/[pcre]/code/trunk/pcre_exec.c
ViewVC logotype

Diff of /code/trunk/pcre_exec.c

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

revision 230 by ph10, Mon Sep 10 13:23:56 2007 UTC revision 336 by ph10, Sat Apr 12 15:59:03 2008 UTC
# Line 6  Line 6 
6  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
7    
8                         Written by Philip Hazel                         Written by Philip Hazel
9             Copyright (c) 1997-2007 University of Cambridge             Copyright (c) 1997-2008 University of Cambridge
10    
11  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
12  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 43  pattern matching using an NFA algorithm, Line 43  pattern matching using an NFA algorithm,
43  possible. There are also some static supporting functions. */  possible. There are also some static supporting functions. */
44    
45  #ifdef HAVE_CONFIG_H  #ifdef HAVE_CONFIG_H
46  #include <config.h>  #include "config.h"
47  #endif  #endif
48    
49  #define NLBLOCK md             /* Block containing newline information */  #define NLBLOCK md             /* Block containing newline information */
# Line 1148  for (;;) Line 1148  for (;;)
1148      do ecode += GET(ecode,1); while (*ecode == OP_ALT);      do ecode += GET(ecode,1); while (*ecode == OP_ALT);
1149      break;      break;
1150    
1151      /* BRAZERO and BRAMINZERO occur just before a bracket group, indicating      /* BRAZERO, BRAMINZERO and SKIPZERO occur just before a bracket group,
1152      that it may occur zero times. It may repeat infinitely, or not at all -      indicating that it may occur zero times. It may repeat infinitely, or not
1153      i.e. it could be ()* or ()? in the pattern. Brackets with fixed upper      at all - i.e. it could be ()* or ()? or even (){0} in the pattern. Brackets
1154      repeat limits are compiled as a number of copies, with the optional ones      with fixed upper repeat limits are compiled as a number of copies, with the
1155      preceded by BRAZERO or BRAMINZERO. */      optional ones preceded by BRAZERO or BRAMINZERO. */
1156    
1157      case OP_BRAZERO:      case OP_BRAZERO:
1158        {        {
# Line 1174  for (;;) Line 1174  for (;;)
1174        }        }
1175      break;      break;
1176    
1177        case OP_SKIPZERO:
1178          {
1179          next = ecode+1;
1180          do next += GET(next,1); while (*next == OP_ALT);
1181          ecode = next + 1 + LINK_SIZE;
1182          }
1183        break;
1184    
1185      /* End of a group, repeated or non-repeating. */      /* End of a group, repeated or non-repeating. */
1186    
1187      case OP_KET:      case OP_KET:
# Line 1526  for (;;) Line 1534  for (;;)
1534        case 0x000d:        case 0x000d:
1535        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1536        break;        break;
1537    
1538        case 0x000a:        case 0x000a:
1539          break;
1540    
1541        case 0x000b:        case 0x000b:
1542        case 0x000c:        case 0x000c:
1543        case 0x0085:        case 0x0085:
1544        case 0x2028:        case 0x2028:
1545        case 0x2029:        case 0x2029:
1546          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1547        break;        break;
1548        }        }
1549      ecode++;      ecode++;
# Line 1719  for (;;) Line 1731  for (;;)
1731      case OP_REF:      case OP_REF:
1732        {        {
1733        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
1734        ecode += 3;                                 /* Advance past item */        ecode += 3;
1735    
1736        /* If the reference is unset, set the length to be longer than the amount        /* If the reference is unset, there are two possibilities:
1737        of subject left; this ensures that every attempt at a match fails. We  
1738        can't just fail here, because of the possibility of quantifiers with zero        (a) In the default, Perl-compatible state, set the length to be longer
1739        minima. */        than the amount of subject left; this ensures that every attempt at a
1740          match fails. We can't just fail here, because of the possibility of
1741        length = (offset >= offset_top || md->offset_vector[offset] < 0)?        quantifiers with zero minima.
1742          md->end_subject - eptr + 1 :  
1743          md->offset_vector[offset+1] - md->offset_vector[offset];        (b) If the JavaScript compatibility flag is set, set the length to zero
1744          so that the back reference matches an empty string.
1745    
1746          Otherwise, set the length to the length of what was matched by the
1747          referenced subpattern. */
1748    
1749          if (offset >= offset_top || md->offset_vector[offset] < 0)
1750            length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
1751          else
1752            length = md->offset_vector[offset+1] - md->offset_vector[offset];
1753    
1754        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
1755    
# Line 2954  for (;;) Line 2975  for (;;)
2975              case 0x000d:              case 0x000d:
2976              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2977              break;              break;
2978    
2979              case 0x000a:              case 0x000a:
2980                break;
2981    
2982              case 0x000b:              case 0x000b:
2983              case 0x000c:              case 0x000c:
2984              case 0x0085:              case 0x0085:
2985              case 0x2028:              case 0x2028:
2986              case 0x2029:              case 0x2029:
2987                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2988              break;              break;
2989              }              }
2990            }            }
# Line 3172  for (;;) Line 3197  for (;;)
3197              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3198              break;              break;
3199              case 0x000a:              case 0x000a:
3200                break;
3201    
3202              case 0x000b:              case 0x000b:
3203              case 0x000c:              case 0x000c:
3204              case 0x0085:              case 0x0085:
3205                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3206              break;              break;
3207              }              }
3208            }            }
# Line 3426  for (;;) Line 3454  for (;;)
3454                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3455                break;                break;
3456                case 0x000a:                case 0x000a:
3457                  break;
3458    
3459                case 0x000b:                case 0x000b:
3460                case 0x000c:                case 0x000c:
3461                case 0x0085:                case 0x0085:
3462                case 0x2028:                case 0x2028:
3463                case 0x2029:                case 0x2029:
3464                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3465                break;                break;
3466                }                }
3467              break;              break;
# Line 3582  for (;;) Line 3613  for (;;)
3613                case 0x000d:                case 0x000d:
3614                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3615                break;                break;
3616    
3617                case 0x000a:                case 0x000a:
3618                  break;
3619    
3620                case 0x000b:                case 0x000b:
3621                case 0x000c:                case 0x000c:
3622                case 0x0085:                case 0x0085:
3623                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3624                break;                break;
3625                }                }
3626              break;              break;
# Line 3883  for (;;) Line 3918  for (;;)
3918                }                }
3919              else              else
3920                {                {
3921                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
3922                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
3923                       (c != 0x000b && c != 0x000c &&
3924                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
3925                  break;                  break;
3926                eptr += len;                eptr += len;
3927                }                }
# Line 4074  for (;;) Line 4111  for (;;)
4111                }                }
4112              else              else
4113                {                {
4114                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4115                      (md->bsr_anycrlf ||
4116                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4117                  break;                  break;
4118                eptr++;                eptr++;
4119                }                }
# Line 4224  HEAP_RETURN: Line 4263  HEAP_RETURN:
4263  switch (frame->Xwhere)  switch (frame->Xwhere)
4264    {    {
4265    LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)    LBL( 1) LBL( 2) LBL( 3) LBL( 4) LBL( 5) LBL( 6) LBL( 7) LBL( 8)
4266    LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(16)    LBL( 9) LBL(10) LBL(11) LBL(12) LBL(13) LBL(14) LBL(15) LBL(17)
4267    LBL(17) LBL(18) LBL(19) LBL(20) LBL(21) LBL(22) LBL(23) LBL(24)    LBL(19) LBL(24) LBL(25) LBL(26) LBL(27) LBL(29) LBL(31) LBL(33)
4268    LBL(25) LBL(26) LBL(27) LBL(28) LBL(29) LBL(30) LBL(31) LBL(32)    LBL(35) LBL(43) LBL(47) LBL(48) LBL(49) LBL(50) LBL(51) LBL(52)
4269    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4270    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47) LBL(48)  #ifdef SUPPORT_UTF8
4271    LBL(49) LBL(50) LBL(51) LBL(52) LBL(53) LBL(54)    LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4272      LBL(32) LBL(34) LBL(42) LBL(46)
4273    #ifdef SUPPORT_UCP
4274      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4275    #endif  /* SUPPORT_UCP */
4276    #endif  /* SUPPORT_UTF8 */
4277    default:    default:
4278    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));    DPRINTF(("jump error in pcre match: label %d non-existent\n", frame->Xwhere));
4279    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 4423  end_subject = md->end_subject; Line 4467  end_subject = md->end_subject;
4467    
4468  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
4469  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
4470    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
4471    
4472  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
4473  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
# Line 4435  md->recursive = NULL; Line 4480  md->recursive = NULL;
4480  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4481  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4482    
4483    /* Handle different \R options. */
4484    
4485    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4486      {
4487      case 0:
4488      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4489        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4490      else
4491    #ifdef BSR_ANYCRLF
4492      md->bsr_anycrlf = TRUE;
4493    #else
4494      md->bsr_anycrlf = FALSE;
4495    #endif
4496      break;
4497    
4498      case PCRE_BSR_ANYCRLF:
4499      md->bsr_anycrlf = TRUE;
4500      break;
4501    
4502      case PCRE_BSR_UNICODE:
4503      md->bsr_anycrlf = FALSE;
4504      break;
4505    
4506      default: return PCRE_ERROR_BADNEWLINE;
4507      }
4508    
4509  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4510  nothing is set at run time, whatever was used at compile time applies. */  nothing is set at run time, whatever was used at compile time applies. */
4511    
4512  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4513          (pcre_uint32)options) & PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4514    {    {
4515    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
# Line 4617  for(;;) Line 4688  for(;;)
4688      if (first_byte_caseless)      if (first_byte_caseless)
4689        while (start_match < end_subject &&        while (start_match < end_subject &&
4690               md->lcc[*start_match] != first_byte)               md->lcc[*start_match] != first_byte)
4691          start_match++;          { NEXTCHAR(start_match); }
4692      else      else
4693        while (start_match < end_subject && *start_match != first_byte)        while (start_match < end_subject && *start_match != first_byte)
4694          start_match++;          { NEXTCHAR(start_match); }
4695      }      }
4696    
4697    /* Or to just after a linebreak for a multiline match if possible */    /* Or to just after a linebreak for a multiline match if possible */
# Line 4630  for(;;) Line 4701  for(;;)
4701      if (start_match > md->start_subject + start_offset)      if (start_match > md->start_subject + start_offset)
4702        {        {
4703        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        while (start_match <= end_subject && !WAS_NEWLINE(start_match))
4704          start_match++;          { NEXTCHAR(start_match); }
4705    
4706        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,        /* If we have just passed a CR and the newline option is ANY or ANYCRLF,
4707        and we are now at a LF, advance the match position by one more character.        and we are now at a LF, advance the match position by one more character.
# Line 4651  for(;;) Line 4722  for(;;)
4722      while (start_match < end_subject)      while (start_match < end_subject)
4723        {        {
4724        register unsigned int c = *start_match;        register unsigned int c = *start_match;
4725        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;        if ((start_bits[c/8] & (1 << (c&7))) == 0)
4726            { NEXTCHAR(start_match); }
4727          else break;
4728        }        }
4729      }      }
4730    

Legend:
Removed from v.230  
changed lines
  Added in v.336

  ViewVC Help
Powered by ViewVC 1.1.5