/[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 227 by ph10, Tue Aug 21 15:00:15 2007 UTC revision 341 by ph10, Sat Apr 19 16:41:04 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 1425  for (;;) Line 1433  for (;;)
1433        {        {
1434        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);        if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH);
1435        }        }
1436        /* Fall through */
1437    
1438        case OP_ALLANY:
1439      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);      if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH);
1440      if (utf8)      if (utf8)
1441        while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;        while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
# Line 1526  for (;;) Line 1537  for (;;)
1537        case 0x000d:        case 0x000d:
1538        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;        if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
1539        break;        break;
1540    
1541        case 0x000a:        case 0x000a:
1542          break;
1543    
1544        case 0x000b:        case 0x000b:
1545        case 0x000c:        case 0x000c:
1546        case 0x0085:        case 0x0085:
1547        case 0x2028:        case 0x2028:
1548        case 0x2029:        case 0x2029:
1549          if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
1550        break;        break;
1551        }        }
1552      ecode++;      ecode++;
# Line 1719  for (;;) Line 1734  for (;;)
1734      case OP_REF:      case OP_REF:
1735        {        {
1736        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */        offset = GET2(ecode, 1) << 1;               /* Doubled ref number */
1737        ecode += 3;                                 /* Advance past item */        ecode += 3;
1738    
1739        /* If the reference is unset, set the length to be longer than the amount        /* If the reference is unset, there are two possibilities:
1740        of subject left; this ensures that every attempt at a match fails. We  
1741        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
1742        minima. */        than the amount of subject left; this ensures that every attempt at a
1743          match fails. We can't just fail here, because of the possibility of
1744        length = (offset >= offset_top || md->offset_vector[offset] < 0)?        quantifiers with zero minima.
1745          md->end_subject - eptr + 1 :  
1746          md->offset_vector[offset+1] - md->offset_vector[offset];        (b) If the JavaScript compatibility flag is set, set the length to zero
1747          so that the back reference matches an empty string.
1748    
1749          Otherwise, set the length to the length of what was matched by the
1750          referenced subpattern. */
1751    
1752          if (offset >= offset_top || md->offset_vector[offset] < 0)
1753            length = (md->jscript_compat)? 0 : md->end_subject - eptr + 1;
1754          else
1755            length = md->offset_vector[offset+1] - md->offset_vector[offset];
1756    
1757        /* Set up for repetition, or handle the non-repeated case */        /* Set up for repetition, or handle the non-repeated case */
1758    
# Line 2939  for (;;) Line 2963  for (;;)
2963            }            }
2964          break;          break;
2965    
2966            case OP_ALLANY:
2967            for (i = 1; i <= min; i++)
2968              {
2969              if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH);
2970              eptr++;
2971              while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
2972              }
2973            break;
2974    
2975          case OP_ANYBYTE:          case OP_ANYBYTE:
2976          eptr += min;          eptr += min;
2977          break;          break;
# Line 2954  for (;;) Line 2987  for (;;)
2987              case 0x000d:              case 0x000d:
2988              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
2989              break;              break;
2990    
2991              case 0x000a:              case 0x000a:
2992                break;
2993    
2994              case 0x000b:              case 0x000b:
2995              case 0x000c:              case 0x000c:
2996              case 0x0085:              case 0x0085:
2997              case 0x2028:              case 0x2028:
2998              case 0x2029:              case 0x2029:
2999                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3000              break;              break;
3001              }              }
3002            }            }
# Line 3154  for (;;) Line 3191  for (;;)
3191          else eptr += min;          else eptr += min;
3192          break;          break;
3193    
3194            case OP_ALLANY:
3195            eptr += min;
3196            break;
3197    
3198          case OP_ANYBYTE:          case OP_ANYBYTE:
3199          eptr += min;          eptr += min;
3200          break;          break;
# Line 3172  for (;;) Line 3213  for (;;)
3213              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;              if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3214              break;              break;
3215              case 0x000a:              case 0x000a:
3216                break;
3217    
3218              case 0x000b:              case 0x000b:
3219              case 0x000c:              case 0x000c:
3220              case 0x0085:              case 0x0085:
3221                if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3222              break;              break;
3223              }              }
3224            }            }
# Line 3413  for (;;) Line 3457  for (;;)
3457            switch(ctype)            switch(ctype)
3458              {              {
3459              case OP_ANY:        /* This is the DOTALL case */              case OP_ANY:        /* This is the DOTALL case */
3460              break;              case OP_ALLANY:
   
3461              case OP_ANYBYTE:              case OP_ANYBYTE:
3462              break;              break;
3463    
# Line 3426  for (;;) Line 3469  for (;;)
3469                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3470                break;                break;
3471                case 0x000a:                case 0x000a:
3472                  break;
3473    
3474                case 0x000b:                case 0x000b:
3475                case 0x000c:                case 0x000c:
3476                case 0x0085:                case 0x0085:
3477                case 0x2028:                case 0x2028:
3478                case 0x2029:                case 0x2029:
3479                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3480                break;                break;
3481                }                }
3482              break;              break;
# Line 3569  for (;;) Line 3615  for (;;)
3615            c = *eptr++;            c = *eptr++;
3616            switch(ctype)            switch(ctype)
3617              {              {
3618              case OP_ANY:   /* This is the DOTALL case */              case OP_ANY:     /* This is the DOTALL case */
3619              break;              case OP_ALLANY:
   
3620              case OP_ANYBYTE:              case OP_ANYBYTE:
3621              break;              break;
3622    
# Line 3582  for (;;) Line 3627  for (;;)
3627                case 0x000d:                case 0x000d:
3628                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;                if (eptr < md->end_subject && *eptr == 0x0a) eptr++;
3629                break;                break;
3630    
3631                case 0x000a:                case 0x000a:
3632                  break;
3633    
3634                case 0x000b:                case 0x000b:
3635                case 0x000c:                case 0x000c:
3636                case 0x0085:                case 0x0085:
3637                  if (md->bsr_anycrlf) RRETURN(MATCH_NOMATCH);
3638                break;                break;
3639                }                }
3640              break;              break;
# Line 3861  for (;;) Line 3910  for (;;)
3910              }              }
3911            break;            break;
3912    
3913              case OP_ALLANY:
3914              if (max < INT_MAX)
3915                {
3916                for (i = min; i < max; i++)
3917                  {
3918                  if (eptr >= md->end_subject) break;
3919                  eptr++;
3920                  while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++;
3921                  }
3922                }
3923              else eptr = md->end_subject;   /* Unlimited UTF-8 repeat */
3924              break;
3925    
3926            /* The byte case is the same as non-UTF8 */            /* The byte case is the same as non-UTF8 */
3927    
3928            case OP_ANYBYTE:            case OP_ANYBYTE:
# Line 3883  for (;;) Line 3945  for (;;)
3945                }                }
3946              else              else
3947                {                {
3948                if (c != 0x000a && c != 0x000b && c != 0x000c &&                if (c != 0x000a &&
3949                    c != 0x0085 && c != 0x2028 && c != 0x2029)                    (md->bsr_anycrlf ||
3950                       (c != 0x000b && c != 0x000c &&
3951                        c != 0x0085 && c != 0x2028 && c != 0x2029)))
3952                  break;                  break;
3953                eptr += len;                eptr += len;
3954                }                }
# Line 4053  for (;;) Line 4117  for (;;)
4117                }                }
4118              break;              break;
4119              }              }
4120            /* For DOTALL case, fall through and treat as \C */            /* For DOTALL case, fall through */
4121    
4122              case OP_ALLANY:
4123            case OP_ANYBYTE:            case OP_ANYBYTE:
4124            c = max - min;            c = max - min;
4125            if (c > (unsigned int)(md->end_subject - eptr))            if (c > (unsigned int)(md->end_subject - eptr))
# Line 4074  for (;;) Line 4139  for (;;)
4139                }                }
4140              else              else
4141                {                {
4142                if (c != 0x000a && c != 0x000b && c != 0x000c && c != 0x0085)                if (c != 0x000a &&
4143                      (md->bsr_anycrlf ||
4144                        (c != 0x000b && c != 0x000c && c != 0x0085)))
4145                  break;                  break;
4146                eptr++;                eptr++;
4147                }                }
# Line 4224  HEAP_RETURN: Line 4291  HEAP_RETURN:
4291  switch (frame->Xwhere)  switch (frame->Xwhere)
4292    {    {
4293    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)
4294    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)
4295    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)
4296    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)
4297    LBL(33) LBL(34) LBL(35) LBL(36) LBL(37) LBL(38) LBL(39) LBL(40)    LBL(53) LBL(54)
4298    LBL(41) LBL(42) LBL(43) LBL(44) LBL(45) LBL(46) LBL(47) LBL(48)  #ifdef SUPPORT_UTF8
4299    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)
4300      LBL(32) LBL(34) LBL(42) LBL(46)
4301    #ifdef SUPPORT_UCP
4302      LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
4303    #endif  /* SUPPORT_UCP */
4304    #endif  /* SUPPORT_UTF8 */
4305    default:    default:
4306    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));
4307    return PCRE_ERROR_INTERNAL;    return PCRE_ERROR_INTERNAL;
# Line 4408  if (re->magic_number != MAGIC_NUMBER) Line 4480  if (re->magic_number != MAGIC_NUMBER)
4480  /* Set up other data */  /* Set up other data */
4481    
4482  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;  anchored = ((re->options | options) & PCRE_ANCHORED) != 0;
4483  startline = (re->options & PCRE_STARTLINE) != 0;  startline = (re->flags & PCRE_STARTLINE) != 0;
4484  firstline = (re->options & PCRE_FIRSTLINE) != 0;  firstline = (re->options & PCRE_FIRSTLINE) != 0;
4485    
4486  /* 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 4423  end_subject = md->end_subject; Line 4495  end_subject = md->end_subject;
4495    
4496  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;  md->endonly = (re->options & PCRE_DOLLAR_ENDONLY) != 0;
4497  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;  utf8 = md->utf8 = (re->options & PCRE_UTF8) != 0;
4498    md->jscript_compat = (re->options & PCRE_JAVASCRIPT_COMPAT) != 0;
4499    
4500  md->notbol = (options & PCRE_NOTBOL) != 0;  md->notbol = (options & PCRE_NOTBOL) != 0;
4501  md->noteol = (options & PCRE_NOTEOL) != 0;  md->noteol = (options & PCRE_NOTEOL) != 0;
# Line 4435  md->recursive = NULL; Line 4508  md->recursive = NULL;
4508  md->lcc = tables + lcc_offset;  md->lcc = tables + lcc_offset;
4509  md->ctypes = tables + ctypes_offset;  md->ctypes = tables + ctypes_offset;
4510    
4511    /* Handle different \R options. */
4512    
4513    switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
4514      {
4515      case 0:
4516      if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0)
4517        md->bsr_anycrlf = (re->options & PCRE_BSR_ANYCRLF) != 0;
4518      else
4519    #ifdef BSR_ANYCRLF
4520      md->bsr_anycrlf = TRUE;
4521    #else
4522      md->bsr_anycrlf = FALSE;
4523    #endif
4524      break;
4525    
4526      case PCRE_BSR_ANYCRLF:
4527      md->bsr_anycrlf = TRUE;
4528      break;
4529    
4530      case PCRE_BSR_UNICODE:
4531      md->bsr_anycrlf = FALSE;
4532      break;
4533    
4534      default: return PCRE_ERROR_BADNEWLINE;
4535      }
4536    
4537  /* Handle different types of newline. The three bits give eight cases. If  /* Handle different types of newline. The three bits give eight cases. If
4538  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. */
4539    
4540  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options : (pcre_uint32)options) &  switch ((((options & PCRE_NEWLINE_BITS) == 0)? re->options :
4541         PCRE_NEWLINE_BITS)          (pcre_uint32)options) & PCRE_NEWLINE_BITS)
4542    {    {
4543    case 0: newline = NEWLINE; break;   /* Compile-time default */    case 0: newline = NEWLINE; break;   /* Compile-time default */
4544    case PCRE_NEWLINE_CR: newline = '\r'; break;    case PCRE_NEWLINE_CR: newline = '\r'; break;
# Line 4478  else Line 4577  else
4577  /* 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
4578  moment. */  moment. */
4579    
4580  if (md->partial && (re->options & PCRE_NOPARTIAL) != 0)  if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0)
4581    return PCRE_ERROR_BADPARTIAL;    return PCRE_ERROR_BADPARTIAL;
4582    
4583  /* 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 4555  studied, there may be a bitmap of possib Line 4654  studied, there may be a bitmap of possib
4654    
4655  if (!anchored)  if (!anchored)
4656    {    {
4657    if ((re->options & PCRE_FIRSTSET) != 0)    if ((re->flags & PCRE_FIRSTSET) != 0)
4658      {      {
4659      first_byte = re->first_byte & 255;      first_byte = re->first_byte & 255;
4660      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)      if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE)
# Line 4570  if (!anchored) Line 4669  if (!anchored)
4669  /* For anchored or unanchored matches, there may be a "last known required  /* For anchored or unanchored matches, there may be a "last known required
4670  character" set. */  character" set. */
4671    
4672  if ((re->options & PCRE_REQCHSET) != 0)  if ((re->flags & PCRE_REQCHSET) != 0)
4673    {    {
4674    req_byte = re->req_byte & 255;    req_byte = re->req_byte & 255;
4675    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;    req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0;
# Line 4617  for(;;) Line 4716  for(;;)
4716      if (first_byte_caseless)      if (first_byte_caseless)
4717        while (start_match < end_subject &&        while (start_match < end_subject &&
4718               md->lcc[*start_match] != first_byte)               md->lcc[*start_match] != first_byte)
4719          start_match++;          { NEXTCHAR(start_match); }
4720      else      else
4721        while (start_match < end_subject && *start_match != first_byte)        while (start_match < end_subject && *start_match != first_byte)
4722          start_match++;          { NEXTCHAR(start_match); }
4723      }      }
4724    
4725    /* 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 4729  for(;;)
4729      if (start_match > md->start_subject + start_offset)      if (start_match > md->start_subject + start_offset)
4730        {        {
4731        while (start_match <= end_subject && !WAS_NEWLINE(start_match))        while (start_match <= end_subject && !WAS_NEWLINE(start_match))
4732          start_match++;          { NEXTCHAR(start_match); }
4733    
4734        /* 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,
4735        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 4750  for(;;)
4750      while (start_match < end_subject)      while (start_match < end_subject)
4751        {        {
4752        register unsigned int c = *start_match;        register unsigned int c = *start_match;
4753        if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break;        if ((start_bits[c/8] & (1 << (c&7))) == 0)
4754            { NEXTCHAR(start_match); }
4755          else break;
4756        }        }
4757      }      }
4758    
# Line 4792  for(;;) Line 4893  for(;;)
4893    if (start_match[-1] == '\r' &&    if (start_match[-1] == '\r' &&
4894        start_match < end_subject &&        start_match < end_subject &&
4895        *start_match == '\n' &&        *start_match == '\n' &&
4896        (re->options & PCRE_HASCRORLF) == 0 &&        (re->flags & PCRE_HASCRORLF) == 0 &&
4897          (md->nltype == NLTYPE_ANY ||          (md->nltype == NLTYPE_ANY ||
4898           md->nltype == NLTYPE_ANYCRLF ||           md->nltype == NLTYPE_ANYCRLF ||
4899           md->nllen == 2))           md->nllen == 2))

Legend:
Removed from v.227  
changed lines
  Added in v.341

  ViewVC Help
Powered by ViewVC 1.1.5