/[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 1365 by ph10, Sun Oct 6 18:33:56 2013 UTC revision 1404 by ph10, Tue Nov 19 15:36:57 2013 UTC
# Line 107  because the offset vector is always a mu Line 107  because the offset vector is always a mu
107    
108  /* Min and max values for the common repeats; for the maxima, 0 => infinity */  /* Min and max values for the common repeats; for the maxima, 0 => infinity */
109    
110  static const char rep_min[] = { 0, 0, 1, 1, 0, 0 };  static const char rep_min[] = { 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, };
111  static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };  static const char rep_max[] = { 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 1, };
112    
113  #ifdef PCRE_DEBUG  #ifdef PCRE_DEBUG
114  /*************************************************  /*************************************************
# Line 312  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM Line 312  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM
312         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
313         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
314         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
315         RM61,  RM62, RM63, RM64, RM65, RM66, RM67, RM68 };         RM61,  RM62, RM63, RM64, RM65, RM66, RM67 };
316    
317  /* 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
318  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 1173  for (;;) Line 1173  for (;;)
1173            ecode = md->start_code + code_offset;            ecode = md->start_code + code_offset;
1174            save_capture_last = md->capture_last;            save_capture_last = md->capture_last;
1175            matched_once = TRUE;            matched_once = TRUE;
1176              mstart = md->start_match_ptr;    /* In case \K changed it */
1177            continue;            continue;
1178            }            }
1179    
# Line 1245  for (;;) Line 1246  for (;;)
1246          eptr = md->end_match_ptr;          eptr = md->end_match_ptr;
1247          ecode = md->start_code + code_offset;          ecode = md->start_code + code_offset;
1248          matched_once = TRUE;          matched_once = TRUE;
1249            mstart = md->start_match_ptr;   /* In case \K reset it */
1250          continue;          continue;
1251          }          }
1252    
# Line 1281  for (;;) Line 1283  for (;;)
1283    
1284      case OP_COND:      case OP_COND:
1285      case OP_SCOND:      case OP_SCOND:
1286    
1287      /* The variable codelink will be added to ecode when the condition is      /* The variable codelink will be added to ecode when the condition is
1288      false, to get to the second branch. Setting it to the offset to the ALT      false, to get to the second branch. Setting it to the offset to the ALT
1289      or KET, then incrementing ecode achieves this effect. We now have ecode      or KET, then incrementing ecode achieves this effect. We now have ecode
1290      pointing to the condition or callout. */      pointing to the condition or callout. */
1291    
1292      codelink = GET(ecode, 1);   /* Offset to the second branch */      codelink = GET(ecode, 1);   /* Offset to the second branch */
1293      ecode += 1 + LINK_SIZE;     /* From this opcode */      ecode += 1 + LINK_SIZE;     /* From this opcode */
1294    
# Line 1322  for (;;) Line 1324  for (;;)
1324          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1325          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1326          }          }
1327    
1328        /* Advance ecode past the callout, so it now points to the condition. We        /* Advance ecode past the callout, so it now points to the condition. We
1329        must adjust codelink so that the value of ecode+codelink is unchanged. */        must adjust codelink so that the value of ecode+codelink is unchanged. */
1330    
1331        ecode += PRIV(OP_lengths)[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1332        codelink -= PRIV(OP_lengths)[OP_CALLOUT];        codelink -= PRIV(OP_lengths)[OP_CALLOUT];
1333        }        }
# Line 1334  for (;;) Line 1336  for (;;)
1336    
1337      condition = FALSE;      condition = FALSE;
1338      switch(condcode = *ecode)      switch(condcode = *ecode)
1339        {        {
1340        case OP_RREF:         /* Numbered group recursion test */        case OP_RREF:         /* Numbered group recursion test */
1341        if (md->recursive != NULL)     /* Not recursing => FALSE */        if (md->recursive != NULL)     /* Not recursing => FALSE */
1342          {          {
# Line 1345  for (;;) Line 1347  for (;;)
1347    
1348        case OP_DNRREF:       /* Duplicate named group recursion test */        case OP_DNRREF:       /* Duplicate named group recursion test */
1349        if (md->recursive != NULL)        if (md->recursive != NULL)
1350          {          {
1351          int count = GET2(ecode, 1 + IMM2_SIZE);          int count = GET2(ecode, 1 + IMM2_SIZE);
1352          pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;          pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;
1353          while (count-- > 0)          while (count-- > 0)
1354            {            {
# Line 1355  for (;;) Line 1357  for (;;)
1357            if (condition) break;            if (condition) break;
1358            slot += md->name_entry_size;            slot += md->name_entry_size;
1359            }            }
1360          }          }
1361        break;        break;
1362    
1363        case OP_CREF:         /* Numbered group used test */        case OP_CREF:         /* Numbered group used test */
# Line 1365  for (;;) Line 1367  for (;;)
1367    
1368        case OP_DNCREF:      /* Duplicate named group used test */        case OP_DNCREF:      /* Duplicate named group used test */
1369          {          {
1370          int count = GET2(ecode, 1 + IMM2_SIZE);          int count = GET2(ecode, 1 + IMM2_SIZE);
1371          pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;          pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;
1372          while (count-- > 0)          while (count-- > 0)
1373            {            {
# Line 1375  for (;;) Line 1377  for (;;)
1377            slot += md->name_entry_size;            slot += md->name_entry_size;
1378            }            }
1379          }          }
1380        break;        break;
1381    
1382        case OP_DEF:     /* DEFINE - always false */        case OP_DEF:     /* DEFINE - always false */
1383        break;        break;
# Line 1383  for (;;) Line 1385  for (;;)
1385        /* The condition is an assertion. Call match() to evaluate it - setting        /* The condition is an assertion. Call match() to evaluate it - setting
1386        md->match_function_type to MATCH_CONDASSERT causes it to stop at the end        md->match_function_type to MATCH_CONDASSERT causes it to stop at the end
1387        of an assertion. */        of an assertion. */
1388    
1389        default:        default:
1390        md->match_function_type = MATCH_CONDASSERT;        md->match_function_type = MATCH_CONDASSERT;
1391        RMATCH(eptr, ecode, offset_top, md, NULL, RM3);        RMATCH(eptr, ecode, offset_top, md, NULL, RM3);
1392        if (rrc == MATCH_MATCH)        if (rrc == MATCH_MATCH)
# Line 1392  for (;;) Line 1394  for (;;)
1394          if (md->end_offset_top > offset_top)          if (md->end_offset_top > offset_top)
1395            offset_top = md->end_offset_top;  /* Captures may have happened */            offset_top = md->end_offset_top;  /* Captures may have happened */
1396          condition = TRUE;          condition = TRUE;
1397    
1398          /* Advance ecode past the assertion to the start of the first branch,          /* Advance ecode past the assertion to the start of the first branch,
1399          but adjust it so that the general choosing code below works. */          but adjust it so that the general choosing code below works. */
1400    
1401          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
1402          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1403          ecode += 1 + LINK_SIZE - PRIV(OP_lengths)[condcode];          ecode += 1 + LINK_SIZE - PRIV(OP_lengths)[condcode];
1404          }          }
1405    
1406        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
1407        assertion; it is therefore treated as NOMATCH. Any other return is an        assertion; it is therefore treated as NOMATCH. Any other return is an
1408        error. */        error. */
1409    
1410        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1411          {          {
1412          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1413          }          }
1414        break;        break;
1415        }        }
1416    
1417      /* Choose branch according to the condition */      /* Choose branch according to the condition */
1418    
1419      ecode += condition? PRIV(OP_lengths)[condcode] : codelink;      ecode += condition? PRIV(OP_lengths)[condcode] : codelink;
1420    
1421      /* We are now at the branch that is to be obeyed. As there is only one, we      /* We are now at the branch that is to be obeyed. As there is only one, we
1422      can use tail recursion to avoid using another stack frame, except when      can use tail recursion to avoid using another stack frame, except when
1423      there is unlimited repeat of a possibly empty group. In the latter case, a      there is unlimited repeat of a possibly empty group. In the latter case, a
# Line 1425  for (;;) Line 1427  for (;;)
1427      creating two alternatives. If a THEN is encountered in the branch, it      creating two alternatives. If a THEN is encountered in the branch, it
1428      propagates out to the enclosing alternative (unless nested in a deeper set      propagates out to the enclosing alternative (unless nested in a deeper set
1429      of alternatives, of course). */      of alternatives, of course). */
1430    
1431      if (condition || ecode[-(1+LINK_SIZE)] == OP_ALT)      if (condition || ecode[-(1+LINK_SIZE)] == OP_ALT)
1432        {        {
1433        if (op != OP_SCOND)        if (op != OP_SCOND)
# Line 2007  for (;;) Line 2009  for (;;)
2009    
2010      if (*ecode == OP_KETRPOS)      if (*ecode == OP_KETRPOS)
2011        {        {
2012          md->start_match_ptr = mstart;    /* In case \K reset it */
2013        md->end_match_ptr = eptr;        md->end_match_ptr = eptr;
2014        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
2015        RRETURN(MATCH_KETRPOS);        RRETURN(MATCH_KETRPOS);
# Line 2577  for (;;) Line 2580  for (;;)
2580          /* Perl space used to exclude VT, but from Perl 5.18 it is included,          /* Perl space used to exclude VT, but from Perl 5.18 it is included,
2581          which means that Perl space and POSIX space are now identical. PCRE          which means that Perl space and POSIX space are now identical. PCRE
2582          was changed at release 8.34. */          was changed at release 8.34. */
2583    
2584          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2585          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2586          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||          switch(c)
2587               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||            {
2588               c == CHAR_FF || c == CHAR_CR)            HSPACE_CASES:
2589                 == (op == OP_NOTPROP))            VSPACE_CASES:
2590            RRETURN(MATCH_NOMATCH);            if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
2591              break;
2592    
2593              default:
2594              if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) ==
2595                (op == OP_NOTPROP)) RRETURN(MATCH_NOMATCH);
2596              break;
2597              }
2598          break;          break;
2599    
2600          case PT_WORD:          case PT_WORD:
# Line 2669  for (;;) Line 2679  for (;;)
2679    
2680      Otherwise, set the length to the length of what was matched by the      Otherwise, set the length to the length of what was matched by the
2681      referenced subpattern.      referenced subpattern.
2682    
2683      The OP_REF and OP_REFI opcodes are used for a reference to a numbered group      The OP_REF and OP_REFI opcodes are used for a reference to a numbered group
2684      or to a non-duplicated named group. For a duplicated named group, OP_DNREF      or to a non-duplicated named group. For a duplicated named group, OP_DNREF
2685      and OP_DNREFI are used. In this case we must scan the list of groups to      and OP_DNREFI are used. In this case we must scan the list of groups to
2686      which the name refers, and use the first one that is set. */      which the name refers, and use the first one that is set. */
2687    
2688      case OP_DNREF:      case OP_DNREF:
2689      case OP_DNREFI:      case OP_DNREFI:
2690      caseless = op == OP_DNREFI;      caseless = op == OP_DNREFI;
2691        {        {
2692        int count = GET2(ecode, 1+IMM2_SIZE);        int count = GET2(ecode, 1+IMM2_SIZE);
2693        pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;        pcre_uchar *slot = md->name_table + GET2(ecode, 1) * md->name_entry_size;
2694        ecode += 1 + 2*IMM2_SIZE;        ecode += 1 + 2*IMM2_SIZE;
2695    
2696        while (count-- > 0)        while (count-- > 0)
2697          {          {
2698          offset = GET2(slot, 0) << 1;          offset = GET2(slot, 0) << 1;
2699          if (offset < offset_top && md->offset_vector[offset] >= 0) break;          if (offset < offset_top && md->offset_vector[offset] >= 0) break;
2700          slot += md->name_entry_size;          slot += md->name_entry_size;
2701          }          }
2702        if (count < 0)        if (count < 0)
2703          length = (md->jscript_compat)? 0 : -1;          length = (md->jscript_compat)? 0 : -1;
2704        else        else
2705          length = md->offset_vector[offset+1] - md->offset_vector[offset];          length = md->offset_vector[offset+1] - md->offset_vector[offset];
# Line 2857  for (;;) Line 2867  for (;;)
2867          case OP_CRMINPLUS:          case OP_CRMINPLUS:
2868          case OP_CRQUERY:          case OP_CRQUERY:
2869          case OP_CRMINQUERY:          case OP_CRMINQUERY:
2870            case OP_CRPOSSTAR:
2871            case OP_CRPOSPLUS:
2872            case OP_CRPOSQUERY:
2873          c = *ecode++ - OP_CRSTAR;          c = *ecode++ - OP_CRSTAR;
2874          minimize = (c & 1) != 0;          if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0;
2875            else possessive = TRUE;
2876          min = rep_min[c];                 /* Pick up values from tables; */          min = rep_min[c];                 /* Pick up values from tables; */
2877          max = rep_max[c];                 /* zero for max => infinity */          max = rep_max[c];                 /* zero for max => infinity */
2878          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 2866  for (;;) Line 2880  for (;;)
2880    
2881          case OP_CRRANGE:          case OP_CRRANGE:
2882          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2883            case OP_CRPOSRANGE:
2884          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2885            possessive = (*ecode == OP_CRPOSRANGE);
2886          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2887          max = GET2(ecode, 1 + IMM2_SIZE);          max = GET2(ecode, 1 + IMM2_SIZE);
2888          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3008  for (;;) Line 3024  for (;;)
3024                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3025              eptr += len;              eptr += len;
3026              }              }
3027    
3028              if (possessive) continue;    /* No backtracking */
3029    
3030            for (;;)            for (;;)
3031              {              {
3032              RMATCH(eptr, ecode, offset_top, md, eptrb, RM18);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM18);
# Line 3038  for (;;) Line 3057  for (;;)
3057                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3058              eptr++;              eptr++;
3059              }              }
3060    
3061              if (possessive) continue;    /* No backtracking */
3062    
3063            while (eptr >= pp)            while (eptr >= pp)
3064              {              {
3065              RMATCH(eptr, ecode, offset_top, md, eptrb, RM19);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM19);
# Line 3071  for (;;) Line 3093  for (;;)
3093          case OP_CRMINPLUS:          case OP_CRMINPLUS:
3094          case OP_CRQUERY:          case OP_CRQUERY:
3095          case OP_CRMINQUERY:          case OP_CRMINQUERY:
3096            case OP_CRPOSSTAR:
3097            case OP_CRPOSPLUS:
3098            case OP_CRPOSQUERY:
3099          c = *ecode++ - OP_CRSTAR;          c = *ecode++ - OP_CRSTAR;
3100          minimize = (c & 1) != 0;          if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0;
3101            else possessive = TRUE;
3102          min = rep_min[c];                 /* Pick up values from tables; */          min = rep_min[c];                 /* Pick up values from tables; */
3103          max = rep_max[c];                 /* zero for max => infinity */          max = rep_max[c];                 /* zero for max => infinity */
3104          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3080  for (;;) Line 3106  for (;;)
3106    
3107          case OP_CRRANGE:          case OP_CRRANGE:
3108          case OP_CRMINRANGE:          case OP_CRMINRANGE:
3109            case OP_CRPOSRANGE:
3110          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
3111            possessive = (*ecode == OP_CRPOSRANGE);
3112          min = GET2(ecode, 1);          min = GET2(ecode, 1);
3113          max = GET2(ecode, 1 + IMM2_SIZE);          max = GET2(ecode, 1 + IMM2_SIZE);
3114          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3152  for (;;) Line 3180  for (;;)
3180            if (!PRIV(xclass)(c, data, utf)) break;            if (!PRIV(xclass)(c, data, utf)) break;
3181            eptr += len;            eptr += len;
3182            }            }
3183    
3184            if (possessive) continue;    /* No backtracking */
3185    
3186          for(;;)          for(;;)
3187            {            {
3188            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
# Line 4200  for (;;) Line 4231  for (;;)
4231            /* Perl space used to exclude VT, but from Perl 5.18 it is included,            /* Perl space used to exclude VT, but from Perl 5.18 it is included,
4232            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
4233            was changed at release 8.34. */            was changed at release 8.34. */
4234    
4235            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
4236            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
4237            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
# Line 4211  for (;;) Line 4242  for (;;)
4242                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4243                }                }
4244              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4245              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
4246                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
4247                     == prop_fail_result)                HSPACE_CASES:
4248                RRETURN(MATCH_NOMATCH);                VSPACE_CASES:
4249                  if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4250                  break;
4251    
4252                  default:
4253                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
4254                    RRETURN(MATCH_NOMATCH);
4255                  break;
4256                  }
4257              }              }
4258            break;            break;
4259    
# Line 4937  for (;;) Line 4976  for (;;)
4976            /* Perl space used to exclude VT, but from Perl 5.18 it is included,            /* Perl space used to exclude VT, but from Perl 5.18 it is included,
4977            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
4978            was changed at release 8.34. */            was changed at release 8.34. */
4979    
4980            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
4981            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
4982            for (fi = min;; fi++)            for (fi = min;; fi++)
# Line 4951  for (;;) Line 4990  for (;;)
4990                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4991                }                }
4992              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4993              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
4994                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
4995                     == prop_fail_result)                HSPACE_CASES:
4996                RRETURN(MATCH_NOMATCH);                VSPACE_CASES:
4997                  if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4998                  break;
4999    
5000                  default:
5001                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
5002                    RRETURN(MATCH_NOMATCH);
5003                  break;
5004                  }
5005              }              }
5006            /* Control never gets here */            /* Control never gets here */
5007    
# Line 5007  for (;;) Line 5054  for (;;)
5054            case PT_UCNC:            case PT_UCNC:
5055            for (fi = min;; fi++)            for (fi = min;; fi++)
5056              {              {
5057              RMATCH(eptr, ecode, offset_top, md, eptrb, RM68);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM60);
5058              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5059              if (fi >= max) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
5060              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
# Line 5441  for (;;) Line 5488  for (;;)
5488            /* Perl space used to exclude VT, but from Perl 5.18 it is included,            /* Perl space used to exclude VT, but from Perl 5.18 it is included,
5489            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
5490            was changed at release 8.34. */            was changed at release 8.34. */
5491    
5492            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
5493            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
5494            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 5453  for (;;) Line 5500  for (;;)
5500                break;                break;
5501                }                }
5502              GETCHARLENTEST(c, eptr, len);              GETCHARLENTEST(c, eptr, len);
5503              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
5504                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
5505                   == prop_fail_result)                HSPACE_CASES:
5506                  VSPACE_CASES:
5507                  if (prop_fail_result) goto ENDLOOP99;  /* Break the loop */
5508                break;                break;
5509    
5510                  default:
5511                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
5512                    goto ENDLOOP99;   /* Break the loop */
5513                  break;
5514                  }
5515              eptr+= len;              eptr+= len;
5516              }              }
5517              ENDLOOP99:
5518            break;            break;
5519    
5520            case PT_WORD:            case PT_WORD:
# Line 5572  for (;;) Line 5628  for (;;)
5628          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
5629    
5630          if (possessive) continue;    /* No backtracking */          if (possessive) continue;    /* No backtracking */
5631    
5632          for(;;)          for(;;)
5633            {            {
5634            int lgb, rgb;            int lgb, rgb;
5635            PCRE_PUCHAR fptr;            PCRE_PUCHAR fptr;
5636    
5637            if (eptr == pp) goto TAIL_RECURSE;   /* At start of char run */            if (eptr == pp) goto TAIL_RECURSE;   /* At start of char run */
5638            RMATCH(eptr, ecode, offset_top, md, eptrb, RM45);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM45);
5639            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 5585  for (;;) Line 5641  for (;;)
5641            /* Backtracking over an extended grapheme cluster involves inspecting            /* Backtracking over an extended grapheme cluster involves inspecting
5642            the previous two characters (if present) to see if a break is            the previous two characters (if present) to see if a break is
5643            permitted between them. */            permitted between them. */
5644    
5645            eptr--;            eptr--;
5646            if (!utf) c = *eptr; else            if (!utf) c = *eptr; else
5647              {              {
# Line 5603  for (;;) Line 5659  for (;;)
5659                BACKCHAR(fptr);                BACKCHAR(fptr);
5660                GETCHAR(c, fptr);                GETCHAR(c, fptr);
5661                }                }
5662              lgb = UCD_GRAPHBREAK(c);              lgb = UCD_GRAPHBREAK(c);
5663              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
5664              eptr = fptr;              eptr = fptr;
5665              rgb = lgb;              rgb = lgb;
# Line 6127  for (;;) Line 6183  for (;;)
6183                eptr[-1] == CHAR_CR) eptr--;                eptr[-1] == CHAR_CR) eptr--;
6184            }            }
6185          }          }
6186    
6187        /* Control never gets here */        /* Control never gets here */
6188        }        }
6189    
# Line 6171  switch (frame->Xwhere) Line 6227  switch (frame->Xwhere)
6227    LBL(32) LBL(34) LBL(42) LBL(46)    LBL(32) LBL(34) LBL(42) LBL(46)
6228  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6229    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)    LBL(36) LBL(37) LBL(38) LBL(39) LBL(40) LBL(41) LBL(44) LBL(45)
6230    LBL(59) LBL(60) LBL(61) LBL(62) LBL(67) LBL(68)    LBL(59) LBL(60) LBL(61) LBL(62) LBL(67)
6231  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
6232  #endif  /* SUPPORT_UTF */  #endif  /* SUPPORT_UTF */
6233    default:    default:

Legend:
Removed from v.1365  
changed lines
  Added in v.1404

  ViewVC Help
Powered by ViewVC 1.1.5