/[pcre]/code/branches/pcre16/pcre_compile.c
ViewVC logotype

Diff of /code/branches/pcre16/pcre_compile.c

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

revision 780 by zherczeg, Thu Dec 1 06:08:45 2011 UTC revision 781 by zherczeg, Sat Dec 3 07:58:30 2011 UTC
# Line 470  static const char error_texts[] = Line 470  static const char error_texts[] =
470    "\\k is not followed by a braced, angle-bracketed, or quoted name\0"    "\\k is not followed by a braced, angle-bracketed, or quoted name\0"
471    /* 70 */    /* 70 */
472    "internal error: unknown opcode in find_fixedlength()\0"    "internal error: unknown opcode in find_fixedlength()\0"
473      "Not allowed UTF-8 / UTF-16 code point (>= 0xd800 && <= 0xdfff)\0"
474    ;    ;
475    
476  /* Table to identify digits and hex digits. This is used when compiling  /* Table to identify digits and hex digits. This is used when compiling
# Line 538  static const pcre_uint8 digitab[] = Line 539  static const pcre_uint8 digitab[] =
539    
540  /* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */  /* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */
541    
542  static const pcre_unit8 digitab[] =  static const pcre_uint8 digitab[] =
543    {    {
544    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7  0 */    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   0-  7  0 */
545    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   8- 15    */    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /*   8- 15    */
# Line 706  static int Line 707  static int
707  check_escape(const pcre_uchar **ptrptr, int *errorcodeptr, int bracount,  check_escape(const pcre_uchar **ptrptr, int *errorcodeptr, int bracount,
708    int options, BOOL isclass)    int options, BOOL isclass)
709  {  {
710  BOOL utf8 = (options & PCRE_UTF8) != 0;  /* PCRE_UTF16 has the same value as PCRE_UTF8. */
711    BOOL utf = (options & PCRE_UTF8) != 0;
712  const pcre_uchar *ptr = *ptrptr + 1;  const pcre_uchar *ptr = *ptrptr + 1;
713  int c, i;  pcre_int32 c;
714    int i;
715    
716  GETCHARINCTEST(c, ptr);           /* Get character value, increment pointer */  GETCHARINCTEST(c, ptr);           /* Get character value, increment pointer */
717  ptr--;                            /* Set pointer back to the last byte */  ptr--;                            /* Set pointer back to the last byte */
# Line 940  else Line 943  else
943      c -= CHAR_0;      c -= CHAR_0;
944      while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7)      while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7)
945          c = c * 8 + *(++ptr) - CHAR_0;          c = c * 8 + *(++ptr) - CHAR_0;
946      if (!utf8 && c > 0xff) *errorcodeptr = ERR51;      if (!utf && c > 0xff) *errorcodeptr = ERR51;
947      break;      break;
948    
949      /* \x is complicated. \x{ddd} is a character number which can be greater      /* \x is complicated. \x{ddd} is a character number which can be greater
950      than 0xff in utf8 mode, but only if the ddd are hex digits. If not, { is      than 0xff in utf or non-8bit mode, but only if the ddd are hex digits.
951      treated as a data character. */      If not, { is treated as a data character. */
952    
953      case CHAR_x:      case CHAR_x:
954      if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)      if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
# Line 974  else Line 977  else
977      if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)      if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
978        {        {
979        const pcre_uchar *pt = ptr + 2;        const pcre_uchar *pt = ptr + 2;
       int count = 0;  
980    
981        c = 0;        c = 0;
982        while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0)        while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0)
983          {          {
984          register int cc = *pt++;          register int cc = *pt++;
985          if (c == 0 && cc == CHAR_0) continue;     /* Leading zeroes */          if (c == 0 && cc == CHAR_0) continue;     /* Leading zeroes */
         count++;  
986    
987  #ifndef EBCDIC  /* ASCII/UTF-8 coding */  #ifndef EBCDIC  /* ASCII/UTF-8 coding */
988          if (cc >= CHAR_a) cc -= 32;               /* Convert to upper case */          if (cc >= CHAR_a) cc -= 32;               /* Convert to upper case */
# Line 990  else Line 991  else
991          if (cc >= CHAR_a && cc <= CHAR_z) cc += 64;  /* Convert to upper case */          if (cc >= CHAR_a && cc <= CHAR_z) cc += 64;  /* Convert to upper case */
992          c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));          c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
993  #endif  #endif
         }  
994    
       if (*pt == CHAR_RIGHT_CURLY_BRACKET)  
         {  
995  #ifdef COMPILE_PCRE8  #ifdef COMPILE_PCRE8
996          if (c < 0 || count > (utf8? 8:2)) *errorcodeptr = ERR34;          if (c > (utf ? 0x10ffff : 0xff)) { c = -1; break; }
997  #else  #else
998  #ifdef COMPILE_PCRE16  #ifdef COMPILE_PCRE16
999          if (c < 0 || count > (utf8? 8:4)) *errorcodeptr = ERR34;          if (c > (utf ? 0x10ffff : 0xffff)) { c = -1; break; }
1000  #endif  #endif
1001  #endif  #endif
1002            }
1003    
1004          if (c < 0)
1005            {
1006            while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0) pt++;
1007            *errorcodeptr = ERR34;
1008            }
1009    
1010          if (*pt == CHAR_RIGHT_CURLY_BRACKET)
1011            {
1012            if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR71;
1013          ptr = pt;          ptr = pt;
1014          break;          break;
1015          }          }
# Line 1281  Arguments: Line 1290  Arguments:
1290    name         name to seek, or NULL if seeking a numbered subpattern    name         name to seek, or NULL if seeking a numbered subpattern
1291    lorn         name length, or subpattern number if name is NULL    lorn         name length, or subpattern number if name is NULL
1292    xmode        TRUE if we are in /x mode    xmode        TRUE if we are in /x mode
1293    utf8         TRUE if we are in UTF-8 mode    utf          TRUE if we are in UTF-8 / UTF-16 mode
1294    count        pointer to the current capturing subpattern number (updated)    count        pointer to the current capturing subpattern number (updated)
1295    
1296  Returns:       the number of the named subpattern, or -1 if not found  Returns:       the number of the named subpattern, or -1 if not found
# Line 1289  Returns:       the number of the named s Line 1298  Returns:       the number of the named s
1298    
1299  static int  static int
1300  find_parens_sub(pcre_uchar **ptrptr, compile_data *cd, const pcre_uchar *name, int lorn,  find_parens_sub(pcre_uchar **ptrptr, compile_data *cd, const pcre_uchar *name, int lorn,
1301    BOOL xmode, BOOL utf8, int *count)    BOOL xmode, BOOL utf, int *count)
1302  {  {
1303  pcre_uchar *ptr = *ptrptr;  pcre_uchar *ptr = *ptrptr;
1304  int start_count = *count;  int start_count = *count;
# Line 1458  for (; ptr < cd->end_pattern; ptr++) Line 1467  for (; ptr < cd->end_pattern; ptr++)
1467        if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }        if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
1468        ptr++;        ptr++;
1469  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1470        if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;        if (utf) while ((*ptr & 0xc0) == 0x80) ptr++;
1471  #endif  #endif
1472        }        }
1473      if (*ptr == 0) goto FAIL_EXIT;      if (*ptr == 0) goto FAIL_EXIT;
# Line 1469  for (; ptr < cd->end_pattern; ptr++) Line 1478  for (; ptr < cd->end_pattern; ptr++)
1478    
1479    if (*ptr == CHAR_LEFT_PARENTHESIS)    if (*ptr == CHAR_LEFT_PARENTHESIS)
1480      {      {
1481      int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf8, count);      int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, count);
1482      if (rc > 0) return rc;      if (rc > 0) return rc;
1483      if (*ptr == 0) goto FAIL_EXIT;      if (*ptr == 0) goto FAIL_EXIT;
1484      }      }
# Line 1515  Arguments: Line 1524  Arguments:
1524    name         name to seek, or NULL if seeking a numbered subpattern    name         name to seek, or NULL if seeking a numbered subpattern
1525    lorn         name length, or subpattern number if name is NULL    lorn         name length, or subpattern number if name is NULL
1526    xmode        TRUE if we are in /x mode    xmode        TRUE if we are in /x mode
1527    utf8         TRUE if we are in UTF-8 mode    utf          TRUE if we are in UTF-8 / UTF-16 mode
1528    
1529  Returns:       the number of the found subpattern, or -1 if not found  Returns:       the number of the found subpattern, or -1 if not found
1530  */  */
1531    
1532  static int  static int
1533  find_parens(compile_data *cd, const pcre_uchar *name, int lorn, BOOL xmode,  find_parens(compile_data *cd, const pcre_uchar *name, int lorn, BOOL xmode,
1534    BOOL utf8)    BOOL utf)
1535  {  {
1536  pcre_uchar *ptr = (pcre_uchar *)cd->start_pattern;  pcre_uchar *ptr = (pcre_uchar *)cd->start_pattern;
1537  int count = 0;  int count = 0;
# Line 1535  matching closing parens. That is why we Line 1544  matching closing parens. That is why we
1544    
1545  for (;;)  for (;;)
1546    {    {
1547    rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf8, &count);    rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, &count);
1548    if (rc > 0 || *ptr++ == 0) break;    if (rc > 0 || *ptr++ == 0) break;
1549    }    }
1550    
# Line 1618  and doing the check at the end; a flag s Line 1627  and doing the check at the end; a flag s
1627    
1628  Arguments:  Arguments:
1629    code     points to the start of the pattern (the bracket)    code     points to the start of the pattern (the bracket)
1630    utf8     TRUE in UTF-8 mode    utf      TRUE in UTF-8 / UTF-16 mode
1631    atend    TRUE if called when the pattern is complete    atend    TRUE if called when the pattern is complete
1632    cd       the "compile data" structure    cd       the "compile data" structure
1633    
# Line 1630  Returns:   the fixed length, Line 1639  Returns:   the fixed length,
1639  */  */
1640    
1641  static int  static int
1642  find_fixedlength(pcre_uchar *code, BOOL utf8, BOOL atend, compile_data *cd)  find_fixedlength(pcre_uchar *code, BOOL utf, BOOL atend, compile_data *cd)
1643  {  {
1644  int length = -1;  int length = -1;
1645    
# Line 1657  for (;;) Line 1666  for (;;)
1666      case OP_ONCE:      case OP_ONCE:
1667      case OP_ONCE_NC:      case OP_ONCE_NC:
1668      case OP_COND:      case OP_COND:
1669      d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf8, atend, cd);      d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf, atend, cd);
1670      if (d < 0) return d;      if (d < 0) return d;
1671      branchlength += d;      branchlength += d;
1672      do cc += GET(cc, 1); while (*cc == OP_ALT);      do cc += GET(cc, 1); while (*cc == OP_ALT);
# Line 1691  for (;;) Line 1700  for (;;)
1700      cs = ce = (pcre_uchar *)cd->start_code + GET(cc, 1);  /* Start subpattern */      cs = ce = (pcre_uchar *)cd->start_code + GET(cc, 1);  /* Start subpattern */
1701      do ce += GET(ce, 1); while (*ce == OP_ALT);           /* End subpattern */      do ce += GET(ce, 1); while (*ce == OP_ALT);           /* End subpattern */
1702      if (cc > cs && cc < ce) return -1;                    /* Recursion */      if (cc > cs && cc < ce) return -1;                    /* Recursion */
1703      d = find_fixedlength(cs + 2, utf8, atend, cd);      d = find_fixedlength(cs + 2, utf, atend, cd);
1704      if (d < 0) return d;      if (d < 0) return d;
1705      branchlength += d;      branchlength += d;
1706      cc += 1 + LINK_SIZE;      cc += 1 + LINK_SIZE;
# Line 1751  for (;;) Line 1760  for (;;)
1760      branchlength++;      branchlength++;
1761      cc += 2;      cc += 2;
1762  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1763      if (utf8 && cc[-1] >= 0xc0) cc += PRIV(utf8_table4)[cc[-1] & 0x3f];      if (utf && cc[-1] >= 0xc0) cc += PRIV(utf8_table4)[cc[-1] & 0x3f];
1764  #endif  #endif
1765      break;      break;
1766    
# Line 1765  for (;;) Line 1774  for (;;)
1774      branchlength += GET2(cc,1);      branchlength += GET2(cc,1);
1775      cc += 2 + IMM2_SIZE;      cc += 2 + IMM2_SIZE;
1776  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
1777      if (utf8 && cc[-1] >= 0xc0) cc += PRIV(utf8_table4)[cc[-1] & 0x3f];      if (utf && cc[-1] >= 0xc0) cc += PRIV(utf8_table4)[cc[-1] & 0x3f];
1778  #endif  #endif
1779      break;      break;
1780    
# Line 1945  length. Line 1954  length.
1954    
1955  Arguments:  Arguments:
1956    code        points to start of expression    code        points to start of expression
1957    utf8        TRUE in UTF-8 mode    utf         TRUE in UTF-8 / UTF-16 mode
1958    number      the required bracket number or negative to find a lookbehind    number      the required bracket number or negative to find a lookbehind
1959    
1960  Returns:      pointer to the opcode for the bracket, or NULL if not found  Returns:      pointer to the opcode for the bracket, or NULL if not found
1961  */  */
1962    
1963  const pcre_uchar *  const pcre_uchar *
1964  PRIV(find_bracket)(const pcre_uchar *code, BOOL utf8, int number)  PRIV(find_bracket)(const pcre_uchar *code, BOOL utf, int number)
1965  {  {
1966  for (;;)  for (;;)
1967    {    {
# Line 2033  for (;;) Line 2042  for (;;)
2042    arrange to skip the extra bytes. */    arrange to skip the extra bytes. */
2043    
2044  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2045      if (utf8) switch(c)      if (utf) switch(c)
2046        {        {
2047        case OP_CHAR:        case OP_CHAR:
2048        case OP_CHARI:        case OP_CHARI:
# Line 2067  for (;;) Line 2076  for (;;)
2076        break;        break;
2077        }        }
2078  #else  #else
2079      (void)(utf8);  /* Keep compiler happy by referencing function argument */      (void)(utf);  /* Keep compiler happy by referencing function argument */
2080  #endif  #endif
2081      }      }
2082    }    }
# Line 2084  instance of OP_RECURSE. Line 2093  instance of OP_RECURSE.
2093    
2094  Arguments:  Arguments:
2095    code        points to start of expression    code        points to start of expression
2096    utf8        TRUE in UTF-8 mode    utf         TRUE in UTF-8 / UTF-16 mode
2097    
2098  Returns:      pointer to the opcode for OP_RECURSE, or NULL if not found  Returns:      pointer to the opcode for OP_RECURSE, or NULL if not found
2099  */  */
2100    
2101  static const pcre_uchar *  static const pcre_uchar *
2102  find_recurse(const pcre_uchar *code, BOOL utf8)  find_recurse(const pcre_uchar *code, BOOL utf)
2103  {  {
2104  for (;;)  for (;;)
2105    {    {
# Line 2153  for (;;) Line 2162  for (;;)
2162      to arrange to skip the extra bytes. */      to arrange to skip the extra bytes. */
2163    
2164  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2165      if (utf8) switch(c)      if (utf) switch(c)
2166        {        {
2167        case OP_CHAR:        case OP_CHAR:
2168        case OP_CHARI:        case OP_CHARI:
# Line 2187  for (;;) Line 2196  for (;;)
2196        break;        break;
2197        }        }
2198  #else  #else
2199      (void)(utf8);  /* Keep compiler happy by referencing function argument */      (void)(utf);  /* Keep compiler happy by referencing function argument */
2200  #endif  #endif
2201      }      }
2202    }    }
# Line 2210  bracket whose current branch will alread Line 2219  bracket whose current branch will alread
2219  Arguments:  Arguments:
2220    code        points to start of search    code        points to start of search
2221    endcode     points to where to stop    endcode     points to where to stop
2222    utf8        TRUE if in UTF8 mode    utf         TRUE if in UTF-8 / UTF-16 mode
2223    cd          contains pointers to tables etc.    cd          contains pointers to tables etc.
2224    
2225  Returns:      TRUE if what is matched could be empty  Returns:      TRUE if what is matched could be empty
# Line 2218  Returns:      TRUE if what is matched co Line 2227  Returns:      TRUE if what is matched co
2227    
2228  static BOOL  static BOOL
2229  could_be_empty_branch(const pcre_uchar *code, const pcre_uchar *endcode,  could_be_empty_branch(const pcre_uchar *code, const pcre_uchar *endcode,
2230    BOOL utf8, compile_data *cd)    BOOL utf, compile_data *cd)
2231  {  {
2232  register int c;  register int c;
2233  for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE);  for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE);
# Line 2266  for (code = first_significant_code(code Line 2275  for (code = first_significant_code(code
2275    
2276      do      do
2277        {        {
2278        if (could_be_empty_branch(scode, endcode, utf8, cd))        if (could_be_empty_branch(scode, endcode, utf, cd))
2279          {          {
2280          empty_branch = TRUE;          empty_branch = TRUE;
2281          break;          break;
# Line 2322  for (code = first_significant_code(code Line 2331  for (code = first_significant_code(code
2331        empty_branch = FALSE;        empty_branch = FALSE;
2332        do        do
2333          {          {
2334          if (!empty_branch && could_be_empty_branch(code, endcode, utf8, cd))          if (!empty_branch && could_be_empty_branch(code, endcode, utf, cd))
2335            empty_branch = TRUE;            empty_branch = TRUE;
2336          code += GET(code, 1);          code += GET(code, 1);
2337          }          }
# Line 2456  for (code = first_significant_code(code Line 2465  for (code = first_significant_code(code
2465      case OP_MINQUERYI:      case OP_MINQUERYI:
2466      case OP_POSQUERY:      case OP_POSQUERY:
2467      case OP_POSQUERYI:      case OP_POSQUERYI:
2468      if (utf8 && code[1] >= 0xc0) code += PRIV(utf8_table4)[code[1] & 0x3f];      if (utf && code[1] >= 0xc0) code += PRIV(utf8_table4)[code[1] & 0x3f];
2469      break;      break;
2470    
2471      case OP_UPTO:      case OP_UPTO:
# Line 2465  for (code = first_significant_code(code Line 2474  for (code = first_significant_code(code
2474      case OP_MINUPTOI:      case OP_MINUPTOI:
2475      case OP_POSUPTO:      case OP_POSUPTO:
2476      case OP_POSUPTOI:      case OP_POSUPTOI:
2477      if (utf8 && code[1 + IMM2_SIZE] >= 0xc0) code += PRIV(utf8_table4)[code[1 + IMM2_SIZE] & 0x3f];      if (utf && code[1 + IMM2_SIZE] >= 0xc0) code += PRIV(utf8_table4)[code[1 + IMM2_SIZE] & 0x3f];
2478      break;      break;
2479  #endif  #endif
2480    
# Line 2509  Arguments: Line 2518  Arguments:
2518    code        points to start of the recursion    code        points to start of the recursion
2519    endcode     points to where to stop (current RECURSE item)    endcode     points to where to stop (current RECURSE item)
2520    bcptr       points to the chain of current (unclosed) branch starts    bcptr       points to the chain of current (unclosed) branch starts
2521    utf8        TRUE if in UTF-8 mode    utf         TRUE if in UTF-8 / UTF-16 mode
2522    cd          pointers to tables etc    cd          pointers to tables etc
2523    
2524  Returns:      TRUE if what is matched could be empty  Returns:      TRUE if what is matched could be empty
# Line 2517  Returns:      TRUE if what is matched co Line 2526  Returns:      TRUE if what is matched co
2526    
2527  static BOOL  static BOOL
2528  could_be_empty(const pcre_uchar *code, const pcre_uchar *endcode,  could_be_empty(const pcre_uchar *code, const pcre_uchar *endcode,
2529    branch_chain *bcptr, BOOL utf8, compile_data *cd)    branch_chain *bcptr, BOOL utf, compile_data *cd)
2530  {  {
2531  while (bcptr != NULL && bcptr->current_branch >= code)  while (bcptr != NULL && bcptr->current_branch >= code)
2532    {    {
2533    if (!could_be_empty_branch(bcptr->current_branch, endcode, utf8, cd))    if (!could_be_empty_branch(bcptr->current_branch, endcode, utf, cd))
2534      return FALSE;      return FALSE;
2535    bcptr = bcptr->outer;    bcptr = bcptr->outer;
2536    }    }
# Line 2656  value in the reference (which is a group Line 2665  value in the reference (which is a group
2665  Arguments:  Arguments:
2666    group      points to the start of the group    group      points to the start of the group
2667    adjust     the amount by which the group is to be moved    adjust     the amount by which the group is to be moved
2668    utf8       TRUE in UTF-8 mode    utf        TRUE in UTF-8 / UTF-16 mode
2669    cd         contains pointers to tables etc.    cd         contains pointers to tables etc.
2670    save_hwm   the hwm forward reference pointer at the start of the group    save_hwm   the hwm forward reference pointer at the start of the group
2671    
# Line 2664  Returns:     nothing Line 2673  Returns:     nothing
2673  */  */
2674    
2675  static void  static void
2676  adjust_recurse(pcre_uchar *group, int adjust, BOOL utf8, compile_data *cd,  adjust_recurse(pcre_uchar *group, int adjust, BOOL utf, compile_data *cd,
2677    pcre_uchar *save_hwm)    pcre_uchar *save_hwm)
2678  {  {
2679  pcre_uchar *ptr = group;  pcre_uchar *ptr = group;
2680    
2681  while ((ptr = (pcre_uchar *)find_recurse(ptr, utf8)) != NULL)  while ((ptr = (pcre_uchar *)find_recurse(ptr, utf)) != NULL)
2682    {    {
2683    int offset;    int offset;
2684    pcre_uchar *hc;    pcre_uchar *hc;
# Line 2875  sense to automatically possessify the re Line 2884  sense to automatically possessify the re
2884    
2885  Arguments:  Arguments:
2886    previous      pointer to the repeated opcode    previous      pointer to the repeated opcode
2887    utf8          TRUE in UTF-8 mode    utf           TRUE in UTF-8 / UTF-16 mode
2888    ptr           next character in pattern    ptr           next character in pattern
2889    options       options bits    options       options bits
2890    cd            contains pointers to tables etc.    cd            contains pointers to tables etc.
# Line 2884  Returns:        TRUE if possessifying is Line 2893  Returns:        TRUE if possessifying is
2893  */  */
2894    
2895  static BOOL  static BOOL
2896  check_auto_possessive(const pcre_uchar *previous, BOOL utf8,  check_auto_possessive(const pcre_uchar *previous, BOOL utf,
2897    const pcre_uchar *ptr, int options, compile_data *cd)    const pcre_uchar *ptr, int options, compile_data *cd)
2898  {  {
2899  int c, next;  int c, next;
# Line 2905  if ((options & PCRE_EXTENDED) != 0) Line 2914  if ((options & PCRE_EXTENDED) != 0)
2914          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
2915          ptr++;          ptr++;
2916  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2917          if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;          if (utf) while ((*ptr & 0xc0) == 0x80) ptr++;
2918  #endif  #endif
2919          }          }
2920        }        }
# Line 2927  if (*ptr == CHAR_BACKSLASH) Line 2936  if (*ptr == CHAR_BACKSLASH)
2936  else if ((cd->ctypes[*ptr] & ctype_meta) == 0)  else if ((cd->ctypes[*ptr] & ctype_meta) == 0)
2937    {    {
2938  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2939    if (utf8) { GETCHARINC(next, ptr); } else    if (utf) { GETCHARINC(next, ptr); } else
2940  #endif  #endif
2941    next = *ptr++;    next = *ptr++;
2942    }    }
# Line 2949  if ((options & PCRE_EXTENDED) != 0) Line 2958  if ((options & PCRE_EXTENDED) != 0)
2958          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
2959          ptr++;          ptr++;
2960  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
2961          if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;          if (utf) while ((*ptr & 0xc0) == 0x80) ptr++;
2962  #endif  #endif
2963          }          }
2964        }        }
# Line 2988  if (next >= 0) switch(op_code) Line 2997  if (next >= 0) switch(op_code)
2997  #endif  #endif
2998    if (c == next) return FALSE;    if (c == next) return FALSE;
2999  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
3000    if (utf8)    if (utf)
3001      {      {
3002      unsigned int othercase;      unsigned int othercase;
3003      if (next < 128) othercase = cd->fcc[next]; else      if (next < 128) othercase = cd->fcc[next]; else
# Line 3013  if (next >= 0) switch(op_code) Line 3022  if (next >= 0) switch(op_code)
3022    case OP_NOTI:    case OP_NOTI:
3023    if ((c = *previous) == next) return TRUE;    if ((c = *previous) == next) return TRUE;
3024  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
3025    if (utf8)    if (utf)
3026      {      {
3027      unsigned int othercase;      unsigned int othercase;
3028      if (next < 128) othercase = cd->fcc[next]; else      if (next < 128) othercase = cd->fcc[next]; else
# Line 3348  must not do this for other options (e.g. Line 3357  must not do this for other options (e.g.
3357  dynamically as we process the pattern. */  dynamically as we process the pattern. */
3358    
3359  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
3360  BOOL utf8 = (options & PCRE_UTF8) != 0;  /* PCRE_UTF16 has the same value as PCRE_UTF8. */
3361  pcre_uint8 utf8_char[6];  BOOL utf = (options & PCRE_UTF8) != 0;
3362    pcre_uchar utf_chars[6];
3363  #else  #else
3364  BOOL utf8 = FALSE;  BOOL utf = FALSE;
3365  #endif  #endif
3366    
3367  /* Helper variables for OP_XCLASS opcode (for characters > 255). */  /* Helper variables for OP_XCLASS opcode (for characters > 255). */
# Line 3459  for (;; ptr++) Line 3469  for (;; ptr++)
3469        }        }
3470    
3471      *lengthptr += (int)(code - last_code);      *lengthptr += (int)(code - last_code);
3472      DPRINTF(("length=%d added %d c=%c\n", *lengthptr, (int)(code - last_code),      DPRINTF(("length=%d added %d c=%c (0x%x)\n", *lengthptr,
3473        c));        (int)(code - last_code), c, c));
3474    
3475      /* If "previous" is set and it is not at the start of the work space, move      /* If "previous" is set and it is not at the start of the work space, move
3476      it back to there, in order to avoid filling up the work space. Otherwise,      it back to there, in order to avoid filling up the work space. Otherwise,
# Line 3547  for (;; ptr++) Line 3557  for (;; ptr++)
3557          if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }          if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
3558          ptr++;          ptr++;
3559  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
3560          if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;          if (utf) while ((*ptr & 0xc0) == 0x80) ptr++;
3561  #endif  #endif
3562          }          }
3563        if (*ptr != 0) continue;        if (*ptr != 0) continue;
# Line 3727  for (;; ptr++) Line 3737  for (;; ptr++)
3737        const pcre_uchar *oldptr;        const pcre_uchar *oldptr;
3738    
3739  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
3740        if (utf8 && c > 127)        if (utf && c > 127)
3741          {                           /* Braces are required because the */          {                           /* Braces are required because the */
3742          GETCHARLEN(c, ptr, ptr);    /* macro generates multiple statements */          GETCHARLEN(c, ptr, ptr);    /* macro generates multiple statements */
3743          }          }
# Line 3945  for (;; ptr++) Line 3955  for (;; ptr++)
3955              SETBIT(classbits, 0x20); /* SPACE */              SETBIT(classbits, 0x20); /* SPACE */
3956              SETBIT(classbits, 0xa0); /* NSBP */              SETBIT(classbits, 0xa0); /* NSBP */
3957  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3958              if (utf8)              if (utf)
3959                {                {
3960                xclass = TRUE;                xclass = TRUE;
3961                *class_uchardata++ = XCL_SINGLE;                *class_uchardata++ = XCL_SINGLE;
3962                class_uchardata += PRIV(ord2utf8)(0x1680, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x1680, class_uchardata);
3963                *class_uchardata++ = XCL_SINGLE;                *class_uchardata++ = XCL_SINGLE;
3964                class_uchardata += PRIV(ord2utf8)(0x180e, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x180e, class_uchardata);
3965                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
3966                class_uchardata += PRIV(ord2utf8)(0x2000, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2000, class_uchardata);
3967                class_uchardata += PRIV(ord2utf8)(0x200A, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x200A, class_uchardata);
3968                *class_uchardata++ = XCL_SINGLE;                *class_uchardata++ = XCL_SINGLE;
3969                class_uchardata += PRIV(ord2utf8)(0x202f, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x202f, class_uchardata);
3970                *class_uchardata++ = XCL_SINGLE;                *class_uchardata++ = XCL_SINGLE;
3971                class_uchardata += PRIV(ord2utf8)(0x205f, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x205f, class_uchardata);
3972                *class_uchardata++ = XCL_SINGLE;                *class_uchardata++ = XCL_SINGLE;
3973                class_uchardata += PRIV(ord2utf8)(0x3000, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x3000, class_uchardata);
3974                }                }
3975  #endif  #endif
3976              continue;              continue;
# Line 3980  for (;; ptr++) Line 3990  for (;; ptr++)
3990                }                }
3991    
3992  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
3993              if (utf8)              if (utf)
3994                {                {
3995                xclass = TRUE;                xclass = TRUE;
3996                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
3997                class_uchardata += PRIV(ord2utf8)(0x0100, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x0100, class_uchardata);
3998                class_uchardata += PRIV(ord2utf8)(0x167f, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x167f, class_uchardata);
3999                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4000                class_uchardata += PRIV(ord2utf8)(0x1681, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x1681, class_uchardata);
4001                class_uchardata += PRIV(ord2utf8)(0x180d, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x180d, class_uchardata);
4002                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4003                class_uchardata += PRIV(ord2utf8)(0x180f, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x180f, class_uchardata);
4004                class_uchardata += PRIV(ord2utf8)(0x1fff, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x1fff, class_uchardata);
4005                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4006                class_uchardata += PRIV(ord2utf8)(0x200B, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x200B, class_uchardata);
4007                class_uchardata += PRIV(ord2utf8)(0x202e, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x202e, class_uchardata);
4008                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4009                class_uchardata += PRIV(ord2utf8)(0x2030, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2030, class_uchardata);
4010                class_uchardata += PRIV(ord2utf8)(0x205e, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x205e, class_uchardata);
4011                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4012                class_uchardata += PRIV(ord2utf8)(0x2060, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2060, class_uchardata);
4013                class_uchardata += PRIV(ord2utf8)(0x2fff, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2fff, class_uchardata);
4014                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4015                class_uchardata += PRIV(ord2utf8)(0x3001, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x3001, class_uchardata);
4016                class_uchardata += PRIV(ord2utf8)(0x7fffffff, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x10ffff, class_uchardata);
4017                }                }
4018  #endif  #endif
4019              continue;              continue;
# Line 4015  for (;; ptr++) Line 4025  for (;; ptr++)
4025              SETBIT(classbits, 0x0d); /* CR */              SETBIT(classbits, 0x0d); /* CR */
4026              SETBIT(classbits, 0x85); /* NEL */              SETBIT(classbits, 0x85); /* NEL */
4027  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4028              if (utf8)              if (utf)
4029                {                {
4030                xclass = TRUE;                xclass = TRUE;
4031                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4032                class_uchardata += PRIV(ord2utf8)(0x2028, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2028, class_uchardata);
4033                class_uchardata += PRIV(ord2utf8)(0x2029, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2029, class_uchardata);
4034                }                }
4035  #endif  #endif
4036              continue;              continue;
# Line 4043  for (;; ptr++) Line 4053  for (;; ptr++)
4053                }                }
4054    
4055  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4056              if (utf8)              if (utf)
4057                {                {
4058                xclass = TRUE;                xclass = TRUE;
4059                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4060                class_uchardata += PRIV(ord2utf8)(0x0100, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x0100, class_uchardata);
4061                class_uchardata += PRIV(ord2utf8)(0x2027, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2027, class_uchardata);
4062                *class_uchardata++ = XCL_RANGE;                *class_uchardata++ = XCL_RANGE;
4063                class_uchardata += PRIV(ord2utf8)(0x2029, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x2029, class_uchardata);
4064                class_uchardata += PRIV(ord2utf8)(0x7fffffff, class_uchardata);                class_uchardata += PRIV(ord2utf)(0x10ffff, class_uchardata);
4065                }                }
4066  #endif  #endif
4067              continue;              continue;
# Line 4139  for (;; ptr++) Line 4149  for (;; ptr++)
4149            }            }
4150    
4151  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4152          if (utf8)          if (utf)
4153            {                           /* Braces are required because the */            {                           /* Braces are required because the */
4154            GETCHARLEN(d, ptr, ptr);    /* macro generates multiple statements */            GETCHARLEN(d, ptr, ptr);    /* macro generates multiple statements */
4155            }            }
# Line 4189  for (;; ptr++) Line 4199  for (;; ptr++)
4199          available. */          available. */
4200    
4201  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4202          if (utf8 && (d > 255 || ((options & PCRE_CASELESS) != 0 && d > 127)))          if (utf && (d > 255 || ((options & PCRE_CASELESS) != 0 && d > 127)))
4203  #endif  #endif
4204  #ifndef COMPILE_PCRE8  #ifndef COMPILE_PCRE8
4205          if (d > 255)          if (d > 255)
# Line 4234  for (;; ptr++) Line 4244  for (;; ptr++)
4244                else                else
4245                  {                  {
4246                  *class_uchardata++ = XCL_RANGE;                  *class_uchardata++ = XCL_RANGE;
4247                  class_uchardata += PRIV(ord2utf8)(occ, class_uchardata);                  class_uchardata += PRIV(ord2utf)(occ, class_uchardata);
4248                  }                  }
4249                class_uchardata += PRIV(ord2utf8)(ocd, class_uchardata);                class_uchardata += PRIV(ord2utf)(ocd, class_uchardata);
4250                }                }
4251              }              }
4252  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
# Line 4246  for (;; ptr++) Line 4256  for (;; ptr++)
4256    
4257            *class_uchardata++ = XCL_RANGE;            *class_uchardata++ = XCL_RANGE;
4258  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4259            class_uchardata += PRIV(ord2utf8)(c, class_uchardata);            class_uchardata += PRIV(ord2utf)(c, class_uchardata);
4260            class_uchardata += PRIV(ord2utf8)(d, class_uchardata);            class_uchardata += PRIV(ord2utf)(d, class_uchardata);
4261  #else  #else
4262            *class_uchardata++ = c;            *class_uchardata++ = c;
4263            *class_uchardata++ = d;            *class_uchardata++ = d;
# Line 4304  for (;; ptr++) Line 4314  for (;; ptr++)
4314        /* Handle a character that cannot go in the bit map */        /* Handle a character that cannot go in the bit map */
4315    
4316  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4317        if (utf8 && (c > 255 || ((options & PCRE_CASELESS) != 0 && c > 127)))        if (utf && (c > 255 || ((options & PCRE_CASELESS) != 0 && c > 127)))
4318  #endif  #endif
4319  #ifndef COMPILE_PCRE8  #ifndef COMPILE_PCRE8
4320        if (c > 255)        if (c > 255)
# Line 4314  for (;; ptr++) Line 4324  for (;; ptr++)
4324          xclass = TRUE;          xclass = TRUE;
4325          *class_uchardata++ = XCL_SINGLE;          *class_uchardata++ = XCL_SINGLE;
4326  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4327          class_uchardata += PRIV(ord2utf8)(c, class_uchardata);          class_uchardata += PRIV(ord2utf)(c, class_uchardata);
4328  #else  #else
4329          *class_uchardata++ = c;          *class_uchardata++ = c;
4330  #endif  #endif
# Line 4326  for (;; ptr++) Line 4336  for (;; ptr++)
4336            if ((othercase = UCD_OTHERCASE(c)) != c)            if ((othercase = UCD_OTHERCASE(c)) != c)
4337              {              {
4338              *class_uchardata++ = XCL_SINGLE;              *class_uchardata++ = XCL_SINGLE;
4339              class_uchardata += PRIV(ord2utf8)(othercase, class_uchardata);              class_uchardata += PRIV(ord2utf)(othercase, class_uchardata);
4340              }              }
4341            }            }
4342  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
# Line 4384  for (;; ptr++) Line 4394  for (;; ptr++)
4394    
4395  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
4396      if (class_charcount == 1 && !xclass &&      if (class_charcount == 1 && !xclass &&
4397        (!utf8 || !negate_class || class_lastchar < 128))        (!utf || !negate_class || class_lastchar < 128))
 #elif defined COMPILE_PCRE8  
     if (class_charcount == 1)  
4398  #else  #else
4399      if (class_charcount == 1 && !xclass)      if (class_charcount == 1)
4400  #endif  #endif
4401        {        {
4402        zeroreqchar = reqchar;        zeroreqchar = reqchar;
# Line 4408  for (;; ptr++) Line 4416  for (;; ptr++)
4416        then we can handle this with the normal one-character code. */        then we can handle this with the normal one-character code. */
4417    
4418  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4419        if (utf8 && class_lastchar > 127)        if (utf && class_lastchar > 127)
4420          mclength = PRIV(ord2utf8)(class_lastchar, mcbuffer);          mclength = PRIV(ord2utf)(class_lastchar, mcbuffer);
4421        else        else
4422  #endif  #endif
4423          {          {
# Line 4599  for (;; ptr++) Line 4607  for (;; ptr++)
4607        length rather than a small character. */        length rather than a small character. */
4608    
4609  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4610        if (utf8 && (code[-1] & 0x80) != 0)        if (utf && (code[-1] & 0x80) != 0)
4611          {          {
4612          pcre_uchar *lastchar = code - 1;          pcre_uchar *lastchar = code - 1;
4613          while((*lastchar & 0xc0) == 0x80) lastchar--;          while((*lastchar & 0xc0) == 0x80) lastchar--;
4614          c = code - lastchar;            /* Length of UTF-8 character */          c = code - lastchar;            /* Length of UTF-8 character */
4615          memcpy(utf8_char, lastchar, c); /* Save the char */          memcpy(utf_chars, lastchar, c); /* Save the char */
4616          c |= 0x80;                      /* Flag c as a length */          c |= 0x80;                      /* Flag c as a length */
4617          }          }
4618        else        else
# Line 4625  for (;; ptr++) Line 4633  for (;; ptr++)
4633    
4634        if (!possessive_quantifier &&        if (!possessive_quantifier &&
4635            repeat_max < 0 &&            repeat_max < 0 &&
4636            check_auto_possessive(previous, utf8, ptr + 1, options, cd))            check_auto_possessive(previous, utf, ptr + 1, options, cd))
4637          {          {
4638          repeat_type = 0;    /* Force greedy */          repeat_type = 0;    /* Force greedy */
4639          possessive_quantifier = TRUE;          possessive_quantifier = TRUE;
# Line 4646  for (;; ptr++) Line 4654  for (;; ptr++)
4654        c = previous[1];        c = previous[1];
4655        if (!possessive_quantifier &&        if (!possessive_quantifier &&
4656            repeat_max < 0 &&            repeat_max < 0 &&
4657            check_auto_possessive(previous, utf8, ptr + 1, options, cd))            check_auto_possessive(previous, utf, ptr + 1, options, cd))
4658          {          {
4659          repeat_type = 0;    /* Force greedy */          repeat_type = 0;    /* Force greedy */
4660          possessive_quantifier = TRUE;          possessive_quantifier = TRUE;
# Line 4670  for (;; ptr++) Line 4678  for (;; ptr++)
4678    
4679        if (!possessive_quantifier &&        if (!possessive_quantifier &&
4680            repeat_max < 0 &&            repeat_max < 0 &&
4681            check_auto_possessive(previous, utf8, ptr + 1, options, cd))            check_auto_possessive(previous, utf, ptr + 1, options, cd))
4682          {          {
4683          repeat_type = 0;    /* Force greedy */          repeat_type = 0;    /* Force greedy */
4684          possessive_quantifier = TRUE;          possessive_quantifier = TRUE;
# Line 4755  for (;; ptr++) Line 4763  for (;; ptr++)
4763          if (repeat_max < 0)          if (repeat_max < 0)
4764            {            {
4765  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4766            if (utf8 && c >= 128)            if (utf && c >= 128)
4767              {              {
4768              memcpy(code, utf8_char, c & 7);              memcpy(code, utf_chars, c & 7);
4769              code += c & 7;              code += c & 7;
4770              }              }
4771            else            else
# Line 4780  for (;; ptr++) Line 4788  for (;; ptr++)
4788          else if (repeat_max != repeat_min)          else if (repeat_max != repeat_min)
4789            {            {
4790  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4791            if (utf8 && c >= 128)            if (utf && c >= 128)
4792              {              {
4793              memcpy(code, utf8_char, c & 7);              memcpy(code, utf_chars, c & 7);
4794              code += c & 7;              code += c & 7;
4795              }              }
4796            else            else
# Line 4810  for (;; ptr++) Line 4818  for (;; ptr++)
4818        /* The character or character type itself comes last in all cases. */        /* The character or character type itself comes last in all cases. */
4819    
4820  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
4821        if (utf8 && c >= 128)        if (utf && c >= 128)
4822          {          {
4823          memcpy(code, utf8_char, c & 7);          memcpy(code, utf_chars, c & 7);
4824          code += c & 7;          code += c & 7;
4825          }          }
4826        else        else
# Line 4939  for (;; ptr++) Line 4947  for (;; ptr++)
4947          if (repeat_max <= 1)    /* Covers 0, 1, and unlimited */          if (repeat_max <= 1)    /* Covers 0, 1, and unlimited */
4948            {            {
4949            *code = OP_END;            *code = OP_END;
4950            adjust_recurse(previous, 1, utf8, cd, save_hwm);            adjust_recurse(previous, 1, utf, cd, save_hwm);
4951            memmove(previous + 1, previous, IN_UCHARS(len));            memmove(previous + 1, previous, IN_UCHARS(len));
4952            code++;            code++;
4953            if (repeat_max == 0)            if (repeat_max == 0)
# Line 4963  for (;; ptr++) Line 4971  for (;; ptr++)
4971            {            {
4972            int offset;            int offset;
4973            *code = OP_END;            *code = OP_END;
4974            adjust_recurse(previous, 2 + LINK_SIZE, utf8, cd, save_hwm);            adjust_recurse(previous, 2 + LINK_SIZE, utf, cd, save_hwm);
4975            memmove(previous + 2 + LINK_SIZE, previous, IN_UCHARS(len));            memmove(previous + 2 + LINK_SIZE, previous, IN_UCHARS(len));
4976            code += 2 + LINK_SIZE;            code += 2 + LINK_SIZE;
4977            *previous++ = OP_BRAZERO + repeat_type;            *previous++ = OP_BRAZERO + repeat_type;
# Line 5165  for (;; ptr++) Line 5173  for (;; ptr++)
5173              pcre_uchar *scode = bracode;              pcre_uchar *scode = bracode;
5174              do              do
5175                {                {
5176                if (could_be_empty_branch(scode, ketcode, utf8, cd))                if (could_be_empty_branch(scode, ketcode, utf, cd))
5177                  {                  {
5178                  *bracode += OP_SBRA - OP_BRA;                  *bracode += OP_SBRA - OP_BRA;
5179                  break;                  break;
# Line 5188  for (;; ptr++) Line 5196  for (;; ptr++)
5196                {                {
5197                int nlen = (int)(code - bracode);                int nlen = (int)(code - bracode);
5198                *code = OP_END;                *code = OP_END;
5199                adjust_recurse(bracode, 1 + LINK_SIZE, utf8, cd, save_hwm);                adjust_recurse(bracode, 1 + LINK_SIZE, utf, cd, save_hwm);
5200                memmove(bracode + 1 + LINK_SIZE, bracode, IN_UCHARS(nlen));                memmove(bracode + 1 + LINK_SIZE, bracode, IN_UCHARS(nlen));
5201                code += 1 + LINK_SIZE;                code += 1 + LINK_SIZE;
5202                nlen += 1 + LINK_SIZE;                nlen += 1 + LINK_SIZE;
# Line 5266  for (;; ptr++) Line 5274  for (;; ptr++)
5274          {          {
5275          tempcode += PRIV(OP_lengths)[*tempcode];          tempcode += PRIV(OP_lengths)[*tempcode];
5276  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
5277          if (utf8 && tempcode[-1] >= 0xc0)          if (utf && tempcode[-1] >= 0xc0)
5278            tempcode += PRIV(utf8_table4)[tempcode[-1] & 0x3f];            tempcode += PRIV(utf8_table4)[tempcode[-1] & 0x3f];
5279  #endif  #endif
5280          }          }
# Line 5304  for (;; ptr++) Line 5312  for (;; ptr++)
5312    
5313          default:          default:
5314          *code = OP_END;          *code = OP_END;
5315          adjust_recurse(tempcode, 1 + LINK_SIZE, utf8, cd, save_hwm);          adjust_recurse(tempcode, 1 + LINK_SIZE, utf, cd, save_hwm);
5316          memmove(tempcode + 1 + LINK_SIZE, tempcode, IN_UCHARS(len));          memmove(tempcode + 1 + LINK_SIZE, tempcode, IN_UCHARS(len));
5317          code += 1 + LINK_SIZE;          code += 1 + LINK_SIZE;
5318          len += 1 + LINK_SIZE;          len += 1 + LINK_SIZE;
# Line 5613  for (;; ptr++) Line 5621  for (;; ptr++)
5621          /* Search the pattern for a forward reference */          /* Search the pattern for a forward reference */
5622    
5623          else if ((i = find_parens(cd, name, namelen,          else if ((i = find_parens(cd, name, namelen,
5624                          (options & PCRE_EXTENDED) != 0, utf8)) > 0)                          (options & PCRE_EXTENDED) != 0, utf)) > 0)
5625            {            {
5626            PUT2(code, 2+LINK_SIZE, i);            PUT2(code, 2+LINK_SIZE, i);
5627            code[1+LINK_SIZE]++;            code[1+LINK_SIZE]++;
# Line 5958  for (;; ptr++) Line 5966  for (;; ptr++)
5966            temp = cd->end_pattern;            temp = cd->end_pattern;
5967            cd->end_pattern = ptr;            cd->end_pattern = ptr;
5968            recno = find_parens(cd, name, namelen,            recno = find_parens(cd, name, namelen,
5969              (options & PCRE_EXTENDED) != 0, utf8);              (options & PCRE_EXTENDED) != 0, utf);
5970            cd->end_pattern = temp;            cd->end_pattern = temp;
5971            if (recno < 0) recno = 0;    /* Forward ref; set dummy number */            if (recno < 0) recno = 0;    /* Forward ref; set dummy number */
5972            }            }
# Line 5985  for (;; ptr++) Line 5993  for (;; ptr++)
5993              }              }
5994            else if ((recno =                /* Forward back reference */            else if ((recno =                /* Forward back reference */
5995                      find_parens(cd, name, namelen,                      find_parens(cd, name, namelen,
5996                        (options & PCRE_EXTENDED) != 0, utf8)) <= 0)                        (options & PCRE_EXTENDED) != 0, utf)) <= 0)
5997              {              {
5998              *errorcodeptr = ERR15;              *errorcodeptr = ERR15;
5999              goto FAILED;              goto FAILED;
# Line 6089  for (;; ptr++) Line 6097  for (;; ptr++)
6097              {              {
6098              *code = OP_END;              *code = OP_END;
6099              if (recno != 0)              if (recno != 0)
6100                called = PRIV(find_bracket)(cd->start_code, utf8, recno);                called = PRIV(find_bracket)(cd->start_code, utf, recno);
6101    
6102              /* Forward reference */              /* Forward reference */
6103    
6104              if (called == NULL)              if (called == NULL)
6105                {                {
6106                if (find_parens(cd, NULL, recno,                if (find_parens(cd, NULL, recno,
6107                      (options & PCRE_EXTENDED) != 0, utf8) < 0)                      (options & PCRE_EXTENDED) != 0, utf) < 0)
6108                  {                  {
6109                  *errorcodeptr = ERR15;                  *errorcodeptr = ERR15;
6110                  goto FAILED;                  goto FAILED;
# Line 6120  for (;; ptr++) Line 6128  for (;; ptr++)
6128              conditional subpatterns will be picked up then. */              conditional subpatterns will be picked up then. */
6129    
6130              else if (GET(called, 1) == 0 && cond_depth <= 0 &&              else if (GET(called, 1) == 0 && cond_depth <= 0 &&
6131                       could_be_empty(called, code, bcptr, utf8, cd))                       could_be_empty(called, code, bcptr, utf, cd))
6132                {                {
6133                *errorcodeptr = ERR40;                *errorcodeptr = ERR40;
6134                goto FAILED;                goto FAILED;
# Line 6618  for (;; ptr++) Line 6626  for (;; ptr++)
6626    
6627            {            {
6628            previous = (-c > ESC_b && -c < ESC_Z)? code : NULL;            previous = (-c > ESC_b && -c < ESC_Z)? code : NULL;
6629            *code++ = (!utf8 && c == -ESC_C)? OP_ALLANY : -c;            *code++ = (!utf && c == -ESC_C)? OP_ALLANY : -c;
6630            }            }
6631          }          }
6632        continue;        continue;
# Line 6629  for (;; ptr++) Line 6637  for (;; ptr++)
6637      handle it as a data character. */      handle it as a data character. */
6638    
6639  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
6640      if (utf8 && c > 127)      if (utf && c > 127)
6641        mclength = PRIV(ord2utf8)(c, mcbuffer);        mclength = PRIV(ord2utf)(c, mcbuffer);
6642      else      else
6643  #endif  #endif
6644    
# Line 6652  for (;; ptr++) Line 6660  for (;; ptr++)
6660      mcbuffer[0] = c;      mcbuffer[0] = c;
6661    
6662  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
6663      if (utf8 && c >= 0xc0)      if (utf && c >= 0xc0)
6664        {        {
6665        while ((ptr[1] & 0xc0) == 0x80)        while ((ptr[1] & 0xc0) == 0x80)
6666          mcbuffer[mclength++] = *(++ptr);          mcbuffer[mclength++] = *(++ptr);
# Line 7360  pcre_int32 firstchar, reqchar; Line 7368  pcre_int32 firstchar, reqchar;
7368  int newline;  int newline;
7369  int errorcode = 0;  int errorcode = 0;
7370  int skipatstart = 0;  int skipatstart = 0;
7371  BOOL utf8;  BOOL utf;
7372  size_t size;  size_t size;
7373  pcre_uchar *code;  pcre_uchar *code;
7374  const pcre_uchar *codestart;  const pcre_uchar *codestart;
# Line 7458  while (ptr[skipatstart] == CHAR_LEFT_PAR Line 7466  while (ptr[skipatstart] == CHAR_LEFT_PAR
7466    else break;    else break;
7467    }    }
7468    
7469  utf8 = (options & PCRE_UTF8) != 0;  /* PCRE_UTF16 has the same value as PCRE_UTF8. */
7470    utf = (options & PCRE_UTF8) != 0;
7471    
7472  /* Can't support UTF8 unless PCRE has been compiled to include the code. The  /* Can't support UTF8 unless PCRE has been compiled to include the code. The
7473  return of an error code from PRIV(valid_utf8)() is a new feature, introduced in  return of an error code from PRIV(valid_utf)() is a new feature, introduced in
7474  release 8.13. It is passed back from pcre_[dfa_]exec(), but at the moment is  release 8.13. It is passed back from pcre_[dfa_]exec(), but at the moment is
7475  not used here. */  not used here. */
7476    
7477  #ifdef SUPPORT_UTF8  #ifdef SUPPORT_UTF8
7478  if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0 &&  if (utf && (options & PCRE_NO_UTF8_CHECK) == 0 &&
7479       (errorcode = PRIV(valid_utf8)((PCRE_PUCHAR)pattern, -1, erroroffset)) != 0)       (errorcode = PRIV(valid_utf)((PCRE_PUCHAR)pattern, -1, erroroffset)) != 0)
7480    {    {
7481    errorcode = ERR44;    errorcode = ERR44;
7482    goto PCRE_EARLY_ERROR_RETURN2;    goto PCRE_EARLY_ERROR_RETURN2;
7483    }    }
7484  #else  #else
7485  if (utf8)  if (utf)
7486    {    {
7487    errorcode = ERR32;    errorcode = ERR32;
7488    goto PCRE_EARLY_ERROR_RETURN;    goto PCRE_EARLY_ERROR_RETURN;
# Line 7688  while (errorcode == 0 && cd->hwm > cwork Line 7697  while (errorcode == 0 && cd->hwm > cwork
7697    cd->hwm -= LINK_SIZE;    cd->hwm -= LINK_SIZE;
7698    offset = GET(cd->hwm, 0);    offset = GET(cd->hwm, 0);
7699    recno = GET(codestart, offset);    recno = GET(codestart, offset);
7700    groupptr = PRIV(find_bracket)(codestart, utf8, recno);    groupptr = PRIV(find_bracket)(codestart, utf, recno);
7701    if (groupptr == NULL) errorcode = ERR53;    if (groupptr == NULL) errorcode = ERR53;
7702      else PUT(((pcre_uchar *)codestart), offset, (int)(groupptr - codestart));      else PUT(((pcre_uchar *)codestart), offset, (int)(groupptr - codestart));
7703    }    }
# Line 7715  if (cd->check_lookbehind) Line 7724  if (cd->check_lookbehind)
7724    of zero, but that is a pathological case, and it does no harm.) When we find    of zero, but that is a pathological case, and it does no harm.) When we find
7725    one, we temporarily terminate the branch it is in while we scan it. */    one, we temporarily terminate the branch it is in while we scan it. */
7726    
7727    for (cc = (pcre_uchar *)PRIV(find_bracket)(codestart, utf8, -1);    for (cc = (pcre_uchar *)PRIV(find_bracket)(codestart, utf, -1);
7728         cc != NULL;         cc != NULL;
7729         cc = (pcre_uchar *)PRIV(find_bracket)(cc, utf8, -1))         cc = (pcre_uchar *)PRIV(find_bracket)(cc, utf, -1))
7730      {      {
7731      if (GET(cc, 1) == 0)      if (GET(cc, 1) == 0)
7732        {        {
Legend:
Removed from v.780  
changed lines
  Added in v.781
  ViewVC Help
Powered by ViewVC 1.1.5