/[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 335 by ph10, Sat Apr 12 14:36:14 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 2050  for (;;) Line 2100  for (;;)
2100            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM21);
2101            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
2102            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
2103            BACKCHAR(eptr)            if (utf8) BACKCHAR(eptr);
2104            }            }
2105          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
2106          }          }
# Line 2916  for (;;) Line 2966  for (;;)
2966              case 0x000d:              case 0x000d:
2967              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2968              break;              break;
2969    
2970              case 0x000a:              case 0x000a:
2971                break;
2972    
2973              case 0x000b:              case 0x000b:
2974              case 0x000c:              case 0x000c:
2975              case 0x0085:              case 0x0085:
2976              case 0x2028:              case 0x2028:
2977              case 0x2029:              case 0x2029:
2978                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
2979              break;              break;
2980              }              }
2981            }            }
# Line 3055  for (;;) Line 3109  for (;;)
3109          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3110            {            {
3111            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3112               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_space) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0))
3113              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3114            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3115            }            }
3116          break;          break;
3117    
# Line 3075  for (;;) Line 3129  for (;;)
3129          for (i = 1; i <= min; i++)          for (i = 1; i <= min; i++)
3130            {            {
3131            if (eptr >= md->end_subject ||            if (eptr >= md->end_subject ||
3132               (*eptr < 128 && (md->ctypes[*eptr++] & ctype_word) != 0))               (*eptr < 128 && (md->ctypes[*eptr] & ctype_word) != 0))
3133              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3134            while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;            while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80);
3135            }            }
3136          break;          break;
3137    
# Line 3134  for (;;) Line 3188  for (;;)
3188              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3189              break;              break;
3190              case 0x000a:              case 0x000a:
3191                break;
3192    
3193              case 0x000b:              case 0x000b:
3194              case 0x000c:              case 0x000c:
3195              case 0x0085:              case 0x0085:
3196                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3197              break;              break;
3198              }              }
3199            }            }
# Line 3388  for (;;) Line 3445  for (;;)
3445                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3446                break;                break;
3447                case 0x000a:                case 0x000a:
3448                  break;
3449    
3450                case 0x000b:                case 0x000b:
3451                case 0x000c:                case 0x000c:
3452                case 0x0085:                case 0x0085:
3453                case 0x2028:                case 0x2028:
3454                case 0x2029:                case 0x2029:
3455                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3456                break;                break;
3457                }                }
3458              break;              break;
# Line 3544  for (;;) Line 3604  for (;;)
3604                case 0x000d:                case 0x000d:
3605                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3606                break;                break;
3607    
3608                case 0x000a:                case 0x000a:
3609                  break;
3610    
3611                case 0x000b:                case 0x000b:
3612                case 0x000c:                case 0x000c:
3613                case 0x0085:                case 0x0085:
3614                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3615                break;                break;
3616                }                }
3617              break;              break;
# Line 3719  for (;;) Line 3783  for (;;)
3783            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);            RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM44);
3784            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
3785            if (eptr-- == pp) break;        /* Stop if tried at original pos */            if (eptr-- == pp) break;        /* Stop if tried at original pos */
3786            BACKCHAR(eptr);            if (utf8) BACKCHAR(eptr);
3787            }            }
3788          }          }
3789    
# Line 3758  for (;;) Line 3822  for (;;)
3822            for (;;)                        /* Move back over one extended */            for (;;)                        /* Move back over one extended */
3823              {              {
3824              int len = 1;              int len = 1;
             BACKCHAR(eptr);  
3825              if (!utf8) c = *eptr; else              if (!utf8) c = *eptr; else
3826                {                {
3827                  BACKCHAR(eptr);
3828                GETCHARLEN(c, eptr, len);                GETCHARLEN(c, eptr, len);
3829                }                }
3830              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);              prop_category = _pcre_ucp_findprop(c, &prop_chartype, &prop_script);
# Line 3845  for (;;) Line 3909  for (;;)
3909                }                }
3910              else              else
3911                {                {
3912                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
3913                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
3914                       (c != 0x000b && c != 0x000c &&
3915                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
3916                  break;                  break;
3917                eptr += len;                eptr += len;
3918                }                }
# Line 3999  for (;;) Line 4065  for (;;)
4065            }            }
4066          }          }
4067        else        else
4068  #endif  #endif  /* SUPPORT_UTF8 */
4069    
4070        /* Not UTF-8 mode */        /* Not UTF-8 mode */
4071          {          {
# Line 4036  for (;;) Line 4102  for (;;)
4102                }                }
4103              else              else
4104                {                {
4105                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4106                      (md->bsr_anycrlf ||
4107                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4108                  break;                  break;
4109                eptr++;                eptr++;
4110                }                }
# Line 4186  HEAP_RETURN: Line 4254  HEAP_RETURN:
4254  switch (frame->Xwhere)  switch (frame->Xwhere)
4255    {    {
4256    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)
4257    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)
4258    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)
4259    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)
4260    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4261    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47)  #ifdef SUPPORT_UTF8
4262      LBL(16) LBL(18) LBL(20) LBL(21) LBL(22) LBL(23) LBL(28) LBL(30)
4263      LBL(32) LBL(34) LBL(42) LBL(46)
4264    #ifdef SUPPORT_UCP
4265      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4266    #endif  /* SUPPORT_UCP */
4267    #endif  /* SUPPORT_UTF8 */
4268    default:    default:
4269    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));
4270    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 4369  if (re->magic_number != MAGIC_NUMBER) Line 4443  if (re->magic_number != MAGIC_NUMBER)
4443  /* Set up other data */  /* Set up other data */
4444    
4445  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4446  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4447  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4448    
4449  /* 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 4396  md->recursive = NULL; Line 4470  md->recursive = NULL;
4470  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4471  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4472    
4473    /* Handle different \R options. */
4474    
4475    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4476      {
4477      case 0:
4478      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4479        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4480      else
4481    #ifdef BSR_ANYCRLF
4482      md->bsr_anycrlf = TRUE;
4483    #else
4484      md->bsr_anycrlf = FALSE;
4485    #endif
4486      break;
4487    
4488      case PCRE_BSR_ANYCRLF:
4489      md->bsr_anycrlf = TRUE;
4490      break;
4491    
4492      case PCRE_BSR_UNICODE:
4493      md->bsr_anycrlf = FALSE;
4494      break;
4495    
4496      default: return PCRE_ERROR_BADNEWLINE;
4497      }
4498    
4499  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4500  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. */
4501    
4502  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4503         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4504    {    {
4505    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4506    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = '\r'; break;
# Line 4439  else Line 4539  else
4539  /* 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
4540  moment. */  moment. */
4541    
4542  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4543    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4544    
4545  /* 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 4616  studied, there may be a bitmap of possib
4616    
4617  if (!anchored)  if (!anchored)
4618    {    {
4619    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
4620      {      {
4621      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
4622      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 4631  if (!anchored)
4631  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
4632  character" set. */  character" set. */
4633    
4634  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
4635    {    {
4636    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
4637    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 4647  the loop runs just once. */
4647  for(;;)  for(;;)
4648    {    {
4649    USPTR save_end_subject = end_subject;    USPTR save_end_subject = end_subject;
4650      USPTR new_start_match;
4651    
4652    /* Reset the maximum number of extractions we might see. */    /* Reset the maximum number of extractions we might see. */
4653    
# Line 4577  for(;;) Line 4678  for(;;)
4678      if (first_byte_caseless)      if (first_byte_caseless)
4679        while (start_match < end_subject &&        while (start_match < end_subject &&
4680               md->lcc[*start_match] != first_byte)               md->lcc[*start_match] != first_byte)
4681          start_match++;          { NEXTCHAR(start_match); }
4682      else      else
4683        while (start_match < end_subject && *start_match != first_byte)        while (start_match < end_subject && *start_match != first_byte)
4684          start_match++;          { NEXTCHAR(start_match); }
4685      }      }
4686    
4687    /* 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 4691  for(;;)
4691      if (start_match > md->start_subject + start_offset)      if (start_match > md->start_subject + start_offset)
4692        {        {
4693        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        while (start_match <= end_subject && !WAS_NEWLINE(start_match))
4694          start_match++;          { NEXTCHAR(start_match); }
4695    
4696        /* 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,
4697        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 4712  for(;;)
4712      while (start_match < end_subject)      while (start_match < end_subject)
4713        {        {
4714        register unsigned int c = *start_match;        register unsigned int c = *start_match;
4715        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;        if ((start_bits[c/8] & (1 << (c&7))) == 0)
4716            { NEXTCHAR(start_match); }
4717          else break;
4718        }        }
4719      }      }
4720    
# Line 4687  for(;;) Line 4790  for(;;)
4790    
4791    /* OK, we can now run the match. */    /* OK, we can now run the match. */
4792    
4793    md->start_match_ptr = start_match;      /* Insurance */    md->start_match_ptr = start_match;
4794    md->match_call_count = 0;    md->match_call_count = 0;
4795    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);
4796    
4797    /* Any return other than MATCH_NOMATCH breaks the loop. */    switch(rc)
4798        {
4799        /* NOMATCH and PRUNE advance by one character. THEN at this level acts
4800        exactly like PRUNE. */
4801    
4802        case MATCH_NOMATCH:
4803        case MATCH_PRUNE:
4804        case MATCH_THEN:
4805        new_start_match = start_match + 1;
4806    #ifdef SUPPORT_UTF8
4807        if (utf8)
4808          while(new_start_match < end_subject && (*new_start_match & 0xc0) == 0x80)
4809            new_start_match++;
4810    #endif
4811        break;
4812    
4813        /* SKIP passes back the next starting point explicitly. */
4814    
4815        case MATCH_SKIP:
4816        new_start_match = md->start_match_ptr;
4817        break;
4818    
4819        /* COMMIT disables the bumpalong, but otherwise behaves as NOMATCH. */
4820    
4821    if (rc != MATCH_NOMATCH) break;      case MATCH_COMMIT:
4822        rc = MATCH_NOMATCH;
4823        goto ENDLOOP;
4824    
4825        /* Any other return is some kind of error. */
4826    
4827        default:
4828        goto ENDLOOP;
4829        }
4830    
4831      /* Control reaches here for the various types of "no match at this point"
4832      result. Reset the code to MATCH_NOMATCH for subsequent checking. */
4833    
4834      rc = MATCH_NOMATCH;
4835    
4836    /* 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
4837    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 4839  for(;;)
4839    
4840    if (firstline && IS_NEWLINE(start_match)) break;    if (firstline && IS_NEWLINE(start_match)) break;
4841    
4842    /* Advance the match position by one character. */    /* Advance to new matching position */
4843    
4844    start_match++;    start_match = new_start_match;
 #ifdef SUPPORT_UTF8  
   if (utf8)  
     while(start_match < end_subject && (*start_match & 0xc0) == 0x80)  
       start_match++;  
 #endif  
4845    
4846    /* 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
4847    the subject. */    the subject. */
4848    
4849    if (anchored || start_match > end_subject) break;    if (anchored || start_match > end_subject) break;
4850    
4851    /* 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
4852    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
4853    character. */    or ANY or ANYCRLF, advance the match position by one more character. */
4854    
4855    if (start_match[-1] == '\r' &&    if (start_match[-1] == '\r' &&
4856         (md->nltype == NLTYPE_ANY ||        start_match < end_subject &&
4857          md->nltype == NLTYPE_ANYCRLF ||        *start_match == '\n' &&
4858          md->nllen == 2) &&        (re->flags & PCRE_HASCRORLF) == 0 &&
4859         start_match < end_subject &&          (md->nltype == NLTYPE_ANY ||
4860         *start_match == '\n')           md->nltype == NLTYPE_ANYCRLF ||
4861             md->nllen == 2))
4862      start_match++;      start_match++;
4863    
4864    }   /* End of for(;;) "bumpalong" loop */    }   /* End of for(;;) "bumpalong" loop */
# Line 4734  for(;;) Line 4868  for(;;)
4868  /* 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
4869  conditions is true:  conditions is true:
4870    
4871  (1) The pattern is anchored;  (1) The pattern is anchored or the match was failed by (*COMMIT);
4872    
4873  (2) We are past the end of the subject;  (2) We are past the end of the subject;
4874    
# Line 4749  processing, copy those that we can. In t Line 4883  processing, copy those that we can. In t
4883  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
4884  capturing parentheses than vector slots. */  capturing parentheses than vector slots. */
4885    
4886    ENDLOOP:
4887    
4888  if (rc == MATCH_MATCH)  if (rc == MATCH_MATCH)
4889    {    {
4890    if (using_temporary_offsets)    if (using_temporary_offsets)

Legend:
Removed from v.199  
changed lines
  Added in v.335

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12