/[pcre]/code/trunk/pcre_internal.h
ViewVC logotype

Diff of /code/trunk/pcre_internal.h

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 459 by ph10, Sun Oct 4 09:21:39 2009 UTC revision 573 by ph10, Fri Nov 19 19:29:44 2010 UTC
# Line 7  Line 7 
7  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.
8    
9                         Written by Philip Hazel                         Written by Philip Hazel
10             Copyright (c) 1997-2009 University of Cambridge             Copyright (c) 1997-2010 University of Cambridge
11    
12  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
13  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 45  functions whose names all begin with "_p Line 45  functions whose names all begin with "_p
45  #ifndef PCRE_INTERNAL_H  #ifndef PCRE_INTERNAL_H
46  #define PCRE_INTERNAL_H  #define PCRE_INTERNAL_H
47    
48  /* Define DEBUG to get debugging output on stdout. */  /* Define PCRE_DEBUG to get debugging output on stdout. */
49    
50  #if 0  #if 0
51  #define DEBUG  #define PCRE_DEBUG
52  #endif  #endif
53    
54  /* We do not support both EBCDIC and UTF-8 at the same time. The "configure"  /* We do not support both EBCDIC and UTF-8 at the same time. The "configure"
# Line 74  It turns out that the Mac Debugging.h he Line 74  It turns out that the Mac Debugging.h he
74  be absolutely sure we get our version. */  be absolutely sure we get our version. */
75    
76  #undef DPRINTF  #undef DPRINTF
77  #ifdef DEBUG  #ifdef PCRE_DEBUG
78  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
79  #else  #else
80  #define DPRINTF(p) /* Nothing */  #define DPRINTF(p) /* Nothing */
# Line 86  setjmp and stdarg are used is when NO_RE Line 86  setjmp and stdarg are used is when NO_RE
86    
87  #include <ctype.h>  #include <ctype.h>
88  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
89  #include <stddef.h>  #include <stddef.h>
90  #include <stdio.h>  #include <stdio.h>
91  #include <stdlib.h>  #include <stdlib.h>
# Line 186  preprocessor time in standard C environm Line 184  preprocessor time in standard C environm
184    #error Cannot determine a type for 32-bit unsigned integers    #error Cannot determine a type for 32-bit unsigned integers
185  #endif  #endif
186    
187    /* When checking for integer overflow in pcre_compile(), we need to handle
188    large integers. If a 64-bit integer type is available, we can use that.
189    Otherwise we have to cast to double, which of course requires floating point
190    arithmetic. Handle this by defining a macro for the appropriate type. If
191    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
192    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
193    by "configure". */
194    
195    #if HAVE_STDINT_H
196    #include <stdint.h>
197    #elif HAVE_INTTYPES_H
198    #include <inttypes.h>
199    #endif
200    
201    #if defined INT64_MAX || defined int64_t
202    #define INT64_OR_DOUBLE int64_t
203    #else
204    #define INT64_OR_DOUBLE double
205    #endif
206    
207  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
208  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
209  However, we leave the interface to the outside world as char *, because that  However, we leave the interface to the outside world as char *, because that
# Line 390  capturing parenthesis numbers in back re Line 408  capturing parenthesis numbers in back re
408    
409  /* When UTF-8 encoding is being used, a character is no longer just a single  /* When UTF-8 encoding is being used, a character is no longer just a single
410  byte. The macros for character handling generate simple sequences when used in  byte. The macros for character handling generate simple sequences when used in
411  byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should  byte-mode, and more complicated ones for UTF-8 characters. GETCHARLENTEST is
412  never be called in byte mode. To make sure it can never even appear when UTF-8  not used when UTF-8 is not supported, so it is not defined, and BACKCHAR should
413  support is omitted, we don't even define it. */  never be called in byte mode. To make sure they can never even appear when
414    UTF-8 support is omitted, we don't even define them. */
415    
416  #ifndef SUPPORT_UTF8  #ifndef SUPPORT_UTF8
417  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
# Line 400  support is omitted, we don't even define Line 419  support is omitted, we don't even define
419  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
420  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
421  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
422    /* #define GETCHARLENTEST(c, eptr, len) */
423  /* #define BACKCHAR(eptr) */  /* #define BACKCHAR(eptr) */
424    
425  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF8 */
426    
427    /* These macros were originally written in the form of loops that used data
428    from the tables whose names start with _pcre_utf8_table. They were rewritten by
429    a user so as not to use loops, because in some environments this gives a
430    significant performance advantage, and it seems never to do any harm. */
431    
432    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
433    advancing the pointer. */
434    
435    #define GETUTF8(c, eptr) \
436        { \
437        if ((c & 0x20) == 0) \
438          c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
439        else if ((c & 0x10) == 0) \
440          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
441        else if ((c & 0x08) == 0) \
442          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
443          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
444        else if ((c & 0x04) == 0) \
445          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
446              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
447              (eptr[4] & 0x3f); \
448        else \
449          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
450              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
451              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
452        }
453    
454  /* Get the next UTF-8 character, not advancing the pointer. This is called when  /* Get the next UTF-8 character, not advancing the pointer. This is called when
455  we know we are in UTF-8 mode. */  we know we are in UTF-8 mode. */
456    
457  #define GETCHAR(c, eptr) \  #define GETCHAR(c, eptr) \
458    c = *eptr; \    c = *eptr; \
459    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8(c, eptr);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     }  
460    
461  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
462  pointer. */  pointer. */
463    
464  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
465    c = *eptr; \    c = *eptr; \
466    if (utf8 && c >= 0xc0) \    if (utf8 && c >= 0xc0) GETUTF8(c, eptr);
467    
468    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
469    the pointer. */
470    
471    #define GETUTF8INC(c, eptr) \
472      { \      { \
473      int gcii; \      if ((c & 0x20) == 0) \
474      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
475      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
476      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        { \
477      for (gcii = 1; gcii <= gcaa; gcii++) \        c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
478          eptr += 2; \
479          } \
480        else if ((c & 0x08) == 0) \
481          { \
482          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
483              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
484          eptr += 3; \
485          } \
486        else if ((c & 0x04) == 0) \
487          { \
488          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
489              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
490              (eptr[3] & 0x3f); \
491          eptr += 4; \
492          } \
493        else \
494        { \        { \
495        gcss -= 6; \        c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
496        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
497              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
498          eptr += 5; \
499        } \        } \
500      }      }
501    
# Line 445  know we are in UTF-8 mode. */ Line 504  know we are in UTF-8 mode. */
504    
505  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
506    c = *eptr++; \    c = *eptr++; \
507    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8INC(c, eptr);
     { \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
       { \  
       gcss -= 6; \  
       c |= (*eptr++ & 0x3f) << gcss; \  
       } \  
     }  
508    
509  /* Get the next character, testing for UTF-8 mode, and advancing the pointer */  /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
510    This is called when we don't know if we are in UTF-8 mode. */
511    
512  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
513    c = *eptr++; \    c = *eptr++; \
514    if (utf8 && c >= 0xc0) \    if (utf8 && c >= 0xc0) GETUTF8INC(c, eptr);
515    
516    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
517    advancing the pointer, incrementing the length. */
518    
519    #define GETUTF8LEN(c, eptr, len) \
520      { \      { \
521      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
522        { \        { \
523        gcss -= 6; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
524        c |= (*eptr++ & 0x3f) << gcss; \        len++; \
525          } \
526        else if ((c & 0x10)  == 0) \
527          { \
528          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
529          len += 2; \
530          } \
531        else if ((c & 0x08)  == 0) \
532          {\
533          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
534              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
535          len += 3; \
536          } \
537        else if ((c & 0x04)  == 0) \
538          { \
539          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
540              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
541              (eptr[4] & 0x3f); \
542          len += 4; \
543          } \
544        else \
545          {\
546          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
547              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
548              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
549          len += 5; \
550        } \        } \
551      }      }
552    
# Line 478  if there are extra bytes. This is called Line 555  if there are extra bytes. This is called
555    
556  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
557    c = *eptr; \    c = *eptr; \
558    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     len += gcaa; \  
     }  
559    
560  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
561  pointer, incrementing length if there are extra bytes. This is called when we  pointer, incrementing length if there are extra bytes. This is called when we
562  know we are in UTF-8 mode. */  do not know if we are in UTF-8 mode. */
563    
564  #define GETCHARLENTEST(c, eptr, len) \  #define GETCHARLENTEST(c, eptr, len) \
565    c = *eptr; \    c = *eptr; \
566    if (utf8 && c >= 0xc0) \    if (utf8 && c >= 0xc0) GETUTF8LEN(c, eptr, len);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     len += gcaa; \  
     }  
567    
568  /* If the pointer is not at the start of a character, move it back until  /* If the pointer is not at the start of a character, move it back until
569  it is. This is called only in UTF-8 mode - we don't put a test within the macro  it is. This is called only in UTF-8 mode - we don't put a test within the macro
# Line 518  because almost all calls are already wit Line 571  because almost all calls are already wit
571    
572  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
573    
574  #endif  #endif  /* SUPPORT_UTF8 */
575    
576    
577  /* In case there is no definition of offsetof() provided - though any proper  /* In case there is no definition of offsetof() provided - though any proper
# Line 562  time, run time, or study time, respectiv Line 615  time, run time, or study time, respectiv
615     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
616     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
617     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
618     PCRE_JAVASCRIPT_COMPAT)     PCRE_JAVASCRIPT_COMPAT|PCRE_UCP)
619    
620  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
621    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
# Line 857  so that PCRE works on both ASCII and EBC Line 910  so that PCRE works on both ASCII and EBC
910  #define STRING_COMMIT0              "COMMIT\0"  #define STRING_COMMIT0              "COMMIT\0"
911  #define STRING_F0                   "F\0"  #define STRING_F0                   "F\0"
912  #define STRING_FAIL0                "FAIL\0"  #define STRING_FAIL0                "FAIL\0"
913    #define STRING_MARK0                "MARK\0"
914  #define STRING_PRUNE0               "PRUNE\0"  #define STRING_PRUNE0               "PRUNE\0"
915  #define STRING_SKIP0                "SKIP\0"  #define STRING_SKIP0                "SKIP\0"
916  #define STRING_THEN                 "THEN"  #define STRING_THEN                 "THEN"
# Line 886  so that PCRE works on both ASCII and EBC Line 940  so that PCRE works on both ASCII and EBC
940  #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"  #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
941  #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"  #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
942  #define STRING_UTF8_RIGHTPAR        "UTF8)"  #define STRING_UTF8_RIGHTPAR        "UTF8)"
943    #define STRING_UCP_RIGHTPAR         "UCP)"
944    
945  #else  /* SUPPORT_UTF8 */  #else  /* SUPPORT_UTF8 */
946    
# Line 1109  only. */ Line 1164  only. */
1164  #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"  #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1165  #define STRING_F0                   STR_F "\0"  #define STRING_F0                   STR_F "\0"
1166  #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"  #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1167    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1168  #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"  #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1169  #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"  #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1170  #define STRING_THEN                 STR_T STR_H STR_E STR_N  #define STRING_THEN                 STR_T STR_H STR_E STR_N
# Line 1138  only. */ Line 1194  only. */
1194  #define STRING_BSR_ANYCRLF_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS  #define STRING_BSR_ANYCRLF_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1195  #define STRING_BSR_UNICODE_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS  #define STRING_BSR_UNICODE_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
1196  #define STRING_UTF8_RIGHTPAR        STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS  #define STRING_UTF8_RIGHTPAR        STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1197    #define STRING_UCP_RIGHTPAR         STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1198    
1199  #endif  /* SUPPORT_UTF8 */  #endif  /* SUPPORT_UTF8 */
1200    
# Line 1170  only. */ Line 1227  only. */
1227    
1228  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1229  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1230  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1231  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1232  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1233    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1234    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1235    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1236    #define PT_WORD       8    /* Word - L plus N plus underscore */
1237    
1238  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1239  contain UTF-8 characters with values greater than 255. */  contain UTF-8 characters with values greater than 255. */
# Line 1189  contain UTF-8 characters with values gre Line 1250  contain UTF-8 characters with values gre
1250  /* These are escaped items that aren't just an encoding of a particular data  /* These are escaped items that aren't just an encoding of a particular data
1251  value such as \n. They must have non-zero values, as check_escape() returns  value such as \n. They must have non-zero values, as check_escape() returns
1252  their negation. Also, they must appear in the same order as in the opcode  their negation. Also, they must appear in the same order as in the opcode
1253  definitions below, up to ESC_z. There's a dummy for OP_ANY because it  definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1254  corresponds to "." rather than an escape sequence, and another for OP_ALLANY  corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1255  (which is used for [^] in JavaScript compatibility mode).  used for [^] in JavaScript compatibility mode. In non-DOTALL mode, "." behaves
1256    like \N.
1257    
1258    The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1259    when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1260    They must be contiguous, and remain in order so that the replacements can be
1261    looked up from a table.
1262    
1263  The final escape must be ESC_REF as subsequent values are used for  The final escape must be ESC_REF as subsequent values are used for
1264  backreferences (\1, \2, \3, etc). There are two tests in the code for an escape  backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
# Line 1201  put in between that don't consume a char Line 1268  put in between that don't consume a char
1268  */  */
1269    
1270  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1271         ESC_W, ESC_w, ESC_dum1, ESC_dum2, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,         ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1272         ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_g, ESC_k,         ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1273           ESC_E, ESC_Q, ESC_g, ESC_k,
1274           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1275         ESC_REF };         ESC_REF };
1276    
   
1277  /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to  /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1278  OP_EOD must correspond in order to the list of escapes immediately above.  OP_EOD must correspond in order to the list of escapes immediately above.
1279    
1280  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1281  that follow must also be updated to match. There is also a table called  that follow must also be updated to match. There are also tables called
1282  "coptable" in pcre_dfa_exec.c that must be updated. */  "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1283    
1284  enum {  enum {
1285    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 1229  enum { Line 1297  enum {
1297    OP_WHITESPACE,         /*  9 \s */    OP_WHITESPACE,         /*  9 \s */
1298    OP_NOT_WORDCHAR,       /* 10 \W */    OP_NOT_WORDCHAR,       /* 10 \W */
1299    OP_WORDCHAR,           /* 11 \w */    OP_WORDCHAR,           /* 11 \w */
1300    OP_ANY,            /* 12 Match any character (subject to DOTALL) */    OP_ANY,            /* 12 Match any character except newline */
1301    OP_ALLANY,         /* 13 Match any character (not subject to DOTALL) */    OP_ALLANY,         /* 13 Match any character */
1302    OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */    OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1303    OP_NOTPROP,        /* 15 \P (not Unicode property) */    OP_NOTPROP,        /* 15 \P (not Unicode property) */
1304    OP_PROP,           /* 16 \p (Unicode property) */    OP_PROP,           /* 16 \p (Unicode property) */
# Line 1348  enum { Line 1416  enum {
1416    OP_SCOND,          /* 99 Conditional group, check empty */    OP_SCOND,          /* 99 Conditional group, check empty */
1417    
1418    /* The next two pairs must (respectively) be kept together. */    /* The next two pairs must (respectively) be kept together. */
1419    
1420    OP_CREF,           /* 100 Used to hold a capture number as condition */    OP_CREF,           /* 100 Used to hold a capture number as condition */
1421    OP_NCREF,          /* 101 Same, but generaged by a name reference*/    OP_NCREF,          /* 101 Same, but generaged by a name reference*/
1422    OP_RREF,           /* 102 Used to hold a recursion number as condition */    OP_RREF,           /* 102 Used to hold a recursion number as condition */
# Line 1360  enum { Line 1428  enum {
1428    
1429    /* These are backtracking control verbs */    /* These are backtracking control verbs */
1430    
1431    OP_PRUNE,          /* 107 */    OP_MARK,           /* 107 always has an argument */
1432    OP_SKIP,           /* 108 */    OP_PRUNE,          /* 108 */
1433    OP_THEN,           /* 109 */    OP_PRUNE_ARG,      /* 109 same, but with argument */
1434    OP_COMMIT,         /* 110 */    OP_SKIP,           /* 110 */
1435      OP_SKIP_ARG,       /* 111 same, but with argument */
1436      OP_THEN,           /* 112 */
1437      OP_THEN_ARG,       /* 113 same, but with argument */
1438      OP_COMMIT,         /* 114 */
1439    
1440    /* These are forced failure and success verbs */    /* These are forced failure and success verbs */
1441    
1442    OP_FAIL,           /* 111 */    OP_FAIL,           /* 115 */
1443    OP_ACCEPT,         /* 112 */    OP_ACCEPT,         /* 116 */
1444    OP_CLOSE,          /* 113 Used before OP_ACCEPT to close open captures */    OP_CLOSE,          /* 117 Used before OP_ACCEPT to close open captures */
1445    
1446    /* This is used to skip a subpattern with a {0} quantifier */    /* This is used to skip a subpattern with a {0} quantifier */
1447    
1448    OP_SKIPZERO        /* 114 */    OP_SKIPZERO,       /* 118 */
1449    
1450      /* This is not an opcode, but is used to check that tables indexed by opcode
1451      are the correct length, in order to catch updating errors - there have been
1452      some in the past. */
1453    
1454      OP_TABLE_LENGTH
1455  };  };
1456    
1457    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1458    definitions that follow must also be updated to match. There are also tables
1459    called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1460    
1461    
1462  /* This macro defines textual names for all the opcodes. These are used only  /* This macro defines textual names for all the opcodes. These are used only
1463  for debugging. The macro is referenced only in pcre_printint.c. */  for debugging. The macro is referenced only in pcre_printint.c. */
# Line 1399  for debugging. The macro is referenced o Line 1481  for debugging. The macro is referenced o
1481    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \
1482    "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \    "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
1483    "Brazero", "Braminzero",                                        \    "Brazero", "Braminzero",                                        \
1484    "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT",      \    "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
1485      "*THEN", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT",                \
1486    "Close", "Skip zero"    "Close", "Skip zero"
1487    
1488    
# Line 1417  in UTF-8 mode. The code that uses this t Line 1500  in UTF-8 mode. The code that uses this t
1500    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
1501    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
1502    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
1503    3, 3, 1,                       /* NOTPROP, PROP, EXTUNI                  */ \    3, 3,                          /* \P, \p                                 */ \
1504    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
1505      1,                             /* \X                                     */ \
1506    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \
1507    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
1508    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Charnc  - the minimum length           */ \
# Line 1464  in UTF-8 mode. The code that uses this t Line 1548  in UTF-8 mode. The code that uses this t
1548    3, 3,                          /* RREF, NRREF                            */ \    3, 3,                          /* RREF, NRREF                            */ \
1549    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
1550    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \
1551    1, 1, 1, 1,                    /* PRUNE, SKIP, THEN, COMMIT,             */ \    3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
1552    1, 1, 3, 1                     /* FAIL, ACCEPT, CLOSE, SKIPZERO          */    1, 3,                          /* SKIP, SKIP_ARG                         */ \
1553      1+LINK_SIZE, 3+LINK_SIZE,      /* THEN, THEN_ARG                         */ \
1554      1, 1, 1, 3, 1                  /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO  */
1555    
1556    
1557  /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"  /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
# Line 1473  condition. */ Line 1559  condition. */
1559    
1560  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
1561    
1562  /* Error code numbers. They are given names so that they can more easily be  /* Compile time error code numbers. They are given names so that they can more
1563  tracked. */  easily be tracked. When a new number is added, the table called eint in
1564    pcreposix.c must be updated. */
1565    
1566  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
1567         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
# Line 1482  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4, Line 1569  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,
1569         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1570         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1571         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1572         ERR60, ERR61, ERR62, ERR63, ERR64, ERR65 };         ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERRCOUNT };
1573    
1574  /* The real format of the start of the pcre block; the index of names and the  /* The real format of the start of the pcre block; the index of names and the
1575  code vector run on as long as necessary after the end. We store an explicit  code vector run on as long as necessary after the end. We store an explicit
# Line 1532  typedef struct pcre_study_data { Line 1619  typedef struct pcre_study_data {
1619    
1620  /* Structure for building a chain of open capturing subpatterns during  /* Structure for building a chain of open capturing subpatterns during
1621  compiling, so that instructions to close them can be compiled when (*ACCEPT) is  compiling, so that instructions to close them can be compiled when (*ACCEPT) is
1622  encountered. */  encountered. This is also used to identify subpatterns that contain recursive
1623    back references to themselves, so that they can be made atomic. */
1624    
1625  typedef struct open_capitem {  typedef struct open_capitem {
1626    struct open_capitem *next;    /* Chain link */    struct open_capitem *next;    /* Chain link */
1627    pcre_uint16 number;           /* Capture number */    pcre_uint16 number;           /* Capture number */
1628      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
1629  } open_capitem;  } open_capitem;
1630    
1631  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
# Line 1575  branches, for testing for left recursion Line 1664  branches, for testing for left recursion
1664    
1665  typedef struct branch_chain {  typedef struct branch_chain {
1666    struct branch_chain *outer;    struct branch_chain *outer;
1667    uschar *current;    uschar *current_branch;
1668  } branch_chain;  } branch_chain;
1669    
1670  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
# Line 1585  typedef struct recursion_info { Line 1674  typedef struct recursion_info {
1674    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
1675    int group_num;                /* Number of group that was called */    int group_num;                /* Number of group that was called */
1676    const uschar *after_call;     /* "Return value": points after the call in the expr */    const uschar *after_call;     /* "Return value": points after the call in the expr */
   USPTR save_start;             /* Old value of mstart */  
1677    int *offset_save;             /* Pointer to start of saved offsets */    int *offset_save;             /* Pointer to start of saved offsets */
1678    int saved_max;                /* Number of saved offsets */    int saved_max;                /* Number of saved offsets */
1679    int offset_top;               /* Current value of offset_top */    int save_offset_top;          /* Current value of offset_top */
1680  } recursion_info;  } recursion_info;
1681    
1682  /* Structure for building a chain of data for holding the values of the subject  /* Structure for building a chain of data for holding the values of the subject
# Line 1615  typedef struct match_data { Line 1703  typedef struct match_data {
1703    int    nllen;                 /* Newline string length */    int    nllen;                 /* Newline string length */
1704    int    name_count;            /* Number of names in name table */    int    name_count;            /* Number of names in name table */
1705    int    name_entry_size;       /* Size of entry in names table */    int    name_entry_size;       /* Size of entry in names table */
1706    uschar *name_table;           /* Table of names */    uschar *name_table;           /* Table of names */
1707    uschar nl[4];                 /* Newline string when fixed */    uschar nl[4];                 /* Newline string when fixed */
1708    const uschar *lcc;            /* Points to lower casing table */    const uschar *lcc;            /* Points to lower casing table */
1709    const uschar *ctypes;         /* Points to table of type maps */    const uschar *ctypes;         /* Points to table of type maps */
# Line 1624  typedef struct match_data { Line 1712  typedef struct match_data {
1712    BOOL   noteol;                /* NOTEOL flag */    BOOL   noteol;                /* NOTEOL flag */
1713    BOOL   utf8;                  /* UTF8 flag */    BOOL   utf8;                  /* UTF8 flag */
1714    BOOL   jscript_compat;        /* JAVASCRIPT_COMPAT flag */    BOOL   jscript_compat;        /* JAVASCRIPT_COMPAT flag */
1715      BOOL   use_ucp;               /* PCRE_UCP flag */
1716    BOOL   endonly;               /* Dollar not before final \n */    BOOL   endonly;               /* Dollar not before final \n */
1717    BOOL   notempty;              /* Empty string match not wanted */    BOOL   notempty;              /* Empty string match not wanted */
1718    BOOL   notempty_atstart;      /* Empty string match at start not wanted */    BOOL   notempty_atstart;      /* Empty string match at start not wanted */
# Line 1643  typedef struct match_data { Line 1732  typedef struct match_data {
1732    int    eptrn;                 /* Next free eptrblock */    int    eptrn;                 /* Next free eptrblock */
1733    recursion_info *recursive;    /* Linked list of recursion data */    recursion_info *recursive;    /* Linked list of recursion data */
1734    void  *callout_data;          /* To pass back to callouts */    void  *callout_data;          /* To pass back to callouts */
1735      const uschar *mark;           /* Mark pointer to pass back */
1736  } match_data;  } match_data;
1737    
1738  /* A similar structure is used for the same purpose by the DFA matching  /* A similar structure is used for the same purpose by the DFA matching
# Line 1734  one of the exported public functions. Th Line 1824  one of the exported public functions. Th
1824  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
1825    
1826  extern const uschar *_pcre_find_bracket(const uschar *, BOOL, int);  extern const uschar *_pcre_find_bracket(const uschar *, BOOL, int);
1827  extern BOOL          _pcre_is_newline(const uschar *, int, const uschar *,  extern BOOL          _pcre_is_newline(USPTR, int, USPTR, int *, BOOL);
                        int *, BOOL);  
1828  extern int           _pcre_ord2utf8(int, uschar *);  extern int           _pcre_ord2utf8(int, uschar *);
1829  extern real_pcre    *_pcre_try_flipped(const real_pcre *, real_pcre *,  extern real_pcre    *_pcre_try_flipped(const real_pcre *, real_pcre *,
1830                         const pcre_study_data *, pcre_study_data *);                         const pcre_study_data *, pcre_study_data *);
1831  extern int           _pcre_valid_utf8(const uschar *, int);  extern int           _pcre_valid_utf8(USPTR, int);
1832  extern BOOL          _pcre_was_newline(const uschar *, int, const uschar *,  extern BOOL          _pcre_was_newline(USPTR, int, USPTR, int *, BOOL);
                        int *, BOOL);  
1833  extern BOOL          _pcre_xclass(int, const uschar *);  extern BOOL          _pcre_xclass(int, const uschar *);
1834    
1835    

Legend:
Removed from v.459  
changed lines
  Added in v.573

  ViewVC Help
Powered by ViewVC 1.1.5