"Fossies" - the Fresh Open Source Software Archive

Member "AutoHotkey_L-1.1.33.09/source/lib_pcre/pcre/pcre_internal.h" (8 May 2021, 94859 Bytes) of package /windows/misc/AutoHotkey_L-1.1.33.09.zip:


As a special service "Fossies" has tried to format the requested source page into HTML format using (guessed) C and C++ source code syntax highlighting (style: standard) with prefixed line numbers and code folding option. Alternatively you can here view or download the uninterpreted source code file. For more information about "pcre_internal.h" see the Fossies "Dox" file reference documentation.

    1 /*************************************************
    2 *      Perl-Compatible Regular Expressions       *
    3 *************************************************/
    4 
    5 
    6 /* PCRE is a library of functions to support regular expressions whose syntax
    7 and semantics are as close as possible to those of the Perl 5 language.
    8 
    9                        Written by Philip Hazel
   10            Copyright (c) 1997-2012 University of Cambridge
   11 
   12 -----------------------------------------------------------------------------
   13 Redistribution and use in source and binary forms, with or without
   14 modification, are permitted provided that the following conditions are met:
   15 
   16     * Redistributions of source code must retain the above copyright notice,
   17       this list of conditions and the following disclaimer.
   18 
   19     * Redistributions in binary form must reproduce the above copyright
   20       notice, this list of conditions and the following disclaimer in the
   21       documentation and/or other materials provided with the distribution.
   22 
   23     * Neither the name of the University of Cambridge nor the names of its
   24       contributors may be used to endorse or promote products derived from
   25       this software without specific prior written permission.
   26 
   27 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
   28 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
   29 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
   30 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
   31 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
   32 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
   33 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
   34 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
   35 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
   36 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
   37 POSSIBILITY OF SUCH DAMAGE.
   38 -----------------------------------------------------------------------------
   39 */
   40 
   41 /* This header contains definitions that are shared between the different
   42 modules, but which are not relevant to the exported API. This includes some
   43 functions whose names all begin with "_pcre_" or "_pcre16_" depending on
   44 the PRIV macro. */
   45 
   46 #ifndef PCRE_INTERNAL_H
   47 #define PCRE_INTERNAL_H
   48 
   49 /* Define PCRE_DEBUG to get debugging output on stdout. */
   50 
   51 #if 0
   52 #define PCRE_DEBUG
   53 #endif
   54 
   55 /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
   56 #ifndef COMPILE_PCRE16
   57 #define COMPILE_PCRE8
   58 #endif
   59 
   60 /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
   61 "configure" script ensures this, but not everybody uses "configure". */
   62 
   63 #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
   64 #define SUPPORT_UTF 1
   65 #endif
   66 
   67 /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
   68 reasons with existing code. */
   69 
   70 #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
   71 #define SUPPORT_UTF 1
   72 #endif
   73 
   74 /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
   75 Until then we define it if SUPPORT_UTF is defined. */
   76 
   77 #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
   78 #define SUPPORT_UTF8 1
   79 #endif
   80 
   81 /* AutoHotkey: UTF mode is always enabled in some builds and never enabled in
   82 others. This macros is used to aid the compiler to reduce code size. */
   83 
   84 #if defined SUPPORT_UTF_OPTION && defined SUPPORT_UTF
   85 #  define UTF_ENABLED(flag) flag
   86 #else
   87 #  ifdef SUPPORT_UTF
   88 #    define UTF_ENABLED(flag) TRUE
   89 #  else
   90 #    define UTF_ENABLED(flag) FALSE
   91 #  endif
   92 #endif
   93 
   94 /* We do not support both EBCDIC and UTF-8/16 at the same time. The "configure"
   95 script prevents both being selected, but not everybody uses "configure". */
   96 
   97 #if defined EBCDIC && defined SUPPORT_UTF
   98 #error The use of both EBCDIC and SUPPORT_UTF8/16 is not supported.
   99 #endif
  100 
  101 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
  102 inline, and there are *still* stupid compilers about that don't like indented
  103 pre-processor statements, or at least there were when I first wrote this. After
  104 all, it had only been about 10 years then...
  105 
  106 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
  107 be absolutely sure we get our version. */
  108 
  109 #undef DPRINTF
  110 #ifdef PCRE_DEBUG
  111 #define DPRINTF(p) printf p
  112 #else
  113 #define DPRINTF(p) /* Nothing */
  114 #endif
  115 
  116 
  117 /* Standard C headers plus the external interface definition. The only time
  118 setjmp and stdarg are used is when NO_RECURSE is set. */
  119 
  120 #include <ctype.h>
  121 #include <limits.h>
  122 #include <stddef.h>
  123 #include <stdio.h>
  124 #include <stdlib.h>
  125 #include <string.h>
  126 
  127 /* When compiling a DLL for Windows, the exported symbols have to be declared
  128 using some MS magic. I found some useful information on this web page:
  129 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
  130 information there, using __declspec(dllexport) without "extern" we have a
  131 definition; with "extern" we have a declaration. The settings here override the
  132 setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
  133 which is all that is needed for applications (they just import the symbols). We
  134 use:
  135 
  136   PCRE_EXP_DECL       for declarations
  137   PCRE_EXP_DEFN       for definitions of exported functions
  138   PCRE_EXP_DATA_DEFN  for definitions of exported variables
  139 
  140 The reason for the two DEFN macros is that in non-Windows environments, one
  141 does not want to have "extern" before variable definitions because it leads to
  142 compiler warnings. So we distinguish between functions and variables. In
  143 Windows, the two should always be the same.
  144 
  145 The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
  146 which is an application, but needs to import this file in order to "peek" at
  147 internals, can #include pcre.h first to get an application's-eye view.
  148 
  149 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
  150 special-purpose environments) might want to stick other stuff in front of
  151 exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
  152 PCRE_EXP_DATA_DEFN only if they are not already set. */
  153 
  154 #ifndef PCRE_EXP_DECL
  155 #  ifdef _WIN32
  156 #    ifndef PCRE_STATIC
  157 #      define PCRE_EXP_DECL       extern __declspec(dllexport)
  158 #      define PCRE_EXP_DEFN       __declspec(dllexport)
  159 #      define PCRE_EXP_DATA_DEFN  __declspec(dllexport)
  160 #    else
  161 #      define PCRE_EXP_DECL       extern
  162 #      define PCRE_EXP_DEFN
  163 #      define PCRE_EXP_DATA_DEFN
  164 #    endif
  165 #  else
  166 #    ifdef __cplusplus
  167 #      define PCRE_EXP_DECL       extern "C"
  168 #    else
  169 #      define PCRE_EXP_DECL       extern
  170 #    endif
  171 #    ifndef PCRE_EXP_DEFN
  172 #      define PCRE_EXP_DEFN       PCRE_EXP_DECL
  173 #    endif
  174 #    ifndef PCRE_EXP_DATA_DEFN
  175 #      define PCRE_EXP_DATA_DEFN
  176 #    endif
  177 #  endif
  178 #endif
  179 
  180 /* When compiling with the MSVC compiler, it is sometimes necessary to include
  181 a "calling convention" before exported function names. (This is secondhand
  182 information; I know nothing about MSVC myself). For example, something like
  183 
  184   void __cdecl function(....)
  185 
  186 might be needed. In order so make this easy, all the exported functions have
  187 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
  188 set, we ensure here that it has no effect. */
  189 
  190 #ifndef PCRE_CALL_CONVENTION
  191 #define PCRE_CALL_CONVENTION
  192 #endif
  193 
  194 /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
  195 cannot determine these outside the compilation (e.g. by running a program as
  196 part of "configure") because PCRE is often cross-compiled for use on other
  197 systems. Instead we make use of the maximum sizes that are available at
  198 preprocessor time in standard C environments. */
  199 
  200 typedef unsigned char pcre_uint8;
  201 
  202 #if USHRT_MAX == 65535
  203   typedef unsigned short pcre_uint16;
  204   typedef short pcre_int16;
  205 #elif UINT_MAX == 65535
  206   typedef unsigned int pcre_uint16;
  207   typedef int pcre_int16;
  208 #else
  209   #error Cannot determine a type for 16-bit unsigned integers
  210 #endif
  211 
  212 #if UINT_MAX == 4294967295
  213   typedef unsigned int pcre_uint32;
  214   typedef int pcre_int32;
  215 #elif ULONG_MAX == 4294967295
  216   typedef unsigned long int pcre_uint32;
  217   typedef long int pcre_int32;
  218 #else
  219   #error Cannot determine a type for 32-bit unsigned integers
  220 #endif
  221 
  222 /* When checking for integer overflow in pcre_compile(), we need to handle
  223 large integers. If a 64-bit integer type is available, we can use that.
  224 Otherwise we have to cast to double, which of course requires floating point
  225 arithmetic. Handle this by defining a macro for the appropriate type. If
  226 stdint.h is available, include it; it may define INT64_MAX. Systems that do not
  227 have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
  228 by "configure". */
  229 
  230 #if HAVE_STDINT_H
  231 #include <stdint.h>
  232 #elif HAVE_INTTYPES_H
  233 #include <inttypes.h>
  234 #endif
  235 
  236 #if defined INT64_MAX || defined int64_t
  237 #define INT64_OR_DOUBLE int64_t
  238 #else
  239 #define INT64_OR_DOUBLE double
  240 #endif
  241 
  242 /* All character handling must be done as unsigned characters. Otherwise there
  243 are problems with top-bit-set characters and functions such as isspace().
  244 However, we leave the interface to the outside world as char * or short *,
  245 because that should make things easier for callers. This character type is
  246 called pcre_uchar.
  247 
  248 The IN_UCHARS macro multiply its argument with the byte size of the current
  249 pcre_uchar type. Useful for memcpy and such operations, whose require the
  250 byte size of their input/output buffers.
  251 
  252 The MAX_255 macro checks whether its pcre_uchar input is less than 256.
  253 
  254 The TABLE_GET macro is designed for accessing elements of tables whose contain
  255 exactly 256 items. When the character is able to contain more than 256
  256 items, some check is needed before accessing these tables.
  257 */
  258 
  259 #ifdef COMPILE_PCRE8
  260 
  261 typedef unsigned char pcre_uchar;
  262 #define IN_UCHARS(x) (x)
  263 #define MAX_255(c) 1
  264 #define TABLE_GET(c, table, default) ((table)[c])
  265 
  266 #else
  267 
  268 #ifdef COMPILE_PCRE16
  269 #if USHRT_MAX != 65535
  270 /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
  271 pcre.h(.in) and disable (comment out) this message. */
  272 #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
  273 #endif
  274 
  275 typedef pcre_uint16 pcre_uchar;
  276 #define IN_UCHARS(x) ((x) << 1)
  277 #define MAX_255(c) ((c) <= 255u)
  278 #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
  279 
  280 #else
  281 #error Unsupported compiling mode
  282 #endif /* COMPILE_PCRE16 */
  283 
  284 #endif /* COMPILE_PCRE8 */
  285 
  286 /* This is an unsigned int value that no character can ever have. UTF-8
  287 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
  288 0x0010ffff). */
  289 
  290 #define NOTACHAR 0xffffffff
  291 
  292 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
  293 "any" and "anycrlf" at present). The following macros are used to package up
  294 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
  295 modules to indicate in which datablock the parameters exist, and what the
  296 start/end of string field names are. */
  297 
  298 #define NLTYPE_FIXED    0     /* Newline is a fixed length string */
  299 #define NLTYPE_ANY      1     /* Newline is any Unicode line ending */
  300 #define NLTYPE_ANYCRLF  2     /* Newline is CR, LF, or CRLF */
  301 
  302 /* This macro checks for a newline at the given position */
  303 
  304 #define IS_NEWLINE(p) \
  305   ((NLBLOCK->nltype != NLTYPE_FIXED)? \
  306     ((p) < NLBLOCK->PSEND && \
  307      PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
  308        &(NLBLOCK->nllen), utf)) \
  309     : \
  310     ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
  311      (p)[0] == NLBLOCK->nl[0] && \
  312      (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
  313     ) \
  314   )
  315 
  316 /* This macro checks for a newline immediately preceding the given position */
  317 
  318 #define WAS_NEWLINE(p) \
  319   ((NLBLOCK->nltype != NLTYPE_FIXED)? \
  320     ((p) > NLBLOCK->PSSTART && \
  321      PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
  322        &(NLBLOCK->nllen), utf)) \
  323     : \
  324     ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
  325      (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
  326      (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
  327     ) \
  328   )
  329 
  330 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
  331 with a custom type. This makes it possible, for example, to allow pcre_exec()
  332 to process subject strings that are discontinuous by using a smart pointer
  333 class. It must always be possible to inspect all of the subject string in
  334 pcre_exec() because of the way it backtracks. Two macros are required in the
  335 normal case, for sign-unspecified and unsigned char pointers. The former is
  336 used for the external interface and appears in pcre.h, which is why its name
  337 must begin with PCRE_. */
  338 
  339 #ifdef CUSTOM_SUBJECT_PTR
  340 #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
  341 #else
  342 #define PCRE_PUCHAR const pcre_uchar *
  343 #endif
  344 
  345 /* Include the public PCRE header and the definitions of UCP character property
  346 values. */
  347 
  348 #include "pcre.h"
  349 #include "ucp.h"
  350 
  351 /* When compiling for use with the Virtual Pascal compiler, these functions
  352 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
  353 option on the command line. */
  354 
  355 #ifdef VPCOMPAT
  356 #define strlen(s)        _strlen(s)
  357 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
  358 #define memcmp(s,c,n)    _memcmp(s,c,n)
  359 #define memcpy(d,s,n)    _memcpy(d,s,n)
  360 #define memmove(d,s,n)   _memmove(d,s,n)
  361 #define memset(s,c,n)    _memset(s,c,n)
  362 #else  /* VPCOMPAT */
  363 
  364 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
  365 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
  366 is set. Otherwise, include an emulating function for those systems that have
  367 neither (there some non-Unix environments where this is the case). */
  368 
  369 #ifndef HAVE_MEMMOVE
  370 #undef  memmove        /* some systems may have a macro */
  371 #ifdef HAVE_BCOPY
  372 #define memmove(a, b, c) bcopy(b, a, c)
  373 #else  /* HAVE_BCOPY */
  374 static void *
  375 pcre_memmove(void *d, const void *s, size_t n)
  376 {
  377 size_t i;
  378 unsigned char *dest = (unsigned char *)d;
  379 const unsigned char *src = (const unsigned char *)s;
  380 if (dest > src)
  381   {
  382   dest += n;
  383   src += n;
  384   for (i = 0; i < n; ++i) *(--dest) = *(--src);
  385   return (void *)dest;
  386   }
  387 else
  388   {
  389   for (i = 0; i < n; ++i) *dest++ = *src++;
  390   return (void *)(dest - n);
  391   }
  392 }
  393 #define memmove(a, b, c) pcre_memmove(a, b, c)
  394 #endif   /* not HAVE_BCOPY */
  395 #endif   /* not HAVE_MEMMOVE */
  396 #endif   /* not VPCOMPAT */
  397 
  398 
  399 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
  400 in big-endian order) by default. These are used, for example, to link from the
  401 start of a subpattern to its alternatives and its end. The use of 2 bytes per
  402 offset limits the size of the compiled regex to around 64K, which is big enough
  403 for almost everybody. However, I received a request for an even bigger limit.
  404 For this reason, and also to make the code easier to maintain, the storing and
  405 loading of offsets from the byte string is now handled by the macros that are
  406 defined here.
  407 
  408 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
  409 the config.h file, but can be overridden by using -D on the command line. This
  410 is automated on Unix systems via the "configure" command. */
  411 
  412 #ifdef COMPILE_PCRE8
  413 
  414 #if LINK_SIZE == 2
  415 
  416 #define PUT(a,n,d)   \
  417   (a[n] = (d) >> 8), \
  418   (a[(n)+1] = (d) & 255)
  419 
  420 #define GET(a,n) \
  421   (((a)[n] << 8) | (a)[(n)+1])
  422 
  423 #define MAX_PATTERN_SIZE (1 << 16)
  424 
  425 
  426 #elif LINK_SIZE == 3
  427 
  428 #define PUT(a,n,d)       \
  429   (a[n] = (d) >> 16),    \
  430   (a[(n)+1] = (d) >> 8), \
  431   (a[(n)+2] = (d) & 255)
  432 
  433 #define GET(a,n) \
  434   (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
  435 
  436 #define MAX_PATTERN_SIZE (1 << 24)
  437 
  438 
  439 #elif LINK_SIZE == 4
  440 
  441 #define PUT(a,n,d)        \
  442   (a[n] = (d) >> 24),     \
  443   (a[(n)+1] = (d) >> 16), \
  444   (a[(n)+2] = (d) >> 8),  \
  445   (a[(n)+3] = (d) & 255)
  446 
  447 #define GET(a,n) \
  448   (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
  449 
  450 /* Keep it positive */
  451 #define MAX_PATTERN_SIZE (1 << 30)
  452 
  453 #else
  454 #error LINK_SIZE must be either 2, 3, or 4
  455 #endif
  456 
  457 #else /* COMPILE_PCRE8 */
  458 
  459 #ifdef COMPILE_PCRE16
  460 
  461 #if LINK_SIZE == 2
  462 
  463 #undef LINK_SIZE
  464 #define LINK_SIZE 1
  465 
  466 #define PUT(a,n,d)   \
  467   (a[n] = (d))
  468 
  469 #define GET(a,n) \
  470   (a[n])
  471 
  472 #define MAX_PATTERN_SIZE (1 << 16)
  473 
  474 #elif LINK_SIZE == 3 || LINK_SIZE == 4
  475 
  476 #undef LINK_SIZE
  477 #define LINK_SIZE 2
  478 
  479 #define PUT(a,n,d)   \
  480   (a[n] = (d) >> 16), \
  481   (a[(n)+1] = (d) & 65535)
  482 
  483 #define GET(a,n) \
  484   (((a)[n] << 16) | (a)[(n)+1])
  485 
  486 /* Keep it positive */
  487 #define MAX_PATTERN_SIZE (1 << 30)
  488 
  489 #else
  490 #error LINK_SIZE must be either 2, 3, or 4
  491 #endif
  492 
  493 #else
  494 #error Unsupported compiling mode
  495 #endif /* COMPILE_PCRE16 */
  496 
  497 #endif /* COMPILE_PCRE8 */
  498 
  499 /* Convenience macro defined in terms of the others */
  500 
  501 #define PUTINC(a,n,d)   PUT(a,n,d), a += LINK_SIZE
  502 
  503 
  504 /* PCRE uses some other 2-byte quantities that do not change when the size of
  505 offsets changes. There are used for repeat counts and for other things such as
  506 capturing parenthesis numbers in back references. */
  507 
  508 #ifdef COMPILE_PCRE8
  509 
  510 #define IMM2_SIZE 2
  511 #define IMMPTR_SIZE sizeof(void *)
  512 
  513 #define PUT2(a,n,d)   \
  514   a[n] = (d) >> 8; \
  515   a[(n)+1] = (d) & 255
  516 
  517 #define GET2(a,n) \
  518   (((a)[n] << 8) | (a)[(n)+1])
  519 
  520 #else /* COMPILE_PCRE8 */
  521 
  522 #ifdef COMPILE_PCRE16
  523 
  524 #define IMM2_SIZE 1
  525 #define IMMPTR_SIZE (sizeof(void *) / 2)
  526 
  527 #define PUT2(a,n,d)   \
  528    a[n] = d
  529 
  530 #define GET2(a,n) \
  531    a[n]
  532 
  533 #else
  534 #error Unsupported compiling mode
  535 #endif /* COMPILE_PCRE16 */
  536 
  537 #endif /* COMPILE_PCRE8 */
  538 
  539 #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
  540 
  541 /* When UTF encoding is being used, a character is no longer just a single
  542 character. The macros for character handling generate simple sequences when
  543 used in character-mode, and more complicated ones for UTF characters.
  544 GETCHARLENTEST and other macros are not used when UTF is not supported,
  545 so they are not defined. To make sure they can never even appear when
  546 UTF support is omitted, we don't even define them. */
  547 
  548 #ifndef SUPPORT_UTF
  549 
  550 /* #define MAX_VALUE_FOR_SINGLE_CHAR */
  551 /* #define HAS_EXTRALEN(c) */
  552 /* #define GET_EXTRALEN(c) */
  553 /* #define NOT_FIRSTCHAR(c) */
  554 #define GETCHAR(c, eptr) c = *eptr;
  555 #define GETCHARTEST(c, eptr) c = *eptr;
  556 #define GETCHARINC(c, eptr) c = *eptr++;
  557 #define GETCHARINCTEST(c, eptr) c = *eptr++;
  558 #define GETCHARLEN(c, eptr, len) c = *eptr;
  559 /* #define GETCHARLENTEST(c, eptr, len) */
  560 /* #define BACKCHAR(eptr) */
  561 /* #define FORWARDCHAR(eptr) */
  562 /* #define ACROSSCHAR(condition, eptr, action) */
  563 
  564 #else   /* SUPPORT_UTF */
  565 
  566 #ifdef COMPILE_PCRE8
  567 
  568 /* These macros were originally written in the form of loops that used data
  569 from the tables whose names start with PRIV(utf8_table). They were rewritten by
  570 a user so as not to use loops, because in some environments this gives a
  571 significant performance advantage, and it seems never to do any harm. */
  572 
  573 /* Tells the biggest code point which can be encoded as a single character. */
  574 
  575 #define MAX_VALUE_FOR_SINGLE_CHAR 127
  576 
  577 /* Tests whether the code point needs extra characters to decode. */
  578 
  579 #define HAS_EXTRALEN(c) ((c) >= 0xc0)
  580 
  581 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
  582 Otherwise it has an undefined behaviour. */
  583 
  584 #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
  585 
  586 /* Returns TRUE, if the given character is not the first character
  587 of a UTF sequence. */
  588 
  589 #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
  590 
  591 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
  592 advancing the pointer. */
  593 
  594 #define GETUTF8(c, eptr) \
  595     { \
  596     if ((c & 0x20) == 0) \
  597       c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
  598     else if ((c & 0x10) == 0) \
  599       c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
  600     else if ((c & 0x08) == 0) \
  601       c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
  602       ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
  603     else if ((c & 0x04) == 0) \
  604       c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
  605           ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
  606           (eptr[4] & 0x3f); \
  607     else \
  608       c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
  609           ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
  610           ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
  611     }
  612 
  613 /* Get the next UTF-8 character, not advancing the pointer. This is called when
  614 we know we are in UTF-8 mode. */
  615 
  616 #define GETCHAR(c, eptr) \
  617   c = *eptr; \
  618   if (c >= 0xc0) GETUTF8(c, eptr);
  619 
  620 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
  621 pointer. */
  622 
  623 #define GETCHARTEST(c, eptr) \
  624   c = *eptr; \
  625   if (utf && c >= 0xc0) GETUTF8(c, eptr);
  626 
  627 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
  628 the pointer. */
  629 
  630 #define GETUTF8INC(c, eptr) \
  631     { \
  632     if ((c & 0x20) == 0) \
  633       c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
  634     else if ((c & 0x10) == 0) \
  635       { \
  636       c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
  637       eptr += 2; \
  638       } \
  639     else if ((c & 0x08) == 0) \
  640       { \
  641       c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
  642           ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
  643       eptr += 3; \
  644       } \
  645     else if ((c & 0x04) == 0) \
  646       { \
  647       c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
  648           ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
  649           (eptr[3] & 0x3f); \
  650       eptr += 4; \
  651       } \
  652     else \
  653       { \
  654       c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
  655           ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
  656           ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
  657       eptr += 5; \
  658       } \
  659     }
  660 
  661 /* Get the next UTF-8 character, advancing the pointer. This is called when we
  662 know we are in UTF-8 mode. */
  663 
  664 #define GETCHARINC(c, eptr) \
  665   c = *eptr++; \
  666   if (c >= 0xc0) GETUTF8INC(c, eptr);
  667 
  668 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
  669 This is called when we don't know if we are in UTF-8 mode. */
  670 
  671 #define GETCHARINCTEST(c, eptr) \
  672   c = *eptr++; \
  673   if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
  674 
  675 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
  676 advancing the pointer, incrementing the length. */
  677 
  678 #define GETUTF8LEN(c, eptr, len) \
  679     { \
  680     if ((c & 0x20) == 0) \
  681       { \
  682       c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
  683       len++; \
  684       } \
  685     else if ((c & 0x10)  == 0) \
  686       { \
  687       c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
  688       len += 2; \
  689       } \
  690     else if ((c & 0x08)  == 0) \
  691       {\
  692       c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
  693           ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
  694       len += 3; \
  695       } \
  696     else if ((c & 0x04)  == 0) \
  697       { \
  698       c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
  699           ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
  700           (eptr[4] & 0x3f); \
  701       len += 4; \
  702       } \
  703     else \
  704       {\
  705       c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
  706           ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
  707           ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
  708       len += 5; \
  709       } \
  710     }
  711 
  712 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
  713 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
  714 
  715 #define GETCHARLEN(c, eptr, len) \
  716   c = *eptr; \
  717   if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
  718 
  719 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
  720 pointer, incrementing length if there are extra bytes. This is called when we
  721 do not know if we are in UTF-8 mode. */
  722 
  723 #define GETCHARLENTEST(c, eptr, len) \
  724   c = *eptr; \
  725   if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
  726 
  727 /* If the pointer is not at the start of a character, move it back until
  728 it is. This is called only in UTF-8 mode - we don't put a test within the macro
  729 because almost all calls are already within a block of UTF-8 only code. */
  730 
  731 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
  732 
  733 /* Same as above, just in the other direction. */
  734 #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
  735 
  736 /* Same as above, but it allows a fully customizable form. */
  737 #define ACROSSCHAR(condition, eptr, action) \
  738   while((condition) && ((eptr) & 0xc0) == 0x80) action
  739 
  740 #else /* COMPILE_PCRE8 */
  741 
  742 #ifdef COMPILE_PCRE16
  743 
  744 /* Tells the biggest code point which can be encoded as a single character. */
  745 
  746 #define MAX_VALUE_FOR_SINGLE_CHAR 65535
  747 
  748 /* Tests whether the code point needs extra characters to decode. */
  749 
  750 #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
  751 
  752 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
  753 Otherwise it has an undefined behaviour. */
  754 
  755 #define GET_EXTRALEN(c) 1
  756 
  757 /* Returns TRUE, if the given character is not the first character
  758 of a UTF sequence. */
  759 
  760 #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
  761 
  762 /* Base macro to pick up the low surrogate of a UTF-16 character, not
  763 advancing the pointer. */
  764 
  765 #define GETUTF16(c, eptr) \
  766    { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
  767 
  768 /* Get the next UTF-16 character, not advancing the pointer. This is called when
  769 we know we are in UTF-16 mode. */
  770 
  771 #define GETCHAR(c, eptr) \
  772   c = *eptr; \
  773   if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
  774 
  775 /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
  776 pointer. */
  777 
  778 #define GETCHARTEST(c, eptr) \
  779   c = *eptr; \
  780   if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
  781 
  782 /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
  783 the pointer. */
  784 
  785 #define GETUTF16INC(c, eptr) \
  786    { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
  787 
  788 /* Get the next UTF-16 character, advancing the pointer. This is called when we
  789 know we are in UTF-16 mode. */
  790 
  791 #define GETCHARINC(c, eptr) \
  792   c = *eptr++; \
  793   if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
  794 
  795 /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
  796 This is called when we don't know if we are in UTF-16 mode. */
  797 
  798 #define GETCHARINCTEST(c, eptr) \
  799   c = *eptr++; \
  800   if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
  801 
  802 /* Base macro to pick up the low surrogate of a UTF-16 character, not
  803 advancing the pointer, incrementing the length. */
  804 
  805 #define GETUTF16LEN(c, eptr, len) \
  806    { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
  807 
  808 /* Get the next UTF-16 character, not advancing the pointer, incrementing
  809 length if there is a low surrogate. This is called when we know we are in
  810 UTF-16 mode. */
  811 
  812 #define GETCHARLEN(c, eptr, len) \
  813   c = *eptr; \
  814   if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
  815 
  816 /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
  817 pointer, incrementing length if there is a low surrogate. This is called when
  818 we do not know if we are in UTF-16 mode. */
  819 
  820 #define GETCHARLENTEST(c, eptr, len) \
  821   c = *eptr; \
  822   if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
  823 
  824 /* If the pointer is not at the start of a character, move it back until
  825 it is. This is called only in UTF-16 mode - we don't put a test within the
  826 macro because almost all calls are already within a block of UTF-16 only
  827 code. */
  828 
  829 #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
  830 
  831 /* Same as above, just in the other direction. */
  832 #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
  833 
  834 /* Same as above, but it allows a fully customizable form. */
  835 #define ACROSSCHAR(condition, eptr, action) \
  836   if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
  837 
  838 #endif
  839 
  840 #endif /* COMPILE_PCRE8 */
  841 
  842 #endif  /* SUPPORT_UTF */
  843 
  844 
  845 /* In case there is no definition of offsetof() provided - though any proper
  846 Standard C system should have one. */
  847 
  848 #ifndef offsetof
  849 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
  850 #endif
  851 
  852 
  853 /* Private flags containing information about the compiled regex. They used to
  854 live at the top end of the options word, but that got almost full, so now they
  855 are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
  856 the restrictions on partial matching have been lifted. It remains for backwards
  857 compatibility. */
  858 
  859 #ifdef COMPILE_PCRE8
  860 #define PCRE_MODE          0x0001  /* compiled in 8 bit mode */
  861 #endif
  862 #ifdef COMPILE_PCRE16
  863 #define PCRE_MODE          0x0002  /* compiled in 16 bit mode */
  864 #endif
  865 #define PCRE_FIRSTSET      0x0010  /* first_char is set */
  866 #define PCRE_FCH_CASELESS  0x0020  /* caseless first char */
  867 #define PCRE_REQCHSET      0x0040  /* req_byte is set */
  868 #define PCRE_RCH_CASELESS  0x0080  /* caseless requested char */
  869 #define PCRE_STARTLINE     0x0100  /* start after \n for multiline */
  870 #define PCRE_NOPARTIAL     0x0200  /* can't use partial with this regex */
  871 #define PCRE_JCHANGED      0x0400  /* j option used in regex */
  872 #define PCRE_HASCRORLF     0x0800  /* explicit \r or \n in pattern */
  873 #define PCRE_HASTHEN       0x1000  /* pattern contains (*THEN) */
  874 
  875 /* Flags for the "extra" block produced by pcre_study(). */
  876 
  877 #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
  878 #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
  879 
  880 /* Masks for identifying the public options that are permitted at compile
  881 time, run time, or study time, respectively. */
  882 
  883 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
  884                            PCRE_NEWLINE_ANYCRLF)
  885 
  886 #define PUBLIC_COMPILE_OPTIONS \
  887   (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
  888    PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
  889    PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
  890    PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
  891    PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
  892 
  893 #define PUBLIC_EXEC_OPTIONS \
  894   (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
  895    PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
  896    PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
  897 
  898 #define PUBLIC_DFA_EXEC_OPTIONS \
  899   (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
  900    PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
  901    PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
  902    PCRE_NO_START_OPTIMIZE)
  903 
  904 #define PUBLIC_STUDY_OPTIONS \
  905    PCRE_STUDY_JIT_COMPILE
  906 
  907 /* Magic number to provide a small check against being handed junk. */
  908 
  909 #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
  910 
  911 /* This variable is used to detect a loaded regular expression
  912 in different endianness. */
  913 
  914 #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
  915 
  916 /* Negative values for the firstchar and reqchar variables */
  917 
  918 #define REQ_UNSET (-2)
  919 #define REQ_NONE  (-1)
  920 
  921 /* The maximum remaining length of subject we are prepared to search for a
  922 req_byte match. */
  923 
  924 #define REQ_BYTE_MAX 1000
  925 
  926 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
  927 environments where these macros are defined elsewhere. Unfortunately, there
  928 is no way to do the same for the typedef. */
  929 
  930 typedef int BOOL;
  931 
  932 #ifndef FALSE
  933 #define FALSE   0
  934 #define TRUE    1
  935 #endif
  936 
  937 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
  938 character constants like '*' because the compiler would emit their EBCDIC code,
  939 which is different from their ASCII/UTF-8 code. Instead we define macros for
  940 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
  941 is enabled. When UTF-8 support is not enabled, the definitions use character
  942 literals. Both character and string versions of each character are needed, and
  943 there are some longer strings as well.
  944 
  945 This means that, on EBCDIC platforms, the PCRE library can handle either
  946 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
  947 would need different lookups depending on whether PCRE_UTF8 was set or not.
  948 This would make it impossible to use characters in switch/case statements,
  949 which would reduce performance. For a theoretical use (which nobody has asked
  950 for) in a minority area (EBCDIC platforms), this is not sensible. Any
  951 application that did need both could compile two versions of the library, using
  952 macros to give the functions distinct names. */
  953 
  954 #ifndef SUPPORT_UTF
  955 
  956 /* UTF-8 support is not enabled; use the platform-dependent character literals
  957 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
  958 
  959 #define CHAR_HT                     '\t'
  960 #define CHAR_VT                     '\v'
  961 #define CHAR_FF                     '\f'
  962 #define CHAR_CR                     '\r'
  963 #define CHAR_NL                     '\n'
  964 #define CHAR_BS                     '\b'
  965 #define CHAR_BEL                    '\a'
  966 #ifdef EBCDIC
  967 #define CHAR_ESC                    '\047'
  968 #define CHAR_DEL                    '\007'
  969 #else
  970 #define CHAR_ESC                    '\033'
  971 #define CHAR_DEL                    '\177'
  972 #endif
  973 
  974 #define CHAR_SPACE                  ' '
  975 #define CHAR_EXCLAMATION_MARK       '!'
  976 #define CHAR_QUOTATION_MARK         '"'
  977 #define CHAR_NUMBER_SIGN            '#'
  978 #define CHAR_DOLLAR_SIGN            '$'
  979 #define CHAR_PERCENT_SIGN           '%'
  980 #define CHAR_AMPERSAND              '&'
  981 #define CHAR_APOSTROPHE             '\''
  982 #define CHAR_LEFT_PARENTHESIS       '('
  983 #define CHAR_RIGHT_PARENTHESIS      ')'
  984 #define CHAR_ASTERISK               '*'
  985 #define CHAR_PLUS                   '+'
  986 #define CHAR_COMMA                  ','
  987 #define CHAR_MINUS                  '-'
  988 #define CHAR_DOT                    '.'
  989 #define CHAR_SLASH                  '/'
  990 #define CHAR_0                      '0'
  991 #define CHAR_1                      '1'
  992 #define CHAR_2                      '2'
  993 #define CHAR_3                      '3'
  994 #define CHAR_4                      '4'
  995 #define CHAR_5                      '5'
  996 #define CHAR_6                      '6'
  997 #define CHAR_7                      '7'
  998 #define CHAR_8                      '8'
  999 #define CHAR_9                      '9'
 1000 #define CHAR_COLON                  ':'
 1001 #define CHAR_SEMICOLON              ';'
 1002 #define CHAR_LESS_THAN_SIGN         '<'
 1003 #define CHAR_EQUALS_SIGN            '='
 1004 #define CHAR_GREATER_THAN_SIGN      '>'
 1005 #define CHAR_QUESTION_MARK          '?'
 1006 #define CHAR_COMMERCIAL_AT          '@'
 1007 #define CHAR_A                      'A'
 1008 #define CHAR_B                      'B'
 1009 #define CHAR_C                      'C'
 1010 #define CHAR_D                      'D'
 1011 #define CHAR_E                      'E'
 1012 #define CHAR_F                      'F'
 1013 #define CHAR_G                      'G'
 1014 #define CHAR_H                      'H'
 1015 #define CHAR_I                      'I'
 1016 #define CHAR_J                      'J'
 1017 #define CHAR_K                      'K'
 1018 #define CHAR_L                      'L'
 1019 #define CHAR_M                      'M'
 1020 #define CHAR_N                      'N'
 1021 #define CHAR_O                      'O'
 1022 #define CHAR_P                      'P'
 1023 #define CHAR_Q                      'Q'
 1024 #define CHAR_R                      'R'
 1025 #define CHAR_S                      'S'
 1026 #define CHAR_T                      'T'
 1027 #define CHAR_U                      'U'
 1028 #define CHAR_V                      'V'
 1029 #define CHAR_W                      'W'
 1030 #define CHAR_X                      'X'
 1031 #define CHAR_Y                      'Y'
 1032 #define CHAR_Z                      'Z'
 1033 #define CHAR_LEFT_SQUARE_BRACKET    '['
 1034 #define CHAR_BACKSLASH              '\\'
 1035 #define CHAR_RIGHT_SQUARE_BRACKET   ']'
 1036 #define CHAR_CIRCUMFLEX_ACCENT      '^'
 1037 #define CHAR_UNDERSCORE             '_'
 1038 #define CHAR_GRAVE_ACCENT           '`'
 1039 #define CHAR_a                      'a'
 1040 #define CHAR_b                      'b'
 1041 #define CHAR_c                      'c'
 1042 #define CHAR_d                      'd'
 1043 #define CHAR_e                      'e'
 1044 #define CHAR_f                      'f'
 1045 #define CHAR_g                      'g'
 1046 #define CHAR_h                      'h'
 1047 #define CHAR_i                      'i'
 1048 #define CHAR_j                      'j'
 1049 #define CHAR_k                      'k'
 1050 #define CHAR_l                      'l'
 1051 #define CHAR_m                      'm'
 1052 #define CHAR_n                      'n'
 1053 #define CHAR_o                      'o'
 1054 #define CHAR_p                      'p'
 1055 #define CHAR_q                      'q'
 1056 #define CHAR_r                      'r'
 1057 #define CHAR_s                      's'
 1058 #define CHAR_t                      't'
 1059 #define CHAR_u                      'u'
 1060 #define CHAR_v                      'v'
 1061 #define CHAR_w                      'w'
 1062 #define CHAR_x                      'x'
 1063 #define CHAR_y                      'y'
 1064 #define CHAR_z                      'z'
 1065 #define CHAR_LEFT_CURLY_BRACKET     '{'
 1066 #define CHAR_VERTICAL_LINE          '|'
 1067 #define CHAR_RIGHT_CURLY_BRACKET    '}'
 1068 #define CHAR_TILDE                  '~'
 1069 
 1070 #define STR_HT                      "\t"
 1071 #define STR_VT                      "\v"
 1072 #define STR_FF                      "\f"
 1073 #define STR_CR                      "\r"
 1074 #define STR_NL                      "\n"
 1075 #define STR_BS                      "\b"
 1076 #define STR_BEL                     "\a"
 1077 #ifdef EBCDIC
 1078 #define STR_ESC                     "\047"
 1079 #define STR_DEL                     "\007"
 1080 #else
 1081 #define STR_ESC                     "\033"
 1082 #define STR_DEL                     "\177"
 1083 #endif
 1084 
 1085 #define STR_SPACE                   " "
 1086 #define STR_EXCLAMATION_MARK        "!"
 1087 #define STR_QUOTATION_MARK          "\""
 1088 #define STR_NUMBER_SIGN             "#"
 1089 #define STR_DOLLAR_SIGN             "$"
 1090 #define STR_PERCENT_SIGN            "%"
 1091 #define STR_AMPERSAND               "&"
 1092 #define STR_APOSTROPHE              "'"
 1093 #define STR_LEFT_PARENTHESIS        "("
 1094 #define STR_RIGHT_PARENTHESIS       ")"
 1095 #define STR_ASTERISK                "*"
 1096 #define STR_PLUS                    "+"
 1097 #define STR_COMMA                   ","
 1098 #define STR_MINUS                   "-"
 1099 #define STR_DOT                     "."
 1100 #define STR_SLASH                   "/"
 1101 #define STR_0                       "0"
 1102 #define STR_1                       "1"
 1103 #define STR_2                       "2"
 1104 #define STR_3                       "3"
 1105 #define STR_4                       "4"
 1106 #define STR_5                       "5"
 1107 #define STR_6                       "6"
 1108 #define STR_7                       "7"
 1109 #define STR_8                       "8"
 1110 #define STR_9                       "9"
 1111 #define STR_COLON                   ":"
 1112 #define STR_SEMICOLON               ";"
 1113 #define STR_LESS_THAN_SIGN          "<"
 1114 #define STR_EQUALS_SIGN             "="
 1115 #define STR_GREATER_THAN_SIGN       ">"
 1116 #define STR_QUESTION_MARK           "?"
 1117 #define STR_COMMERCIAL_AT           "@"
 1118 #define STR_A                       "A"
 1119 #define STR_B                       "B"
 1120 #define STR_C                       "C"
 1121 #define STR_D                       "D"
 1122 #define STR_E                       "E"
 1123 #define STR_F                       "F"
 1124 #define STR_G                       "G"
 1125 #define STR_H                       "H"
 1126 #define STR_I                       "I"
 1127 #define STR_J                       "J"
 1128 #define STR_K                       "K"
 1129 #define STR_L                       "L"
 1130 #define STR_M                       "M"
 1131 #define STR_N                       "N"
 1132 #define STR_O                       "O"
 1133 #define STR_P                       "P"
 1134 #define STR_Q                       "Q"
 1135 #define STR_R                       "R"
 1136 #define STR_S                       "S"
 1137 #define STR_T                       "T"
 1138 #define STR_U                       "U"
 1139 #define STR_V                       "V"
 1140 #define STR_W                       "W"
 1141 #define STR_X                       "X"
 1142 #define STR_Y                       "Y"
 1143 #define STR_Z                       "Z"
 1144 #define STR_LEFT_SQUARE_BRACKET     "["
 1145 #define STR_BACKSLASH               "\\"
 1146 #define STR_RIGHT_SQUARE_BRACKET    "]"
 1147 #define STR_CIRCUMFLEX_ACCENT       "^"
 1148 #define STR_UNDERSCORE              "_"
 1149 #define STR_GRAVE_ACCENT            "`"
 1150 #define STR_a                       "a"
 1151 #define STR_b                       "b"
 1152 #define STR_c                       "c"
 1153 #define STR_d                       "d"
 1154 #define STR_e                       "e"
 1155 #define STR_f                       "f"
 1156 #define STR_g                       "g"
 1157 #define STR_h                       "h"
 1158 #define STR_i                       "i"
 1159 #define STR_j                       "j"
 1160 #define STR_k                       "k"
 1161 #define STR_l                       "l"
 1162 #define STR_m                       "m"
 1163 #define STR_n                       "n"
 1164 #define STR_o                       "o"
 1165 #define STR_p                       "p"
 1166 #define STR_q                       "q"
 1167 #define STR_r                       "r"
 1168 #define STR_s                       "s"
 1169 #define STR_t                       "t"
 1170 #define STR_u                       "u"
 1171 #define STR_v                       "v"
 1172 #define STR_w                       "w"
 1173 #define STR_x                       "x"
 1174 #define STR_y                       "y"
 1175 #define STR_z                       "z"
 1176 #define STR_LEFT_CURLY_BRACKET      "{"
 1177 #define STR_VERTICAL_LINE           "|"
 1178 #define STR_RIGHT_CURLY_BRACKET     "}"
 1179 #define STR_TILDE                   "~"
 1180 
 1181 #define STRING_ACCEPT0              "ACCEPT\0"
 1182 #define STRING_COMMIT0              "COMMIT\0"
 1183 #define STRING_F0                   "F\0"
 1184 #define STRING_FAIL0                "FAIL\0"
 1185 #define STRING_MARK0                "MARK\0"
 1186 #define STRING_PRUNE0               "PRUNE\0"
 1187 #define STRING_SKIP0                "SKIP\0"
 1188 #define STRING_THEN                 "THEN"
 1189 
 1190 #define STRING_alpha0               "alpha\0"
 1191 #define STRING_lower0               "lower\0"
 1192 #define STRING_upper0               "upper\0"
 1193 #define STRING_alnum0               "alnum\0"
 1194 #define STRING_ascii0               "ascii\0"
 1195 #define STRING_blank0               "blank\0"
 1196 #define STRING_cntrl0               "cntrl\0"
 1197 #define STRING_digit0               "digit\0"
 1198 #define STRING_graph0               "graph\0"
 1199 #define STRING_print0               "print\0"
 1200 #define STRING_punct0               "punct\0"
 1201 #define STRING_space0               "space\0"
 1202 #define STRING_word0                "word\0"
 1203 #define STRING_xdigit               "xdigit"
 1204 
 1205 #define STRING_DEFINE               "DEFINE"
 1206 
 1207 #define STRING_CR_RIGHTPAR             "CR)"
 1208 #define STRING_LF_RIGHTPAR             "LF)"
 1209 #define STRING_CRLF_RIGHTPAR           "CRLF)"
 1210 #define STRING_ANY_RIGHTPAR            "ANY)"
 1211 #define STRING_ANYCRLF_RIGHTPAR        "ANYCRLF)"
 1212 #define STRING_BSR_ANYCRLF_RIGHTPAR    "BSR_ANYCRLF)"
 1213 #define STRING_BSR_UNICODE_RIGHTPAR    "BSR_UNICODE)"
 1214 #ifdef COMPILE_PCRE8
 1215 #define STRING_UTF_RIGHTPAR            "UTF8)"
 1216 #endif
 1217 #ifdef COMPILE_PCRE16
 1218 #define STRING_UTF_RIGHTPAR            "UTF16)"
 1219 #endif
 1220 #define STRING_UCP_RIGHTPAR            "UCP)"
 1221 #define STRING_NO_START_OPT_RIGHTPAR   "NO_START_OPT)"
 1222 
 1223 #else  /* SUPPORT_UTF */
 1224 
 1225 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
 1226 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
 1227 only. */
 1228 
 1229 #define CHAR_HT                     '\011'
 1230 #define CHAR_VT                     '\013'
 1231 #define CHAR_FF                     '\014'
 1232 #define CHAR_CR                     '\015'
 1233 #define CHAR_NL                     '\012'
 1234 #define CHAR_BS                     '\010'
 1235 #define CHAR_BEL                    '\007'
 1236 #define CHAR_ESC                    '\033'
 1237 #define CHAR_DEL                    '\177'
 1238 
 1239 #define CHAR_SPACE                  '\040'
 1240 #define CHAR_EXCLAMATION_MARK       '\041'
 1241 #define CHAR_QUOTATION_MARK         '\042'
 1242 #define CHAR_NUMBER_SIGN            '\043'
 1243 #define CHAR_DOLLAR_SIGN            '\044'
 1244 #define CHAR_PERCENT_SIGN           '\045'
 1245 #define CHAR_AMPERSAND              '\046'
 1246 #define CHAR_APOSTROPHE             '\047'
 1247 #define CHAR_LEFT_PARENTHESIS       '\050'
 1248 #define CHAR_RIGHT_PARENTHESIS      '\051'
 1249 #define CHAR_ASTERISK               '\052'
 1250 #define CHAR_PLUS                   '\053'
 1251 #define CHAR_COMMA                  '\054'
 1252 #define CHAR_MINUS                  '\055'
 1253 #define CHAR_DOT                    '\056'
 1254 #define CHAR_SLASH                  '\057'
 1255 #define CHAR_0                      '\060'
 1256 #define CHAR_1                      '\061'
 1257 #define CHAR_2                      '\062'
 1258 #define CHAR_3                      '\063'
 1259 #define CHAR_4                      '\064'
 1260 #define CHAR_5                      '\065'
 1261 #define CHAR_6                      '\066'
 1262 #define CHAR_7                      '\067'
 1263 #define CHAR_8                      '\070'
 1264 #define CHAR_9                      '\071'
 1265 #define CHAR_COLON                  '\072'
 1266 #define CHAR_SEMICOLON              '\073'
 1267 #define CHAR_LESS_THAN_SIGN         '\074'
 1268 #define CHAR_EQUALS_SIGN            '\075'
 1269 #define CHAR_GREATER_THAN_SIGN      '\076'
 1270 #define CHAR_QUESTION_MARK          '\077'
 1271 #define CHAR_COMMERCIAL_AT          '\100'
 1272 #define CHAR_A                      '\101'
 1273 #define CHAR_B                      '\102'
 1274 #define CHAR_C                      '\103'
 1275 #define CHAR_D                      '\104'
 1276 #define CHAR_E                      '\105'
 1277 #define CHAR_F                      '\106'
 1278 #define CHAR_G                      '\107'
 1279 #define CHAR_H                      '\110'
 1280 #define CHAR_I                      '\111'
 1281 #define CHAR_J                      '\112'
 1282 #define CHAR_K                      '\113'
 1283 #define CHAR_L                      '\114'
 1284 #define CHAR_M                      '\115'
 1285 #define CHAR_N                      '\116'
 1286 #define CHAR_O                      '\117'
 1287 #define CHAR_P                      '\120'
 1288 #define CHAR_Q                      '\121'
 1289 #define CHAR_R                      '\122'
 1290 #define CHAR_S                      '\123'
 1291 #define CHAR_T                      '\124'
 1292 #define CHAR_U                      '\125'
 1293 #define CHAR_V                      '\126'
 1294 #define CHAR_W                      '\127'
 1295 #define CHAR_X                      '\130'
 1296 #define CHAR_Y                      '\131'
 1297 #define CHAR_Z                      '\132'
 1298 #define CHAR_LEFT_SQUARE_BRACKET    '\133'
 1299 #define CHAR_BACKSLASH              '\134'
 1300 #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
 1301 #define CHAR_CIRCUMFLEX_ACCENT      '\136'
 1302 #define CHAR_UNDERSCORE             '\137'
 1303 #define CHAR_GRAVE_ACCENT           '\140'
 1304 #define CHAR_a                      '\141'
 1305 #define CHAR_b                      '\142'
 1306 #define CHAR_c                      '\143'
 1307 #define CHAR_d                      '\144'
 1308 #define CHAR_e                      '\145'
 1309 #define CHAR_f                      '\146'
 1310 #define CHAR_g                      '\147'
 1311 #define CHAR_h                      '\150'
 1312 #define CHAR_i                      '\151'
 1313 #define CHAR_j                      '\152'
 1314 #define CHAR_k                      '\153'
 1315 #define CHAR_l                      '\154'
 1316 #define CHAR_m                      '\155'
 1317 #define CHAR_n                      '\156'
 1318 #define CHAR_o                      '\157'
 1319 #define CHAR_p                      '\160'
 1320 #define CHAR_q                      '\161'
 1321 #define CHAR_r                      '\162'
 1322 #define CHAR_s                      '\163'
 1323 #define CHAR_t                      '\164'
 1324 #define CHAR_u                      '\165'
 1325 #define CHAR_v                      '\166'
 1326 #define CHAR_w                      '\167'
 1327 #define CHAR_x                      '\170'
 1328 #define CHAR_y                      '\171'
 1329 #define CHAR_z                      '\172'
 1330 #define CHAR_LEFT_CURLY_BRACKET     '\173'
 1331 #define CHAR_VERTICAL_LINE          '\174'
 1332 #define CHAR_RIGHT_CURLY_BRACKET    '\175'
 1333 #define CHAR_TILDE                  '\176'
 1334 
 1335 #define STR_HT                      "\011"
 1336 #define STR_VT                      "\013"
 1337 #define STR_FF                      "\014"
 1338 #define STR_CR                      "\015"
 1339 #define STR_NL                      "\012"
 1340 #define STR_BS                      "\010"
 1341 #define STR_BEL                     "\007"
 1342 #define STR_ESC                     "\033"
 1343 #define STR_DEL                     "\177"
 1344 
 1345 #define STR_SPACE                   "\040"
 1346 #define STR_EXCLAMATION_MARK        "\041"
 1347 #define STR_QUOTATION_MARK          "\042"
 1348 #define STR_NUMBER_SIGN             "\043"
 1349 #define STR_DOLLAR_SIGN             "\044"
 1350 #define STR_PERCENT_SIGN            "\045"
 1351 #define STR_AMPERSAND               "\046"
 1352 #define STR_APOSTROPHE              "\047"
 1353 #define STR_LEFT_PARENTHESIS        "\050"
 1354 #define STR_RIGHT_PARENTHESIS       "\051"
 1355 #define STR_ASTERISK                "\052"
 1356 #define STR_PLUS                    "\053"
 1357 #define STR_COMMA                   "\054"
 1358 #define STR_MINUS                   "\055"
 1359 #define STR_DOT                     "\056"
 1360 #define STR_SLASH                   "\057"
 1361 #define STR_0                       "\060"
 1362 #define STR_1                       "\061"
 1363 #define STR_2                       "\062"
 1364 #define STR_3                       "\063"
 1365 #define STR_4                       "\064"
 1366 #define STR_5                       "\065"
 1367 #define STR_6                       "\066"
 1368 #define STR_7                       "\067"
 1369 #define STR_8                       "\070"
 1370 #define STR_9                       "\071"
 1371 #define STR_COLON                   "\072"
 1372 #define STR_SEMICOLON               "\073"
 1373 #define STR_LESS_THAN_SIGN          "\074"
 1374 #define STR_EQUALS_SIGN             "\075"
 1375 #define STR_GREATER_THAN_SIGN       "\076"
 1376 #define STR_QUESTION_MARK           "\077"
 1377 #define STR_COMMERCIAL_AT           "\100"
 1378 #define STR_A                       "\101"
 1379 #define STR_B                       "\102"
 1380 #define STR_C                       "\103"
 1381 #define STR_D                       "\104"
 1382 #define STR_E                       "\105"
 1383 #define STR_F                       "\106"
 1384 #define STR_G                       "\107"
 1385 #define STR_H                       "\110"
 1386 #define STR_I                       "\111"
 1387 #define STR_J                       "\112"
 1388 #define STR_K                       "\113"
 1389 #define STR_L                       "\114"
 1390 #define STR_M                       "\115"
 1391 #define STR_N                       "\116"
 1392 #define STR_O                       "\117"
 1393 #define STR_P                       "\120"
 1394 #define STR_Q                       "\121"
 1395 #define STR_R                       "\122"
 1396 #define STR_S                       "\123"
 1397 #define STR_T                       "\124"
 1398 #define STR_U                       "\125"
 1399 #define STR_V                       "\126"
 1400 #define STR_W                       "\127"
 1401 #define STR_X                       "\130"
 1402 #define STR_Y                       "\131"
 1403 #define STR_Z                       "\132"
 1404 #define STR_LEFT_SQUARE_BRACKET     "\133"
 1405 #define STR_BACKSLASH               "\134"
 1406 #define STR_RIGHT_SQUARE_BRACKET    "\135"
 1407 #define STR_CIRCUMFLEX_ACCENT       "\136"
 1408 #define STR_UNDERSCORE              "\137"
 1409 #define STR_GRAVE_ACCENT            "\140"
 1410 #define STR_a                       "\141"
 1411 #define STR_b                       "\142"
 1412 #define STR_c                       "\143"
 1413 #define STR_d                       "\144"
 1414 #define STR_e                       "\145"
 1415 #define STR_f                       "\146"
 1416 #define STR_g                       "\147"
 1417 #define STR_h                       "\150"
 1418 #define STR_i                       "\151"
 1419 #define STR_j                       "\152"
 1420 #define STR_k                       "\153"
 1421 #define STR_l                       "\154"
 1422 #define STR_m                       "\155"
 1423 #define STR_n                       "\156"
 1424 #define STR_o                       "\157"
 1425 #define STR_p                       "\160"
 1426 #define STR_q                       "\161"
 1427 #define STR_r                       "\162"
 1428 #define STR_s                       "\163"
 1429 #define STR_t                       "\164"
 1430 #define STR_u                       "\165"
 1431 #define STR_v                       "\166"
 1432 #define STR_w                       "\167"
 1433 #define STR_x                       "\170"
 1434 #define STR_y                       "\171"
 1435 #define STR_z                       "\172"
 1436 #define STR_LEFT_CURLY_BRACKET      "\173"
 1437 #define STR_VERTICAL_LINE           "\174"
 1438 #define STR_RIGHT_CURLY_BRACKET     "\175"
 1439 #define STR_TILDE                   "\176"
 1440 
 1441 #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
 1442 #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
 1443 #define STRING_F0                   STR_F "\0"
 1444 #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
 1445 #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
 1446 #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
 1447 #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
 1448 #define STRING_THEN                 STR_T STR_H STR_E STR_N
 1449 
 1450 #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
 1451 #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
 1452 #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
 1453 #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
 1454 #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
 1455 #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
 1456 #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
 1457 #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
 1458 #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
 1459 #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
 1460 #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
 1461 #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
 1462 #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
 1463 #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
 1464 
 1465 #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
 1466 
 1467 #define STRING_CR_RIGHTPAR             STR_C STR_R STR_RIGHT_PARENTHESIS
 1468 #define STRING_LF_RIGHTPAR             STR_L STR_F STR_RIGHT_PARENTHESIS
 1469 #define STRING_CRLF_RIGHTPAR           STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
 1470 #define STRING_ANY_RIGHTPAR            STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
 1471 #define STRING_ANYCRLF_RIGHTPAR        STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
 1472 #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
 1473 #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
 1474 #ifdef COMPILE_PCRE8
 1475 #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
 1476 #endif
 1477 #ifdef COMPILE_PCRE16
 1478 #define STRING_UTF_RIGHTPAR            STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
 1479 #endif
 1480 #define STRING_UCP_RIGHTPAR            STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
 1481 #define STRING_NO_START_OPT_RIGHTPAR   STR_N STR_O STR_UNDERSCORE STR_S STR_T STR_A STR_R STR_T STR_UNDERSCORE STR_O STR_P STR_T STR_RIGHT_PARENTHESIS
 1482 
 1483 #endif  /* SUPPORT_UTF */
 1484 
 1485 /* Escape items that are just an encoding of a particular data value. */
 1486 
 1487 #ifndef ESC_e
 1488 #define ESC_e CHAR_ESC
 1489 #endif
 1490 
 1491 #ifndef ESC_f
 1492 #define ESC_f CHAR_FF
 1493 #endif
 1494 
 1495 #ifndef ESC_n
 1496 #define ESC_n CHAR_NL
 1497 #endif
 1498 
 1499 #ifndef ESC_r
 1500 #define ESC_r CHAR_CR
 1501 #endif
 1502 
 1503 /* We can't officially use ESC_t because it is a POSIX reserved identifier
 1504 (presumably because of all the others like size_t). */
 1505 
 1506 #ifndef ESC_tee
 1507 #define ESC_tee CHAR_HT
 1508 #endif
 1509 
 1510 /* Codes for different types of Unicode property */
 1511 
 1512 #define PT_ANY        0    /* Any property - matches all chars */
 1513 #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
 1514 #define PT_GC         2    /* Specified general characteristic (e.g. L) */
 1515 #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
 1516 #define PT_SC         4    /* Script (e.g. Han) */
 1517 #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
 1518 #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
 1519 #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
 1520 #define PT_WORD       8    /* Word - L plus N plus underscore */
 1521 
 1522 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
 1523 contain characters with values greater than 255. */
 1524 
 1525 #define XCL_NOT    0x01    /* Flag: this is a negative class */
 1526 #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */
 1527 
 1528 #define XCL_END       0    /* Marks end of individual items */
 1529 #define XCL_SINGLE    1    /* Single item (one multibyte char) follows */
 1530 #define XCL_RANGE     2    /* A range (two multibyte chars) follows */
 1531 #define XCL_PROP      3    /* Unicode property (2-byte property code follows) */
 1532 #define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */
 1533 
 1534 /* These are escaped items that aren't just an encoding of a particular data
 1535 value such as \n. They must have non-zero values, as check_escape() returns
 1536 their negation. Also, they must appear in the same order as in the opcode
 1537 definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
 1538 corresponds to "." in DOTALL mode rather than an escape sequence. It is also
 1539 used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
 1540 non-DOTALL mode, "." behaves like \N.
 1541 
 1542 The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
 1543 when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
 1544 They must be contiguous, and remain in order so that the replacements can be
 1545 looked up from a table.
 1546 
 1547 The final escape must be ESC_REF as subsequent values are used for
 1548 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
 1549 greater than ESC_b and less than ESC_Z to detect the types that may be
 1550 repeated. These are the types that consume characters. If any new escapes are
 1551 put in between that don't consume a character, that code will have to change.
 1552 */
 1553 
 1554 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
 1555        ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
 1556        ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
 1557        ESC_E, ESC_Q, ESC_g, ESC_k,
 1558        ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
 1559        ESC_REF };
 1560 
 1561 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
 1562 OP_EOD must correspond in order to the list of escapes immediately above.
 1563 
 1564 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
 1565 that follow must also be updated to match. There are also tables called
 1566 "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
 1567 
 1568 enum {
 1569   OP_END,            /* 0 End of pattern */
 1570 
 1571   /* Values corresponding to backslashed metacharacters */
 1572 
 1573   OP_SOD,            /* 1 Start of data: \A */
 1574   OP_SOM,            /* 2 Start of match (subject + offset): \G */
 1575   OP_SET_SOM,        /* 3 Set start of match (\K) */
 1576   OP_NOT_WORD_BOUNDARY,  /*  4 \B */
 1577   OP_WORD_BOUNDARY,      /*  5 \b */
 1578   OP_NOT_DIGIT,          /*  6 \D */
 1579   OP_DIGIT,              /*  7 \d */
 1580   OP_NOT_WHITESPACE,     /*  8 \S */
 1581   OP_WHITESPACE,         /*  9 \s */
 1582   OP_NOT_WORDCHAR,       /* 10 \W */
 1583   OP_WORDCHAR,           /* 11 \w */
 1584 
 1585   OP_ANY,            /* 12 Match any character except newline */
 1586   OP_ALLANY,         /* 13 Match any character */
 1587   OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
 1588   OP_NOTPROP,        /* 15 \P (not Unicode property) */
 1589   OP_PROP,           /* 16 \p (Unicode property) */
 1590   OP_ANYNL,          /* 17 \R (any newline sequence) */
 1591   OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
 1592   OP_HSPACE,         /* 19 \h (horizontal whitespace) */
 1593   OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
 1594   OP_VSPACE,         /* 21 \v (vertical whitespace) */
 1595   OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
 1596   OP_EODN,           /* 23 End of data or \n at end of data: \Z. */
 1597   OP_EOD,            /* 24 End of data: \z */
 1598 
 1599   OP_CIRC,           /* 25 Start of line - not multiline */
 1600   OP_CIRCM,          /* 26 Start of line - multiline */
 1601   OP_DOLL,           /* 27 End of line - not multiline */
 1602   OP_DOLLM,          /* 28 End of line - multiline */
 1603   OP_CHAR,           /* 29 Match one character, casefully */
 1604   OP_CHARI,          /* 30 Match one character, caselessly */
 1605   OP_NOT,            /* 31 Match one character, not the given one, casefully */
 1606   OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
 1607 
 1608   /* The following sets of 13 opcodes must always be kept in step because
 1609   the offset from the first one is used to generate the others. */
 1610 
 1611   /**** Single characters, caseful, must precede the caseless ones ****/
 1612 
 1613   OP_STAR,           /* 33 The maximizing and minimizing versions of */
 1614   OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
 1615   OP_PLUS,           /* 35 the minimizing one second. */
 1616   OP_MINPLUS,        /* 36 */
 1617   OP_QUERY,          /* 37 */
 1618   OP_MINQUERY,       /* 38 */
 1619 
 1620   OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
 1621   OP_MINUPTO,        /* 40 */
 1622   OP_EXACT,          /* 41 Exactly n matches */
 1623 
 1624   OP_POSSTAR,        /* 42 Possessified star, caseful */
 1625   OP_POSPLUS,        /* 43 Possessified plus, caseful */
 1626   OP_POSQUERY,       /* 44 Posesssified query, caseful */
 1627   OP_POSUPTO,        /* 45 Possessified upto, caseful */
 1628 
 1629   /**** Single characters, caseless, must follow the caseful ones */
 1630 
 1631   OP_STARI,          /* 46 */
 1632   OP_MINSTARI,       /* 47 */
 1633   OP_PLUSI,          /* 48 */
 1634   OP_MINPLUSI,       /* 49 */
 1635   OP_QUERYI,         /* 50 */
 1636   OP_MINQUERYI,      /* 51 */
 1637 
 1638   OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
 1639   OP_MINUPTOI,       /* 53 */
 1640   OP_EXACTI,         /* 54 */
 1641 
 1642   OP_POSSTARI,       /* 55 Possessified star, caseless */
 1643   OP_POSPLUSI,       /* 56 Possessified plus, caseless */
 1644   OP_POSQUERYI,      /* 57 Posesssified query, caseless */
 1645   OP_POSUPTOI,       /* 58 Possessified upto, caseless */
 1646 
 1647   /**** The negated ones must follow the non-negated ones, and match them ****/
 1648   /**** Negated single character, caseful; must precede the caseless ones ****/
 1649 
 1650   OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
 1651   OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
 1652   OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
 1653   OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
 1654   OP_NOTQUERY,       /* 63 */
 1655   OP_NOTMINQUERY,    /* 64 */
 1656 
 1657   OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
 1658   OP_NOTMINUPTO,     /* 66 */
 1659   OP_NOTEXACT,       /* 67 Exactly n matches */
 1660 
 1661   OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
 1662   OP_NOTPOSPLUS,     /* 69 */
 1663   OP_NOTPOSQUERY,    /* 70 */
 1664   OP_NOTPOSUPTO,     /* 71 */
 1665 
 1666   /**** Negated single character, caseless; must follow the caseful ones ****/
 1667 
 1668   OP_NOTSTARI,       /* 72 */
 1669   OP_NOTMINSTARI,    /* 73 */
 1670   OP_NOTPLUSI,       /* 74 */
 1671   OP_NOTMINPLUSI,    /* 75 */
 1672   OP_NOTQUERYI,      /* 76 */
 1673   OP_NOTMINQUERYI,   /* 77 */
 1674 
 1675   OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
 1676   OP_NOTMINUPTOI,    /* 79 */
 1677   OP_NOTEXACTI,      /* 80 Exactly n matches */
 1678 
 1679   OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
 1680   OP_NOTPOSPLUSI,    /* 82 */
 1681   OP_NOTPOSQUERYI,   /* 83 */
 1682   OP_NOTPOSUPTOI,    /* 84 */
 1683 
 1684   /**** Character types ****/
 1685 
 1686   OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
 1687   OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
 1688   OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
 1689   OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
 1690   OP_TYPEQUERY,      /* 89 */
 1691   OP_TYPEMINQUERY,   /* 90 */
 1692 
 1693   OP_TYPEUPTO,       /* 91 From 0 to n matches */
 1694   OP_TYPEMINUPTO,    /* 92 */
 1695   OP_TYPEEXACT,      /* 93 Exactly n matches */
 1696 
 1697   OP_TYPEPOSSTAR,    /* 94 Possessified versions */
 1698   OP_TYPEPOSPLUS,    /* 95 */
 1699   OP_TYPEPOSQUERY,   /* 96 */
 1700   OP_TYPEPOSUPTO,    /* 97 */
 1701 
 1702   /* These are used for character classes and back references; only the
 1703   first six are the same as the sets above. */
 1704 
 1705   OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
 1706   OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
 1707   OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
 1708   OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
 1709   OP_CRQUERY,        /* 102 */
 1710   OP_CRMINQUERY,     /* 103 */
 1711 
 1712   OP_CRRANGE,        /* 104 These are different to the three sets above. */
 1713   OP_CRMINRANGE,     /* 105 */
 1714 
 1715   /* End of quantifier opcodes */
 1716 
 1717   OP_CLASS,          /* 106 Match a character class, chars < 256 only */
 1718   OP_NCLASS,         /* 107 Same, but the bitmap was created from a negative
 1719                               class - the difference is relevant only when a
 1720                               character > 255 is encountered. */
 1721   OP_XCLASS,         /* 108 Extended class for handling > 255 chars within the
 1722                               class. This does both positive and negative. */
 1723   OP_REF,            /* 109 Match a back reference, casefully */
 1724   OP_REFI,           /* 110 Match a back reference, caselessly */
 1725   OP_RECURSE,        /* 111 Match a numbered subpattern (possibly recursive) */
 1726   OP_CALLOUT,        /* 112 Call out to external function if provided */
 1727 
 1728   OP_ALT,            /* 113 Start of alternation */
 1729   OP_KET,            /* 114 End of group that doesn't have an unbounded repeat */
 1730   OP_KETRMAX,        /* 115 These two must remain together and in this */
 1731   OP_KETRMIN,        /* 116 order. They are for groups the repeat for ever. */
 1732   OP_KETRPOS,        /* 117 Possessive unlimited repeat. */
 1733 
 1734   /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
 1735   asserts must remain in order. */
 1736 
 1737   OP_REVERSE,        /* 118 Move pointer back - used in lookbehind assertions */
 1738   OP_ASSERT,         /* 119 Positive lookahead */
 1739   OP_ASSERT_NOT,     /* 120 Negative lookahead */
 1740   OP_ASSERTBACK,     /* 121 Positive lookbehind */
 1741   OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
 1742 
 1743   /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
 1744   after the assertions, with ONCE first, as there's a test for >= ONCE for a
 1745   subpattern that isn't an assertion. The POS versions must immediately follow
 1746   the non-POS versions in each case. */
 1747 
 1748   OP_ONCE,           /* 123 Atomic group, contains captures */
 1749   OP_ONCE_NC,        /* 124 Atomic group containing no captures */
 1750   OP_BRA,            /* 125 Start of non-capturing bracket */
 1751   OP_BRAPOS,         /* 126 Ditto, with unlimited, possessive repeat */
 1752   OP_CBRA,           /* 127 Start of capturing bracket */
 1753   OP_CBRAPOS,        /* 128 Ditto, with unlimited, possessive repeat */
 1754   OP_COND,           /* 129 Conditional group */
 1755 
 1756   /* These five must follow the previous five, in the same order. There's a
 1757   check for >= SBRA to distinguish the two sets. */
 1758 
 1759   OP_SBRA,           /* 130 Start of non-capturing bracket, check empty  */
 1760   OP_SBRAPOS,        /* 131 Ditto, with unlimited, possessive repeat */
 1761   OP_SCBRA,          /* 132 Start of capturing bracket, check empty */
 1762   OP_SCBRAPOS,       /* 133 Ditto, with unlimited, possessive repeat */
 1763   OP_SCOND,          /* 134 Conditional group, check empty */
 1764 
 1765   /* The next two pairs must (respectively) be kept together. */
 1766 
 1767   OP_CREF,           /* 135 Used to hold a capture number as condition */
 1768   OP_NCREF,          /* 136 Same, but generated by a name reference*/
 1769   OP_RREF,           /* 137 Used to hold a recursion number as condition */
 1770   OP_NRREF,          /* 138 Same, but generated by a name reference*/
 1771   OP_DEF,            /* 139 The DEFINE condition */
 1772 
 1773   OP_BRAZERO,        /* 140 These two must remain together and in this */
 1774   OP_BRAMINZERO,     /* 141 order. */
 1775   OP_BRAPOSZERO,     /* 142 */
 1776 
 1777   /* These are backtracking control verbs */
 1778 
 1779   OP_MARK,           /* 143 always has an argument */
 1780   OP_PRUNE,          /* 144 */
 1781   OP_PRUNE_ARG,      /* 145 same, but with argument */
 1782   OP_SKIP,           /* 146 */
 1783   OP_SKIP_ARG,       /* 147 same, but with argument */
 1784   OP_THEN,           /* 148 */
 1785   OP_THEN_ARG,       /* 149 same, but with argument */
 1786   OP_COMMIT,         /* 150 */
 1787 
 1788   /* These are forced failure and success verbs */
 1789 
 1790   OP_FAIL,           /* 151 */
 1791   OP_ACCEPT,         /* 152 */
 1792   OP_ASSERT_ACCEPT,  /* 153 Used inside assertions */
 1793   OP_CLOSE,          /* 154 Used before OP_ACCEPT to close open captures */
 1794 
 1795   /* This is used to skip a subpattern with a {0} quantifier */
 1796 
 1797   OP_SKIPZERO,       /* 155 */
 1798 
 1799   /* This is not an opcode, but is used to check that tables indexed by opcode
 1800   are the correct length, in order to catch updating errors - there have been
 1801   some in the past. */
 1802 
 1803   OP_TABLE_LENGTH
 1804 };
 1805 
 1806 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
 1807 definitions that follow must also be updated to match. There are also tables
 1808 called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
 1809 
 1810 
 1811 /* This macro defines textual names for all the opcodes. These are used only
 1812 for debugging, and some of them are only partial names. The macro is referenced
 1813 only in pcre_printint.c, which fills out the full names in many cases (and in
 1814 some cases doesn't actually use these names at all). */
 1815 
 1816 #define OP_NAME_LIST \
 1817   "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
 1818   "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
 1819   "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
 1820   "extuni",  "\\Z", "\\z",                                        \
 1821   "^", "^", "$", "$", "char", "chari", "not", "noti",             \
 1822   "*", "*?", "+", "+?", "?", "??",                                \
 1823   "{", "{", "{",                                                  \
 1824   "*+","++", "?+", "{",                                           \
 1825   "*", "*?", "+", "+?", "?", "??",                                \
 1826   "{", "{", "{",                                                  \
 1827   "*+","++", "?+", "{",                                           \
 1828   "*", "*?", "+", "+?", "?", "??",                                \
 1829   "{", "{", "{",                                                  \
 1830   "*+","++", "?+", "{",                                           \
 1831   "*", "*?", "+", "+?", "?", "??",                                \
 1832   "{", "{", "{",                                                  \
 1833   "*+","++", "?+", "{",                                           \
 1834   "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
 1835   "*+","++", "?+", "{",                                           \
 1836   "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
 1837   "class", "nclass", "xclass", "Ref", "Refi",                     \
 1838   "Recurse", "Callout",                                           \
 1839   "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
 1840   "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
 1841   "Once", "Once_NC",                                              \
 1842   "Bra", "BraPos", "CBra", "CBraPos",                             \
 1843   "Cond",                                                         \
 1844   "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
 1845   "SCond",                                                        \
 1846   "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def",   \
 1847   "Brazero", "Braminzero", "Braposzero",                          \
 1848   "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
 1849   "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
 1850   "*ACCEPT", "*ASSERT_ACCEPT",                                    \
 1851   "Close", "Skip zero"
 1852 
 1853 
 1854 /* This macro defines the length of fixed length operations in the compiled
 1855 regex. The lengths are used when searching for specific things, and also in the
 1856 debugging printing of a compiled regex. We use a macro so that it can be
 1857 defined close to the definitions of the opcodes themselves.
 1858 
 1859 As things have been extended, some of these are no longer fixed lenths, but are
 1860 minima instead. For example, the length of a single-character repeat may vary
 1861 in UTF-8 mode. The code that uses this table must know about such things. */
 1862 
 1863 #define OP_LENGTHS \
 1864   1,                             /* End                                    */ \
 1865   1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
 1866   1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
 1867   1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
 1868   3, 3,                          /* \P, \p                                 */ \
 1869   1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
 1870   1,                             /* \X                                     */ \
 1871   1, 1, 1, 1, 1, 1,              /* \Z, \z, ^, ^M, $, $M                   */ \
 1872   2,                             /* Char  - the minimum length             */ \
 1873   2,                             /* Chari  - the minimum length            */ \
 1874   2,                             /* not                                    */ \
 1875   2,                             /* noti                                   */ \
 1876   /* Positive single-char repeats                             ** These are */ \
 1877   2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
 1878   2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
 1879   2+IMM2_SIZE,                   /* exact                                  */ \
 1880   2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
 1881   2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
 1882   2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
 1883   2+IMM2_SIZE,                   /* exact I                                */ \
 1884   2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
 1885   /* Negative single-char repeats - only for chars < 256                   */ \
 1886   2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
 1887   2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
 1888   2+IMM2_SIZE,                   /* NOT exact                              */ \
 1889   2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
 1890   2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
 1891   2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
 1892   2+IMM2_SIZE,                   /* NOT exact I                            */ \
 1893   2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
 1894   /* Positive type repeats                                                 */ \
 1895   2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
 1896   2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
 1897   2+IMM2_SIZE,                   /* Type exact                             */ \
 1898   2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
 1899   /* Character class & ref repeats                                         */ \
 1900   1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
 1901   1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
 1902   1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
 1903   1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
 1904   0,                             /* XCLASS - variable length               */ \
 1905   1+IMM2_SIZE,                   /* REF                                    */ \
 1906   1+IMM2_SIZE,                   /* REFI                                   */ \
 1907   1+LINK_SIZE,                   /* RECURSE                                */ \
 1908   2+IMMPTR_SIZE+2*LINK_SIZE,     /* CALLOUT                                */ \
 1909   1+LINK_SIZE,                   /* Alt                                    */ \
 1910   1+LINK_SIZE,                   /* Ket                                    */ \
 1911   1+LINK_SIZE,                   /* KetRmax                                */ \
 1912   1+LINK_SIZE,                   /* KetRmin                                */ \
 1913   1+LINK_SIZE,                   /* KetRpos                                */ \
 1914   1+LINK_SIZE,                   /* Reverse                                */ \
 1915   1+LINK_SIZE,                   /* Assert                                 */ \
 1916   1+LINK_SIZE,                   /* Assert not                             */ \
 1917   1+LINK_SIZE,                   /* Assert behind                          */ \
 1918   1+LINK_SIZE,                   /* Assert behind not                      */ \
 1919   1+LINK_SIZE,                   /* ONCE                                   */ \
 1920   1+LINK_SIZE,                   /* ONCE_NC                                */ \
 1921   1+LINK_SIZE,                   /* BRA                                    */ \
 1922   1+LINK_SIZE,                   /* BRAPOS                                 */ \
 1923   1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
 1924   1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
 1925   1+LINK_SIZE,                   /* COND                                   */ \
 1926   1+LINK_SIZE,                   /* SBRA                                   */ \
 1927   1+LINK_SIZE,                   /* SBRAPOS                                */ \
 1928   1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
 1929   1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
 1930   1+LINK_SIZE,                   /* SCOND                                  */ \
 1931   1+IMM2_SIZE, 1+IMM2_SIZE,      /* CREF, NCREF                            */ \
 1932   1+IMM2_SIZE, 1+IMM2_SIZE,      /* RREF, NRREF                            */ \
 1933   1,                             /* DEF                                    */ \
 1934   1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
 1935   3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
 1936   1, 3,                          /* SKIP, SKIP_ARG                         */ \
 1937   1, 3,                          /* THEN, THEN_ARG                         */ \
 1938   1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
 1939   1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
 1940 
 1941 /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
 1942 condition. */
 1943 
 1944 #define RREF_ANY  0xffff
 1945 
 1946 /* Compile time error code numbers. They are given names so that they can more
 1947 easily be tracked. When a new number is added, the table called eint in
 1948 pcreposix.c must be updated. */
 1949 
 1950 enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
 1951        ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
 1952        ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
 1953        ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
 1954        ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
 1955        ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
 1956        ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
 1957        ERR70, ERR71, ERR72, ERR73, ERR74, ERRCOUNT };
 1958 
 1959 /* The real format of the start of the pcre block; the index of names and the
 1960 code vector run on as long as necessary after the end. We store an explicit
 1961 offset to the name table so that if a regex is compiled on one host, saved, and
 1962 then run on another where the size of pointers is different, all might still
 1963 be well. For the case of compiled-on-4 and run-on-8, we include an extra
 1964 pointer that is always NULL. For future-proofing, a few dummy fields were
 1965 originally included - even though you can never get this planning right - but
 1966 there is only one left now.
 1967 
 1968 NOTE NOTE NOTE:
 1969 Because people can now save and re-use compiled patterns, any additions to this
 1970 structure should be made at the end, and something earlier (e.g. a new
 1971 flag in the options or one of the dummy fields) should indicate that the new
 1972 fields are present. Currently PCRE always sets the dummy fields to zero.
 1973 NOTE NOTE NOTE
 1974 */
 1975 
 1976 #ifdef COMPILE_PCRE8
 1977 #define REAL_PCRE real_pcre
 1978 #else
 1979 #define REAL_PCRE real_pcre16
 1980 #endif
 1981 
 1982 typedef struct REAL_PCRE {
 1983   pcre_uint32 magic_number;
 1984   pcre_uint32 size;               /* Total that was malloced */
 1985   pcre_uint32 options;            /* Public options */
 1986   pcre_uint16 flags;              /* Private flags */
 1987   pcre_uint16 dummy1;             /* For future use */
 1988   pcre_uint16 top_bracket;
 1989   pcre_uint16 top_backref;
 1990   pcre_uint16 first_char;         /* Starting character */
 1991   pcre_uint16 req_char;           /* This character must be seen */
 1992   pcre_uint16 name_table_offset;  /* Offset to name table that follows */
 1993   pcre_uint16 name_entry_size;    /* Size of any name items */
 1994   pcre_uint16 name_count;         /* Number of name items */
 1995   pcre_uint16 ref_count;          /* Reference count */
 1996 
 1997   const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
 1998   const pcre_uint8 *nullpad;      /* NULL padding */
 1999 } REAL_PCRE;
 2000 
 2001 /* The format of the block used to store data from pcre_study(). The same
 2002 remark (see NOTE above) about extending this structure applies. */
 2003 
 2004 typedef struct pcre_study_data {
 2005   pcre_uint32 size;               /* Total that was malloced */
 2006   pcre_uint32 flags;              /* Private flags */
 2007   pcre_uint8 start_bits[32];      /* Starting char bits */
 2008   pcre_uint32 minlength;          /* Minimum subject length */
 2009 } pcre_study_data;
 2010 
 2011 /* Structure for building a chain of open capturing subpatterns during
 2012 compiling, so that instructions to close them can be compiled when (*ACCEPT) is
 2013 encountered. This is also used to identify subpatterns that contain recursive
 2014 back references to themselves, so that they can be made atomic. */
 2015 
 2016 typedef struct open_capitem {
 2017   struct open_capitem *next;    /* Chain link */
 2018   pcre_uint16 number;           /* Capture number */
 2019   pcre_uint16 flag;             /* Set TRUE if recursive back ref */
 2020 } open_capitem;
 2021 
 2022 /* Structure for passing "static" information around between the functions
 2023 doing the compiling, so that they are thread-safe. */
 2024 
 2025 typedef struct compile_data {
 2026   const pcre_uint8 *lcc;            /* Points to lower casing table */
 2027   const pcre_uint8 *fcc;            /* Points to case-flipping table */
 2028   const pcre_uint8 *cbits;          /* Points to character type table */
 2029   const pcre_uint8 *ctypes;         /* Points to table of type maps */
 2030   const pcre_uchar *start_workspace;/* The start of working space */
 2031   const pcre_uchar *start_code;     /* The start of the compiled code */
 2032   const pcre_uchar *start_pattern;  /* The start of the pattern */
 2033   const pcre_uchar *end_pattern;    /* The end of the pattern */
 2034   open_capitem *open_caps;          /* Chain of open capture items */
 2035   pcre_uchar *hwm;                  /* High watermark of workspace */
 2036   pcre_uchar *name_table;           /* The name/number table */
 2037   int  names_found;                 /* Number of entries so far */
 2038   int  name_entry_size;             /* Size of each entry */
 2039   int  workspace_size;              /* Size of workspace */
 2040   int  bracount;                    /* Count of capturing parens as we compile */
 2041   int  final_bracount;              /* Saved value after first pass */
 2042   int  top_backref;                 /* Maximum back reference */
 2043   unsigned int backref_map;         /* Bitmap of low back refs */
 2044   int  assert_depth;                /* Depth of nested assertions */
 2045   int  external_options;            /* External (initial) options */
 2046   int  external_flags;              /* External flag bits to be set */
 2047   int  req_varyopt;                 /* "After variable item" flag for reqbyte */
 2048   BOOL had_accept;                  /* (*ACCEPT) encountered */
 2049   BOOL check_lookbehind;            /* Lookbehinds need later checking */
 2050   int  nltype;                      /* Newline type */
 2051   int  nllen;                       /* Newline string length */
 2052   pcre_uchar nl[4];                 /* Newline string when fixed length */
 2053 } compile_data;
 2054 
 2055 /* Structure for maintaining a chain of pointers to the currently incomplete
 2056 branches, for testing for left recursion while compiling. */
 2057 
 2058 typedef struct branch_chain {
 2059   struct branch_chain *outer;
 2060   pcre_uchar *current_branch;
 2061 } branch_chain;
 2062 
 2063 /* Structure for items in a linked list that represents an explicit recursive
 2064 call within the pattern; used by pcre_exec(). */
 2065 
 2066 typedef struct recursion_info {
 2067   struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
 2068   int group_num;                  /* Number of group that was called */
 2069   int *offset_save;               /* Pointer to start of saved offsets */
 2070   int saved_max;                  /* Number of saved offsets */
 2071   PCRE_PUCHAR subject_position;   /* Position at start of recursion */
 2072 } recursion_info;
 2073 
 2074 /* A similar structure for pcre_dfa_exec(). */
 2075 
 2076 typedef struct dfa_recursion_info {
 2077   struct dfa_recursion_info *prevrec;
 2078   int group_num;
 2079   PCRE_PUCHAR subject_position;
 2080 } dfa_recursion_info;
 2081 
 2082 /* Structure for building a chain of data for holding the values of the subject
 2083 pointer at the start of each subpattern, so as to detect when an empty string
 2084 has been matched by a subpattern - to break infinite loops; used by
 2085 pcre_exec(). */
 2086 
 2087 typedef struct eptrblock {
 2088   struct eptrblock *epb_prev;
 2089   PCRE_PUCHAR epb_saved_eptr;
 2090 } eptrblock;
 2091 
 2092 
 2093 /* Structure for passing "static" information around between the functions
 2094 doing traditional NFA matching, so that they are thread-safe. */
 2095 
 2096 typedef struct match_data {
 2097   unsigned long int match_call_count;      /* As it says */
 2098   unsigned long int match_limit;           /* As it says */
 2099   unsigned long int match_limit_recursion; /* As it says */
 2100   int   *offset_vector;           /* Offset vector */
 2101   int    offset_end;              /* One past the end */
 2102   int    offset_max;              /* The maximum usable for return data */
 2103   int    nltype;                  /* Newline type */
 2104   int    nllen;                   /* Newline string length */
 2105   int    name_count;              /* Number of names in name table */
 2106   int    name_entry_size;         /* Size of entry in names table */
 2107   pcre_uchar *name_table;         /* Table of names */
 2108   pcre_uchar nl[4];               /* Newline string when fixed */
 2109   const  pcre_uint8 *lcc;         /* Points to lower casing table */
 2110   const  pcre_uint8 *fcc;         /* Points to case-flipping table */
 2111   const  pcre_uint8 *ctypes;      /* Points to table of type maps */
 2112   BOOL   offset_overflow;         /* Set if too many extractions */
 2113   BOOL   notbol;                  /* NOTBOL flag */
 2114   BOOL   noteol;                  /* NOTEOL flag */
 2115   BOOL   utf;                     /* UTF-8 / UTF-16 flag */
 2116   BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
 2117   BOOL   use_ucp;                 /* PCRE_UCP flag */
 2118   BOOL   endonly;                 /* Dollar not before final \n */
 2119   BOOL   notempty;                /* Empty string match not wanted */
 2120   BOOL   notempty_atstart;        /* Empty string match at start not wanted */
 2121   BOOL   hitend;                  /* Hit the end of the subject at some point */
 2122   BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
 2123   BOOL   hasthen;                 /* Pattern contains (*THEN) */
 2124   BOOL   ignore_skip_arg;         /* For re-run when SKIP name not found */
 2125   const  pcre_uchar *start_code;  /* For use when recursing */
 2126   PCRE_PUCHAR start_subject;      /* Start of the subject string */
 2127   PCRE_PUCHAR end_subject;        /* End of the subject string */
 2128   PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
 2129   PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
 2130   PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
 2131   int    partial;                 /* PARTIAL options */
 2132   int    end_offset_top;          /* Highwater mark at end of match */
 2133   int    capture_last;            /* Most recent capture number */
 2134   int    start_offset;            /* The start offset value */
 2135   int    match_function_type;     /* Set for certain special calls of MATCH() */
 2136   eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
 2137   int    eptrn;                   /* Next free eptrblock */
 2138   recursion_info *recursive;      /* Linked list of recursion data */
 2139   void  *callout_data;            /* To pass back to callouts */
 2140   const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
 2141   const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
 2142   const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
 2143 } match_data;
 2144 
 2145 /* A similar structure is used for the same purpose by the DFA matching
 2146 functions. */
 2147 
 2148 typedef struct dfa_match_data {
 2149   const pcre_uchar *start_code;     /* Start of the compiled pattern */
 2150   const pcre_uchar *start_subject ; /* Start of the subject string */
 2151   const pcre_uchar *end_subject;    /* End of subject string */
 2152   const pcre_uchar *start_used_ptr; /* Earliest consulted character */
 2153   const pcre_uint8 *tables;         /* Character tables */
 2154   int   start_offset;               /* The start offset value */
 2155   int   moptions;                   /* Match options */
 2156   int   poptions;                   /* Pattern options */
 2157   int   nltype;                     /* Newline type */
 2158   int   nllen;                      /* Newline string length */
 2159   pcre_uchar nl[4];                 /* Newline string when fixed */
 2160   void *callout_data;               /* To pass back to callouts */
 2161   dfa_recursion_info *recursive;    /* Linked list of recursion data */
 2162 } dfa_match_data;
 2163 
 2164 /* Bit definitions for entries in the pcre_ctypes table. */
 2165 
 2166 #define ctype_space   0x01
 2167 #define ctype_letter  0x02
 2168 #define ctype_digit   0x04
 2169 #define ctype_xdigit  0x08
 2170 #define ctype_word    0x10   /* alphanumeric or '_' */
 2171 #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
 2172 
 2173 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
 2174 of bits for a class map. Some classes are built by combining these tables. */
 2175 
 2176 #define cbit_space     0      /* [:space:] or \s */
 2177 #define cbit_xdigit   32      /* [:xdigit:] */
 2178 #define cbit_digit    64      /* [:digit:] or \d */
 2179 #define cbit_upper    96      /* [:upper:] */
 2180 #define cbit_lower   128      /* [:lower:] */
 2181 #define cbit_word    160      /* [:word:] or \w */
 2182 #define cbit_graph   192      /* [:graph:] */
 2183 #define cbit_print   224      /* [:print:] */
 2184 #define cbit_punct   256      /* [:punct:] */
 2185 #define cbit_cntrl   288      /* [:cntrl:] */
 2186 #define cbit_length  320      /* Length of the cbits table */
 2187 
 2188 /* Offsets of the various tables from the base tables pointer, and
 2189 total length. */
 2190 
 2191 #define lcc_offset      0
 2192 #define fcc_offset    256
 2193 #define cbits_offset  512
 2194 #define ctypes_offset (cbits_offset + cbit_length)
 2195 #define tables_length (ctypes_offset + 256)
 2196 
 2197 /* Internal function prefix */
 2198 
 2199 #ifdef COMPILE_PCRE8
 2200 #ifndef PUBL
 2201 #define PUBL(name) pcre_##name
 2202 #endif
 2203 #ifndef PRIV
 2204 #define PRIV(name) _pcre_##name
 2205 #endif
 2206 #else /* COMPILE_PCRE8 */
 2207 #ifdef COMPILE_PCRE16
 2208 #ifndef PUBL
 2209 #define PUBL(name) pcre16_##name
 2210 #endif
 2211 #ifndef PRIV
 2212 #define PRIV(name) _pcre16_##name
 2213 #endif
 2214 #else
 2215 #error Unsupported compiling mode
 2216 #endif /* COMPILE_PCRE16 */
 2217 #endif /* COMPILE_PCRE8 */
 2218 
 2219 /* Layout of the UCP type table that translates property names into types and
 2220 codes. Each entry used to point directly to a name, but to reduce the number of
 2221 relocations in shared libraries, it now has an offset into a single string
 2222 instead. */
 2223 
 2224 typedef struct {
 2225   pcre_uint16 name_offset;
 2226   pcre_uint16 type;
 2227   pcre_uint16 value;
 2228 } ucp_type_table;
 2229 
 2230 
 2231 /* Internal shared data tables. These are tables that are used by more than one
 2232 of the exported public functions. They have to be "external" in the C sense,
 2233 but are not part of the PCRE public API. The data for these tables is in the
 2234 pcre_tables.c module. */
 2235 
 2236 #ifdef COMPILE_PCRE8
 2237 
 2238 extern const int            PRIV(utf8_table1)[];
 2239 extern const int            PRIV(utf8_table1_size);
 2240 extern const int            PRIV(utf8_table2)[];
 2241 extern const int            PRIV(utf8_table3)[];
 2242 extern const pcre_uint8     PRIV(utf8_table4)[];
 2243 
 2244 #endif /* COMPILE_PCRE8 */
 2245 
 2246 extern const char           PRIV(utt_names)[];
 2247 extern const ucp_type_table PRIV(utt)[];
 2248 extern const int            PRIV(utt_size);
 2249 
 2250 extern const pcre_uint8     PRIV(default_tables)[];
 2251 
 2252 extern const pcre_uint8     PRIV(OP_lengths)[];
 2253 
 2254 
 2255 /* Internal shared functions. These are functions that are used by more than
 2256 one of the exported public functions. They have to be "external" in the C
 2257 sense, but are not part of the PCRE public API. */
 2258 
 2259 /* String comparison functions. */
 2260 #ifdef COMPILE_PCRE8
 2261 
 2262 #define STRCMP_UC_UC(str1, str2) \
 2263   strcmp((char *)(str1), (char *)(str2))
 2264 #define STRCMP_UC_C8(str1, str2) \
 2265   strcmp((char *)(str1), (str2))
 2266 #define STRNCMP_UC_UC(str1, str2, num) \
 2267   strncmp((char *)(str1), (char *)(str2), (num))
 2268 #define STRNCMP_UC_C8(str1, str2, num) \
 2269   strncmp((char *)(str1), (str2), (num))
 2270 #define STRLEN_UC(str) strlen((const char *)str)
 2271 
 2272 #else
 2273 
 2274 extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
 2275                            const pcre_uchar *);
 2276 extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
 2277                            const char *);
 2278 extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
 2279                            const pcre_uchar *, unsigned int num);
 2280 extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
 2281                            const char *, unsigned int num);
 2282 extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
 2283 
 2284 #define STRCMP_UC_UC(str1, str2) \
 2285   PRIV(strcmp_uc_uc)((str1), (str2))
 2286 #define STRCMP_UC_C8(str1, str2) \
 2287   PRIV(strcmp_uc_c8)((str1), (str2))
 2288 #define STRNCMP_UC_UC(str1, str2, num) \
 2289   PRIV(strncmp_uc_uc)((str1), (str2), (num))
 2290 #define STRNCMP_UC_C8(str1, str2, num) \
 2291   PRIV(strncmp_uc_c8)((str1), (str2), (num))
 2292 #define STRLEN_UC(str) PRIV(strlen_uc)(str)
 2293 
 2294 #endif /* COMPILE_PCRE8 */
 2295 
 2296 extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
 2297 extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
 2298                            int *, BOOL);
 2299 extern int               PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
 2300 extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
 2301 extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
 2302                            int *, BOOL);
 2303 extern BOOL              PRIV(xclass)(int, const pcre_uchar *, BOOL);
 2304 
 2305 #ifdef SUPPORT_JIT
 2306 extern void              PRIV(jit_compile)(const REAL_PCRE *, PUBL(extra) *);
 2307 extern int               PRIV(jit_exec)(const REAL_PCRE *, void *,
 2308                            const pcre_uchar *, int, int, int, int, int *, int);
 2309 extern void              PRIV(jit_free)(void *);
 2310 extern int               PRIV(jit_get_size)(void *);
 2311 extern const char*       PRIV(jit_get_target)(void);
 2312 #endif
 2313 
 2314 /* Unicode character database (UCD) */
 2315 
 2316 typedef struct {
 2317   pcre_uint8 script;
 2318   pcre_uint8 chartype;
 2319   pcre_int32 other_case;
 2320 } ucd_record;
 2321 
 2322 extern const ucd_record  PRIV(ucd_records)[];
 2323 extern const pcre_uint8  PRIV(ucd_stage1)[];
 2324 extern const pcre_uint16 PRIV(ucd_stage2)[];
 2325 extern const int         PRIV(ucp_gentype)[];
 2326 #ifdef SUPPORT_JIT
 2327 extern const int         PRIV(ucp_typerange)[];
 2328 #endif
 2329 
 2330 #ifdef SUPPORT_UCP
 2331 /* UCD access macros */
 2332 
 2333 #define UCD_BLOCK_SIZE 128
 2334 #define GET_UCD(ch) (PRIV(ucd_records) + \
 2335         PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
 2336         UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
 2337 
 2338 #define UCD_CHARTYPE(ch)  GET_UCD(ch)->chartype
 2339 #define UCD_SCRIPT(ch)    GET_UCD(ch)->script
 2340 #define UCD_CATEGORY(ch)  PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
 2341 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
 2342 
 2343 #endif /* SUPPORT_UCP */
 2344 
 2345 #endif
 2346 
 2347 /* End of pcre_internal.h */