0) return FALSE;    if (temperrorcode != 0) return FALSE;
3096    ptr++;    /* Point after the escape sequence */    ptr++;    /* Point after the escape sequence */
3097    }    }
# Line 3117  if ((options & PCRE_EXTENDED) != 0) Line 3115  if ((options & PCRE_EXTENDED) != 0)
3115      if (*ptr == CHAR_NUMBER_SIGN)      if (*ptr == CHAR_NUMBER_SIGN)
3116        {        {
3117        ptr++;        ptr++;
3118        while (*ptr != 0)        while (*ptr != CHAR_NULL)
3119          {          {
3120          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
3121          ptr++;          ptr++;
# Line 3138  if (*ptr == CHAR_ASTERISK || *ptr == CHA Line 3136  if (*ptr == CHAR_ASTERISK || *ptr == CHA
3136    
3137  /* If the previous item is a character, get its value. */  /* If the previous item is a character, get its value. */
3138    
3139  if (op_code == OP_CHAR || op_code == OP_CHARI ||  if (op_code == OP_CHAR || op_code == OP_CHARI ||
3140      op_code == OP_NOT || op_code == OP_NOTI)      op_code == OP_NOT || op_code == OP_NOTI)
3141    {    {
3142  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
# Line 3155  if (escape == 0) Line 3153  if (escape == 0)
3153    {    {
3154    /* For a caseless UTF match, the next character may have more than one other    /* For a caseless UTF match, the next character may have more than one other
3155    case, which maps to the special PT_CLIST property. Check this first. */    case, which maps to the special PT_CLIST property. Check this first. */
3156    
3157  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3158    if (utf && c != NOTACHAR && (options & PCRE_CASELESS) != 0)    if (utf && c != NOTACHAR && (options & PCRE_CASELESS) != 0)
3159      {      {
# Line 3171  if (escape == 0) Line 3169  if (escape == 0)
3169    
3170      /* For CHARI (caseless character) we must check the other case. If we have      /* For CHARI (caseless character) we must check the other case. If we have
3171      Unicode property support, we can use it to test the other case of      Unicode property support, we can use it to test the other case of
3172      high-valued characters. We know that next can have only one other case,      high-valued characters. We know that next can have only one other case,
3173      because multi-other-case characters are dealt with above. */      because multi-other-case characters are dealt with above. */
3174    
3175      case OP_CHARI:      case OP_CHARI:
# Line 3191  if (escape == 0) Line 3189  if (escape == 0)
3189      else      else
3190  #endif  /* SUPPORT_UTF */  #endif  /* SUPPORT_UTF */
3191      return (c != TABLE_GET(next, cd->fcc, next));  /* Not UTF */      return (c != TABLE_GET(next, cd->fcc, next));  /* Not UTF */
3192    
3193      case OP_NOT:      case OP_NOT:
3194      return c == next;      return c == next;
3195    
3196      case OP_NOTI:      case OP_NOTI:
3197      if (c == next) return TRUE;      if (c == next) return TRUE;
3198  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
# Line 3238  if (escape == 0) Line 3236  if (escape == 0)
3236      case OP_NOT_HSPACE:      case OP_NOT_HSPACE:
3237      switch(next)      switch(next)
3238        {        {
3239        HSPACE_CASES:        HSPACE_CASES:
3240        return op_code == OP_NOT_HSPACE;        return op_code == OP_NOT_HSPACE;
3241    
3242        default:        default:
# Line 3250  if (escape == 0) Line 3248  if (escape == 0)
3248      case OP_NOT_VSPACE:      case OP_NOT_VSPACE:
3249      switch(next)      switch(next)
3250        {        {
3251        VSPACE_CASES:        VSPACE_CASES:
3252        return op_code == OP_NOT_VSPACE;        return op_code == OP_NOT_VSPACE;
3253    
3254        default:        default:
# Line 3304  switch(op_code) Line 3302  switch(op_code)
3302      case ESC_H:      case ESC_H:
3303      switch(c)      switch(c)
3304        {        {
3305        HSPACE_CASES:        HSPACE_CASES:
3306        return escape != ESC_h;        return escape != ESC_h;
3307    
3308        default:        default:
3309        return escape == ESC_h;        return escape == ESC_h;
3310        }        }
# Line 3315  switch(op_code) Line 3313  switch(op_code)
3313      case ESC_V:      case ESC_V:
3314      switch(c)      switch(c)
3315        {        {
3316        VSPACE_CASES:        VSPACE_CASES:
3317        return escape != ESC_v;        return escape != ESC_v;
3318    
3319        default:        default:
# Line 3427  switch(op_code) Line 3425  switch(op_code)
3425  *************************************************/  *************************************************/
3426    
3427  /* This function packages up the logic of adding a character or range of  /* This function packages up the logic of adding a character or range of
3428  characters to a class. The character values in the arguments will be within the  characters to a class. The character values in the arguments will be within the
3429  valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is  valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is
3430  mutually recursive with the function immediately below.  mutually recursive with the function immediately below.
3431    
3432  Arguments:  Arguments:
3433    classbits     the bit map for characters < 256    classbits     the bit map for characters < 256
3434    uchardptr     points to the pointer for extra data    uchardptr     points to the pointer for extra data
3435    options       the options word    options       the options word
3436    cd            contains pointers to tables etc.    cd            contains pointers to tables etc.
3437    start         start of range character    start         start of range character
3438    end           end of range character    end           end of range character
3439    
3440  Returns:        the number of < 256 characters added  Returns:        the number of < 256 characters added
3441                  the pointer to extra data is updated                  the pointer to extra data is updated
3442  */  */
# Line 3450  add_to_class(pcre_uint8 *classbits, pcre Line 3448  add_to_class(pcre_uint8 *classbits, pcre
3448  pcre_uint32 c;  pcre_uint32 c;
3449  int n8 = 0;  int n8 = 0;
3450    
3451  /* If caseless matching is required, scan the range and process alternate  /* If caseless matching is required, scan the range and process alternate
3452  cases. In Unicode, there are 8-bit characters that have alternate cases that  cases. In Unicode, there are 8-bit characters that have alternate cases that
3453  are greater than 255 and vice-versa. Sometimes we can just extend the original  are greater than 255 and vice-versa. Sometimes we can just extend the original
3454  range. */  range. */
3455    
3456  if ((options & PCRE_CASELESS) != 0)  if ((options & PCRE_CASELESS) != 0)
3457    {    {
3458  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
3459    if ((options & PCRE_UTF8) != 0)    if ((options & PCRE_UTF8) != 0)
3460      {      {
3461      int rc;      int rc;
3462      pcre_uint32 oc, od;      pcre_uint32 oc, od;
3463    
3464      options &= ~PCRE_CASELESS;   /* Remove for recursive calls */      options &= ~PCRE_CASELESS;   /* Remove for recursive calls */
3465      c = start;      c = start;
3466    
3467      while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0)      while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0)
3468        {        {
3469        /* Handle a single character that has more than one other case. */        /* Handle a single character that has more than one other case. */
3470    
3471        if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd,        if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd,
3472          PRIV(ucd_caseless_sets) + rc, oc);          PRIV(ucd_caseless_sets) + rc, oc);
3473    
3474        /* Do nothing if the other case range is within the original range. */        /* Do nothing if the other case range is within the original range. */
3475    
3476        else if (oc >= start && od <= end) continue;        else if (oc >= start && od <= end) continue;
3477    
3478        /* Extend the original range if there is overlap, noting that if oc < c, we        /* Extend the original range if there is overlap, noting that if oc < c, we
3479        can't have od > end because a subrange is always shorter than the basic        can't have od > end because a subrange is always shorter than the basic
3480        range. Otherwise, use a recursive call to add the additional range. */        range. Otherwise, use a recursive call to add the additional range. */
3481    
3482        else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */        else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */
3483        else if (od > end && oc <= end + 1) end = od;       /* Extend upwards */        else if (od > end && oc <= end + 1) end = od;       /* Extend upwards */
3484        else n8 += add_to_class(classbits, uchardptr, options, cd, oc, od);        else n8 += add_to_class(classbits, uchardptr, options, cd, oc, od);
# Line 3490  if ((options & PCRE_CASELESS) != 0) Line 3488  if ((options & PCRE_CASELESS) != 0)
3488  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
3489    
3490    /* Not UTF-mode, or no UCP */    /* Not UTF-mode, or no UCP */
3491    
3492    for (c = start; c <= end && c < 256; c++)    for (c = start; c <= end && c < 256; c++)
3493      {      {
3494      SETBIT(classbits, cd->fcc[c]);      SETBIT(classbits, cd->fcc[c]);
3495      n8++;      n8++;
3496      }      }
3497    }    }
3498    
3499  /* Now handle the original range. Adjust the final value according to the bit  /* Now handle the original range. Adjust the final value according to the bit
3500  length - this means that the same lists of (e.g.) horizontal spaces can be used  length - this means that the same lists of (e.g.) horizontal spaces can be used
3501  in all cases. */  in all cases. */
# Line 3523  if (end < 0x100) Line 3521  if (end < 0x100)
3521    {    {
3522    for (c = start; c <= end; c++)    for (c = start; c <= end; c++)
3523      {      {
3524      n8++;      n8++;
3525      SETBIT(classbits, c);      SETBIT(classbits, c);
3526      }      }
3527    }    }
3528    
3529  else  else
3530    {    {
3531    pcre_uchar *uchardata = *uchardptr;    pcre_uchar *uchardata = *uchardptr;
3532    
3533  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3534    if ((options & PCRE_UTF8) != 0)  /* All UTFs use the same flag bit */    if ((options & PCRE_UTF8) != 0)  /* All UTFs use the same flag bit */
3535      {      {
3536      if (start < end)      if (start < end)
3537        {        {
3538        *uchardata++ = XCL_RANGE;        *uchardata++ = XCL_RANGE;
3539        uchardata += PRIV(ord2utf)(start, uchardata);        uchardata += PRIV(ord2utf)(start, uchardata);
3540        uchardata += PRIV(ord2utf)(end, uchardata);        uchardata += PRIV(ord2utf)(end, uchardata);
3541        }        }
3542      else if (start == end)      else if (start == end)
3543        {        {
3544        *uchardata++ = XCL_SINGLE;        *uchardata++ = XCL_SINGLE;
3545        uchardata += PRIV(ord2utf)(start, uchardata);        uchardata += PRIV(ord2utf)(start, uchardata);
3546        }        }
3547      }      }
3548    else    else
3549  #endif  /* SUPPORT_UTF */  #endif  /* SUPPORT_UTF */
3550    
3551    /* Without UTF support, character values are constrained by the bit length,    /* Without UTF support, character values are constrained by the bit length,
3552    and can only be > 256 for 16-bit and 32-bit libraries. */    and can only be > 256 for 16-bit and 32-bit libraries. */
3553    
3554  #ifdef COMPILE_PCRE8  #ifdef COMPILE_PCRE8
3555      {}      {}
3556  #else  #else
3557    if (start < end)    if (start < end)
3558      {      {
3559      *uchardata++ = XCL_RANGE;      *uchardata++ = XCL_RANGE;
# Line 3566  else Line 3564  else
3564      {      {
3565      *uchardata++ = XCL_SINGLE;      *uchardata++ = XCL_SINGLE;
3566      *uchardata++ = start;      *uchardata++ = start;
3567      }      }
3568  #endif  #endif
3569    
3570    *uchardptr = uchardata;   /* Updata extra data pointer */    *uchardptr = uchardata;   /* Updata extra data pointer */
3571    }    }
3572    
3573  return n8;    /* Number of 8-bit characters */  return n8;    /* Number of 8-bit characters */
3574  }  }
3575    
3576    
3577    
3578    
3579  /*************************************************  /*************************************************
3580  *        Add a list of characters to a class     *  *        Add a list of characters to a class     *
3581  *************************************************/  *************************************************/
3582    
3583  /* This function is used for adding a list of case-equivalent characters to a  /* This function is used for adding a list of case-equivalent characters to a
3584  class, and also for adding a list of horizontal or vertical whitespace. If the  class, and also for adding a list of horizontal or vertical whitespace. If the
3585  list is in order (which it should be), ranges of characters are detected and  list is in order (which it should be), ranges of characters are detected and
3586  handled appropriately. This function is mutually recursive with the function  handled appropriately. This function is mutually recursive with the function
# Line 3592  Arguments: Line 3590  Arguments:
3590    classbits     the bit map for characters < 256    classbits     the bit map for characters < 256
3591    uchardptr     points to the pointer for extra data    uchardptr     points to the pointer for extra data
3592    options       the options word    options       the options word
3593    cd            contains pointers to tables etc.    cd            contains pointers to tables etc.
3594    p             points to row of 32-bit values, terminated by NOTACHAR    p             points to row of 32-bit values, terminated by NOTACHAR
3595    except        character to omit; this is used when adding lists of    except        character to omit; this is used when adding lists of
3596                    case-equivalent characters to avoid including the one we                    case-equivalent characters to avoid including the one we
3597                    already know about                    already know about
3598    
3599  Returns:        the number of < 256 characters added  Returns:        the number of < 256 characters added
3600                  the pointer to extra data is updated                  the pointer to extra data is updated
3601  */  */
# Line 3611  while (p[0] < NOTACHAR) Line 3609  while (p[0] < NOTACHAR)
3609    {    {
3610    int n = 0;    int n = 0;
3611    if (p[0] != except)    if (p[0] != except)
3612      {      {
3613      while(p[n+1] == p[0] + n + 1) n++;      while(p[n+1] == p[0] + n + 1) n++;
3614      n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]);      n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]);
3615      }      }
3616    p += n + 1;    p += n + 1;
3617    }    }
3618  return n8;  return n8;
3619  }  }
3620    
3621    
3622    
# Line 3633  Arguments: Line 3631  Arguments:
3631    classbits     the bit map for characters < 256    classbits     the bit map for characters < 256
3632    uchardptr     points to the pointer for extra data    uchardptr     points to the pointer for extra data
3633    options       the options word    options       the options word
3634    cd            contains pointers to tables etc.    cd            contains pointers to tables etc.
3635    p             points to row of 32-bit values, terminated by NOTACHAR    p             points to row of 32-bit values, terminated by NOTACHAR
3636    
3637  Returns:        the number of < 256 characters added  Returns:        the number of < 256 characters added
3638                  the pointer to extra data is updated                  the pointer to extra data is updated
3639  */  */
3640    
3641  static int  static int
3642  add_not_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr,  add_not_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr,
3643    int options, compile_data *cd, const pcre_uint32 *p)    int options, compile_data *cd, const pcre_uint32 *p)
3644  {  {
3645  BOOL utf = (options & PCRE_UTF8) != 0;  BOOL utf = (options & PCRE_UTF8) != 0;
# Line 3653  while (p[0] < NOTACHAR) Line 3651  while (p[0] < NOTACHAR)
3651    while (p[1] == p[0] + 1) p++;    while (p[1] == p[0] + 1) p++;
3652    n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1,    n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1,
3653      (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1);      (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1);
3654    p++;    p++;
3655    }    }
3656  return n8;  return n8;
3657  }  }
3658    
3659    
3660    
# Line 3740  BOOL utf = FALSE; Line 3738  BOOL utf = FALSE;
3738  #endif  #endif
3739    
3740  /* Helper variables for OP_XCLASS opcode (for characters > 255). We define  /* Helper variables for OP_XCLASS opcode (for characters > 255). We define
3741  class_uchardata always so that it can be passed to add_to_class() always,  class_uchardata always so that it can be passed to add_to_class() always,
3742  though it will not be used in non-UTF 8-bit cases. This avoids having to supply  though it will not be used in non-UTF 8-bit cases. This avoids having to supply
3743  alternative calls for the different cases. */  alternative calls for the different cases. */
3744    
3745  pcre_uchar *class_uchardata;  pcre_uchar *class_uchardata;
# Line 3811  for (;; ptr++) Line 3809  for (;; ptr++)
3809    /* If we are at the end of a nested substitution, revert to the outer level    /* If we are at the end of a nested substitution, revert to the outer level
3810    string. Nesting only happens one level deep. */    string. Nesting only happens one level deep. */
3811    
3812    if (c == 0 && nestptr != NULL)    if (c == CHAR_NULL && nestptr != NULL)
3813      {      {
3814      ptr = nestptr;      ptr = nestptr;
3815      nestptr = NULL;      nestptr = NULL;
# Line 3886  for (;; ptr++) Line 3884  for (;; ptr++)
3884    
3885    /* If in \Q...\E, check for the end; if not, we have a literal */    /* If in \Q...\E, check for the end; if not, we have a literal */
3886    
3887    if (inescq && c != 0)    if (inescq && c != CHAR_NULL)
3888      {      {
3889      if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)      if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)
3890        {        {
# Line 3934  for (;; ptr++) Line 3932  for (;; ptr++)
3932      if (c == CHAR_NUMBER_SIGN)      if (c == CHAR_NUMBER_SIGN)
3933        {        {
3934        ptr++;        ptr++;
3935        while (*ptr != 0)        while (*ptr != CHAR_NULL)
3936          {          {
3937          if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
3938          ptr++;          ptr++;
# Line 3942  for (;; ptr++) Line 3940  for (;; ptr++)
3940          if (utf) FORWARDCHAR(ptr);          if (utf) FORWARDCHAR(ptr);
3941  #endif  #endif
3942          }          }
3943        if (*ptr != 0) continue;        if (*ptr != CHAR_NULL) continue;
3944    
3945        /* Else fall through to handle end of string */        /* Else fall through to handle end of string */
3946        c = 0;        c = 0;
# Line 4120  for (;; ptr++) Line 4118  for (;; ptr++)
4118      means that an initial ] is taken as a data character. At the start of the      means that an initial ] is taken as a data character. At the start of the
4119      loop, c contains the first byte of the character. */      loop, c contains the first byte of the character. */
4120    
4121      if (c != 0) do      if (c != CHAR_NULL) do
4122        {        {
4123        const pcre_uchar *oldptr;        const pcre_uchar *oldptr;
4124    
# Line 4135  for (;; ptr++) Line 4133  for (;; ptr++)
4133        /* In the pre-compile phase, accumulate the length of any extra        /* In the pre-compile phase, accumulate the length of any extra
4134        data and reset the pointer. This is so that very large classes that        data and reset the pointer. This is so that very large classes that
4135        contain a zillion > 255 characters no longer overwrite the work space        contain a zillion > 255 characters no longer overwrite the work space
4136        (which is on the stack). We have to remember that there was XCLASS data,        (which is on the stack). We have to remember that there was XCLASS data,
4137        however. */        however. */
4138    
4139        if (lengthptr != NULL && class_uchardata > class_uchardata_base)        if (lengthptr != NULL && class_uchardata > class_uchardata_base)
# Line 4199  for (;; ptr++) Line 4197  for (;; ptr++)
4197          alpha. This relies on the fact that the class table starts with          alpha. This relies on the fact that the class table starts with
4198          alpha, lower, upper as the first 3 entries. */          alpha, lower, upper as the first 3 entries. */
4199    
4200          if ((options & PCRE_CASELESS) != 0 && posix_class <= 2)          if ((options & PCRE_CASELESS) != 0 && posix_class <= 2)
4201            posix_class = 0;            posix_class = 0;
4202    
4203          /* When PCRE_UCP is set, some of the POSIX classes are converted to          /* When PCRE_UCP is set, some of the POSIX classes are converted to
# Line 4276  for (;; ptr++) Line 4274  for (;; ptr++)
4274    
4275        if (c == CHAR_BACKSLASH)        if (c == CHAR_BACKSLASH)
4276          {          {
4277          escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, TRUE);          escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options,
4278              TRUE);
4279          if (*errorcodeptr != 0) goto FAILED;          if (*errorcodeptr != 0) goto FAILED;
4280            if (escape == 0) c = ec;
         if (escape == 0)  
           c = ec;  
4281          else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */          else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */
4282          else if (escape == ESC_N)            /* \N is not supported in a class */          else if (escape == ESC_N)          /* \N is not supported in a class */
4283            {            {
4284            *errorcodeptr = ERR71;            *errorcodeptr = ERR71;
4285            goto FAILED;            goto FAILED;
# Line 4355  for (;; ptr++) Line 4351  for (;; ptr++)
4351              for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space];              for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space];
4352              classbits[1] |= 0x08;    /* Perl 5.004 onwards omits VT from \s */              classbits[1] |= 0x08;    /* Perl 5.004 onwards omits VT from \s */
4353              continue;              continue;
4354    
4355              /* The rest apply in both UCP and non-UCP cases. */              /* The rest apply in both UCP and non-UCP cases. */
4356    
4357              case ESC_h:              case ESC_h:
4358              (void)add_list_to_class(classbits, &class_uchardata, options, cd,              (void)add_list_to_class(classbits, &class_uchardata, options, cd,
4359                PRIV(hspace_list), NOTACHAR);                PRIV(hspace_list), NOTACHAR);
4360              continue;              continue;
4361    
4362              case ESC_H:              case ESC_H:
4363              (void)add_not_list_to_class(classbits, &class_uchardata, options,              (void)add_not_list_to_class(classbits, &class_uchardata, options,
4364                cd, PRIV(hspace_list));                cd, PRIV(hspace_list));
4365              continue;              continue;
4366    
4367              case ESC_v:              case ESC_v:
4368              (void)add_list_to_class(classbits, &class_uchardata, options, cd,              (void)add_list_to_class(classbits, &class_uchardata, options, cd,
4369                PRIV(vspace_list), NOTACHAR);                PRIV(vspace_list), NOTACHAR);
4370              continue;              continue;
4371    
4372              case ESC_V:              case ESC_V:
4373              (void)add_not_list_to_class(classbits, &class_uchardata, options,              (void)add_not_list_to_class(classbits, &class_uchardata, options,
4374                cd, PRIV(vspace_list));                cd, PRIV(vspace_list));
4375              continue;              continue;
4376    
4377  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
# Line 4413  for (;; ptr++) Line 4409  for (;; ptr++)
4409    
4410          /* Fall through if the escape just defined a single character (c >= 0).          /* Fall through if the escape just defined a single character (c >= 0).
4411          This may be greater than 256. */          This may be greater than 256. */
4412    
4413          escape = 0;          escape = 0;
4414    
4415          }   /* End of backslash handling */          }   /* End of backslash handling */
# Line 4454  for (;; ptr++) Line 4450  for (;; ptr++)
4450            inescq = TRUE;            inescq = TRUE;
4451            break;            break;
4452            }            }
4453    
4454          /* Minus (hyphen) at the end of a class is treated as a literal, so put          /* Minus (hyphen) at the end of a class is treated as a literal, so put
4455          back the pointer and jump to handle the character that preceded it. */          back the pointer and jump to handle the character that preceded it. */
4456    
4457          if (*ptr == 0 || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET))          if (*ptr == CHAR_NULL || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET))
4458            {            {
4459            ptr = oldptr;            ptr = oldptr;
4460            goto CLASS_SINGLE_CHARACTER;            goto CLASS_SINGLE_CHARACTER;
4461            }            }
4462    
4463          /* Otherwise, we have a potential range; pick up the next character */          /* Otherwise, we have a potential range; pick up the next character */
4464    
4465  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
# Line 4510  for (;; ptr++) Line 4506  for (;; ptr++)
4506          /* We have found a character range, so single character optimizations          /* We have found a character range, so single character optimizations
4507          cannot be done anymore. Any value greater than 1 indicates that there          cannot be done anymore. Any value greater than 1 indicates that there
4508          is more than one character. */          is more than one character. */
4509    
4510          class_one_char = 2;          class_one_char = 2;
4511    
4512          /* Remember an explicit \r or \n, and add the range to the class. */          /* Remember an explicit \r or \n, and add the range to the class. */
4513    
4514          if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;          if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
4515    
4516          class_has_8bitchar +=          class_has_8bitchar +=
4517            add_to_class(classbits, &class_uchardata, options, cd, c, d);            add_to_class(classbits, &class_uchardata, options, cd, c, d);
4518    
4519          continue;   /* Go get the next char in the class */          continue;   /* Go get the next char in the class */
4520          }          }
4521    
4522        /* Handle a single character - we can get here for a normal non-escape        /* Handle a single character - we can get here for a normal non-escape
4523        char, or after \ that introduces a single character or for an apparent        char, or after \ that introduces a single character or for an apparent
4524        range that isn't. Only the value 1 matters for class_one_char, so don't        range that isn't. Only the value 1 matters for class_one_char, so don't
4525        increase it if it is already 2 or more ... just in case there's a class        increase it if it is already 2 or more ... just in case there's a class
4526        with a zillion characters in it. */        with a zillion characters in it. */
4527    
4528        CLASS_SINGLE_CHARACTER:        CLASS_SINGLE_CHARACTER:
# Line 4549  for (;; ptr++) Line 4545  for (;; ptr++)
4545    
4546          if (negate_class)          if (negate_class)
4547            {            {
4548  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
4549            int d;            int d;
4550  #endif  #endif
4551            if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;            if (firstcharflags == REQ_UNSET) firstcharflags = REQ_NONE;
4552            zerofirstchar = firstchar;            zerofirstchar = firstchar;
4553            zerofirstcharflags = firstcharflags;            zerofirstcharflags = firstcharflags;
# Line 4559  for (;; ptr++) Line 4555  for (;; ptr++)
4555            /* For caseless UTF-8 mode when UCP support is available, check            /* For caseless UTF-8 mode when UCP support is available, check
4556            whether this character has more than one other case. If so, generate            whether this character has more than one other case. If so, generate
4557            a special OP_NOTPROP item instead of OP_NOTI. */            a special OP_NOTPROP item instead of OP_NOTI. */
4558    
4559  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
4560            if (utf && (options & PCRE_CASELESS) != 0 &&            if (utf && (options & PCRE_CASELESS) != 0 &&
4561                (d = UCD_CASESET(c)) != 0)                (d = UCD_CASESET(c)) != 0)
4562              {              {
4563              *code++ = OP_NOTPROP;              *code++ = OP_NOTPROP;
4564              *code++ = PT_CLIST;              *code++ = PT_CLIST;
4565              *code++ = d;              *code++ = d;
4566              }              }
4567            else            else
4568  #endif  #endif
4569            /* Char has only one other case, or UCP not available */            /* Char has only one other case, or UCP not available */
4570    
# Line 4581  for (;; ptr++) Line 4577  for (;; ptr++)
4577  #endif  #endif
4578                *code++ = c;                *code++ = c;
4579              }              }
4580    
4581            /* We are finished with this character class */            /* We are finished with this character class */
4582    
4583            goto END_CLASS;            goto END_CLASS;
4584            }            }
4585    
# Line 4601  for (;; ptr++) Line 4597  for (;; ptr++)
4597            }            }
4598          goto ONE_CHAR;          goto ONE_CHAR;
4599          }       /* End of 1-char optimization */          }       /* End of 1-char optimization */
4600    
4601        /* There is more than one character in the class, or an XCLASS item        /* There is more than one character in the class, or an XCLASS item
4602        has been generated. Add this character to the class. */        has been generated. Add this character to the class. */
4603    
4604        class_has_8bitchar +=        class_has_8bitchar +=
4605          add_to_class(classbits, &class_uchardata, options, cd, c, c);          add_to_class(classbits, &class_uchardata, options, cd, c, c);
4606        }        }
4607    
# Line 4613  for (;; ptr++) Line 4609  for (;; ptr++)
4609      If we are at the end of an internal nested string, revert to the outer      If we are at the end of an internal nested string, revert to the outer
4610      string. */      string. */
4611    
4612      while (((c = *(++ptr)) != 0 ||      while (((c = *(++ptr)) != CHAR_NULL ||
4613             (nestptr != NULL &&             (nestptr != NULL &&
4614               (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != 0)) &&               (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != CHAR_NULL)) &&
4615             (c != CHAR_RIGHT_SQUARE_BRACKET || inescq));             (c != CHAR_RIGHT_SQUARE_BRACKET || inescq));
4616    
4617      /* Check for missing terminating ']' */      /* Check for missing terminating ']' */
4618    
4619      if (c == 0)      if (c == CHAR_NULL)
4620        {        {
4621        *errorcodeptr = ERR6;        *errorcodeptr = ERR6;
4622        goto FAILED;        goto FAILED;
4623        }        }
4624    
4625      /* We will need an XCLASS if data has been placed in class_uchardata. In      /* We will need an XCLASS if data has been placed in class_uchardata. In
4626      the second phase this is a sufficient test. However, in the pre-compile      the second phase this is a sufficient test. However, in the pre-compile
4627      phase, class_uchardata gets emptied to prevent workspace overflow, so it      phase, class_uchardata gets emptied to prevent workspace overflow, so it
4628      only if the very last character in the class needs XCLASS will it contain      only if the very last character in the class needs XCLASS will it contain
4629      anything at this point. For this reason, xclass gets set TRUE above when      anything at this point. For this reason, xclass gets set TRUE above when
4630      uchar_classdata is emptied, and that's why this code is the way it is here      uchar_classdata is emptied, and that's why this code is the way it is here
4631      instead of just doing a test on class_uchardata below. */      instead of just doing a test on class_uchardata below. */
4632    
4633  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
4634      if (class_uchardata > class_uchardata_base) xclass = TRUE;      if (class_uchardata > class_uchardata_base) xclass = TRUE;
4635  #endif  #endif
# Line 4702  for (;; ptr++) Line 4698  for (;; ptr++)
4698        memcpy(code, classbits, 32);        memcpy(code, classbits, 32);
4699        }        }
4700      code += 32 / sizeof(pcre_uchar);      code += 32 / sizeof(pcre_uchar);
4701    
4702      END_CLASS:      END_CLASS:
4703      break;      break;
4704    
# Line 4906  for (;; ptr++) Line 4902  for (;; ptr++)
4902    
4903        if (repeat_max == 0) goto END_REPEAT;        if (repeat_max == 0) goto END_REPEAT;
4904    
       /*--------------------------------------------------------------------*/  
       /* This code is obsolete from release 8.00; the restriction was finally  
       removed: */  
   
       /* All real repeats make it impossible to handle partial matching (maybe  
       one day we will be able to remove this restriction). */  
   
       /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */  
       /*--------------------------------------------------------------------*/  
   
4905        /* Combine the op_type with the repeat_type */        /* Combine the op_type with the repeat_type */
4906    
4907        repeat_type += op_type;        repeat_type += op_type;
# Line 5062  for (;; ptr++) Line 5048  for (;; ptr++)
5048          goto END_REPEAT;          goto END_REPEAT;
5049          }          }
5050    
       /*--------------------------------------------------------------------*/  
       /* This code is obsolete from release 8.00; the restriction was finally  
       removed: */  
   
       /* All real repeats make it impossible to handle partial matching (maybe  
       one day we will be able to remove this restriction). */  
   
       /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */  
       /*--------------------------------------------------------------------*/  
   
5051        if (repeat_min == 0 && repeat_max == -1)        if (repeat_min == 0 && repeat_max == -1)
5052          *code++ = OP_CRSTAR + repeat_type;          *code++ = OP_CRSTAR + repeat_type;
5053        else if (repeat_min == 1 && repeat_max == -1)        else if (repeat_min == 1 && repeat_max == -1)
# Line 5610  for (;; ptr++) Line 5586  for (;; ptr++)
5586        if (*ptr == CHAR_COLON)        if (*ptr == CHAR_COLON)
5587          {          {
5588          arg = ++ptr;          arg = ++ptr;
5589          while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;          while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5590          arglen = (int)(ptr - arg);          arglen = (int)(ptr - arg);
5591          if ((unsigned int)arglen > MAX_MARK)          if ((unsigned int)arglen > MAX_MARK)
5592            {            {
# Line 5724  for (;; ptr++) Line 5700  for (;; ptr++)
5700          {          {
5701          case CHAR_NUMBER_SIGN:                 /* Comment; skip to ket */          case CHAR_NUMBER_SIGN:                 /* Comment; skip to ket */
5702          ptr++;          ptr++;
5703          while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;          while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5704          if (*ptr == 0)          if (*ptr == CHAR_NULL)
5705            {            {
5706            *errorcodeptr = ERR18;            *errorcodeptr = ERR18;
5707            goto FAILED;            goto FAILED;
# Line 5748  for (;; ptr++) Line 5724  for (;; ptr++)
5724          /* ------------------------------------------------------------ */          /* ------------------------------------------------------------ */
5725          case CHAR_LEFT_PARENTHESIS:          case CHAR_LEFT_PARENTHESIS:
5726          bravalue = OP_COND;       /* Conditional group */          bravalue = OP_COND;       /* Conditional group */
5727            tempptr = ptr;
5728    
5729          /* A condition can be an assertion, a number (referring to a numbered          /* A condition can be an assertion, a number (referring to a numbered
5730          group), a name (referring to a named group), or 'R', referring to          group), a name (referring to a named group), or 'R', referring to
# Line 5760  for (;; ptr++) Line 5737  for (;; ptr++)
5737          be the recursive thing or the name 'R' (and similarly for 'R' followed          be the recursive thing or the name 'R' (and similarly for 'R' followed
5738          by digits), and (b) a number could be a name that consists of digits.          by digits), and (b) a number could be a name that consists of digits.
5739          In both cases, we look for a name first; if not found, we try the other          In both cases, we look for a name first; if not found, we try the other
5740          cases. */          cases.
5741    
5742            For compatibility with auto-callouts, we allow a callout to be
5743            specified before a condition that is an assertion. First, check for the
5744            syntax of a callout; if found, adjust the temporary pointer that is
5745            used to check for an assertion condition. That's all that is needed! */
5746    
5747            if (ptr[1] == CHAR_QUESTION_MARK && ptr[2] == CHAR_C)
5748              {
5749              for (i = 3;; i++) if (!IS_DIGIT(ptr[i])) break;
5750              if (ptr[i] == CHAR_RIGHT_PARENTHESIS)
5751                tempptr += i + 1;
5752              }
5753    
5754          /* For conditions that are assertions, check the syntax, and then exit          /* For conditions that are assertions, check the syntax, and then exit
5755          the switch. This will take control down to where bracketed groups,          the switch. This will take control down to where bracketed groups,
5756          including assertions, are processed. */          including assertions, are processed. */
5757    
5758          if (ptr[1] == CHAR_QUESTION_MARK && (ptr[2] == CHAR_EQUALS_SIGN ||          if (tempptr[1] == CHAR_QUESTION_MARK &&
5759              ptr[2] == CHAR_EXCLAMATION_MARK || ptr[2] == CHAR_LESS_THAN_SIGN))                (tempptr[2] == CHAR_EQUALS_SIGN ||
5760                   tempptr[2] == CHAR_EXCLAMATION_MARK ||
5761                   tempptr[2] == CHAR_LESS_THAN_SIGN))
5762            break;            break;
5763    
5764          /* Most other conditions use OP_CREF (a couple change to OP_RREF          /* Most other conditions use OP_CREF (a couple change to OP_RREF
# Line 5801  for (;; ptr++) Line 5792  for (;; ptr++)
5792            }            }
5793          else          else
5794            {            {
5795            terminator = 0;            terminator = CHAR_NULL;
5796            if (ptr[1] == CHAR_MINUS || ptr[1] == CHAR_PLUS) refsign = *(++ptr);            if (ptr[1] == CHAR_MINUS || ptr[1] == CHAR_PLUS) refsign = *(++ptr);
5797            }            }
5798    
# Line 5891  for (;; ptr++) Line 5882  for (;; ptr++)
5882            code[1+LINK_SIZE]++;            code[1+LINK_SIZE]++;
5883            }            }
5884    
5885          /* If terminator == 0 it means that the name followed directly after          /* If terminator == CHAR_NULL it means that the name followed directly
5886          the opening parenthesis [e.g. (?(abc)...] and in this case there are          after the opening parenthesis [e.g. (?(abc)...] and in this case there
5887          some further alternatives to try. For the cases where terminator != 0          are some further alternatives to try. For the cases where terminator !=
5888          [things like (?(<name>... or (?('name')... or (?(R&name)... ] we have          0 [things like (?(<name>... or (?('name')... or (?(R&name)... ] we have
5889          now checked all the possibilities, so give an error. */          now checked all the possibilities, so give an error. */
5890    
5891          else if (terminator != 0)          else if (terminator != CHAR_NULL)
5892            {            {
5893            *errorcodeptr = ERR15;            *errorcodeptr = ERR15;
5894            goto FAILED;            goto FAILED;
# Line 6702  for (;; ptr++) Line 6693  for (;; ptr++)
6693        /* If the subpattern set a required byte (or set a first byte that isn't        /* If the subpattern set a required byte (or set a first byte that isn't
6694        really the first byte - see above), set it. */        really the first byte - see above), set it. */
6695    
6696        if (subreqcharflags >= 0)        if (subreqcharflags >= 0)
6697          {          {
6698          reqchar = subreqchar;          reqchar = subreqchar;
6699          reqcharflags = subreqcharflags;          reqcharflags = subreqcharflags;
# Line 6737  for (;; ptr++) Line 6728  for (;; ptr++)
6728      case CHAR_BACKSLASH:      case CHAR_BACKSLASH:
6729      tempptr = ptr;      tempptr = ptr;
6730      escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, FALSE);      escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, FALSE);
   
6731      if (*errorcodeptr != 0) goto FAILED;      if (*errorcodeptr != 0) goto FAILED;
6732    
6733      if (escape == 0)      if (escape == 0)                  /* The escape coded a single character */
6734        c = ec;        c = ec;
6735      else      else
6736        {        {
# Line 6794  for (;; ptr++) Line 6784  for (;; ptr++)
6784          if (ptr[1] != CHAR_PLUS && ptr[1] != CHAR_MINUS)          if (ptr[1] != CHAR_PLUS && ptr[1] != CHAR_MINUS)
6785            {            {
6786            BOOL is_a_number = TRUE;            BOOL is_a_number = TRUE;
6787            for (p = ptr + 1; *p != 0 && *p != (pcre_uchar)terminator; p++)            for (p = ptr + 1; *p != CHAR_NULL && *p != (pcre_uchar)terminator; p++)
6788              {              {
6789              if (!MAX_255(*p)) { is_a_number = FALSE; break; }              if (!MAX_255(*p)) { is_a_number = FALSE; break; }
6790              if ((cd->ctypes[*p] & ctype_digit) == 0) is_a_number = FALSE;              if ((cd->ctypes[*p] & ctype_digit) == 0) is_a_number = FALSE;
# Line 6906  for (;; ptr++) Line 6896  for (;; ptr++)
6896        can obtain the OP value by negating the escape value in the default        can obtain the OP value by negating the escape value in the default
6897        situation when PCRE_UCP is not set. When it *is* set, we substitute        situation when PCRE_UCP is not set. When it *is* set, we substitute
6898        Unicode property tests. Note that \b and \B do a one-character        Unicode property tests. Note that \b and \B do a one-character
6899        lookbehind. */        lookbehind, and \A also behaves as if it does. */
6900    
6901        else        else
6902          {          {
6903          if ((escape == ESC_b || escape == ESC_B) && cd->max_lookbehind == 0)          if ((escape == ESC_b || escape == ESC_B || escape == ESC_A) &&
6904                 cd->max_lookbehind == 0)
6905            cd->max_lookbehind = 1;            cd->max_lookbehind = 1;
6906  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6907          if (escape >= ESC_DU && escape <= ESC_wu)          if (escape >= ESC_DU && escape <= ESC_wu)
# Line 6968  for (;; ptr++) Line 6959  for (;; ptr++)
6959    
6960      ONE_CHAR:      ONE_CHAR:
6961      previous = code;      previous = code;
6962    
6963      /* For caseless UTF-8 mode when UCP support is available, check whether      /* For caseless UTF-8 mode when UCP support is available, check whether
6964      this character has more than one other case. If so, generate a special      this character has more than one other case. If so, generate a special
6965      OP_PROP item instead of OP_CHARI. */      OP_PROP item instead of OP_CHARI. */
6966    
6967  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6968      if (utf && (options & PCRE_CASELESS) != 0)      if (utf && (options & PCRE_CASELESS) != 0)
6969        {        {
# Line 6982  for (;; ptr++) Line 6973  for (;; ptr++)
6973          *code++ = OP_PROP;          *code++ = OP_PROP;
6974          *code++ = PT_CLIST;          *code++ = PT_CLIST;
6975          *code++ = c;          *code++ = c;
6976          if (firstcharflags == REQ_UNSET) firstcharflags = zerofirstcharflags = REQ_NONE;          if (firstcharflags == REQ_UNSET) firstcharflags = zerofirstcharflags = REQ_NONE;
6977          break;          break;
6978          }          }
6979        }        }
6980  #endif  #endif
6981    
6982      /* Caseful matches, or not one of the multicase characters. */      /* Caseful matches, or not one of the multicase characters. */
6983    
6984      *code++ = ((options & PCRE_CASELESS) != 0)? OP_CHARI : OP_CHAR;      *code++ = ((options & PCRE_CASELESS) != 0)? OP_CHARI : OP_CHAR;
6985      for (c = 0; c < mclength; c++) *code++ = mcbuffer[c];      for (c = 0; c < mclength; c++) *code++ = mcbuffer[c];
6986    
# Line 7179  for (;;) Line 7170  for (;;)
7170    into the length. */    into the length. */
7171    
7172    if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstchar,    if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstchar,
7173          &branchfirstcharflags, &branchreqchar, &branchreqcharflags, &bc,          &branchfirstcharflags, &branchreqchar, &branchreqcharflags, &bc,
7174          cond_depth, cd, (lengthptr == NULL)? NULL : &length))          cond_depth, cd, (lengthptr == NULL)? NULL : &length))
7175      {      {
7176      *ptrptr = ptr;      *ptrptr = ptr;
# Line 7217  for (;;) Line 7208  for (;;)
7208        we have to abandon the firstchar for the regex, but if there was        we have to abandon the firstchar for the regex, but if there was
7209        previously no reqchar, it takes on the value of the old firstchar. */        previously no reqchar, it takes on the value of the old firstchar. */
7210    
7211        if (firstcharflags >= 0 &&        if (firstcharflags >= 0 &&
7212            (firstcharflags != branchfirstcharflags || firstchar != branchfirstchar))            (firstcharflags != branchfirstcharflags || firstchar != branchfirstchar))
7213          {          {
7214          if (reqcharflags < 0)          if (reqcharflags < 0)
7215            {            {
7216            reqchar = firstchar;            reqchar = firstchar;
7217            reqcharflags = firstcharflags;            reqcharflags = firstcharflags;
# Line 7579  do { Line 7570  do {
7570       {       {
7571       if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE;       if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE;
7572       }       }
7573    
7574     /* Atomic brackets */     /* Atomic brackets */
7575    
7576     else if (op == OP_ONCE || op == OP_ONCE_NC)     else if (op == OP_ONCE || op == OP_ONCE_NC)
# Line 7848  while (ptr[skipatstart] == CHAR_LEFT_PAR Line 7839  while (ptr[skipatstart] == CHAR_LEFT_PAR
7839    int newnl = 0;    int newnl = 0;
7840    int newbsr = 0;    int newbsr = 0;
7841    
7842    /* For completeness and backward compatibility, (*UTFn) is supported in the
7843    relevant libraries, but (*UTF) is generic and always supported. Note that
7844    PCRE_UTF8 == PCRE_UTF16 == PCRE_UTF32. */
7845    
7846  #ifdef COMPILE_PCRE8  #ifdef COMPILE_PCRE8
7847    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF_RIGHTPAR, 5) == 0)    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF8_RIGHTPAR, 5) == 0)
7848      { skipatstart += 7; options |= PCRE_UTF8; continue; }      { skipatstart += 7; options |= PCRE_UTF8; continue; }
7849  #endif  #endif
7850  #ifdef COMPILE_PCRE16  #ifdef COMPILE_PCRE16
7851    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF_RIGHTPAR, 6) == 0)    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF16_RIGHTPAR, 6) == 0)
7852      { skipatstart += 8; options |= PCRE_UTF16; continue; }      { skipatstart += 8; options |= PCRE_UTF16; continue; }
7853  #endif  #endif
7854  #ifdef COMPILE_PCRE32  #ifdef COMPILE_PCRE32
7855    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF_RIGHTPAR, 6) == 0)    if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF32_RIGHTPAR, 6) == 0)
7856      { skipatstart += 8; options |= PCRE_UTF32; continue; }      { skipatstart += 8; options |= PCRE_UTF32; continue; }
7857  #endif  #endif
7858    
7859      else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UTF_RIGHTPAR, 4) == 0)
7860        { skipatstart += 6; options |= PCRE_UTF8; continue; }
7861    else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UCP_RIGHTPAR, 4) == 0)    else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_UCP_RIGHTPAR, 4) == 0)
7862      { skipatstart += 6; options |= PCRE_UCP; continue; }      { skipatstart += 6; options |= PCRE_UCP; continue; }
7863    else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_NO_START_OPT_RIGHTPAR, 13) == 0)    else if (STRNCMP_UC_C8(ptr+skipatstart+2, STRING_NO_START_OPT_RIGHTPAR, 13) == 0)
# Line 8118  if (cd->had_accept) Line 8116  if (cd->had_accept)
8116    
8117  /* If not reached end of pattern on success, there's an excess bracket. */  /* If not reached end of pattern on success, there's an excess bracket. */
8118    
8119  if (errorcode == 0 && *ptr != 0) errorcode = ERR22;  if (errorcode == 0 && *ptr != CHAR_NULL) errorcode = ERR22;
8120    
8121  /* Fill in the terminating state and check for disastrous overflow, but  /* Fill in the terminating state and check for disastrous overflow, but
8122  if debugging, leave the test till after things are printed out. */  if debugging, leave the test till after things are printed out. */
# Line 8131  if (code - codestart > length) errorcode Line 8129  if (code - codestart > length) errorcode
8129    
8130  #ifdef SUPPORT_VALGRIND  #ifdef SUPPORT_VALGRIND
8131  /* If the estimated length exceeds the really used length, mark the extra  /* If the estimated length exceeds the really used length, mark the extra
8132  allocated memory as unadressable, so that any out-of-bound reads can be  allocated memory as unadressable, so that any out-of-bound reads can be
8133  detected. */  detected. */
8134  VALGRIND_MAKE_MEM_NOACCESS(code, (length - (code - codestart)) * sizeof(pcre_uchar));  VALGRIND_MAKE_MEM_NOACCESS(code, (length - (code - codestart)) * sizeof(pcre_uchar));
8135  #endif  #endif

Legend:
Removed from v.1152  
changed lines
  Added in v.1305

  ViewVC Help
Powered by ViewVC 1.1.5