/[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 1431 by ph10, Thu Jan 2 17:41:28 2014 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) 1997-2013 University of Cambridge             Copyright (c) 1997-2014 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 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 134  pcre_uint32 c; Line 134  pcre_uint32 c;
134  BOOL utf = md->utf;  BOOL utf = md->utf;
135  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;  if (is_subject && length > md->end_subject - p) length = md->end_subject - p;
136  while (length-- > 0)  while (length-- > 0)
137    if (isprint(c = RAWUCHARINCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c);    if (isprint(c = UCHAR21INCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c);
138  }  }
139  #endif  #endif
140    
# Line 167  match_ref(int offset, register PCRE_PUCH Line 167  match_ref(int offset, register PCRE_PUCH
167  {  {
168  PCRE_PUCHAR eptr_start = eptr;  PCRE_PUCHAR eptr_start = eptr;
169  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];  register PCRE_PUCHAR p = md->start_subject + md->offset_vector[offset];
170  #ifdef SUPPORT_UTF  #if defined SUPPORT_UTF && defined SUPPORT_UCP
171  BOOL utf = md->utf;  BOOL utf = md->utf;
172  #endif  #endif
173    
# Line 195  ASCII characters. */ Line 195  ASCII characters. */
195    
196  if (caseless)  if (caseless)
197    {    {
198  #ifdef SUPPORT_UTF  #if defined SUPPORT_UTF && defined SUPPORT_UCP
 #ifdef SUPPORT_UCP  
199    if (utf)    if (utf)
200      {      {
201      /* Match characters up to the end of the reference. NOTE: the number of      /* Match characters up to the end of the reference. NOTE: the number of
# Line 230  if (caseless) Line 229  if (caseless)
229      }      }
230    else    else
231  #endif  #endif
 #endif  
232    
233    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there    /* The same code works when not in UTF-8 mode and in UTF-8 mode when there
234    is no UCP support. */    is no UCP support. */
# Line 239  if (caseless) Line 237  if (caseless)
237        {        {
238        pcre_uint32 cc, cp;        pcre_uint32 cc, cp;
239        if (eptr >= md->end_subject) return -2;   /* Partial match */        if (eptr >= md->end_subject) return -2;   /* Partial match */
240        cc = RAWUCHARTEST(eptr);        cc = UCHAR21TEST(eptr);
241        cp = RAWUCHARTEST(p);        cp = UCHAR21TEST(p);
242        if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;        if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;
243        p++;        p++;
244        eptr++;        eptr++;
# Line 256  else Line 254  else
254    while (length-- > 0)    while (length-- > 0)
255      {      {
256      if (eptr >= md->end_subject) return -2;   /* Partial match */      if (eptr >= md->end_subject) return -2;   /* Partial match */
257      if (RAWUCHARINCTEST(p) != RAWUCHARINCTEST(eptr)) return -1;      if (UCHAR21INCTEST(p) != UCHAR21INCTEST(eptr)) return -1;
258      }      }
259    }    }
260    
# Line 312  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM Line 310  enum { RM1=1, RM2,  RM3,  RM4,  RM5,  RM
310         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,         RM31,  RM32, RM33, RM34, RM35, RM36, RM37, RM38, RM39, RM40,
311         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,         RM41,  RM42, RM43, RM44, RM45, RM46, RM47, RM48, RM49, RM50,
312         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,         RM51,  RM52, RM53, RM54, RM55, RM56, RM57, RM58, RM59, RM60,
313         RM61,  RM62, RM63, RM64, RM65, RM66, RM67, RM68 };         RM61,  RM62, RM63, RM64, RM65, RM66, RM67 };
314    
315  /* 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
316  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 1171  for (;;)
1171            ecode = md->start_code + code_offset;            ecode = md->start_code + code_offset;
1172            save_capture_last = md->capture_last;            save_capture_last = md->capture_last;
1173            matched_once = TRUE;            matched_once = TRUE;
1174              mstart = md->start_match_ptr;    /* In case \K changed it */
1175            continue;            continue;
1176            }            }
1177    
# Line 1245  for (;;) Line 1244  for (;;)
1244          eptr = md->end_match_ptr;          eptr = md->end_match_ptr;
1245          ecode = md->start_code + code_offset;          ecode = md->start_code + code_offset;
1246          matched_once = TRUE;          matched_once = TRUE;
1247            mstart = md->start_match_ptr;   /* In case \K reset it */
1248          continue;          continue;
1249          }          }
1250    
# Line 1281  for (;;) Line 1281  for (;;)
1281    
1282      case OP_COND:      case OP_COND:
1283      case OP_SCOND:      case OP_SCOND:
1284    
1285      /* The variable codelink will be added to ecode when the condition is      /* The variable codelink will be added to ecode when the condition is
1286      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
1287      or KET, then incrementing ecode achieves this effect. We now have ecode      or KET, then incrementing ecode achieves this effect. We now have ecode
1288      pointing to the condition or callout. */      pointing to the condition or callout. */
1289    
1290      codelink = GET(ecode, 1);   /* Offset to the second branch */      codelink = GET(ecode, 1);   /* Offset to the second branch */
1291      ecode += 1 + LINK_SIZE;     /* From this opcode */      ecode += 1 + LINK_SIZE;     /* From this opcode */
1292    
# Line 1322  for (;;) Line 1322  for (;;)
1322          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);          if ((rrc = (*PUBL(callout))(&cb)) > 0) RRETURN(MATCH_NOMATCH);
1323          if (rrc < 0) RRETURN(rrc);          if (rrc < 0) RRETURN(rrc);
1324          }          }
1325    
1326        /* 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
1327        must adjust codelink so that the value of ecode+codelink is unchanged. */        must adjust codelink so that the value of ecode+codelink is unchanged. */
1328    
1329        ecode += PRIV(OP_lengths)[OP_CALLOUT];        ecode += PRIV(OP_lengths)[OP_CALLOUT];
1330        codelink -= PRIV(OP_lengths)[OP_CALLOUT];        codelink -= PRIV(OP_lengths)[OP_CALLOUT];
1331        }        }
# Line 1334  for (;;) Line 1334  for (;;)
1334    
1335      condition = FALSE;      condition = FALSE;
1336      switch(condcode = *ecode)      switch(condcode = *ecode)
1337        {        {
1338        case OP_RREF:         /* Numbered group recursion test */        case OP_RREF:         /* Numbered group recursion test */
1339        if (md->recursive != NULL)     /* Not recursing => FALSE */        if (md->recursive != NULL)     /* Not recursing => FALSE */
1340          {          {
# Line 1345  for (;;) Line 1345  for (;;)
1345    
1346        case OP_DNRREF:       /* Duplicate named group recursion test */        case OP_DNRREF:       /* Duplicate named group recursion test */
1347        if (md->recursive != NULL)        if (md->recursive != NULL)
1348          {          {
1349          int count = GET2(ecode, 1 + IMM2_SIZE);          int count = GET2(ecode, 1 + IMM2_SIZE);
1350          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;
1351          while (count-- > 0)          while (count-- > 0)
1352            {            {
# Line 1355  for (;;) Line 1355  for (;;)
1355            if (condition) break;            if (condition) break;
1356            slot += md->name_entry_size;            slot += md->name_entry_size;
1357            }            }
1358          }          }
1359        break;        break;
1360    
1361        case OP_CREF:         /* Numbered group used test */        case OP_CREF:         /* Numbered group used test */
# Line 1365  for (;;) Line 1365  for (;;)
1365    
1366        case OP_DNCREF:      /* Duplicate named group used test */        case OP_DNCREF:      /* Duplicate named group used test */
1367          {          {
1368          int count = GET2(ecode, 1 + IMM2_SIZE);          int count = GET2(ecode, 1 + IMM2_SIZE);
1369          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;
1370          while (count-- > 0)          while (count-- > 0)
1371            {            {
# Line 1375  for (;;) Line 1375  for (;;)
1375            slot += md->name_entry_size;            slot += md->name_entry_size;
1376            }            }
1377          }          }
1378        break;        break;
1379    
1380        case OP_DEF:     /* DEFINE - always false */        case OP_DEF:     /* DEFINE - always false */
1381        break;        break;
# Line 1383  for (;;) Line 1383  for (;;)
1383        /* The condition is an assertion. Call match() to evaluate it - setting        /* The condition is an assertion. Call match() to evaluate it - setting
1384        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
1385        of an assertion. */        of an assertion. */
1386    
1387        default:        default:
1388        md->match_function_type = MATCH_CONDASSERT;        md->match_function_type = MATCH_CONDASSERT;
1389        RMATCH(eptr, ecode, offset_top, md, NULL, RM3);        RMATCH(eptr, ecode, offset_top, md, NULL, RM3);
1390        if (rrc == MATCH_MATCH)        if (rrc == MATCH_MATCH)
# Line 1392  for (;;) Line 1392  for (;;)
1392          if (md->end_offset_top > offset_top)          if (md->end_offset_top > offset_top)
1393            offset_top = md->end_offset_top;  /* Captures may have happened */            offset_top = md->end_offset_top;  /* Captures may have happened */
1394          condition = TRUE;          condition = TRUE;
1395    
1396          /* Advance ecode past the assertion to the start of the first branch,          /* Advance ecode past the assertion to the start of the first branch,
1397          but adjust it so that the general choosing code below works. */          but adjust it so that the general choosing code below works. */
1398    
1399          ecode += GET(ecode, 1);          ecode += GET(ecode, 1);
1400          while (*ecode == OP_ALT) ecode += GET(ecode, 1);          while (*ecode == OP_ALT) ecode += GET(ecode, 1);
1401          ecode += 1 + LINK_SIZE - PRIV(OP_lengths)[condcode];          ecode += 1 + LINK_SIZE - PRIV(OP_lengths)[condcode];
1402          }          }
1403    
1404        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an        /* PCRE doesn't allow the effect of (*THEN) to escape beyond an
1405        assertion; it is therefore treated as NOMATCH. Any other return is an        assertion; it is therefore treated as NOMATCH. Any other return is an
1406        error. */        error. */
1407    
1408        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)        else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN)
1409          {          {
1410          RRETURN(rrc);         /* Need braces because of following else */          RRETURN(rrc);         /* Need braces because of following else */
1411          }          }
1412        break;        break;
1413        }        }
1414    
1415      /* Choose branch according to the condition */      /* Choose branch according to the condition */
1416    
1417      ecode += condition? PRIV(OP_lengths)[condcode] : codelink;      ecode += condition? PRIV(OP_lengths)[condcode] : codelink;
1418    
1419      /* 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
1420      can use tail recursion to avoid using another stack frame, except when      can use tail recursion to avoid using another stack frame, except when
1421      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 1425  for (;;)
1425      creating two alternatives. If a THEN is encountered in the branch, it      creating two alternatives. If a THEN is encountered in the branch, it
1426      propagates out to the enclosing alternative (unless nested in a deeper set      propagates out to the enclosing alternative (unless nested in a deeper set
1427      of alternatives, of course). */      of alternatives, of course). */
1428    
1429      if (condition || ecode[-(1+LINK_SIZE)] == OP_ALT)      if (condition || ecode[-(1+LINK_SIZE)] == OP_ALT)
1430        {        {
1431        if (op != OP_SCOND)        if (op != OP_SCOND)
# Line 2007  for (;;) Line 2007  for (;;)
2007    
2008      if (*ecode == OP_KETRPOS)      if (*ecode == OP_KETRPOS)
2009        {        {
2010          md->start_match_ptr = mstart;    /* In case \K reset it */
2011        md->end_match_ptr = eptr;        md->end_match_ptr = eptr;
2012        md->end_offset_top = offset_top;        md->end_offset_top = offset_top;
2013        RRETURN(MATCH_KETRPOS);        RRETURN(MATCH_KETRPOS);
# Line 2102  for (;;) Line 2103  for (;;)
2103              eptr + 1 >= md->end_subject &&              eptr + 1 >= md->end_subject &&
2104              NLBLOCK->nltype == NLTYPE_FIXED &&              NLBLOCK->nltype == NLTYPE_FIXED &&
2105              NLBLOCK->nllen == 2 &&              NLBLOCK->nllen == 2 &&
2106              RAWUCHARTEST(eptr) == NLBLOCK->nl[0])              UCHAR21TEST(eptr) == NLBLOCK->nl[0])
2107            {            {
2108            md->hitend = TRUE;            md->hitend = TRUE;
2109            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);            if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 2146  for (;;) Line 2147  for (;;)
2147            eptr + 1 >= md->end_subject &&            eptr + 1 >= md->end_subject &&
2148            NLBLOCK->nltype == NLTYPE_FIXED &&            NLBLOCK->nltype == NLTYPE_FIXED &&
2149            NLBLOCK->nllen == 2 &&            NLBLOCK->nllen == 2 &&
2150            RAWUCHARTEST(eptr) == NLBLOCK->nl[0])            UCHAR21TEST(eptr) == NLBLOCK->nl[0])
2151          {          {
2152          md->hitend = TRUE;          md->hitend = TRUE;
2153          if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);          if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 2289  for (;;) Line 2290  for (;;)
2290          eptr + 1 >= md->end_subject &&          eptr + 1 >= md->end_subject &&
2291          NLBLOCK->nltype == NLTYPE_FIXED &&          NLBLOCK->nltype == NLTYPE_FIXED &&
2292          NLBLOCK->nllen == 2 &&          NLBLOCK->nllen == 2 &&
2293          RAWUCHARTEST(eptr) == NLBLOCK->nl[0])          UCHAR21TEST(eptr) == NLBLOCK->nl[0])
2294        {        {
2295        md->hitend = TRUE;        md->hitend = TRUE;
2296        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);        if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 2443  for (;;) Line 2444  for (;;)
2444          {          {
2445          SCHECK_PARTIAL();          SCHECK_PARTIAL();
2446          }          }
2447        else if (RAWUCHARTEST(eptr) == CHAR_LF) eptr++;        else if (UCHAR21TEST(eptr) == CHAR_LF) eptr++;
2448        break;        break;
2449    
2450        case CHAR_LF:        case CHAR_LF:
# Line 2577  for (;;) Line 2578  for (;;)
2578          /* 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,
2579          which means that Perl space and POSIX space are now identical. PCRE          which means that Perl space and POSIX space are now identical. PCRE
2580          was changed at release 8.34. */          was changed at release 8.34. */
2581    
2582          case PT_SPACE:    /* Perl space */          case PT_SPACE:    /* Perl space */
2583          case PT_PXSPACE:  /* POSIX space */          case PT_PXSPACE:  /* POSIX space */
2584          if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||          switch(c)
2585               c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||            {
2586               c == CHAR_FF || c == CHAR_CR)            HSPACE_CASES:
2587                 == (op == OP_NOTPROP))            VSPACE_CASES:
2588            RRETURN(MATCH_NOMATCH);            if (op == OP_NOTPROP) RRETURN(MATCH_NOMATCH);
2589              break;
2590    
2591              default:
2592              if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) ==
2593                (op == OP_NOTPROP)) RRETURN(MATCH_NOMATCH);
2594              break;
2595              }
2596          break;          break;
2597    
2598          case PT_WORD:          case PT_WORD:
# Line 2669  for (;;) Line 2677  for (;;)
2677    
2678      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
2679      referenced subpattern.      referenced subpattern.
2680    
2681      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
2682      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
2683      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
2684      which the name refers, and use the first one that is set. */      which the name refers, and use the first one that is set. */
2685    
2686      case OP_DNREF:      case OP_DNREF:
2687      case OP_DNREFI:      case OP_DNREFI:
2688      caseless = op == OP_DNREFI;      caseless = op == OP_DNREFI;
2689        {        {
2690        int count = GET2(ecode, 1+IMM2_SIZE);        int count = GET2(ecode, 1+IMM2_SIZE);
2691        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;
2692        ecode += 1 + 2*IMM2_SIZE;        ecode += 1 + 2*IMM2_SIZE;
2693    
2694          /* Setting the default length first and initializing 'offset' avoids
2695          compiler warnings in the REF_REPEAT code. */
2696    
2697          length = (md->jscript_compat)? 0 : -1;
2698          offset = 0;
2699    
2700        while (count-- > 0)        while (count-- > 0)
2701          {          {
2702          offset = GET2(slot, 0) << 1;          offset = GET2(slot, 0) << 1;
2703          if (offset < offset_top && md->offset_vector[offset] >= 0) break;          if (offset < offset_top && md->offset_vector[offset] >= 0)
2704              {
2705              length = md->offset_vector[offset+1] - md->offset_vector[offset];
2706              break;
2707              }
2708          slot += md->name_entry_size;          slot += md->name_entry_size;
2709          }          }
       if (count < 0)  
         length = (md->jscript_compat)? 0 : -1;  
       else  
         length = md->offset_vector[offset+1] - md->offset_vector[offset];  
2710        }        }
2711      goto REF_REPEAT;      goto REF_REPEAT;
2712    
# Line 2857  for (;;) Line 2871  for (;;)
2871          case OP_CRMINPLUS:          case OP_CRMINPLUS:
2872          case OP_CRQUERY:          case OP_CRQUERY:
2873          case OP_CRMINQUERY:          case OP_CRMINQUERY:
2874            case OP_CRPOSSTAR:
2875            case OP_CRPOSPLUS:
2876            case OP_CRPOSQUERY:
2877          c = *ecode++ - OP_CRSTAR;          c = *ecode++ - OP_CRSTAR;
2878          minimize = (c & 1) != 0;          if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0;
2879            else possessive = TRUE;
2880          min = rep_min[c];                 /* Pick up values from tables; */          min = rep_min[c];                 /* Pick up values from tables; */
2881          max = rep_max[c];                 /* zero for max => infinity */          max = rep_max[c];                 /* zero for max => infinity */
2882          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 2866  for (;;) Line 2884  for (;;)
2884    
2885          case OP_CRRANGE:          case OP_CRRANGE:
2886          case OP_CRMINRANGE:          case OP_CRMINRANGE:
2887            case OP_CRPOSRANGE:
2888          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
2889            possessive = (*ecode == OP_CRPOSRANGE);
2890          min = GET2(ecode, 1);          min = GET2(ecode, 1);
2891          max = GET2(ecode, 1 + IMM2_SIZE);          max = GET2(ecode, 1 + IMM2_SIZE);
2892          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3008  for (;;) Line 3028  for (;;)
3028                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3029              eptr += len;              eptr += len;
3030              }              }
3031    
3032              if (possessive) continue;    /* No backtracking */
3033    
3034            for (;;)            for (;;)
3035              {              {
3036              RMATCH(eptr, ecode, offset_top, md, eptrb, RM18);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM18);
# Line 3038  for (;;) Line 3061  for (;;)
3061                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;                if ((BYTE_MAP[c/8] & (1 << (c&7))) == 0) break;
3062              eptr++;              eptr++;
3063              }              }
3064    
3065              if (possessive) continue;    /* No backtracking */
3066    
3067            while (eptr >= pp)            while (eptr >= pp)
3068              {              {
3069              RMATCH(eptr, ecode, offset_top, md, eptrb, RM19);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM19);
# Line 3053  for (;;) Line 3079  for (;;)
3079      /* Control never gets here */      /* Control never gets here */
3080    
3081    
3082      /* Match an extended character class. This opcode is encountered only      /* Match an extended character class. In the 8-bit library, this opcode is
3083      when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8      encountered only when UTF-8 mode mode is supported. In the 16-bit and
3084      mode, because Unicode properties are supported in non-UTF-8 mode. */      32-bit libraries, codepoints greater than 255 may be encountered even when
3085        UTF is not supported. */
3086    
3087  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
3088      case OP_XCLASS:      case OP_XCLASS:
# Line 3071  for (;;) Line 3098  for (;;)
3098          case OP_CRMINPLUS:          case OP_CRMINPLUS:
3099          case OP_CRQUERY:          case OP_CRQUERY:
3100          case OP_CRMINQUERY:          case OP_CRMINQUERY:
3101            case OP_CRPOSSTAR:
3102            case OP_CRPOSPLUS:
3103            case OP_CRPOSQUERY:
3104          c = *ecode++ - OP_CRSTAR;          c = *ecode++ - OP_CRSTAR;
3105          minimize = (c & 1) != 0;          if (c < OP_CRPOSSTAR - OP_CRSTAR) minimize = (c & 1) != 0;
3106            else possessive = TRUE;
3107          min = rep_min[c];                 /* Pick up values from tables; */          min = rep_min[c];                 /* Pick up values from tables; */
3108          max = rep_max[c];                 /* zero for max => infinity */          max = rep_max[c];                 /* zero for max => infinity */
3109          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3080  for (;;) Line 3111  for (;;)
3111    
3112          case OP_CRRANGE:          case OP_CRRANGE:
3113          case OP_CRMINRANGE:          case OP_CRMINRANGE:
3114            case OP_CRPOSRANGE:
3115          minimize = (*ecode == OP_CRMINRANGE);          minimize = (*ecode == OP_CRMINRANGE);
3116            possessive = (*ecode == OP_CRPOSRANGE);
3117          min = GET2(ecode, 1);          min = GET2(ecode, 1);
3118          max = GET2(ecode, 1 + IMM2_SIZE);          max = GET2(ecode, 1 + IMM2_SIZE);
3119          if (max == 0) max = INT_MAX;          if (max == 0) max = INT_MAX;
# Line 3152  for (;;) Line 3185  for (;;)
3185            if (!PRIV(xclass)(c, data, utf)) break;            if (!PRIV(xclass)(c, data, utf)) break;
3186            eptr += len;            eptr += len;
3187            }            }
3188    
3189            if (possessive) continue;    /* No backtracking */
3190    
3191          for(;;)          for(;;)
3192            {            {
3193            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM21);
# Line 3182  for (;;) Line 3218  for (;;)
3218          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */          CHECK_PARTIAL();             /* Not SCHECK_PARTIAL() */
3219          RRETURN(MATCH_NOMATCH);          RRETURN(MATCH_NOMATCH);
3220          }          }
3221        while (length-- > 0) if (*ecode++ != RAWUCHARINC(eptr)) RRETURN(MATCH_NOMATCH);        while (length-- > 0) if (*ecode++ != UCHAR21INC(eptr)) RRETURN(MATCH_NOMATCH);
3222        }        }
3223      else      else
3224  #endif  #endif
# Line 3222  for (;;) Line 3258  for (;;)
3258    
3259        if (fc < 128)        if (fc < 128)
3260          {          {
3261          pcre_uint32 cc = RAWUCHAR(eptr);          pcre_uint32 cc = UCHAR21(eptr);
3262          if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);          if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);
3263          ecode++;          ecode++;
3264          eptr++;          eptr++;
# Line 3491  for (;;) Line 3527  for (;;)
3527            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3528            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3529            }            }
3530          cc = RAWUCHARTEST(eptr);          cc = UCHAR21TEST(eptr);
3531          if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);          if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3532          eptr++;          eptr++;
3533          }          }
# Line 3509  for (;;) Line 3545  for (;;)
3545              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3546              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3547              }              }
3548            cc = RAWUCHARTEST(eptr);            cc = UCHAR21TEST(eptr);
3549            if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);            if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
3550            eptr++;            eptr++;
3551            }            }
# Line 3526  for (;;) Line 3562  for (;;)
3562              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3563              break;              break;
3564              }              }
3565            cc = RAWUCHARTEST(eptr);            cc = UCHAR21TEST(eptr);
3566            if (fc != cc && foc != cc) break;            if (fc != cc && foc != cc) break;
3567            eptr++;            eptr++;
3568            }            }
# Line 3553  for (;;) Line 3589  for (;;)
3589            SCHECK_PARTIAL();            SCHECK_PARTIAL();
3590            RRETURN(MATCH_NOMATCH);            RRETURN(MATCH_NOMATCH);
3591            }            }
3592          if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);          if (fc != UCHAR21INCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3593          }          }
3594    
3595        if (min == max) continue;        if (min == max) continue;
# Line 3570  for (;;) Line 3606  for (;;)
3606              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3607              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
3608              }              }
3609            if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);            if (fc != UCHAR21INCTEST(eptr)) RRETURN(MATCH_NOMATCH);
3610            }            }
3611          /* Control never gets here */          /* Control never gets here */
3612          }          }
# Line 3584  for (;;) Line 3620  for (;;)
3620              SCHECK_PARTIAL();              SCHECK_PARTIAL();
3621              break;              break;
3622              }              }
3623            if (fc != RAWUCHARTEST(eptr)) break;            if (fc != UCHAR21TEST(eptr)) break;
3624            eptr++;            eptr++;
3625            }            }
3626          if (possessive) continue;    /* No backtracking */          if (possessive) continue;    /* No backtracking */
# Line 4200  for (;;) Line 4236  for (;;)
4236            /* 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,
4237            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
4238            was changed at release 8.34. */            was changed at release 8.34. */
4239    
4240            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
4241            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
4242            for (i = 1; i <= min; i++)            for (i = 1; i <= min; i++)
# Line 4211  for (;;) Line 4247  for (;;)
4247                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4248                }                }
4249              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4250              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
4251                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
4252                     == prop_fail_result)                HSPACE_CASES:
4253                RRETURN(MATCH_NOMATCH);                VSPACE_CASES:
4254                  if (prop_fail_result) RRETURN(MATCH_NOMATCH);
4255                  break;
4256    
4257                  default:
4258                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
4259                    RRETURN(MATCH_NOMATCH);
4260                  break;
4261                  }
4262              }              }
4263            break;            break;
4264    
# Line 4331  for (;;) Line 4375  for (;;)
4375                eptr + 1 >= md->end_subject &&                eptr + 1 >= md->end_subject &&
4376                NLBLOCK->nltype == NLTYPE_FIXED &&                NLBLOCK->nltype == NLTYPE_FIXED &&
4377                NLBLOCK->nllen == 2 &&                NLBLOCK->nllen == 2 &&
4378                RAWUCHAR(eptr) == NLBLOCK->nl[0])                UCHAR21(eptr) == NLBLOCK->nl[0])
4379              {              {
4380              md->hitend = TRUE;              md->hitend = TRUE;
4381              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);              if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 4373  for (;;) Line 4417  for (;;)
4417              default: RRETURN(MATCH_NOMATCH);              default: RRETURN(MATCH_NOMATCH);
4418    
4419              case CHAR_CR:              case CHAR_CR:
4420              if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;              if (eptr < md->end_subject && UCHAR21(eptr) == CHAR_LF) eptr++;
4421              break;              break;
4422    
4423              case CHAR_LF:              case CHAR_LF:
# Line 4483  for (;;) Line 4527  for (;;)
4527              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4528              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4529              }              }
4530            cc = RAWUCHAR(eptr);            cc = UCHAR21(eptr);
4531            if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)            if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)
4532              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4533            eptr++;            eptr++;
# Line 4500  for (;;) Line 4544  for (;;)
4544              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4545              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4546              }              }
4547            cc = RAWUCHAR(eptr);            cc = UCHAR21(eptr);
4548            if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)            if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)
4549              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4550            eptr++;            eptr++;
# Line 4517  for (;;) Line 4561  for (;;)
4561              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4562              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4563              }              }
4564            cc = RAWUCHAR(eptr);            cc = UCHAR21(eptr);
4565            if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)            if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)
4566              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4567            eptr++;            eptr++;
# Line 4534  for (;;) Line 4578  for (;;)
4578              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4579              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4580              }              }
4581            cc = RAWUCHAR(eptr);            cc = UCHAR21(eptr);
4582            if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)            if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)
4583              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4584            eptr++;            eptr++;
# Line 4551  for (;;) Line 4595  for (;;)
4595              SCHECK_PARTIAL();              SCHECK_PARTIAL();
4596              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4597              }              }
4598            cc = RAWUCHAR(eptr);            cc = UCHAR21(eptr);
4599            if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)            if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)
4600              RRETURN(MATCH_NOMATCH);              RRETURN(MATCH_NOMATCH);
4601            eptr++;            eptr++;
# Line 4937  for (;;) Line 4981  for (;;)
4981            /* 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,
4982            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
4983            was changed at release 8.34. */            was changed at release 8.34. */
4984    
4985            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
4986            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
4987            for (fi = min;; fi++)            for (fi = min;; fi++)
# Line 4951  for (;;) Line 4995  for (;;)
4995                RRETURN(MATCH_NOMATCH);                RRETURN(MATCH_NOMATCH);
4996                }                }
4997              GETCHARINCTEST(c, eptr);              GETCHARINCTEST(c, eptr);
4998              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
4999                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
5000                     == prop_fail_result)                HSPACE_CASES:
5001                RRETURN(MATCH_NOMATCH);                VSPACE_CASES:
5002                  if (prop_fail_result) RRETURN(MATCH_NOMATCH);
5003                  break;
5004    
5005                  default:
5006                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
5007                    RRETURN(MATCH_NOMATCH);
5008                  break;
5009                  }
5010              }              }
5011            /* Control never gets here */            /* Control never gets here */
5012    
# Line 5007  for (;;) Line 5059  for (;;)
5059            case PT_UCNC:            case PT_UCNC:
5060            for (fi = min;; fi++)            for (fi = min;; fi++)
5061              {              {
5062              RMATCH(eptr, ecode, offset_top, md, eptrb, RM68);              RMATCH(eptr, ecode, offset_top, md, eptrb, RM60);
5063              if (rrc != MATCH_NOMATCH) RRETURN(rrc);              if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5064              if (fi >= max) RRETURN(MATCH_NOMATCH);              if (fi >= max) RRETURN(MATCH_NOMATCH);
5065              if (eptr >= md->end_subject)              if (eptr >= md->end_subject)
# Line 5104  for (;;) Line 5156  for (;;)
5156                {                {
5157                default: RRETURN(MATCH_NOMATCH);                default: RRETURN(MATCH_NOMATCH);
5158                case CHAR_CR:                case CHAR_CR:
5159                if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;                if (eptr < md->end_subject && UCHAR21(eptr) == CHAR_LF) eptr++;
5160                break;                break;
5161    
5162                case CHAR_LF:                case CHAR_LF:
# Line 5441  for (;;) Line 5493  for (;;)
5493            /* 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,
5494            which means that Perl space and POSIX space are now identical. PCRE            which means that Perl space and POSIX space are now identical. PCRE
5495            was changed at release 8.34. */            was changed at release 8.34. */
5496    
5497            case PT_SPACE:    /* Perl space */            case PT_SPACE:    /* Perl space */
5498            case PT_PXSPACE:  /* POSIX space */            case PT_PXSPACE:  /* POSIX space */
5499            for (i = min; i < max; i++)            for (i = min; i < max; i++)
# Line 5453  for (;;) Line 5505  for (;;)
5505                break;                break;
5506                }                }
5507              GETCHARLENTEST(c, eptr, len);              GETCHARLENTEST(c, eptr, len);
5508              if ((UCD_CATEGORY(c) == ucp_Z || c == CHAR_HT || c == CHAR_NL ||              switch(c)
5509                   c == CHAR_VT || c == CHAR_FF || c == CHAR_CR)                {
5510                   == prop_fail_result)                HSPACE_CASES:
5511                  VSPACE_CASES:
5512                  if (prop_fail_result) goto ENDLOOP99;  /* Break the loop */
5513                  break;
5514    
5515                  default:
5516                  if ((UCD_CATEGORY(c) == ucp_Z) == prop_fail_result)
5517                    goto ENDLOOP99;   /* Break the loop */
5518                break;                break;
5519                  }
5520              eptr+= len;              eptr+= len;
5521              }              }
5522              ENDLOOP99:
5523            break;            break;
5524    
5525            case PT_WORD:            case PT_WORD:
# Line 5572  for (;;) Line 5633  for (;;)
5633          /* eptr is now past the end of the maximum run */          /* eptr is now past the end of the maximum run */
5634    
5635          if (possessive) continue;    /* No backtracking */          if (possessive) continue;    /* No backtracking */
5636    
5637          for(;;)          for(;;)
5638            {            {
5639            int lgb, rgb;            int lgb, rgb;
5640            PCRE_PUCHAR fptr;            PCRE_PUCHAR fptr;
5641    
5642            if (eptr == pp) goto TAIL_RECURSE;   /* At start of char run */            if (eptr == pp) goto TAIL_RECURSE;   /* At start of char run */
5643            RMATCH(eptr, ecode, offset_top, md, eptrb, RM45);            RMATCH(eptr, ecode, offset_top, md, eptrb, RM45);
5644            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
# Line 5585  for (;;) Line 5646  for (;;)
5646            /* Backtracking over an extended grapheme cluster involves inspecting            /* Backtracking over an extended grapheme cluster involves inspecting
5647            the previous two characters (if present) to see if a break is            the previous two characters (if present) to see if a break is
5648            permitted between them. */            permitted between them. */
5649    
5650            eptr--;            eptr--;
5651            if (!utf) c = *eptr; else            if (!utf) c = *eptr; else
5652              {              {
# Line 5603  for (;;) Line 5664  for (;;)
5664                BACKCHAR(fptr);                BACKCHAR(fptr);
5665                GETCHAR(c, fptr);                GETCHAR(c, fptr);
5666                }                }
5667              lgb = UCD_GRAPHBREAK(c);              lgb = UCD_GRAPHBREAK(c);
5668              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;              if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break;
5669              eptr = fptr;              eptr = fptr;
5670              rgb = lgb;              rgb = lgb;
# Line 5634  for (;;) Line 5695  for (;;)
5695                    eptr + 1 >= md->end_subject &&                    eptr + 1 >= md->end_subject &&
5696                    NLBLOCK->nltype == NLTYPE_FIXED &&                    NLBLOCK->nltype == NLTYPE_FIXED &&
5697                    NLBLOCK->nllen == 2 &&                    NLBLOCK->nllen == 2 &&
5698                    RAWUCHAR(eptr) == NLBLOCK->nl[0])                    UCHAR21(eptr) == NLBLOCK->nl[0])
5699                  {                  {
5700                  md->hitend = TRUE;                  md->hitend = TRUE;
5701                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 5660  for (;;) Line 5721  for (;;)
5721                    eptr + 1 >= md->end_subject &&                    eptr + 1 >= md->end_subject &&
5722                    NLBLOCK->nltype == NLTYPE_FIXED &&                    NLBLOCK->nltype == NLTYPE_FIXED &&
5723                    NLBLOCK->nllen == 2 &&                    NLBLOCK->nllen == 2 &&
5724                    RAWUCHAR(eptr) == NLBLOCK->nl[0])                    UCHAR21(eptr) == NLBLOCK->nl[0])
5725                  {                  {
5726                  md->hitend = TRUE;                  md->hitend = TRUE;
5727                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);                  if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
# Line 5717  for (;;) Line 5778  for (;;)
5778              if (c == CHAR_CR)              if (c == CHAR_CR)
5779                {                {
5780                if (++eptr >= md->end_subject) break;                if (++eptr >= md->end_subject) break;
5781                if (RAWUCHAR(eptr) == CHAR_LF) eptr++;                if (UCHAR21(eptr) == CHAR_LF) eptr++;
5782                }                }
5783              else              else
5784                {                {
# Line 5880  for (;;) Line 5941  for (;;)
5941            if (rrc != MATCH_NOMATCH) RRETURN(rrc);            if (rrc != MATCH_NOMATCH) RRETURN(rrc);
5942            eptr--;            eptr--;
5943            BACKCHAR(eptr);            BACKCHAR(eptr);
5944            if (ctype == OP_ANYNL && eptr > pp  && RAWUCHAR(eptr) == CHAR_NL &&            if (ctype == OP_ANYNL && eptr > pp  && UCHAR21(eptr) == CHAR_NL &&
5945                RAWUCHAR(eptr - 1) == CHAR_CR) eptr--;                UCHAR21(eptr - 1) == CHAR_CR) eptr--;
5946            }            }
5947          }          }
5948        else        else
# Line 6127  for (;;) Line 6188  for (;;)
6188                eptr[-1] == CHAR_CR) eptr--;                eptr[-1] == CHAR_CR) eptr--;
6189            }            }
6190          }          }
6191    
6192        /* Control never gets here */        /* Control never gets here */
6193        }        }
6194    
# Line 6163  switch (frame->Xwhere) Line 6224  switch (frame->Xwhere)
6224    LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64)    LBL(53) LBL(54) LBL(55) LBL(56) LBL(57) LBL(58) LBL(63) LBL(64)
6225    LBL(65) LBL(66)    LBL(65) LBL(66)
6226  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8  #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
6227    LBL(21)    LBL(20) LBL(21)
6228  #endif  #endif
6229  #ifdef SUPPORT_UTF  #ifdef SUPPORT_UTF
6230    LBL(16) LBL(18) LBL(20)    LBL(16) LBL(18)
6231    LBL(22) LBL(23) LBL(28) LBL(30)    LBL(22) LBL(23) LBL(28) LBL(30)
6232    LBL(32) LBL(34) LBL(42) LBL(46)    LBL(32) LBL(34) LBL(42) LBL(46)
6233  #ifdef SUPPORT_UCP  #ifdef SUPPORT_UCP
6234    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)
6235    LBL(59) LBL(60) LBL(61) LBL(62) LBL(67) LBL(68)    LBL(59) LBL(60) LBL(61) LBL(62) LBL(67)
6236  #endif  /* SUPPORT_UCP */  #endif  /* SUPPORT_UCP */
6237  #endif  /* SUPPORT_UTF */  #endif  /* SUPPORT_UTF */
6238    default:    default:
# Line 6728  for(;;) Line 6789  for(;;)
6789    
6790        if (first_char != first_char2)        if (first_char != first_char2)
6791          while (start_match < end_subject &&          while (start_match < end_subject &&
6792            (smc = RAWUCHARTEST(start_match)) != first_char && smc != first_char2)            (smc = UCHAR21TEST(start_match)) != first_char && smc != first_char2)
6793            start_match++;            start_match++;
6794        else        else
6795          while (start_match < end_subject && RAWUCHARTEST(start_match) != first_char)          while (start_match < end_subject && UCHAR21TEST(start_match) != first_char)
6796            start_match++;            start_match++;
6797        }        }
6798    
# Line 6763  for(;;) Line 6824  for(;;)
6824          if (start_match[-1] == CHAR_CR &&          if (start_match[-1] == CHAR_CR &&
6825               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&               (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
6826               start_match < end_subject &&               start_match < end_subject &&
6827               RAWUCHARTEST(start_match) == CHAR_NL)               UCHAR21TEST(start_match) == CHAR_NL)
6828            start_match++;            start_match++;
6829          }          }
6830        }        }
# Line 6774  for(;;) Line 6835  for(;;)
6835        {        {
6836        while (start_match < end_subject)        while (start_match < end_subject)
6837          {          {
6838          register pcre_uint32 c = RAWUCHARTEST(start_match);          register pcre_uint32 c = UCHAR21TEST(start_match);
6839  #ifndef COMPILE_PCRE8  #ifndef COMPILE_PCRE8
6840          if (c > 255) c = 255;          if (c > 255) c = 255;
6841  #endif  #endif
6842          if ((start_bits[c/8] & (1 << (c&7))) == 0)          if ((start_bits[c/8] & (1 << (c&7))) != 0) break;
6843            {          start_match++;
           start_match++;  
 #if defined SUPPORT_UTF && defined COMPILE_PCRE8  
           /* In non 8-bit mode, the iteration will stop for  
           characters > 255 at the beginning or not stop at all. */  
           if (utf)  
             ACROSSCHAR(start_match < end_subject, *start_match,  
               start_match++);  
 #endif  
           }  
         else break;  
6844          }          }
6845        }        }
6846      }   /* Starting optimizations */      }   /* Starting optimizations */
# Line 6842  for(;;) Line 6893  for(;;)
6893            {            {
6894            while (p < end_subject)            while (p < end_subject)
6895              {              {
6896              register pcre_uint32 pp = RAWUCHARINCTEST(p);              register pcre_uint32 pp = UCHAR21INCTEST(p);
6897              if (pp == req_char || pp == req_char2) { p--; break; }              if (pp == req_char || pp == req_char2) { p--; break; }
6898              }              }
6899            }            }
# Line 6850  for(;;) Line 6901  for(;;)
6901            {            {
6902            while (p < end_subject)            while (p < end_subject)
6903              {              {
6904              if (RAWUCHARINCTEST(p) == req_char) { p--; break; }              if (UCHAR21INCTEST(p) == req_char) { p--; break; }
6905              }              }
6906            }            }
6907    

Legend:
Removed from v.1365  
changed lines
  Added in v.1431

  ViewVC Help
Powered by ViewVC 1.1.5