Ruby: regint.h Source File

00001 #ifndef ONIGURUMA_REGINT_H
00002 #define ONIGURUMA_REGINT_H
00003 /**********************************************************************
00004   regint.h -  Oniguruma (regular expression library)
00005 **********************************************************************/
00006 /*-
00007  * Copyright (c) 2002-2008  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>
00008  * All rights reserved.
00009  *
00010  * Redistribution and use in source and binary forms, with or without
00011  * modification, are permitted provided that the following conditions
00012  * are met:
00013  * 1. Redistributions of source code must retain the above copyright
00014  *    notice, this list of conditions and the following disclaimer.
00015  * 2. Redistributions in binary form must reproduce the above copyright
00016  *    notice, this list of conditions and the following disclaimer in the
00017  *    documentation and/or other materials provided with the distribution.
00018  *
00019  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
00020  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
00021  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
00022  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
00023  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
00024  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
00025  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
00026  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
00027  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
00028  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
00029  * SUCH DAMAGE.
00030  */
00031 
00032 /* for debug */
00033 /* #define ONIG_DEBUG_PARSE_TREE */
00034 /* #define ONIG_DEBUG_COMPILE */
00035 /* #define ONIG_DEBUG_SEARCH */
00036 /* #define ONIG_DEBUG_MATCH */
00037 /* #define ONIG_DONT_OPTIMIZE */
00038 
00039 /* for byte-code statistical data. */
00040 /* #define ONIG_DEBUG_STATISTICS */
00041 
00042 #if defined(ONIG_DEBUG_PARSE_TREE) || defined(ONIG_DEBUG_MATCH) || \
00043     defined(ONIG_DEBUG_SEARCH) || defined(ONIG_DEBUG_COMPILE) || \
00044     defined(ONIG_DEBUG_STATISTICS)
00045 #ifndef ONIG_DEBUG
00046 #define ONIG_DEBUG
00047 #endif
00048 #endif
00049 
00050 #if defined(__i386) || defined(__i386__) || defined(_M_IX86) || \
00051     defined(__x86_64) || defined(__x86_64__) || defined(_M_AMD86) || \
00052     defined(__mc68020__)
00053 #define PLATFORM_UNALIGNED_WORD_ACCESS
00054 #endif
00055 
00056 /* config */
00057 /* spec. config */
00058 #define USE_NAMED_GROUP
00059 #define USE_SUBEXP_CALL
00060 #define USE_BACKREF_WITH_LEVEL        /* \k<name+n>, \k<name-n> */
00061 #define USE_MONOMANIAC_CHECK_CAPTURES_IN_ENDLESS_REPEAT  /* /(?:()|())*\2/ */
00062 #define USE_NEWLINE_AT_END_OF_STRING_HAS_EMPTY_LINE     /* /\n$/ =~ "\n" */
00063 #define USE_WARNING_REDUNDANT_NESTED_REPEAT_OPERATOR
00064 /* #define USE_RECOMPILE_API */
00065 /* !!! moved to regenc.h. */ /* #define USE_CRNL_AS_LINE_TERMINATOR */
00066 
00067 /* internal config */
00068 #define USE_PARSE_TREE_NODE_RECYCLE
00069 #define USE_OP_PUSH_OR_JUMP_EXACT
00070 #define USE_QTFR_PEEK_NEXT
00071 #define USE_ST_LIBRARY
00072 #define USE_SHARED_CCLASS_TABLE
00073 
00074 #define INIT_MATCH_STACK_SIZE                     160
00075 #define DEFAULT_MATCH_STACK_LIMIT_SIZE              0 /* unlimited */
00076 
00077 #if defined(__GNUC__)
00078 #  define ARG_UNUSED  __attribute__ ((unused))
00079 #else
00080 #  define ARG_UNUSED
00081 #endif
00082 
00083 /* */
00084 /* escape other system UChar definition */
00085 #ifndef RUBY_DEFINES_H
00086 #include "ruby/ruby.h"
00087 #undef xmalloc
00088 #undef xrealloc
00089 #undef xcalloc
00090 #undef xfree
00091 #endif
00092 #ifdef ONIG_ESCAPE_UCHAR_COLLISION
00093 #undef ONIG_ESCAPE_UCHAR_COLLISION
00094 #endif
00095 #undef USE_MATCH_RANGE_IS_COMPLETE_RANGE
00096 #undef USE_CAPTURE_HISTORY
00097 #define USE_VARIABLE_META_CHARS
00098 #define USE_WORD_BEGIN_END          /* "<": word-begin, ">": word-end */
00099 #define USE_POSIX_REGION_OPTION     /* needed for POSIX API support */
00100 #define USE_FIND_LONGEST_SEARCH_ALL_OF_RANGE
00101 /* #define USE_COMBINATION_EXPLOSION_CHECK */     /* (X*)* */
00102 /* #define USE_MULTI_THREAD_SYSTEM */
00103 #define THREAD_SYSTEM_INIT      /* depend on thread system */
00104 #define THREAD_SYSTEM_END       /* depend on thread system */
00105 #define THREAD_ATOMIC_START     /* depend on thread system */
00106 #define THREAD_ATOMIC_END       /* depend on thread system */
00107 #define THREAD_PASS             /* depend on thread system */
00108 #ifndef xmalloc
00109 #define xmalloc     malloc
00110 #define xrealloc    realloc
00111 #define xcalloc     calloc
00112 #define xfree       free
00113 #endif
00114 
00115 #ifdef RUBY
00116 
00117 #define CHECK_INTERRUPT_IN_MATCH_AT rb_thread_check_ints()
00118 #define onig_st_init_table                  st_init_table
00119 #define onig_st_init_table_with_size        st_init_table_with_size
00120 #define onig_st_init_numtable               st_init_numtable
00121 #define onig_st_init_numtable_with_size     st_init_numtable_with_size
00122 #define onig_st_init_strtable               st_init_strtable
00123 #define onig_st_init_strtable_with_size     st_init_strtable_with_size
00124 #define onig_st_delete                      st_delete
00125 #define onig_st_delete_safe                 st_delete_safe
00126 #define onig_st_insert                      st_insert
00127 #define onig_st_lookup                      st_lookup
00128 #define onig_st_foreach                     st_foreach
00129 #define onig_st_add_direct                  st_add_direct
00130 #define onig_st_free_table                  st_free_table
00131 #define onig_st_cleanup_safe                st_cleanup_safe
00132 #define onig_st_copy                        st_copy
00133 #define onig_st_nothing_key_clone           st_nothing_key_clone
00134 #define onig_st_nothing_key_free            st_nothing_key_free
00135 #define onig_st_is_member                   st_is_member
00136 
00137 #define USE_UPPER_CASE_TABLE
00138 #else
00139 
00140 #define st_init_table                  onig_st_init_table
00141 #define st_init_table_with_size        onig_st_init_table_with_size
00142 #define st_init_numtable               onig_st_init_numtable
00143 #define st_init_numtable_with_size     onig_st_init_numtable_with_size
00144 #define st_init_strtable               onig_st_init_strtable
00145 #define st_init_strtable_with_size     onig_st_init_strtable_with_size
00146 #define st_delete                      onig_st_delete
00147 #define st_delete_safe                 onig_st_delete_safe
00148 #define st_insert                      onig_st_insert
00149 #define st_lookup                      onig_st_lookup
00150 #define st_foreach                     onig_st_foreach
00151 #define st_add_direct                  onig_st_add_direct
00152 #define st_free_table                  onig_st_free_table
00153 #define st_cleanup_safe                onig_st_cleanup_safe
00154 #define st_copy                        onig_st_copy
00155 #define st_nothing_key_clone           onig_st_nothing_key_clone
00156 #define st_nothing_key_free            onig_st_nothing_key_free
00157 /* */
00158 #define onig_st_is_member              st_is_member
00159 
00160 #define CHECK_INTERRUPT_IN_MATCH_AT
00161 
00162 #endif
00163 
00164 #define STATE_CHECK_STRING_THRESHOLD_LEN             7
00165 #define STATE_CHECK_BUFF_MAX_SIZE               0x4000
00166 
00167 #define THREAD_PASS_LIMIT_COUNT     8
00168 #define xmemset     memset
00169 #define xmemcpy     memcpy
00170 #define xmemmove    memmove
00171 
00172 #if defined(_WIN32) && !defined(__GNUC__)
00173 #define xalloca     _alloca
00174 #define xvsnprintf  _vsnprintf
00175 #else
00176 #define xalloca     alloca
00177 #define xvsnprintf  vsnprintf
00178 #endif
00179 
00180 
00181 #if defined(USE_RECOMPILE_API) && defined(USE_MULTI_THREAD_SYSTEM)
00182 #define ONIG_STATE_INC(reg) (reg)->state++
00183 #define ONIG_STATE_DEC(reg) (reg)->state--
00184 
00185 #define ONIG_STATE_INC_THREAD(reg) do {\
00186   THREAD_ATOMIC_START;\
00187   (reg)->state++;\
00188   THREAD_ATOMIC_END;\
00189 } while(0)
00190 #define ONIG_STATE_DEC_THREAD(reg) do {\
00191   THREAD_ATOMIC_START;\
00192   (reg)->state--;\
00193   THREAD_ATOMIC_END;\
00194 } while(0)
00195 #else
00196 #define ONIG_STATE_INC(reg)         /* Nothing */
00197 #define ONIG_STATE_DEC(reg)         /* Nothing */
00198 #define ONIG_STATE_INC_THREAD(reg)  /* Nothing */
00199 #define ONIG_STATE_DEC_THREAD(reg)  /* Nothing */
00200 #endif /* USE_RECOMPILE_API && USE_MULTI_THREAD_SYSTEM */
00201 
00202 #ifdef HAVE_STDLIB_H
00203 #include <stdlib.h>
00204 #endif
00205 
00206 #if defined(HAVE_ALLOCA_H) && (defined(_AIX) || !defined(__GNUC__))
00207 #include <alloca.h>
00208 #endif
00209 
00210 #ifdef HAVE_STRING_H
00211 # include <string.h>
00212 #else
00213 # include <strings.h>
00214 #endif
00215 
00216 #include <ctype.h>
00217 #ifdef HAVE_SYS_TYPES_H
00218 #include <sys/types.h>
00219 #endif
00220 
00221 #ifdef ONIG_DEBUG
00222 # include <stdio.h>
00223 #endif
00224 
00225 #include "regenc.h"
00226 
00227 #if defined __GNUC__ && __GNUC__ >= 4
00228 #pragma GCC visibility push(default)
00229 #endif
00230 
00231 #ifdef MIN
00232 #undef MIN
00233 #endif
00234 #ifdef MAX
00235 #undef MAX
00236 #endif
00237 #define MIN(a,b) (((a)>(b))?(b):(a))
00238 #define MAX(a,b) (((a)<(b))?(b):(a))
00239 
00240 #define IS_NULL(p)                    (((void*)(p)) == (void*)0)
00241 #define IS_NOT_NULL(p)                (((void*)(p)) != (void*)0)
00242 #define CHECK_NULL_RETURN(p)          if (IS_NULL(p)) return NULL
00243 #define CHECK_NULL_RETURN_MEMERR(p)   if (IS_NULL(p)) return ONIGERR_MEMORY
00244 #define NULL_UCHARP                   ((UChar* )0)
00245 
00246 #ifdef PLATFORM_UNALIGNED_WORD_ACCESS
00247 
00248 #define PLATFORM_GET_INC(val,p,type) do{\
00249   val  = *(type* )p;\
00250   (p) += sizeof(type);\
00251 } while(0)
00252 
00253 #else
00254 
00255 #define PLATFORM_GET_INC(val,p,type) do{\
00256   xmemcpy(&val, (p), sizeof(type));\
00257   (p) += sizeof(type);\
00258 } while(0)
00259 
00260 /* sizeof(OnigCodePoint) */
00261 #define WORD_ALIGNMENT_SIZE     SIZEOF_LONG
00262 
00263 #define GET_ALIGNMENT_PAD_SIZE(addr,pad_size) do {\
00264   (pad_size) = WORD_ALIGNMENT_SIZE \
00265                - ((uintptr_t )(addr) % WORD_ALIGNMENT_SIZE);\
00266   if ((pad_size) == WORD_ALIGNMENT_SIZE) (pad_size) = 0;\
00267 } while (0)
00268 
00269 #define ALIGNMENT_RIGHT(addr) do {\
00270   (addr) += (WORD_ALIGNMENT_SIZE - 1);\
00271   (addr) -= ((uintptr_t )(addr) % WORD_ALIGNMENT_SIZE);\
00272 } while (0)
00273 
00274 #endif /* PLATFORM_UNALIGNED_WORD_ACCESS */
00275 
00276 /* stack pop level */
00277 #define STACK_POP_LEVEL_FREE        0
00278 #define STACK_POP_LEVEL_MEM_START   1
00279 #define STACK_POP_LEVEL_ALL         2
00280 
00281 /* optimize flags */
00282 #define ONIG_OPTIMIZE_NONE              0
00283 #define ONIG_OPTIMIZE_EXACT             1   /* Slow Search */
00284 #define ONIG_OPTIMIZE_EXACT_BM          2   /* Boyer Moore Search */
00285 #define ONIG_OPTIMIZE_EXACT_BM_NOT_REV  3   /* BM   (but not simple match) */
00286 #define ONIG_OPTIMIZE_EXACT_IC          4   /* Slow Search (ignore case) */
00287 #define ONIG_OPTIMIZE_MAP               5   /* char map */
00288 
00289 /* bit status */
00290 typedef unsigned int  BitStatusType;
00291 
00292 #define BIT_STATUS_BITS_NUM          (sizeof(BitStatusType) * 8)
00293 #define BIT_STATUS_CLEAR(stats)      (stats) = 0
00294 #define BIT_STATUS_ON_ALL(stats)     (stats) = ~((BitStatusType )0)
00295 #define BIT_STATUS_AT(stats,n) \
00296   ((n) < (int )BIT_STATUS_BITS_NUM  ?  ((stats) & (1 << n)) : ((stats) & 1))
00297 
00298 #define BIT_STATUS_ON_AT(stats,n) do {\
00299     if ((n) < (int )BIT_STATUS_BITS_NUM)        \
00300     (stats) |= (1 << (n));\
00301   else\
00302     (stats) |= 1;\
00303 } while (0)
00304 
00305 #define BIT_STATUS_ON_AT_SIMPLE(stats,n) do {\
00306     if ((n) < (int )BIT_STATUS_BITS_NUM)\
00307     (stats) |= (1 << (n));\
00308 } while (0)
00309 
00310 
00311 #define INT_MAX_LIMIT           ((1UL << (SIZEOF_INT * 8 - 1)) - 1)
00312 
00313 #define DIGITVAL(code)    ((code) - '0')
00314 #define ODIGITVAL(code)   DIGITVAL(code)
00315 #define XDIGITVAL(enc,code) \
00316   (ONIGENC_IS_CODE_DIGIT(enc,code) ? DIGITVAL(code) \
00317    : (ONIGENC_IS_CODE_UPPER(enc,code) ? (code) - 'A' + 10 : (code) - 'a' + 10))
00318 
00319 #define IS_SINGLELINE(option)     ((option) & ONIG_OPTION_SINGLELINE)
00320 #define IS_MULTILINE(option)      ((option) & ONIG_OPTION_MULTILINE)
00321 #define IS_IGNORECASE(option)     ((option) & ONIG_OPTION_IGNORECASE)
00322 #define IS_EXTEND(option)         ((option) & ONIG_OPTION_EXTEND)
00323 #define IS_FIND_LONGEST(option)   ((option) & ONIG_OPTION_FIND_LONGEST)
00324 #define IS_FIND_NOT_EMPTY(option) ((option) & ONIG_OPTION_FIND_NOT_EMPTY)
00325 #define IS_FIND_CONDITION(option) ((option) & \
00326           (ONIG_OPTION_FIND_LONGEST | ONIG_OPTION_FIND_NOT_EMPTY))
00327 #define IS_NOTBOL(option)         ((option) & ONIG_OPTION_NOTBOL)
00328 #define IS_NOTEOL(option)         ((option) & ONIG_OPTION_NOTEOL)
00329 #define IS_POSIX_REGION(option)   ((option) & ONIG_OPTION_POSIX_REGION)
00330 
00331 /* OP_SET_OPTION is required for these options.
00332 #define IS_DYNAMIC_OPTION(option) \
00333   (((option) & (ONIG_OPTION_MULTILINE | ONIG_OPTION_IGNORECASE)) != 0)
00334 */
00335 /* ignore-case and multibyte status are included in compiled code. */
00336 #define IS_DYNAMIC_OPTION(option)  0
00337 
00338 #define DISABLE_CASE_FOLD_MULTI_CHAR(case_fold_flag) \
00339   ((case_fold_flag) & ~INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR)
00340 
00341 #define REPEAT_INFINITE         -1
00342 #define IS_REPEAT_INFINITE(n)   ((n) == REPEAT_INFINITE)
00343 
00344 /* bitset */
00345 #define BITS_PER_BYTE      8
00346 #define SINGLE_BYTE_SIZE   (1 << BITS_PER_BYTE)
00347 #define BITS_IN_ROOM       (sizeof(Bits) * BITS_PER_BYTE)
00348 #define BITSET_SIZE        (SINGLE_BYTE_SIZE / BITS_IN_ROOM)
00349 
00350 #ifdef PLATFORM_UNALIGNED_WORD_ACCESS
00351 typedef unsigned int   Bits;
00352 #else
00353 typedef unsigned char  Bits;
00354 #endif
00355 typedef Bits           BitSet[BITSET_SIZE];
00356 typedef Bits*          BitSetRef;
00357 
00358 #define SIZE_BITSET        (int)sizeof(BitSet)
00359 
00360 #define BITSET_CLEAR(bs) do {\
00361   int i;\
00362   for (i = 0; i < (int )BITSET_SIZE; i++) { (bs)[i] = 0; }      \
00363 } while (0)
00364 
00365 #define BS_ROOM(bs,pos)            (bs)[pos / BITS_IN_ROOM]
00366 #define BS_BIT(pos)                (1 << (pos % BITS_IN_ROOM))
00367 
00368 #define BITSET_AT(bs, pos)         (BS_ROOM(bs,pos) & BS_BIT(pos))
00369 #define BITSET_SET_BIT(bs, pos)     BS_ROOM(bs,pos) |= BS_BIT(pos)
00370 #define BITSET_CLEAR_BIT(bs, pos)   BS_ROOM(bs,pos) &= ~(BS_BIT(pos))
00371 #define BITSET_INVERT_BIT(bs, pos)  BS_ROOM(bs,pos) ^= BS_BIT(pos)
00372 
00373 /* bytes buffer */
00374 typedef struct _BBuf {
00375   UChar* p;
00376   unsigned int used;
00377   unsigned int alloc;
00378 } BBuf;
00379 
00380 #define BBUF_INIT(buf,size)    onig_bbuf_init((BBuf* )(buf), (size))
00381 
00382 #define BBUF_SIZE_INC(buf,inc) do{\
00383   (buf)->alloc += (inc);\
00384   (buf)->p = (UChar* )xrealloc((buf)->p, (buf)->alloc);\
00385   if (IS_NULL((buf)->p)) return(ONIGERR_MEMORY);\
00386 } while (0)
00387 
00388 #define BBUF_EXPAND(buf,low) do{\
00389   do { (buf)->alloc *= 2; } while ((buf)->alloc < (unsigned int )low);\
00390   (buf)->p = (UChar* )xrealloc((buf)->p, (buf)->alloc);\
00391   if (IS_NULL((buf)->p)) return(ONIGERR_MEMORY);\
00392 } while (0)
00393 
00394 #define BBUF_ENSURE_SIZE(buf,size) do{\
00395   unsigned int new_alloc = (buf)->alloc;\
00396   while (new_alloc < (unsigned int )(size)) { new_alloc *= 2; }\
00397   if ((buf)->alloc != new_alloc) {\
00398     (buf)->p = (UChar* )xrealloc((buf)->p, new_alloc);\
00399     if (IS_NULL((buf)->p)) return(ONIGERR_MEMORY);\
00400     (buf)->alloc = new_alloc;\
00401   }\
00402 } while (0)
00403 
00404 #define BBUF_WRITE(buf,pos,bytes,n) do{\
00405   int used = (pos) + (int)(n);\
00406   if ((buf)->alloc < (unsigned int )used) BBUF_EXPAND((buf),used);\
00407   xmemcpy((buf)->p + (pos), (bytes), (n));\
00408   if ((buf)->used < (unsigned int )used) (buf)->used = used;\
00409 } while (0)
00410 
00411 #define BBUF_WRITE1(buf,pos,byte) do{\
00412   int used = (pos) + 1;\
00413   if ((buf)->alloc < (unsigned int )used) BBUF_EXPAND((buf),used);\
00414   (buf)->p[(pos)] = (byte);\
00415   if ((buf)->used < (unsigned int )used) (buf)->used = used;\
00416 } while (0)
00417 
00418 #define BBUF_ADD(buf,bytes,n)       BBUF_WRITE((buf),(buf)->used,(bytes),(n))
00419 #define BBUF_ADD1(buf,byte)         BBUF_WRITE1((buf),(buf)->used,(byte))
00420 #define BBUF_GET_ADD_ADDRESS(buf)   ((buf)->p + (buf)->used)
00421 #define BBUF_GET_OFFSET_POS(buf)    ((buf)->used)
00422 
00423 /* from < to */
00424 #define BBUF_MOVE_RIGHT(buf,from,to,n) do {\
00425   if ((unsigned int )((to)+(n)) > (buf)->alloc) BBUF_EXPAND((buf),(to) + (n));\
00426   xmemmove((buf)->p + (to), (buf)->p + (from), (n));\
00427   if ((unsigned int )((to)+(n)) > (buf)->used) (buf)->used = (to) + (n);\
00428 } while (0)
00429 
00430 /* from > to */
00431 #define BBUF_MOVE_LEFT(buf,from,to,n) do {\
00432   xmemmove((buf)->p + (to), (buf)->p + (from), (n));\
00433 } while (0)
00434 
00435 /* from > to */
00436 #define BBUF_MOVE_LEFT_REDUCE(buf,from,to) do {\
00437   xmemmove((buf)->p + (to), (buf)->p + (from), (buf)->used - (from));\
00438   (buf)->used -= (from - to);\
00439 } while (0)
00440 
00441 #define BBUF_INSERT(buf,pos,bytes,n) do {\
00442   if (pos >= (buf)->used) {\
00443     BBUF_WRITE(buf,pos,bytes,n);\
00444   }\
00445   else {\
00446     BBUF_MOVE_RIGHT((buf),(pos),(pos) + (n),((buf)->used - (pos)));\
00447     xmemcpy((buf)->p + (pos), (bytes), (n));\
00448   }\
00449 } while (0)
00450 
00451 #define BBUF_GET_BYTE(buf, pos) (buf)->p[(pos)]
00452 
00453 
00454 #define ANCHOR_BEGIN_BUF        (1<<0)
00455 #define ANCHOR_BEGIN_LINE       (1<<1)
00456 #define ANCHOR_BEGIN_POSITION   (1<<2)
00457 #define ANCHOR_END_BUF          (1<<3)
00458 #define ANCHOR_SEMI_END_BUF     (1<<4)
00459 #define ANCHOR_END_LINE         (1<<5)
00460 
00461 #define ANCHOR_WORD_BOUND       (1<<6)
00462 #define ANCHOR_NOT_WORD_BOUND   (1<<7)
00463 #define ANCHOR_WORD_BEGIN       (1<<8)
00464 #define ANCHOR_WORD_END         (1<<9)
00465 #define ANCHOR_PREC_READ        (1<<10)
00466 #define ANCHOR_PREC_READ_NOT    (1<<11)
00467 #define ANCHOR_LOOK_BEHIND      (1<<12)
00468 #define ANCHOR_LOOK_BEHIND_NOT  (1<<13)
00469 
00470 #define ANCHOR_ANYCHAR_STAR     (1<<14)   /* ".*" optimize info */
00471 #define ANCHOR_ANYCHAR_STAR_ML  (1<<15)   /* ".*" optimize info (multi-line) */
00472 
00473 /* operation code */
00474 enum OpCode {
00475   OP_FINISH = 0,        /* matching process terminator (no more alternative) */
00476   OP_END    = 1,        /* pattern code terminator (success end) */
00477 
00478   OP_EXACT1 = 2,        /* single byte, N = 1 */
00479   OP_EXACT2,            /* single byte, N = 2 */
00480   OP_EXACT3,            /* single byte, N = 3 */
00481   OP_EXACT4,            /* single byte, N = 4 */
00482   OP_EXACT5,            /* single byte, N = 5 */
00483   OP_EXACTN,            /* single byte */
00484   OP_EXACTMB2N1,        /* mb-length = 2 N = 1 */
00485   OP_EXACTMB2N2,        /* mb-length = 2 N = 2 */
00486   OP_EXACTMB2N3,        /* mb-length = 2 N = 3 */
00487   OP_EXACTMB2N,         /* mb-length = 2 */
00488   OP_EXACTMB3N,         /* mb-length = 3 */
00489   OP_EXACTMBN,          /* other length */
00490 
00491   OP_EXACT1_IC,         /* single byte, N = 1, ignore case */
00492   OP_EXACTN_IC,         /* single byte,        ignore case */
00493 
00494   OP_CCLASS,
00495   OP_CCLASS_MB,
00496   OP_CCLASS_MIX,
00497   OP_CCLASS_NOT,
00498   OP_CCLASS_MB_NOT,
00499   OP_CCLASS_MIX_NOT,
00500   OP_CCLASS_NODE,       /* pointer to CClassNode node */
00501 
00502   OP_ANYCHAR,                 /* "."  */
00503   OP_ANYCHAR_ML,              /* "."  multi-line */
00504   OP_ANYCHAR_STAR,            /* ".*" */
00505   OP_ANYCHAR_ML_STAR,         /* ".*" multi-line */
00506   OP_ANYCHAR_STAR_PEEK_NEXT,
00507   OP_ANYCHAR_ML_STAR_PEEK_NEXT,
00508 
00509   OP_WORD,
00510   OP_NOT_WORD,
00511   OP_WORD_BOUND,
00512   OP_NOT_WORD_BOUND,
00513   OP_WORD_BEGIN,
00514   OP_WORD_END,
00515 
00516   OP_BEGIN_BUF,
00517   OP_END_BUF,
00518   OP_BEGIN_LINE,
00519   OP_END_LINE,
00520   OP_SEMI_END_BUF,
00521   OP_BEGIN_POSITION,
00522 
00523   OP_BACKREF1,
00524   OP_BACKREF2,
00525   OP_BACKREFN,
00526   OP_BACKREFN_IC,
00527   OP_BACKREF_MULTI,
00528   OP_BACKREF_MULTI_IC,
00529   OP_BACKREF_WITH_LEVEL,    /* \k<xxx+n>, \k<xxx-n> */
00530 
00531   OP_MEMORY_START,
00532   OP_MEMORY_START_PUSH,   /* push back-tracker to stack */
00533   OP_MEMORY_END_PUSH,     /* push back-tracker to stack */
00534   OP_MEMORY_END_PUSH_REC, /* push back-tracker to stack */
00535   OP_MEMORY_END,
00536   OP_MEMORY_END_REC,      /* push marker to stack */
00537 
00538   OP_FAIL,               /* pop stack and move */
00539   OP_JUMP,
00540   OP_PUSH,
00541   OP_POP,
00542   OP_PUSH_OR_JUMP_EXACT1,  /* if match exact then push, else jump. */
00543   OP_PUSH_IF_PEEK_NEXT,    /* if match exact then push, else none. */
00544   OP_REPEAT,               /* {n,m} */
00545   OP_REPEAT_NG,            /* {n,m}? (non greedy) */
00546   OP_REPEAT_INC,
00547   OP_REPEAT_INC_NG,        /* non greedy */
00548   OP_REPEAT_INC_SG,        /* search and get in stack */
00549   OP_REPEAT_INC_NG_SG,     /* search and get in stack (non greedy) */
00550   OP_NULL_CHECK_START,     /* null loop checker start */
00551   OP_NULL_CHECK_END,       /* null loop checker end   */
00552   OP_NULL_CHECK_END_MEMST, /* null loop checker end (with capture status) */
00553   OP_NULL_CHECK_END_MEMST_PUSH, /* with capture status and push check-end */
00554 
00555   OP_PUSH_POS,             /* (?=...)  start */
00556   OP_POP_POS,              /* (?=...)  end   */
00557   OP_PUSH_POS_NOT,         /* (?!...)  start */
00558   OP_FAIL_POS,             /* (?!...)  end   */
00559   OP_PUSH_STOP_BT,         /* (?>...)  start */
00560   OP_POP_STOP_BT,          /* (?>...)  end   */
00561   OP_LOOK_BEHIND,          /* (?<=...) start (no needs end opcode) */
00562   OP_PUSH_LOOK_BEHIND_NOT, /* (?<!...) start */
00563   OP_FAIL_LOOK_BEHIND_NOT, /* (?<!...) end   */
00564 
00565   OP_CALL,                 /* \g<name> */
00566   OP_RETURN,
00567 
00568   OP_STATE_CHECK_PUSH,         /* combination explosion check and push */
00569   OP_STATE_CHECK_PUSH_OR_JUMP, /* check ok -> push, else jump  */
00570   OP_STATE_CHECK,              /* check only */
00571   OP_STATE_CHECK_ANYCHAR_STAR,
00572   OP_STATE_CHECK_ANYCHAR_ML_STAR,
00573 
00574   /* no need: IS_DYNAMIC_OPTION() == 0 */
00575   OP_SET_OPTION_PUSH,    /* set option and push recover option */
00576   OP_SET_OPTION          /* set option */
00577 };
00578 
00579 typedef int RelAddrType;
00580 typedef int AbsAddrType;
00581 typedef int LengthType;
00582 typedef int RepeatNumType;
00583 typedef short int MemNumType;
00584 typedef short int StateCheckNumType;
00585 typedef void* PointerType;
00586 
00587 #define SIZE_OPCODE           1
00588 #define SIZE_RELADDR          (int)sizeof(RelAddrType)
00589 #define SIZE_ABSADDR          (int)sizeof(AbsAddrType)
00590 #define SIZE_LENGTH           (int)sizeof(LengthType)
00591 #define SIZE_MEMNUM           (int)sizeof(MemNumType)
00592 #define SIZE_STATE_CHECK_NUM  (int)sizeof(StateCheckNumType)
00593 #define SIZE_REPEATNUM        (int)sizeof(RepeatNumType)
00594 #define SIZE_OPTION           (int)sizeof(OnigOptionType)
00595 #define SIZE_CODE_POINT       (int)sizeof(OnigCodePoint)
00596 #define SIZE_POINTER          (int)sizeof(PointerType)
00597 
00598 
00599 #define GET_RELADDR_INC(addr,p)    PLATFORM_GET_INC(addr,   p, RelAddrType)
00600 #define GET_ABSADDR_INC(addr,p)    PLATFORM_GET_INC(addr,   p, AbsAddrType)
00601 #define GET_LENGTH_INC(len,p)      PLATFORM_GET_INC(len,    p, LengthType)
00602 #define GET_MEMNUM_INC(num,p)      PLATFORM_GET_INC(num,    p, MemNumType)
00603 #define GET_REPEATNUM_INC(num,p)   PLATFORM_GET_INC(num,    p, RepeatNumType)
00604 #define GET_OPTION_INC(option,p)   PLATFORM_GET_INC(option, p, OnigOptionType)
00605 #define GET_POINTER_INC(ptr,p)     PLATFORM_GET_INC(ptr,    p, PointerType)
00606 #define GET_STATE_CHECK_NUM_INC(num,p)  PLATFORM_GET_INC(num, p, StateCheckNumType)
00607 
00608 /* code point's address must be aligned address. */
00609 #define GET_CODE_POINT(code,p)   code = *((OnigCodePoint* )(p))
00610 #define GET_BYTE_INC(byte,p) do{\
00611   byte = *(p);\
00612   (p)++;\
00613 } while(0)
00614 
00615 
00616 /* op-code + arg size */
00617 #define SIZE_OP_ANYCHAR_STAR            SIZE_OPCODE
00618 #define SIZE_OP_ANYCHAR_STAR_PEEK_NEXT (SIZE_OPCODE + 1)
00619 #define SIZE_OP_JUMP                   (SIZE_OPCODE + SIZE_RELADDR)
00620 #define SIZE_OP_PUSH                   (SIZE_OPCODE + SIZE_RELADDR)
00621 #define SIZE_OP_POP                     SIZE_OPCODE
00622 #define SIZE_OP_PUSH_OR_JUMP_EXACT1    (SIZE_OPCODE + SIZE_RELADDR + 1)
00623 #define SIZE_OP_PUSH_IF_PEEK_NEXT      (SIZE_OPCODE + SIZE_RELADDR + 1)
00624 #define SIZE_OP_REPEAT_INC             (SIZE_OPCODE + SIZE_MEMNUM)
00625 #define SIZE_OP_REPEAT_INC_NG          (SIZE_OPCODE + SIZE_MEMNUM)
00626 #define SIZE_OP_PUSH_POS                SIZE_OPCODE
00627 #define SIZE_OP_PUSH_POS_NOT           (SIZE_OPCODE + SIZE_RELADDR)
00628 #define SIZE_OP_POP_POS                 SIZE_OPCODE
00629 #define SIZE_OP_FAIL_POS                SIZE_OPCODE
00630 #define SIZE_OP_SET_OPTION             (SIZE_OPCODE + SIZE_OPTION)
00631 #define SIZE_OP_SET_OPTION_PUSH        (SIZE_OPCODE + SIZE_OPTION)
00632 #define SIZE_OP_FAIL                    SIZE_OPCODE
00633 #define SIZE_OP_MEMORY_START           (SIZE_OPCODE + SIZE_MEMNUM)
00634 #define SIZE_OP_MEMORY_START_PUSH      (SIZE_OPCODE + SIZE_MEMNUM)
00635 #define SIZE_OP_MEMORY_END_PUSH        (SIZE_OPCODE + SIZE_MEMNUM)
00636 #define SIZE_OP_MEMORY_END_PUSH_REC    (SIZE_OPCODE + SIZE_MEMNUM)
00637 #define SIZE_OP_MEMORY_END             (SIZE_OPCODE + SIZE_MEMNUM)
00638 #define SIZE_OP_MEMORY_END_REC         (SIZE_OPCODE + SIZE_MEMNUM)
00639 #define SIZE_OP_PUSH_STOP_BT            SIZE_OPCODE
00640 #define SIZE_OP_POP_STOP_BT             SIZE_OPCODE
00641 #define SIZE_OP_NULL_CHECK_START       (SIZE_OPCODE + SIZE_MEMNUM)
00642 #define SIZE_OP_NULL_CHECK_END         (SIZE_OPCODE + SIZE_MEMNUM)
00643 #define SIZE_OP_LOOK_BEHIND            (SIZE_OPCODE + SIZE_LENGTH)
00644 #define SIZE_OP_PUSH_LOOK_BEHIND_NOT   (SIZE_OPCODE + SIZE_RELADDR + SIZE_LENGTH)
00645 #define SIZE_OP_FAIL_LOOK_BEHIND_NOT    SIZE_OPCODE
00646 #define SIZE_OP_CALL                   (SIZE_OPCODE + SIZE_ABSADDR)
00647 #define SIZE_OP_RETURN                  SIZE_OPCODE
00648 
00649 #ifdef USE_COMBINATION_EXPLOSION_CHECK
00650 #define SIZE_OP_STATE_CHECK            (SIZE_OPCODE + SIZE_STATE_CHECK_NUM)
00651 #define SIZE_OP_STATE_CHECK_PUSH       (SIZE_OPCODE + SIZE_STATE_CHECK_NUM + SIZE_RELADDR)
00652 #define SIZE_OP_STATE_CHECK_PUSH_OR_JUMP (SIZE_OPCODE + SIZE_STATE_CHECK_NUM + SIZE_RELADDR)
00653 #define SIZE_OP_STATE_CHECK_ANYCHAR_STAR (SIZE_OPCODE + SIZE_STATE_CHECK_NUM)
00654 #endif
00655 
00656 #define MC_ESC(syn)               (syn)->meta_char_table.esc
00657 #define MC_ANYCHAR(syn)           (syn)->meta_char_table.anychar
00658 #define MC_ANYTIME(syn)           (syn)->meta_char_table.anytime
00659 #define MC_ZERO_OR_ONE_TIME(syn)  (syn)->meta_char_table.zero_or_one_time
00660 #define MC_ONE_OR_MORE_TIME(syn)  (syn)->meta_char_table.one_or_more_time
00661 #define MC_ANYCHAR_ANYTIME(syn)   (syn)->meta_char_table.anychar_anytime
00662 
00663 #define IS_MC_ESC_CODE(code, syn) \
00664   ((code) == MC_ESC(syn) && \
00665    !IS_SYNTAX_OP2((syn), ONIG_SYN_OP2_INEFFECTIVE_ESCAPE))
00666 
00667 
00668 #define SYN_POSIX_COMMON_OP \
00669  ( ONIG_SYN_OP_DOT_ANYCHAR | ONIG_SYN_OP_POSIX_BRACKET | \
00670    ONIG_SYN_OP_DECIMAL_BACKREF | \
00671    ONIG_SYN_OP_BRACKET_CC | ONIG_SYN_OP_ASTERISK_ZERO_INF | \
00672    ONIG_SYN_OP_LINE_ANCHOR | \
00673    ONIG_SYN_OP_ESC_CONTROL_CHARS )
00674 
00675 #define SYN_GNU_REGEX_OP \
00676   ( ONIG_SYN_OP_DOT_ANYCHAR | ONIG_SYN_OP_BRACKET_CC | \
00677     ONIG_SYN_OP_POSIX_BRACKET | ONIG_SYN_OP_DECIMAL_BACKREF | \
00678     ONIG_SYN_OP_BRACE_INTERVAL | ONIG_SYN_OP_LPAREN_SUBEXP | \
00679     ONIG_SYN_OP_VBAR_ALT | \
00680     ONIG_SYN_OP_ASTERISK_ZERO_INF | ONIG_SYN_OP_PLUS_ONE_INF | \
00681     ONIG_SYN_OP_QMARK_ZERO_ONE | \
00682     ONIG_SYN_OP_ESC_AZ_BUF_ANCHOR | ONIG_SYN_OP_ESC_CAPITAL_G_BEGIN_ANCHOR | \
00683     ONIG_SYN_OP_ESC_W_WORD | \
00684     ONIG_SYN_OP_ESC_B_WORD_BOUND | ONIG_SYN_OP_ESC_LTGT_WORD_BEGIN_END | \
00685     ONIG_SYN_OP_ESC_S_WHITE_SPACE | ONIG_SYN_OP_ESC_D_DIGIT | \
00686     ONIG_SYN_OP_LINE_ANCHOR )
00687 
00688 #define SYN_GNU_REGEX_BV \
00689   ( ONIG_SYN_CONTEXT_INDEP_ANCHORS | ONIG_SYN_CONTEXT_INDEP_REPEAT_OPS | \
00690     ONIG_SYN_CONTEXT_INVALID_REPEAT_OPS | ONIG_SYN_ALLOW_INVALID_INTERVAL | \
00691     ONIG_SYN_BACKSLASH_ESCAPE_IN_CC | ONIG_SYN_ALLOW_DOUBLE_RANGE_OP_IN_CC )
00692 
00693 
00694 #define NCCLASS_FLAGS(cc)           ((cc)->flags)
00695 #define NCCLASS_FLAG_SET(cc,flag)    (NCCLASS_FLAGS(cc) |= (flag))
00696 #define NCCLASS_FLAG_CLEAR(cc,flag)  (NCCLASS_FLAGS(cc) &= ~(flag))
00697 #define IS_NCCLASS_FLAG_ON(cc,flag) ((NCCLASS_FLAGS(cc) & (flag)) != 0)
00698 
00699 /* cclass node */
00700 #define FLAG_NCCLASS_NOT           (1<<0)
00701 #define FLAG_NCCLASS_SHARE         (1<<1)
00702 
00703 #define NCCLASS_SET_NOT(nd)     NCCLASS_FLAG_SET(nd, FLAG_NCCLASS_NOT)
00704 #define NCCLASS_SET_SHARE(nd)   NCCLASS_FLAG_SET(nd, FLAG_NCCLASS_SHARE)
00705 #define NCCLASS_CLEAR_NOT(nd)   NCCLASS_FLAG_CLEAR(nd, FLAG_NCCLASS_NOT)
00706 #define IS_NCCLASS_NOT(nd)      IS_NCCLASS_FLAG_ON(nd, FLAG_NCCLASS_NOT)
00707 #define IS_NCCLASS_SHARE(nd)    IS_NCCLASS_FLAG_ON(nd, FLAG_NCCLASS_SHARE)
00708 
00709 typedef struct {
00710   int type;
00711   /* struct _Node* next; */
00712   /* unsigned int flags; */
00713 } NodeBase;
00714 
00715 typedef struct {
00716   NodeBase base;
00717   unsigned int flags;
00718   BitSet bs;
00719   BBuf*  mbuf;   /* multi-byte info or NULL */
00720 } CClassNode;
00721 
00722 typedef intptr_t OnigStackIndex;
00723 
00724 typedef struct _OnigStackType {
00725   unsigned int type;
00726   union {
00727     struct {
00728       UChar *pcode;      /* byte code position */
00729       UChar *pstr;       /* string position */
00730       UChar *pstr_prev;  /* previous char position of pstr */
00731 #ifdef USE_COMBINATION_EXPLOSION_CHECK
00732       unsigned int state_check;
00733 #endif
00734     } state;
00735     struct {
00736       int   count;       /* for OP_REPEAT_INC, OP_REPEAT_INC_NG */
00737       UChar *pcode;      /* byte code position (head of repeated target) */
00738       int   num;         /* repeat id */
00739     } repeat;
00740     struct {
00741       OnigStackIndex si;     /* index of stack */
00742     } repeat_inc;
00743     struct {
00744       int num;           /* memory num */
00745       UChar *pstr;       /* start/end position */
00746       /* Following information is setted, if this stack type is MEM-START */
00747       OnigStackIndex start;  /* prev. info (for backtrack  "(...)*" ) */
00748       OnigStackIndex end;    /* prev. info (for backtrack  "(...)*" ) */
00749     } mem;
00750     struct {
00751       int num;           /* null check id */
00752       UChar *pstr;       /* start position */
00753     } null_check;
00754 #ifdef USE_SUBEXP_CALL
00755     struct {
00756       UChar *ret_addr;   /* byte code position */
00757       int    num;        /* null check id */
00758       UChar *pstr;       /* string position */
00759     } call_frame;
00760 #endif
00761   } u;
00762 } OnigStackType;
00763 
00764 typedef struct {
00765   void* stack_p;
00766   size_t stack_n;
00767   OnigOptionType options;
00768   OnigRegion*    region;
00769   const UChar* start;   /* search start position (for \G: BEGIN_POSITION) */
00770 #ifdef USE_FIND_LONGEST_SEARCH_ALL_OF_RANGE
00771   int    best_len;      /* for ONIG_OPTION_FIND_LONGEST */
00772   UChar* best_s;
00773 #endif
00774 #ifdef USE_COMBINATION_EXPLOSION_CHECK
00775   void* state_check_buff;
00776   int   state_check_buff_size;
00777 #endif
00778 } OnigMatchArg;
00779 
00780 
00781 #define IS_CODE_SB_WORD(enc,code) \
00782   (ONIGENC_IS_CODE_ASCII(code) && ONIGENC_IS_CODE_WORD(enc,code))
00783 
00784 #ifdef ONIG_DEBUG
00785 
00786 typedef struct {
00787   short int opcode;
00788   const char* name;
00789   short int arg_type;
00790 } OnigOpInfoType;
00791 
00792 extern OnigOpInfoType OnigOpInfo[];
00793 
00794 /* extern void onig_print_compiled_byte_code P_((FILE* f, UChar* bp, UChar* bpend, UChar** nextp, OnigEncoding enc)); */
00795 
00796 #ifdef ONIG_DEBUG_STATISTICS
00797 extern void onig_statistics_init P_((void));
00798 extern void onig_print_statistics P_((FILE* f));
00799 #endif
00800 #endif
00801 
00802 extern UChar* onig_error_code_to_format P_((int code));
00803 extern void  onig_snprintf_with_pattern PV_((UChar buf[], int bufsize, OnigEncoding enc, UChar* pat, UChar* pat_end, const UChar *fmt, ...));
00804 extern int  onig_bbuf_init P_((BBuf* buf, OnigDistance size));
00805 extern int  onig_compile P_((regex_t* reg, const UChar* pattern, const UChar* pattern_end, OnigErrorInfo* einfo, const char *sourcefile, int sourceline));
00806 extern void onig_chain_reduce P_((regex_t* reg));
00807 extern void onig_chain_link_add P_((regex_t* to, regex_t* add));
00808 extern void onig_transfer P_((regex_t* to, regex_t* from));
00809 extern int  onig_is_code_in_cc P_((OnigEncoding enc, OnigCodePoint code, CClassNode* cc));
00810 extern int  onig_is_code_in_cc_len P_((int enclen, OnigCodePoint code, CClassNode* cc));
00811 
00812 /* strend hash */
00813 typedef void hash_table_type;
00814 #ifdef RUBY
00815 #include "ruby/st.h"
00816 typedef st_data_t hash_data_type;
00817 #else
00818 typedef unsigned long hash_data_type;
00819 #endif
00820 
00821 extern hash_table_type* onig_st_init_strend_table_with_size P_((st_index_t size));
00822 extern int onig_st_lookup_strend P_((hash_table_type* table, const UChar* str_key, const UChar* end_key, hash_data_type *value));
00823 extern int onig_st_insert_strend P_((hash_table_type* table, const UChar* str_key, const UChar* end_key, hash_data_type value));
00824 
00825 /* encoding property management */
00826 #define PROPERTY_LIST_ADD_PROP(Name, CR) \
00827   r = onigenc_property_list_add_property((UChar* )Name, CR,\
00828               &PropertyNameTable, &PropertyList, &PropertyListNum,\
00829               &PropertyListSize);\
00830   if (r != 0) goto end
00831 
00832 #define PROPERTY_LIST_INIT_CHECK \
00833   if (PropertyInited == 0) {\
00834     int r = onigenc_property_list_init(init_property_list);\
00835     if (r != 0) return r;\
00836   }
00837 
00838 extern int onigenc_property_list_add_property P_((UChar* name, const OnigCodePoint* prop, hash_table_type **table, const OnigCodePoint*** plist, int *pnum, int *psize));
00839 
00840 typedef int (*ONIGENC_INIT_PROPERTY_LIST_FUNC_TYPE)(void);
00841 
00842 extern int onigenc_property_list_init P_((ONIGENC_INIT_PROPERTY_LIST_FUNC_TYPE));
00843 
00844 extern size_t onig_memsize P_((const regex_t *reg));
00845 
00846 #if defined __GNUC__ && __GNUC__ >= 4
00847 #pragma GCC visibility pop
00848 #endif
00849 
00850 #endif /* ONIGURUMA_REGINT_H */
00851