/[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 199 by ph10, Tue Jul 31 14:39:09 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 42  POSSIBILITY OF SUCH DAMAGE. Line 42  POSSIBILITY OF SUCH DAMAGE.
42  pattern matching using an NFA algorithm, trying to mimic Perl as closely as  pattern matching using an NFA algorithm, trying to mimic Perl as closely as
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 */
50  #define PSSTART start_subject  /* Field containing processed string start */  #define PSSTART start_subject  /* Field containing processed string start */
# Line 68  defined PCRE_ERROR_xxx codes, which are Line 68  defined PCRE_ERROR_xxx codes, which are
68  #define MATCH_MATCH        1  #define MATCH_MATCH        1
69  #define MATCH_NOMATCH      0  #define MATCH_NOMATCH      0
70    
71    /* Special internal returns from the match() function. Make them sufficiently
72    negative to avoid the external error codes. */
73    
74    #define MATCH_COMMIT       (-999)
75    #define MATCH_PRUNE        (-998)
76    #define MATCH_SKIP         (-997)
77    #define MATCH_THEN         (-996)
78    
79  /* Maximum number of ints of offset to save on the stack for recursive calls.  /* Maximum number of ints of offset to save on the stack for recursive calls.
80  If the offset vector is bigger, malloc is used. This should be a multiple of 3,  If the offset vector is bigger, malloc is used. This should be a multiple of 3,
81  because the offset vector is always a multiple of 3 long. */  because the offset vector is always a multiple of 3 long. */
# Line 203  variable instead of being passed in the Line 211  variable instead of being passed in the
211  ****************************************************************************  ****************************************************************************
212  ***************************************************************************/  ***************************************************************************/
213    
214    /* Numbers for RMATCH calls. When this list is changed, the code at HEAP_RETURN
215  /* Numbers for RMATCH calls */  below must be updated in sync.  */
216    
217  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM6,  RM7,  RM8,  RM9,  RM10,
218         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,         RM11,  RM12, RM13, RM14, RM15, RM16, RM17, RM18, RM19, RM20,
219         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,         RM21,  RM22, RM23, RM24, RM25, RM26, RM27, RM28, RM29, RM30,
220         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
221         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50 };         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
222           RM51,  RM52, RM53, RM54 };
223    
224  /* These versions of the macros use the stack, as normal. There are debugging  /* These versions of the macros use the stack, as normal. There are debugging
225  versions and production versions. Note that the "rw" argument of RMATCH isn't  versions and production versions. Note that the "rw" argument of RMATCH isn't
# Line 612  for (;;) Line 620  for (;;)
620    
621    switch(op)    switch(op)
622      {      {
623        case OP_FAIL:
624        RRETURN(MATCH_NOMATCH);
625    
626        case OP_PRUNE:
627        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
628          ims, eptrb, flags, RM51);
629        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
630        RRETURN(MATCH_PRUNE);
631    
632        case OP_COMMIT:
633        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
634          ims, eptrb, flags, RM52);
635        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
636        RRETURN(MATCH_COMMIT);
637    
638        case OP_SKIP:
639        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
640          ims, eptrb, flags, RM53);
641        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
642        md->start_match_ptr = eptr;   /* Pass back current position */
643        RRETURN(MATCH_SKIP);
644    
645        case OP_THEN:
646        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
647          ims, eptrb, flags, RM54);
648        if (rrc != MATCH_NOMATCH) RRETURN(rrc);
649        RRETURN(MATCH_THEN);
650    
651      /* Handle a capturing bracket. If there is space in the offset vector, save      /* Handle a capturing bracket. If there is space in the offset vector, save
652      the current subject position in the working slot at the top of the vector.      the current subject position in the working slot at the top of the vector.
653      We mustn't change the current values of the data slot, because they may be      We mustn't change the current values of the data slot, because they may be
# Line 653  for (;;) Line 689  for (;;)
689          {          {
690          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,          RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md,
691            ims, eptrb, flags, RM1);            ims, eptrb, flags, RM1);
692          if (rrc != MATCH_NOMATCH) RRETURN(rrc);          if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
693          md->capture_last = save_capture_last;          md->capture_last = save_capture_last;
694          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
695          }          }
# Line 712  for (;;) Line 748  for (;;)
748    
749        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,        RMATCH(eptr, ecode + _pcre_OP_lengths[*ecode], offset_top, md, ims,
750          eptrb, flags, RM2);          eptrb, flags, RM2);
751        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
752        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
753        }        }
754      /* Control never reaches here. */      /* Control never reaches here. */
# Line 760  for (;;) Line 796  for (;;)
796          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);          ecode += 1 + LINK_SIZE + GET(ecode, LINK_SIZE + 2);
797          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
798          }          }
799        else if (rrc != MATCH_NOMATCH)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
800          {          {
801          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
802          }          }
# Line 797  for (;;) Line 833  for (;;)
833      break;      break;
834    
835    
836      /* End of the pattern. If we are in a top-level recursion, we should      /* End of the pattern, either real or forced. If we are in a top-level
837      restore the offsets appropriately and continue from after the call. */      recursion, we should restore the offsets appropriately and continue from
838        after the call. */
839    
840        case OP_ACCEPT:
841      case OP_END:      case OP_END:
842      if (md->recursive != NULL && md->recursive->group_num == 0)      if (md->recursive != NULL && md->recursive->group_num == 0)
843        {        {
# Line 820  for (;;) Line 858  for (;;)
858      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);      if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH);
859      md->end_match_ptr = eptr;           /* Record where we ended */      md->end_match_ptr = eptr;           /* Record where we ended */
860      md->end_offset_top = offset_top;    /* and how many extracts were taken */      md->end_offset_top = offset_top;    /* and how many extracts were taken */
861      md->start_match_ptr = mstart;  /* and the start (\K can modify) */      md->start_match_ptr = mstart;       /* and the start (\K can modify) */
862      RRETURN(MATCH_MATCH);      RRETURN(MATCH_MATCH);
863    
864      /* Change option settings */      /* Change option settings */
# Line 844  for (;;) Line 882  for (;;)
882        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
883          RM4);          RM4);
884        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
885        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
886        ecode += GET(ecode, 1);        ecode += GET(ecode, 1);
887        }        }
888      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 871  for (;;) Line 909  for (;;)
909        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, NULL, 0,
910          RM5);          RM5);
911        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);        if (rrc == MATCH_MATCH) RRETURN(MATCH_NOMATCH);
912        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
913        ecode += GET(ecode,1);        ecode += GET(ecode,1);
914        }        }
915      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 895  for (;;) Line 933  for (;;)
933          {          {
934          eptr--;          eptr--;
935          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);          if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH);
936          BACKCHAR(eptr)          BACKCHAR(eptr);
937          }          }
938        }        }
939      else      else
# Line 1008  for (;;) Line 1046  for (;;)
1046              (pcre_free)(new_recursive.offset_save);              (pcre_free)(new_recursive.offset_save);
1047            RRETURN(MATCH_MATCH);            RRETURN(MATCH_MATCH);
1048            }            }
1049          else if (rrc != MATCH_NOMATCH)          else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1050            {            {
1051            DPRINTF(("Recursion gave error %d\n", rrc));            DPRINTF(("Recursion gave error %d\n", rrc));
1052            RRETURN(rrc);            RRETURN(rrc);
# Line 1044  for (;;) Line 1082  for (;;)
1082        {        {
1083        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);        RMATCH(eptr, ecode + 1 + LINK_SIZE, offset_top, md, ims, eptrb, 0, RM7);
1084        if (rrc == MATCH_MATCH) break;        if (rrc == MATCH_MATCH) break;
1085        if (rrc != MATCH_NOMATCH) RRETURN(rrc);        if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) RRETURN(rrc);
1086        ecode += GET(ecode,1);        ecode += GET(ecode,1);
1087        }        }
1088      while (*ecode == OP_ALT);      while (*ecode == OP_ALT);
# Line 1110  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 1136  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 1488  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 1681  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 2050  for (;;) Line 2109  for (;;)
2109            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2110            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2111            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2112            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2113            }            }
2114          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2115          }          }
# Line 2916  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 3055  for (;;) Line 3118  for (;;)
3118          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3119            {            {
3120            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3121               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))
3122              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3123            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3124            }            }
3125          break;          break;
3126    
# Line 3075  for (;;) Line 3138  for (;;)
3138          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3139            {            {
3140            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3141               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3142              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3143            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3144            }            }
3145          break;          break;
3146    
# Line 3134  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 3388  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 3544  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 3719  for (;;) Line 3792  for (;;)
3792            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
3793            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3794            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3795            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
3796            }            }
3797          }          }
3798    
# Line 3758  for (;;) Line 3831  for (;;)
3831            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
3832              {              {
3833              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
3834              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
3835                {                {
3836                  BACKCHAR(eptr);
3837                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
3838                }                }
3839              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
# Line 3845  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 3999  for (;;) Line 4074  for (;;)
4074            }            }
4075          }          }
4076        else        else
4077  #endif  #endif  /* SUPPORT_UTF8 */
4078    
4079        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4080          {          {
# Line 4036  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 4186  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)  #ifdef SUPPORT_UTF8
4271      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 4369  if (re->magic_number != MAGIC_NUMBER) Line 4452  if (re->magic_number != MAGIC_NUMBER)
4452  /* Set up other data */  /* Set up other data */
4453    
4454  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4455  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4456  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4457    
4458  /* The code starts after the real_pcre block and the capture name table. */  /* The code starts after the real_pcre block and the capture name table. */
# Line 4384  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 4396  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 : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4513         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 */
4516    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = '\r'; break;
# Line 4439  else Line 4549  else
4549  /* Partial matching is supported only for a restricted set of regexes at the  /* Partial matching is supported only for a restricted set of regexes at the
4550  moment. */  moment. */
4551    
4552  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4553    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4554    
4555  /* Check a UTF-8 string if required. Unfortunately there's no way of passing  /* Check a UTF-8 string if required. Unfortunately there's no way of passing
# Line 4516  studied, there may be a bitmap of possib Line 4626  studied, there may be a bitmap of possib
4626    
4627  if (!anchored)  if (!anchored)
4628    {    {
4629    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
4630      {      {
4631      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
4632      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)
# Line 4531  if (!anchored) Line 4641  if (!anchored)
4641  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
4642  character" set. */  character" set. */
4643    
4644  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
4645    {    {
4646    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
4647    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;
# Line 4547  the loop runs just once. */ Line 4657  the loop runs just once. */
4657  for(;;)  for(;;)
4658    {    {
4659    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
4660      USPTR new_start_match;
4661    
4662    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
4663    
# Line 4577  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 4590  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 4611  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    
# Line 4687  for(;;) Line 4800  for(;;)
4800    
4801    /* OK, we can now run the match. */    /* OK, we can now run the match. */
4802    
4803    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
4804    md->match_call_count = 0;    md->match_call_count = 0;
4805    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);    rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0);
4806    
4807    /* Any return other than MATCH_NOMATCH breaks the loop. */    switch(rc)
4808        {
4809        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
4810        exactly like PRUNE. */
4811    
4812    if (rc != MATCH_NOMATCH) break;      case MATCH_NOMATCH:
4813        case MATCH_PRUNE:
4814        case MATCH_THEN:
4815        new_start_match = start_match + 1;
4816    #ifdef SUPPORT_UTF8
4817        if (utf8)
4818          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
4819            new_start_match++;
4820    #endif
4821        break;
4822    
4823        /* SKIP passes back the next starting point explicitly. */
4824    
4825        case MATCH_SKIP:
4826        new_start_match = md->start_match_ptr;
4827        break;
4828    
4829        /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
4830    
4831        case MATCH_COMMIT:
4832        rc = MATCH_NOMATCH;
4833        goto ENDLOOP;
4834    
4835        /* Any other return is some kind of error. */
4836    
4837        default:
4838        goto ENDLOOP;
4839        }
4840    
4841      /* Control reaches here for the various types of "no match at this point"
4842      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
4843    
4844      rc = MATCH_NOMATCH;
4845    
4846    /* If PCRE_FIRSTLINE is set, the match must happen before or at the first    /* If PCRE_FIRSTLINE is set, the match must happen before or at the first
4847    newline in the subject (though it may continue over the newline). Therefore,    newline in the subject (though it may continue over the newline). Therefore,
# Line 4701  for(;;) Line 4849  for(;;)
4849    
4850    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
4851    
4852    /* Advance the match position by one character. */    /* Advance to new matching position */
4853    
4854    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
4855    
4856    /* Break the loop if the pattern is anchored or if we have passed the end of    /* Break the loop if the pattern is anchored or if we have passed the end of
4857    the subject. */    the subject. */
4858    
4859    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
4860    
4861    /* If we have just passed a CR and the newline option is CRLF or ANY or    /* If we have just passed a CR and we are now at a LF, and the pattern does
4862    ANYCRLF, and we are now at a LF, advance the match position by one more    not contain any explicit matches for \r or \n, and the newline option is CRLF
4863    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
4864    
4865    if (start_match[-1] == '\r' &&    if (start_match[-1] == '\r' &&
4866         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
4867          md->nltype == NLTYPE_ANYCRLF ||        *start_match == '\n' &&
4868          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
4869         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
4870         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
4871             md->nllen == 2))
4872      start_match++;      start_match++;
4873    
4874    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4734  for(;;) Line 4878  for(;;)
4878  /* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping  /* We reach here when rc is not MATCH_NOMATCH, or if one of the stopping
4879  conditions is true:  conditions is true:
4880    
4881  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
4882    
4883  (2) We are past the end of the subject;  (2) We are past the end of the subject;
4884    
# Line 4749  processing, copy those that we can. In t Line 4893  processing, copy those that we can. In t
4893  certain parts of the pattern were not used, even though there are more  certain parts of the pattern were not used, even though there are more
4894  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
4895    
4896    ENDLOOP:
4897    
4898  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
4899    {    {
4900    if (using_temporary_offsets)    if (using_temporary_offsets)

Legend:
Removed from v.199  
changed lines
  Added in v.336

  ViewVC Help
Powered by ViewVC 1.1.5