1 /************************************************* 2 * Perl-Compatible Regular Expressions * 3 *************************************************/ 4 5 /* PCRE2 is a library of functions to support regular expressions whose syntax 6 and semantics are as close as possible to those of the Perl 5 language. 7 8 Written by Philip Hazel 9 Original API code Copyright (c) 1997-2012 University of Cambridge 10 New API code Copyright (c) 2016-2019 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 #ifndef PCRE2_INTERNAL_H_IDEMPOTENT_GUARD 42 #define PCRE2_INTERNAL_H_IDEMPOTENT_GUARD 43 44 /* We do not support both EBCDIC and Unicode at the same time. The "configure" 45 script prevents both being selected, but not everybody uses "configure". EBCDIC 46 is only supported for the 8-bit library, but the check for this has to be later 47 in this file, because the first part is not width-dependent, and is included by 48 pcre2test.c with CODE_UNIT_WIDTH == 0. */ 49 50 #if defined EBCDIC && defined SUPPORT_UNICODE 51 #error The use of both EBCDIC and SUPPORT_UNICODE is not supported. 52 #endif 53 54 /* Standard C headers */ 55 56 #include <ctype.h> 57 #include <limits.h> 58 #include <stddef.h> 59 #include <stdio.h> 60 #include <stdlib.h> 61 #include <string.h> 62 63 /* Macros to make boolean values more obvious. The #ifndef is to pacify 64 compiler warnings in environments where these macros are defined elsewhere. 65 Unfortunately, there is no way to do the same for the typedef. */ 66 67 typedef int BOOL; 68 #ifndef FALSE 69 #define FALSE 0 70 #define TRUE 1 71 #endif 72 73 /* Valgrind (memcheck) support */ 74 75 #ifdef SUPPORT_VALGRIND 76 #include <valgrind/memcheck.h> 77 #endif 78 79 /* -ftrivial-auto-var-init support supports initializing all local variables 80 to avoid some classes of bug, but this can cause an unacceptable slowdown 81 for large on-stack arrays in hot functions. This macro lets us annotate 82 such arrays. */ 83 84 #ifdef HAVE_ATTRIBUTE_UNINITIALIZED 85 #define PCRE2_KEEP_UNINITIALIZED __attribute__((uninitialized)) 86 #else 87 #define PCRE2_KEEP_UNINITIALIZED 88 #endif 89 90 /* Older versions of MSVC lack snprintf(). This define allows for 91 warning/error-free compilation and testing with MSVC compilers back to at least 92 MSVC 10/2010. Except for VC6 (which is missing some fundamentals and fails). */ 93 94 #if defined(_MSC_VER) && (_MSC_VER < 1900) 95 #define snprintf _snprintf 96 #endif 97 98 /* When compiling a DLL for Windows, the exported symbols have to be declared 99 using some MS magic. I found some useful information on this web page: 100 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the 101 information there, using __declspec(dllexport) without "extern" we have a 102 definition; with "extern" we have a declaration. The settings here override the 103 setting in pcre2.h (which is included below); it defines only PCRE2_EXP_DECL, 104 which is all that is needed for applications (they just import the symbols). We 105 use: 106 107 PCRE2_EXP_DECL for declarations 108 PCRE2_EXP_DEFN for definitions 109 110 The reason for wrapping this in #ifndef PCRE2_EXP_DECL is so that pcre2test, 111 which is an application, but needs to import this file in order to "peek" at 112 internals, can #include pcre2.h first to get an application's-eye view. 113 114 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon, 115 special-purpose environments) might want to stick other stuff in front of 116 exported symbols. That's why, in the non-Windows case, we set PCRE2_EXP_DEFN 117 only if it is not already set. */ 118 119 #ifndef PCRE2_EXP_DECL 120 # ifdef _WIN32 121 # ifndef PCRE2_STATIC 122 # define PCRE2_EXP_DECL extern __declspec(dllexport) 123 # define PCRE2_EXP_DEFN __declspec(dllexport) 124 # else 125 # define PCRE2_EXP_DECL extern 126 # define PCRE2_EXP_DEFN 127 # endif 128 # else 129 # ifdef __cplusplus 130 # define PCRE2_EXP_DECL extern "C" 131 # else 132 # define PCRE2_EXP_DECL extern 133 # endif 134 # ifndef PCRE2_EXP_DEFN 135 # define PCRE2_EXP_DEFN PCRE2_EXP_DECL 136 # endif 137 # endif 138 #endif 139 140 /* Include the public PCRE2 header and the definitions of UCP character 141 property values. This must follow the setting of PCRE2_EXP_DECL above. */ 142 143 #include "pcre2.h" 144 #include "pcre2_ucp.h" 145 146 /* When PCRE2 is compiled as a C++ library, the subject pointer can be replaced 147 with a custom type. This makes it possible, for example, to allow pcre2_match() 148 to process subject strings that are discontinuous by using a smart pointer 149 class. It must always be possible to inspect all of the subject string in 150 pcre2_match() because of the way it backtracks. */ 151 152 /* WARNING: This is as yet untested for PCRE2. */ 153 154 #ifdef CUSTOM_SUBJECT_PTR 155 #undef PCRE2_SPTR 156 #define PCRE2_SPTR CUSTOM_SUBJECT_PTR 157 #endif 158 159 /* When checking for integer overflow in pcre2_compile(), we need to handle 160 large integers. If a 64-bit integer type is available, we can use that. 161 Otherwise we have to cast to double, which of course requires floating point 162 arithmetic. Handle this by defining a macro for the appropriate type. */ 163 164 #if defined INT64_MAX || defined int64_t 165 #define INT64_OR_DOUBLE int64_t 166 #else 167 #define INT64_OR_DOUBLE double 168 #endif 169 170 /* External (in the C sense) functions and tables that are private to the 171 libraries are always referenced using the PRIV macro. This makes it possible 172 for pcre2test.c to include some of the source files from the libraries using a 173 different PRIV definition to avoid name clashes. It also makes it clear in the 174 code that a non-static object is being referenced. */ 175 176 #ifndef PRIV 177 #define PRIV(name) _pcre2_##name 178 #endif 179 180 /* When compiling for use with the Virtual Pascal compiler, these functions 181 need to have their names changed. PCRE2 must be compiled with the -DVPCOMPAT 182 option on the command line. */ 183 184 #ifdef VPCOMPAT 185 #define strlen(s) _strlen(s) 186 #define strncmp(s1,s2,m) _strncmp(s1,s2,m) 187 #define memcmp(s,c,n) _memcmp(s,c,n) 188 #define memcpy(d,s,n) _memcpy(d,s,n) 189 #define memmove(d,s,n) _memmove(d,s,n) 190 #define memset(s,c,n) _memset(s,c,n) 191 #else /* VPCOMPAT */ 192 193 /* Otherwise, to cope with SunOS4 and other systems that lack memmove(), define 194 a macro that calls an emulating function. */ 195 196 #ifndef HAVE_MEMMOVE 197 #undef memmove /* Some systems may have a macro */ 198 #define memmove(a, b, c) PRIV(memmove)(a, b, c) 199 #endif /* not HAVE_MEMMOVE */ 200 #endif /* not VPCOMPAT */ 201 202 /* This is an unsigned int value that no UTF character can ever have, as 203 Unicode doesn't go beyond 0x0010ffff. */ 204 205 #define NOTACHAR 0xffffffff 206 207 /* This is the largest valid UTF/Unicode code point. */ 208 209 #define MAX_UTF_CODE_POINT 0x10ffff 210 211 /* Compile-time positive error numbers (all except UTF errors, which are 212 negative) start at this value. It should probably never be changed, in case 213 some application is checking for specific numbers. There is a copy of this 214 #define in pcre2posix.c (which now no longer includes this file). Ideally, a 215 way of having a single definition should be found, but as the number is 216 unlikely to change, this is not a pressing issue. The original reason for 217 having a base other than 0 was to keep the absolute values of compile-time and 218 run-time error numbers numerically different, but in the event the code does 219 not rely on this. */ 220 221 #define COMPILE_ERROR_BASE 100 222 223 /* The initial frames vector for remembering backtracking points in 224 pcre2_match() is allocated on the system stack, of this size (bytes). The size 225 must be a multiple of sizeof(PCRE2_SPTR) in all environments, so making it a 226 multiple of 8 is best. Typical frame sizes are a few hundred bytes (it depends 227 on the number of capturing parentheses) so 20KiB handles quite a few frames. A 228 larger vector on the heap is obtained for patterns that need more frames. The 229 maximum size of this can be limited. */ 230 231 #define START_FRAMES_SIZE 20480 232 233 /* Similarly, for DFA matching, an initial internal workspace vector is 234 allocated on the stack. */ 235 236 #define DFA_START_RWS_SIZE 30720 237 238 /* Define the default BSR convention. */ 239 240 #ifdef BSR_ANYCRLF 241 #define BSR_DEFAULT PCRE2_BSR_ANYCRLF 242 #else 243 #define BSR_DEFAULT PCRE2_BSR_UNICODE 244 #endif 245 246 247 /* ---------------- Basic UTF-8 macros ---------------- */ 248 249 /* These UTF-8 macros are always defined because they are used in pcre2test for 250 handling wide characters in 16-bit and 32-bit modes, even if an 8-bit library 251 is not supported. */ 252 253 /* Tests whether a UTF-8 code point needs extra bytes to decode. */ 254 255 #define HASUTF8EXTRALEN(c) ((c) >= 0xc0) 256 257 /* The following macros were originally written in the form of loops that used 258 data from the tables whose names start with PRIV(utf8_table). They were 259 rewritten by a user so as not to use loops, because in some environments this 260 gives a significant performance advantage, and it seems never to do any harm. 261 */ 262 263 /* Base macro to pick up the remaining bytes of a UTF-8 character, not 264 advancing the pointer. */ 265 266 #define GETUTF8(c, eptr) \ 267 { \ 268 if ((c & 0x20u) == 0) \ 269 c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \ 270 else if ((c & 0x10u) == 0) \ 271 c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 272 else if ((c & 0x08u) == 0) \ 273 c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \ 274 ((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \ 275 else if ((c & 0x04u) == 0) \ 276 c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \ 277 ((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \ 278 (eptr[4] & 0x3fu); \ 279 else \ 280 c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \ 281 ((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \ 282 ((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \ 283 } 284 285 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing 286 the pointer. */ 287 288 #define GETUTF8INC(c, eptr) \ 289 { \ 290 if ((c & 0x20u) == 0) \ 291 c = ((c & 0x1fu) << 6) | (*eptr++ & 0x3fu); \ 292 else if ((c & 0x10u) == 0) \ 293 { \ 294 c = ((c & 0x0fu) << 12) | ((*eptr & 0x3fu) << 6) | (eptr[1] & 0x3fu); \ 295 eptr += 2; \ 296 } \ 297 else if ((c & 0x08u) == 0) \ 298 { \ 299 c = ((c & 0x07u) << 18) | ((*eptr & 0x3fu) << 12) | \ 300 ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 301 eptr += 3; \ 302 } \ 303 else if ((c & 0x04u) == 0) \ 304 { \ 305 c = ((c & 0x03u) << 24) | ((*eptr & 0x3fu) << 18) | \ 306 ((eptr[1] & 0x3fu) << 12) | ((eptr[2] & 0x3fu) << 6) | \ 307 (eptr[3] & 0x3fu); \ 308 eptr += 4; \ 309 } \ 310 else \ 311 { \ 312 c = ((c & 0x01u) << 30) | ((*eptr & 0x3fu) << 24) | \ 313 ((eptr[1] & 0x3fu) << 18) | ((eptr[2] & 0x3fu) << 12) | \ 314 ((eptr[3] & 0x3fu) << 6) | (eptr[4] & 0x3fu); \ 315 eptr += 5; \ 316 } \ 317 } 318 319 /* Base macro to pick up the remaining bytes of a UTF-8 character, not 320 advancing the pointer, incrementing the length. */ 321 322 #define GETUTF8LEN(c, eptr, len) \ 323 { \ 324 if ((c & 0x20u) == 0) \ 325 { \ 326 c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \ 327 len++; \ 328 } \ 329 else if ((c & 0x10u) == 0) \ 330 { \ 331 c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 332 len += 2; \ 333 } \ 334 else if ((c & 0x08u) == 0) \ 335 {\ 336 c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \ 337 ((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \ 338 len += 3; \ 339 } \ 340 else if ((c & 0x04u) == 0) \ 341 { \ 342 c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \ 343 ((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \ 344 (eptr[4] & 0x3fu); \ 345 len += 4; \ 346 } \ 347 else \ 348 {\ 349 c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \ 350 ((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \ 351 ((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \ 352 len += 5; \ 353 } \ 354 } 355 356 /* --------------- Whitespace macros ---------------- */ 357 358 /* Tests for Unicode horizontal and vertical whitespace characters must check a 359 number of different values. Using a switch statement for this generates the 360 fastest code (no loop, no memory access), and there are several places in the 361 interpreter code where this happens. In order to ensure that all the case lists 362 remain in step, we use macros so that there is only one place where the lists 363 are defined. 364 365 These values are also required as lists in pcre2_compile.c when processing \h, 366 \H, \v and \V in a character class. The lists are defined in pcre2_tables.c, 367 but macros that define the values are here so that all the definitions are 368 together. The lists must be in ascending character order, terminated by 369 NOTACHAR (which is 0xffffffff). 370 371 Any changes should ensure that the various macros are kept in step with each 372 other. NOTE: The values also appear in pcre2_jit_compile.c. */ 373 374 /* -------------- ASCII/Unicode environments -------------- */ 375 376 #ifndef EBCDIC 377 378 /* Character U+180E (Mongolian Vowel Separator) is not included in the list of 379 spaces in the Unicode file PropList.txt, and Perl does not recognize it as a 380 space. However, in many other sources it is listed as a space and has been in 381 PCRE (both APIs) for a long time. */ 382 383 #define HSPACE_LIST \ 384 CHAR_HT, CHAR_SPACE, CHAR_NBSP, \ 385 0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \ 386 0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \ 387 NOTACHAR 388 389 #define HSPACE_MULTIBYTE_CASES \ 390 case 0x1680: /* OGHAM SPACE MARK */ \ 391 case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ \ 392 case 0x2000: /* EN QUAD */ \ 393 case 0x2001: /* EM QUAD */ \ 394 case 0x2002: /* EN SPACE */ \ 395 case 0x2003: /* EM SPACE */ \ 396 case 0x2004: /* THREE-PER-EM SPACE */ \ 397 case 0x2005: /* FOUR-PER-EM SPACE */ \ 398 case 0x2006: /* SIX-PER-EM SPACE */ \ 399 case 0x2007: /* FIGURE SPACE */ \ 400 case 0x2008: /* PUNCTUATION SPACE */ \ 401 case 0x2009: /* THIN SPACE */ \ 402 case 0x200A: /* HAIR SPACE */ \ 403 case 0x202f: /* NARROW NO-BREAK SPACE */ \ 404 case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ \ 405 case 0x3000 /* IDEOGRAPHIC SPACE */ 406 407 #define HSPACE_BYTE_CASES \ 408 case CHAR_HT: \ 409 case CHAR_SPACE: \ 410 case CHAR_NBSP 411 412 #define HSPACE_CASES \ 413 HSPACE_BYTE_CASES: \ 414 HSPACE_MULTIBYTE_CASES 415 416 #define VSPACE_LIST \ 417 CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR 418 419 #define VSPACE_MULTIBYTE_CASES \ 420 case 0x2028: /* LINE SEPARATOR */ \ 421 case 0x2029 /* PARAGRAPH SEPARATOR */ 422 423 #define VSPACE_BYTE_CASES \ 424 case CHAR_LF: \ 425 case CHAR_VT: \ 426 case CHAR_FF: \ 427 case CHAR_CR: \ 428 case CHAR_NEL 429 430 #define VSPACE_CASES \ 431 VSPACE_BYTE_CASES: \ 432 VSPACE_MULTIBYTE_CASES 433 434 /* -------------- EBCDIC environments -------------- */ 435 436 #else 437 #define HSPACE_LIST CHAR_HT, CHAR_SPACE, CHAR_NBSP, NOTACHAR 438 439 #define HSPACE_BYTE_CASES \ 440 case CHAR_HT: \ 441 case CHAR_SPACE: \ 442 case CHAR_NBSP 443 444 #define HSPACE_CASES HSPACE_BYTE_CASES 445 446 #ifdef EBCDIC_NL25 447 #define VSPACE_LIST \ 448 CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR 449 #else 450 #define VSPACE_LIST \ 451 CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR 452 #endif 453 454 #define VSPACE_BYTE_CASES \ 455 case CHAR_LF: \ 456 case CHAR_VT: \ 457 case CHAR_FF: \ 458 case CHAR_CR: \ 459 case CHAR_NEL 460 461 #define VSPACE_CASES VSPACE_BYTE_CASES 462 #endif /* EBCDIC */ 463 464 /* -------------- End of whitespace macros -------------- */ 465 466 467 /* PCRE2 is able to support several different kinds of newline (CR, LF, CRLF, 468 "any" and "anycrlf" at present). The following macros are used to package up 469 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various 470 modules to indicate in which datablock the parameters exist, and what the 471 start/end of string field names are. */ 472 473 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */ 474 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */ 475 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */ 476 477 /* This macro checks for a newline at the given position */ 478 479 #define IS_NEWLINE(p) \ 480 ((NLBLOCK->nltype != NLTYPE_FIXED)? \ 481 ((p) < NLBLOCK->PSEND && \ 482 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \ 483 &(NLBLOCK->nllen), utf)) \ 484 : \ 485 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ 486 UCHAR21TEST(p) == NLBLOCK->nl[0] && \ 487 (NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1]) \ 488 ) \ 489 ) 490 491 /* This macro checks for a newline immediately preceding the given position */ 492 493 #define WAS_NEWLINE(p) \ 494 ((NLBLOCK->nltype != NLTYPE_FIXED)? \ 495 ((p) > NLBLOCK->PSSTART && \ 496 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ 497 &(NLBLOCK->nllen), utf)) \ 498 : \ 499 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ 500 UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \ 501 (NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \ 502 ) \ 503 ) 504 505 /* Private flags containing information about the compiled pattern. The first 506 three must not be changed, because whichever is set is actually the number of 507 bytes in a code unit in that mode. */ 508 509 #define PCRE2_MODE8 0x00000001 /* compiled in 8 bit mode */ 510 #define PCRE2_MODE16 0x00000002 /* compiled in 16 bit mode */ 511 #define PCRE2_MODE32 0x00000004 /* compiled in 32 bit mode */ 512 #define PCRE2_FIRSTSET 0x00000010 /* first_code unit is set */ 513 #define PCRE2_FIRSTCASELESS 0x00000020 /* caseless first code unit */ 514 #define PCRE2_FIRSTMAPSET 0x00000040 /* bitmap of first code units is set */ 515 #define PCRE2_LASTSET 0x00000080 /* last code unit is set */ 516 #define PCRE2_LASTCASELESS 0x00000100 /* caseless last code unit */ 517 #define PCRE2_STARTLINE 0x00000200 /* start after \n for multiline */ 518 #define PCRE2_JCHANGED 0x00000400 /* j option used in pattern */ 519 #define PCRE2_HASCRORLF 0x00000800 /* explicit \r or \n in pattern */ 520 #define PCRE2_HASTHEN 0x00001000 /* pattern contains (*THEN) */ 521 #define PCRE2_MATCH_EMPTY 0x00002000 /* pattern can match empty string */ 522 #define PCRE2_BSR_SET 0x00004000 /* BSR was set in the pattern */ 523 #define PCRE2_NL_SET 0x00008000 /* newline was set in the pattern */ 524 #define PCRE2_NOTEMPTY_SET 0x00010000 /* (*NOTEMPTY) used ) keep */ 525 #define PCRE2_NE_ATST_SET 0x00020000 /* (*NOTEMPTY_ATSTART) used) together */ 526 #define PCRE2_DEREF_TABLES 0x00040000 /* release character tables */ 527 #define PCRE2_NOJIT 0x00080000 /* (*NOJIT) used */ 528 #define PCRE2_HASBKPORX 0x00100000 /* contains \P, \p, or \X */ 529 #define PCRE2_DUPCAPUSED 0x00200000 /* contains (?| */ 530 #define PCRE2_HASBKC 0x00400000 /* contains \C */ 531 532 #define PCRE2_MODE_MASK (PCRE2_MODE8 | PCRE2_MODE16 | PCRE2_MODE32) 533 534 /* Values for the matchedby field in a match data block. */ 535 536 enum { PCRE2_MATCHEDBY_INTERPRETER, /* pcre2_match() */ 537 PCRE2_MATCHEDBY_DFA_INTERPRETER, /* pcre2_dfa_match() */ 538 PCRE2_MATCHEDBY_JIT }; /* pcre2_jit_match() */ 539 540 /* Values for the flags field in a match data block. */ 541 542 #define PCRE2_MD_COPIED_SUBJECT 0x01u 543 544 /* Magic number to provide a small check against being handed junk. */ 545 546 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ 547 548 /* The maximum remaining length of subject we are prepared to search for a 549 req_unit match. In 8-bit mode, memchr() is used and is much faster than the 550 search loop that has to be used in 16-bit and 32-bit modes. */ 551 552 #if PCRE2_CODE_UNIT_WIDTH == 8 553 #define REQ_CU_MAX 2000 554 #else 555 #define REQ_CU_MAX 1000 556 #endif 557 558 /* Offsets for the bitmap tables in the cbits set of tables. Each table 559 contains a set of bits for a class map. Some classes are built by combining 560 these tables. */ 561 562 #define cbit_space 0 /* [:space:] or \s */ 563 #define cbit_xdigit 32 /* [:xdigit:] */ 564 #define cbit_digit 64 /* [:digit:] or \d */ 565 #define cbit_upper 96 /* [:upper:] */ 566 #define cbit_lower 128 /* [:lower:] */ 567 #define cbit_word 160 /* [:word:] or \w */ 568 #define cbit_graph 192 /* [:graph:] */ 569 #define cbit_print 224 /* [:print:] */ 570 #define cbit_punct 256 /* [:punct:] */ 571 #define cbit_cntrl 288 /* [:cntrl:] */ 572 #define cbit_length 320 /* Length of the cbits table */ 573 574 /* Bit definitions for entries in the ctypes table. Do not change these values 575 without checking pcre2_jit_compile.c, which has an assertion to ensure that 576 ctype_word has the value 16. */ 577 578 #define ctype_space 0x01 579 #define ctype_letter 0x02 580 #define ctype_lcletter 0x04 581 #define ctype_digit 0x08 582 #define ctype_word 0x10 /* alphanumeric or '_' */ 583 584 /* Offsets of the various tables from the base tables pointer, and 585 total length of the tables. */ 586 587 #define lcc_offset 0 /* Lower case */ 588 #define fcc_offset 256 /* Flip case */ 589 #define cbits_offset 512 /* Character classes */ 590 #define ctypes_offset (cbits_offset + cbit_length) /* Character types */ 591 #define tables_length (ctypes_offset + 256) 592 593 594 /* -------------------- Character and string names ------------------------ */ 595 596 /* If PCRE2 is to support UTF-8 on EBCDIC platforms, we cannot use normal 597 character constants like '*' because the compiler would emit their EBCDIC code, 598 which is different from their ASCII/UTF-8 code. Instead we define macros for 599 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support 600 is enabled. When UTF-8 support is not enabled, the definitions use character 601 literals. Both character and string versions of each character are needed, and 602 there are some longer strings as well. 603 604 This means that, on EBCDIC platforms, the PCRE2 library can handle either 605 EBCDIC, or UTF-8, but not both. To support both in the same compiled library 606 would need different lookups depending on whether PCRE2_UTF was set or not. 607 This would make it impossible to use characters in switch/case statements, 608 which would reduce performance. For a theoretical use (which nobody has asked 609 for) in a minority area (EBCDIC platforms), this is not sensible. Any 610 application that did need both could compile two versions of the library, using 611 macros to give the functions distinct names. */ 612 613 #ifndef SUPPORT_UNICODE 614 615 /* UTF-8 support is not enabled; use the platform-dependent character literals 616 so that PCRE2 works in both ASCII and EBCDIC environments, but only in non-UTF 617 mode. Newline characters are problematic in EBCDIC. Though it has CR and LF 618 characters, a common practice has been to use its NL (0x15) character as the 619 line terminator in C-like processing environments. However, sometimes the LF 620 (0x25) character is used instead, according to this Unicode document: 621 622 http://unicode.org/standard/reports/tr13/tr13-5.html 623 624 PCRE2 defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25 625 instead. Whichever is *not* chosen is defined as NEL. 626 627 In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the 628 same code point. */ 629 630 #ifdef EBCDIC 631 632 #ifndef EBCDIC_NL25 633 #define CHAR_NL '\x15' 634 #define CHAR_NEL '\x25' 635 #define STR_NL "\x15" 636 #define STR_NEL "\x25" 637 #else 638 #define CHAR_NL '\x25' 639 #define CHAR_NEL '\x15' 640 #define STR_NL "\x25" 641 #define STR_NEL "\x15" 642 #endif 643 644 #define CHAR_LF CHAR_NL 645 #define STR_LF STR_NL 646 647 #define CHAR_ESC '\047' 648 #define CHAR_DEL '\007' 649 #define CHAR_NBSP ((unsigned char)'\x41') 650 #define STR_ESC "\047" 651 #define STR_DEL "\007" 652 653 #else /* Not EBCDIC */ 654 655 /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for 656 compatibility. NEL is the Unicode newline character; make sure it is 657 a positive value. */ 658 659 #define CHAR_LF '\n' 660 #define CHAR_NL CHAR_LF 661 #define CHAR_NEL ((unsigned char)'\x85') 662 #define CHAR_ESC '\033' 663 #define CHAR_DEL '\177' 664 #define CHAR_NBSP ((unsigned char)'\xa0') 665 666 #define STR_LF "\n" 667 #define STR_NL STR_LF 668 #define STR_NEL "\x85" 669 #define STR_ESC "\033" 670 #define STR_DEL "\177" 671 672 #endif /* EBCDIC */ 673 674 /* The remaining definitions work in both environments. */ 675 676 #define CHAR_NUL '\0' 677 #define CHAR_HT '\t' 678 #define CHAR_VT '\v' 679 #define CHAR_FF '\f' 680 #define CHAR_CR '\r' 681 #define CHAR_BS '\b' 682 #define CHAR_BEL '\a' 683 684 #define CHAR_SPACE ' ' 685 #define CHAR_EXCLAMATION_MARK '!' 686 #define CHAR_QUOTATION_MARK '"' 687 #define CHAR_NUMBER_SIGN '#' 688 #define CHAR_DOLLAR_SIGN '$' 689 #define CHAR_PERCENT_SIGN '%' 690 #define CHAR_AMPERSAND '&' 691 #define CHAR_APOSTROPHE '\'' 692 #define CHAR_LEFT_PARENTHESIS '(' 693 #define CHAR_RIGHT_PARENTHESIS ')' 694 #define CHAR_ASTERISK '*' 695 #define CHAR_PLUS '+' 696 #define CHAR_COMMA ',' 697 #define CHAR_MINUS '-' 698 #define CHAR_DOT '.' 699 #define CHAR_SLASH '/' 700 #define CHAR_0 '0' 701 #define CHAR_1 '1' 702 #define CHAR_2 '2' 703 #define CHAR_3 '3' 704 #define CHAR_4 '4' 705 #define CHAR_5 '5' 706 #define CHAR_6 '6' 707 #define CHAR_7 '7' 708 #define CHAR_8 '8' 709 #define CHAR_9 '9' 710 #define CHAR_COLON ':' 711 #define CHAR_SEMICOLON ';' 712 #define CHAR_LESS_THAN_SIGN '<' 713 #define CHAR_EQUALS_SIGN '=' 714 #define CHAR_GREATER_THAN_SIGN '>' 715 #define CHAR_QUESTION_MARK '?' 716 #define CHAR_COMMERCIAL_AT '@' 717 #define CHAR_A 'A' 718 #define CHAR_B 'B' 719 #define CHAR_C 'C' 720 #define CHAR_D 'D' 721 #define CHAR_E 'E' 722 #define CHAR_F 'F' 723 #define CHAR_G 'G' 724 #define CHAR_H 'H' 725 #define CHAR_I 'I' 726 #define CHAR_J 'J' 727 #define CHAR_K 'K' 728 #define CHAR_L 'L' 729 #define CHAR_M 'M' 730 #define CHAR_N 'N' 731 #define CHAR_O 'O' 732 #define CHAR_P 'P' 733 #define CHAR_Q 'Q' 734 #define CHAR_R 'R' 735 #define CHAR_S 'S' 736 #define CHAR_T 'T' 737 #define CHAR_U 'U' 738 #define CHAR_V 'V' 739 #define CHAR_W 'W' 740 #define CHAR_X 'X' 741 #define CHAR_Y 'Y' 742 #define CHAR_Z 'Z' 743 #define CHAR_LEFT_SQUARE_BRACKET '[' 744 #define CHAR_BACKSLASH '\\' 745 #define CHAR_RIGHT_SQUARE_BRACKET ']' 746 #define CHAR_CIRCUMFLEX_ACCENT '^' 747 #define CHAR_UNDERSCORE '_' 748 #define CHAR_GRAVE_ACCENT '`' 749 #define CHAR_a 'a' 750 #define CHAR_b 'b' 751 #define CHAR_c 'c' 752 #define CHAR_d 'd' 753 #define CHAR_e 'e' 754 #define CHAR_f 'f' 755 #define CHAR_g 'g' 756 #define CHAR_h 'h' 757 #define CHAR_i 'i' 758 #define CHAR_j 'j' 759 #define CHAR_k 'k' 760 #define CHAR_l 'l' 761 #define CHAR_m 'm' 762 #define CHAR_n 'n' 763 #define CHAR_o 'o' 764 #define CHAR_p 'p' 765 #define CHAR_q 'q' 766 #define CHAR_r 'r' 767 #define CHAR_s 's' 768 #define CHAR_t 't' 769 #define CHAR_u 'u' 770 #define CHAR_v 'v' 771 #define CHAR_w 'w' 772 #define CHAR_x 'x' 773 #define CHAR_y 'y' 774 #define CHAR_z 'z' 775 #define CHAR_LEFT_CURLY_BRACKET '{' 776 #define CHAR_VERTICAL_LINE '|' 777 #define CHAR_RIGHT_CURLY_BRACKET '}' 778 #define CHAR_TILDE '~' 779 780 #define STR_HT "\t" 781 #define STR_VT "\v" 782 #define STR_FF "\f" 783 #define STR_CR "\r" 784 #define STR_BS "\b" 785 #define STR_BEL "\a" 786 787 #define STR_SPACE " " 788 #define STR_EXCLAMATION_MARK "!" 789 #define STR_QUOTATION_MARK "\"" 790 #define STR_NUMBER_SIGN "#" 791 #define STR_DOLLAR_SIGN "$" 792 #define STR_PERCENT_SIGN "%" 793 #define STR_AMPERSAND "&" 794 #define STR_APOSTROPHE "'" 795 #define STR_LEFT_PARENTHESIS "(" 796 #define STR_RIGHT_PARENTHESIS ")" 797 #define STR_ASTERISK "*" 798 #define STR_PLUS "+" 799 #define STR_COMMA "," 800 #define STR_MINUS "-" 801 #define STR_DOT "." 802 #define STR_SLASH "/" 803 #define STR_0 "0" 804 #define STR_1 "1" 805 #define STR_2 "2" 806 #define STR_3 "3" 807 #define STR_4 "4" 808 #define STR_5 "5" 809 #define STR_6 "6" 810 #define STR_7 "7" 811 #define STR_8 "8" 812 #define STR_9 "9" 813 #define STR_COLON ":" 814 #define STR_SEMICOLON ";" 815 #define STR_LESS_THAN_SIGN "<" 816 #define STR_EQUALS_SIGN "=" 817 #define STR_GREATER_THAN_SIGN ">" 818 #define STR_QUESTION_MARK "?" 819 #define STR_COMMERCIAL_AT "@" 820 #define STR_A "A" 821 #define STR_B "B" 822 #define STR_C "C" 823 #define STR_D "D" 824 #define STR_E "E" 825 #define STR_F "F" 826 #define STR_G "G" 827 #define STR_H "H" 828 #define STR_I "I" 829 #define STR_J "J" 830 #define STR_K "K" 831 #define STR_L "L" 832 #define STR_M "M" 833 #define STR_N "N" 834 #define STR_O "O" 835 #define STR_P "P" 836 #define STR_Q "Q" 837 #define STR_R "R" 838 #define STR_S "S" 839 #define STR_T "T" 840 #define STR_U "U" 841 #define STR_V "V" 842 #define STR_W "W" 843 #define STR_X "X" 844 #define STR_Y "Y" 845 #define STR_Z "Z" 846 #define STR_LEFT_SQUARE_BRACKET "[" 847 #define STR_BACKSLASH "\\" 848 #define STR_RIGHT_SQUARE_BRACKET "]" 849 #define STR_CIRCUMFLEX_ACCENT "^" 850 #define STR_UNDERSCORE "_" 851 #define STR_GRAVE_ACCENT "`" 852 #define STR_a "a" 853 #define STR_b "b" 854 #define STR_c "c" 855 #define STR_d "d" 856 #define STR_e "e" 857 #define STR_f "f" 858 #define STR_g "g" 859 #define STR_h "h" 860 #define STR_i "i" 861 #define STR_j "j" 862 #define STR_k "k" 863 #define STR_l "l" 864 #define STR_m "m" 865 #define STR_n "n" 866 #define STR_o "o" 867 #define STR_p "p" 868 #define STR_q "q" 869 #define STR_r "r" 870 #define STR_s "s" 871 #define STR_t "t" 872 #define STR_u "u" 873 #define STR_v "v" 874 #define STR_w "w" 875 #define STR_x "x" 876 #define STR_y "y" 877 #define STR_z "z" 878 #define STR_LEFT_CURLY_BRACKET "{" 879 #define STR_VERTICAL_LINE "|" 880 #define STR_RIGHT_CURLY_BRACKET "}" 881 #define STR_TILDE "~" 882 883 #define STRING_ACCEPT0 "ACCEPT\0" 884 #define STRING_COMMIT0 "COMMIT\0" 885 #define STRING_F0 "F\0" 886 #define STRING_FAIL0 "FAIL\0" 887 #define STRING_MARK0 "MARK\0" 888 #define STRING_PRUNE0 "PRUNE\0" 889 #define STRING_SKIP0 "SKIP\0" 890 #define STRING_THEN "THEN" 891 892 #define STRING_atomic0 "atomic\0" 893 #define STRING_pla0 "pla\0" 894 #define STRING_plb0 "plb\0" 895 #define STRING_nla0 "nla\0" 896 #define STRING_nlb0 "nlb\0" 897 #define STRING_sr0 "sr\0" 898 #define STRING_asr0 "asr\0" 899 #define STRING_positive_lookahead0 "positive_lookahead\0" 900 #define STRING_positive_lookbehind0 "positive_lookbehind\0" 901 #define STRING_negative_lookahead0 "negative_lookahead\0" 902 #define STRING_negative_lookbehind0 "negative_lookbehind\0" 903 #define STRING_script_run0 "script_run\0" 904 #define STRING_atomic_script_run "atomic_script_run" 905 906 #define STRING_alpha0 "alpha\0" 907 #define STRING_lower0 "lower\0" 908 #define STRING_upper0 "upper\0" 909 #define STRING_alnum0 "alnum\0" 910 #define STRING_ascii0 "ascii\0" 911 #define STRING_blank0 "blank\0" 912 #define STRING_cntrl0 "cntrl\0" 913 #define STRING_digit0 "digit\0" 914 #define STRING_graph0 "graph\0" 915 #define STRING_print0 "print\0" 916 #define STRING_punct0 "punct\0" 917 #define STRING_space0 "space\0" 918 #define STRING_word0 "word\0" 919 #define STRING_xdigit "xdigit" 920 921 #define STRING_DEFINE "DEFINE" 922 #define STRING_VERSION "VERSION" 923 #define STRING_WEIRD_STARTWORD "[:<:]]" 924 #define STRING_WEIRD_ENDWORD "[:>:]]" 925 926 #define STRING_CR_RIGHTPAR "CR)" 927 #define STRING_LF_RIGHTPAR "LF)" 928 #define STRING_CRLF_RIGHTPAR "CRLF)" 929 #define STRING_ANY_RIGHTPAR "ANY)" 930 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)" 931 #define STRING_NUL_RIGHTPAR "NUL)" 932 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)" 933 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)" 934 #define STRING_UTF8_RIGHTPAR "UTF8)" 935 #define STRING_UTF16_RIGHTPAR "UTF16)" 936 #define STRING_UTF32_RIGHTPAR "UTF32)" 937 #define STRING_UTF_RIGHTPAR "UTF)" 938 #define STRING_UCP_RIGHTPAR "UCP)" 939 #define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)" 940 #define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR "NO_DOTSTAR_ANCHOR)" 941 #define STRING_NO_JIT_RIGHTPAR "NO_JIT)" 942 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)" 943 #define STRING_NOTEMPTY_RIGHTPAR "NOTEMPTY)" 944 #define STRING_NOTEMPTY_ATSTART_RIGHTPAR "NOTEMPTY_ATSTART)" 945 #define STRING_LIMIT_HEAP_EQ "LIMIT_HEAP=" 946 #define STRING_LIMIT_MATCH_EQ "LIMIT_MATCH=" 947 #define STRING_LIMIT_DEPTH_EQ "LIMIT_DEPTH=" 948 #define STRING_LIMIT_RECURSION_EQ "LIMIT_RECURSION=" 949 #define STRING_MARK "MARK" 950 951 #else /* SUPPORT_UNICODE */ 952 953 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This 954 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode 955 only. */ 956 957 #define CHAR_HT '\011' 958 #define CHAR_VT '\013' 959 #define CHAR_FF '\014' 960 #define CHAR_CR '\015' 961 #define CHAR_LF '\012' 962 #define CHAR_NL CHAR_LF 963 #define CHAR_NEL ((unsigned char)'\x85') 964 #define CHAR_BS '\010' 965 #define CHAR_BEL '\007' 966 #define CHAR_ESC '\033' 967 #define CHAR_DEL '\177' 968 969 #define CHAR_NUL '\0' 970 #define CHAR_SPACE '\040' 971 #define CHAR_EXCLAMATION_MARK '\041' 972 #define CHAR_QUOTATION_MARK '\042' 973 #define CHAR_NUMBER_SIGN '\043' 974 #define CHAR_DOLLAR_SIGN '\044' 975 #define CHAR_PERCENT_SIGN '\045' 976 #define CHAR_AMPERSAND '\046' 977 #define CHAR_APOSTROPHE '\047' 978 #define CHAR_LEFT_PARENTHESIS '\050' 979 #define CHAR_RIGHT_PARENTHESIS '\051' 980 #define CHAR_ASTERISK '\052' 981 #define CHAR_PLUS '\053' 982 #define CHAR_COMMA '\054' 983 #define CHAR_MINUS '\055' 984 #define CHAR_DOT '\056' 985 #define CHAR_SLASH '\057' 986 #define CHAR_0 '\060' 987 #define CHAR_1 '\061' 988 #define CHAR_2 '\062' 989 #define CHAR_3 '\063' 990 #define CHAR_4 '\064' 991 #define CHAR_5 '\065' 992 #define CHAR_6 '\066' 993 #define CHAR_7 '\067' 994 #define CHAR_8 '\070' 995 #define CHAR_9 '\071' 996 #define CHAR_COLON '\072' 997 #define CHAR_SEMICOLON '\073' 998 #define CHAR_LESS_THAN_SIGN '\074' 999 #define CHAR_EQUALS_SIGN '\075' 1000 #define CHAR_GREATER_THAN_SIGN '\076' 1001 #define CHAR_QUESTION_MARK '\077' 1002 #define CHAR_COMMERCIAL_AT '\100' 1003 #define CHAR_A '\101' 1004 #define CHAR_B '\102' 1005 #define CHAR_C '\103' 1006 #define CHAR_D '\104' 1007 #define CHAR_E '\105' 1008 #define CHAR_F '\106' 1009 #define CHAR_G '\107' 1010 #define CHAR_H '\110' 1011 #define CHAR_I '\111' 1012 #define CHAR_J '\112' 1013 #define CHAR_K '\113' 1014 #define CHAR_L '\114' 1015 #define CHAR_M '\115' 1016 #define CHAR_N '\116' 1017 #define CHAR_O '\117' 1018 #define CHAR_P '\120' 1019 #define CHAR_Q '\121' 1020 #define CHAR_R '\122' 1021 #define CHAR_S '\123' 1022 #define CHAR_T '\124' 1023 #define CHAR_U '\125' 1024 #define CHAR_V '\126' 1025 #define CHAR_W '\127' 1026 #define CHAR_X '\130' 1027 #define CHAR_Y '\131' 1028 #define CHAR_Z '\132' 1029 #define CHAR_LEFT_SQUARE_BRACKET '\133' 1030 #define CHAR_BACKSLASH '\134' 1031 #define CHAR_RIGHT_SQUARE_BRACKET '\135' 1032 #define CHAR_CIRCUMFLEX_ACCENT '\136' 1033 #define CHAR_UNDERSCORE '\137' 1034 #define CHAR_GRAVE_ACCENT '\140' 1035 #define CHAR_a '\141' 1036 #define CHAR_b '\142' 1037 #define CHAR_c '\143' 1038 #define CHAR_d '\144' 1039 #define CHAR_e '\145' 1040 #define CHAR_f '\146' 1041 #define CHAR_g '\147' 1042 #define CHAR_h '\150' 1043 #define CHAR_i '\151' 1044 #define CHAR_j '\152' 1045 #define CHAR_k '\153' 1046 #define CHAR_l '\154' 1047 #define CHAR_m '\155' 1048 #define CHAR_n '\156' 1049 #define CHAR_o '\157' 1050 #define CHAR_p '\160' 1051 #define CHAR_q '\161' 1052 #define CHAR_r '\162' 1053 #define CHAR_s '\163' 1054 #define CHAR_t '\164' 1055 #define CHAR_u '\165' 1056 #define CHAR_v '\166' 1057 #define CHAR_w '\167' 1058 #define CHAR_x '\170' 1059 #define CHAR_y '\171' 1060 #define CHAR_z '\172' 1061 #define CHAR_LEFT_CURLY_BRACKET '\173' 1062 #define CHAR_VERTICAL_LINE '\174' 1063 #define CHAR_RIGHT_CURLY_BRACKET '\175' 1064 #define CHAR_TILDE '\176' 1065 #define CHAR_NBSP ((unsigned char)'\xa0') 1066 1067 #define STR_HT "\011" 1068 #define STR_VT "\013" 1069 #define STR_FF "\014" 1070 #define STR_CR "\015" 1071 #define STR_NL "\012" 1072 #define STR_BS "\010" 1073 #define STR_BEL "\007" 1074 #define STR_ESC "\033" 1075 #define STR_DEL "\177" 1076 1077 #define STR_SPACE "\040" 1078 #define STR_EXCLAMATION_MARK "\041" 1079 #define STR_QUOTATION_MARK "\042" 1080 #define STR_NUMBER_SIGN "\043" 1081 #define STR_DOLLAR_SIGN "\044" 1082 #define STR_PERCENT_SIGN "\045" 1083 #define STR_AMPERSAND "\046" 1084 #define STR_APOSTROPHE "\047" 1085 #define STR_LEFT_PARENTHESIS "\050" 1086 #define STR_RIGHT_PARENTHESIS "\051" 1087 #define STR_ASTERISK "\052" 1088 #define STR_PLUS "\053" 1089 #define STR_COMMA "\054" 1090 #define STR_MINUS "\055" 1091 #define STR_DOT "\056" 1092 #define STR_SLASH "\057" 1093 #define STR_0 "\060" 1094 #define STR_1 "\061" 1095 #define STR_2 "\062" 1096 #define STR_3 "\063" 1097 #define STR_4 "\064" 1098 #define STR_5 "\065" 1099 #define STR_6 "\066" 1100 #define STR_7 "\067" 1101 #define STR_8 "\070" 1102 #define STR_9 "\071" 1103 #define STR_COLON "\072" 1104 #define STR_SEMICOLON "\073" 1105 #define STR_LESS_THAN_SIGN "\074" 1106 #define STR_EQUALS_SIGN "\075" 1107 #define STR_GREATER_THAN_SIGN "\076" 1108 #define STR_QUESTION_MARK "\077" 1109 #define STR_COMMERCIAL_AT "\100" 1110 #define STR_A "\101" 1111 #define STR_B "\102" 1112 #define STR_C "\103" 1113 #define STR_D "\104" 1114 #define STR_E "\105" 1115 #define STR_F "\106" 1116 #define STR_G "\107" 1117 #define STR_H "\110" 1118 #define STR_I "\111" 1119 #define STR_J "\112" 1120 #define STR_K "\113" 1121 #define STR_L "\114" 1122 #define STR_M "\115" 1123 #define STR_N "\116" 1124 #define STR_O "\117" 1125 #define STR_P "\120" 1126 #define STR_Q "\121" 1127 #define STR_R "\122" 1128 #define STR_S "\123" 1129 #define STR_T "\124" 1130 #define STR_U "\125" 1131 #define STR_V "\126" 1132 #define STR_W "\127" 1133 #define STR_X "\130" 1134 #define STR_Y "\131" 1135 #define STR_Z "\132" 1136 #define STR_LEFT_SQUARE_BRACKET "\133" 1137 #define STR_BACKSLASH "\134" 1138 #define STR_RIGHT_SQUARE_BRACKET "\135" 1139 #define STR_CIRCUMFLEX_ACCENT "\136" 1140 #define STR_UNDERSCORE "\137" 1141 #define STR_GRAVE_ACCENT "\140" 1142 #define STR_a "\141" 1143 #define STR_b "\142" 1144 #define STR_c "\143" 1145 #define STR_d "\144" 1146 #define STR_e "\145" 1147 #define STR_f "\146" 1148 #define STR_g "\147" 1149 #define STR_h "\150" 1150 #define STR_i "\151" 1151 #define STR_j "\152" 1152 #define STR_k "\153" 1153 #define STR_l "\154" 1154 #define STR_m "\155" 1155 #define STR_n "\156" 1156 #define STR_o "\157" 1157 #define STR_p "\160" 1158 #define STR_q "\161" 1159 #define STR_r "\162" 1160 #define STR_s "\163" 1161 #define STR_t "\164" 1162 #define STR_u "\165" 1163 #define STR_v "\166" 1164 #define STR_w "\167" 1165 #define STR_x "\170" 1166 #define STR_y "\171" 1167 #define STR_z "\172" 1168 #define STR_LEFT_CURLY_BRACKET "\173" 1169 #define STR_VERTICAL_LINE "\174" 1170 #define STR_RIGHT_CURLY_BRACKET "\175" 1171 #define STR_TILDE "\176" 1172 1173 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0" 1174 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0" 1175 #define STRING_F0 STR_F "\0" 1176 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0" 1177 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0" 1178 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0" 1179 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0" 1180 #define STRING_THEN STR_T STR_H STR_E STR_N 1181 1182 #define STRING_atomic0 STR_a STR_t STR_o STR_m STR_i STR_c "\0" 1183 #define STRING_pla0 STR_p STR_l STR_a "\0" 1184 #define STRING_plb0 STR_p STR_l STR_b "\0" 1185 #define STRING_nla0 STR_n STR_l STR_a "\0" 1186 #define STRING_nlb0 STR_n STR_l STR_b "\0" 1187 #define STRING_sr0 STR_s STR_r "\0" 1188 #define STRING_asr0 STR_a STR_s STR_r "\0" 1189 #define STRING_positive_lookahead0 STR_p STR_o STR_s STR_i STR_t STR_i STR_v STR_e STR_UNDERSCORE STR_l STR_o STR_o STR_k STR_a STR_h STR_e STR_a STR_d "\0" 1190 #define STRING_positive_lookbehind0 STR_p STR_o STR_s STR_i STR_t STR_i STR_v STR_e STR_UNDERSCORE STR_l STR_o STR_o STR_k STR_b STR_e STR_h STR_i STR_n STR_d "\0" 1191 #define STRING_negative_lookahead0 STR_n STR_e STR_g STR_a STR_t STR_i STR_v STR_e STR_UNDERSCORE STR_l STR_o STR_o STR_k STR_a STR_h STR_e STR_a STR_d "\0" 1192 #define STRING_negative_lookbehind0 STR_n STR_e STR_g STR_a STR_t STR_i STR_v STR_e STR_UNDERSCORE STR_l STR_o STR_o STR_k STR_b STR_e STR_h STR_i STR_n STR_d "\0" 1193 #define STRING_script_run0 STR_s STR_c STR_r STR_i STR_p STR_t STR_UNDERSCORE STR_r STR_u STR_n "\0" 1194 #define STRING_atomic_script_run STR_a STR_t STR_o STR_m STR_i STR_c STR_UNDERSCORE STR_s STR_c STR_r STR_i STR_p STR_t STR_UNDERSCORE STR_r STR_u STR_n 1195 1196 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0" 1197 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0" 1198 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0" 1199 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0" 1200 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0" 1201 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0" 1202 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0" 1203 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0" 1204 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0" 1205 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0" 1206 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0" 1207 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0" 1208 #define STRING_word0 STR_w STR_o STR_r STR_d "\0" 1209 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t 1210 1211 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E 1212 #define STRING_VERSION STR_V STR_E STR_R STR_S STR_I STR_O STR_N 1213 #define STRING_WEIRD_STARTWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1214 #define STRING_WEIRD_ENDWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1215 1216 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS 1217 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS 1218 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1219 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS 1220 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1221 #define STRING_NUL_RIGHTPAR STR_N STR_U STR_L STR_RIGHT_PARENTHESIS 1222 #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 1223 #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 1224 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS 1225 #define STRING_UTF16_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS 1226 #define STRING_UTF32_RIGHTPAR STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS 1227 #define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_RIGHT_PARENTHESIS 1228 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS 1229 #define STRING_NO_AUTO_POSSESS_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_A STR_U STR_T STR_O STR_UNDERSCORE STR_P STR_O STR_S STR_S STR_E STR_S STR_S STR_RIGHT_PARENTHESIS 1230 #define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_D STR_O STR_T STR_S STR_T STR_A STR_R STR_UNDERSCORE STR_A STR_N STR_C STR_H STR_O STR_R STR_RIGHT_PARENTHESIS 1231 #define STRING_NO_JIT_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_J STR_I STR_T STR_RIGHT_PARENTHESIS 1232 #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 1233 #define STRING_NOTEMPTY_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_RIGHT_PARENTHESIS 1234 #define STRING_NOTEMPTY_ATSTART_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_UNDERSCORE STR_A STR_T STR_S STR_T STR_A STR_R STR_T STR_RIGHT_PARENTHESIS 1235 #define STRING_LIMIT_HEAP_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_H STR_E STR_A STR_P STR_EQUALS_SIGN 1236 #define STRING_LIMIT_MATCH_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_M STR_A STR_T STR_C STR_H STR_EQUALS_SIGN 1237 #define STRING_LIMIT_DEPTH_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_D STR_E STR_P STR_T STR_H STR_EQUALS_SIGN 1238 #define STRING_LIMIT_RECURSION_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_R STR_E STR_C STR_U STR_R STR_S STR_I STR_O STR_N STR_EQUALS_SIGN 1239 #define STRING_MARK STR_M STR_A STR_R STR_K 1240 1241 #endif /* SUPPORT_UNICODE */ 1242 1243 /* -------------------- End of character and string names -------------------*/ 1244 1245 /* -------------------- Definitions for compiled patterns -------------------*/ 1246 1247 /* Codes for different types of Unicode property */ 1248 1249 #define PT_ANY 0 /* Any property - matches all chars */ 1250 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */ 1251 #define PT_GC 2 /* Specified general characteristic (e.g. L) */ 1252 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */ 1253 #define PT_SC 4 /* Script (e.g. Han) */ 1254 #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */ 1255 #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */ 1256 #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */ 1257 #define PT_WORD 8 /* Word - L plus N plus underscore */ 1258 #define PT_CLIST 9 /* Pseudo-property: match character list */ 1259 #define PT_UCNC 10 /* Universal Character nameable character */ 1260 #define PT_TABSIZE 11 /* Size of square table for autopossessify tests */ 1261 1262 /* The following special properties are used only in XCLASS items, when POSIX 1263 classes are specified and PCRE2_UCP is set - in other words, for Unicode 1264 handling of these classes. They are not available via the \p or \P escapes like 1265 those in the above list, and so they do not take part in the autopossessifying 1266 table. */ 1267 1268 #define PT_PXGRAPH 11 /* [:graph:] - characters that mark the paper */ 1269 #define PT_PXPRINT 12 /* [:print:] - [:graph:] plus non-control spaces */ 1270 #define PT_PXPUNCT 13 /* [:punct:] - punctuation characters */ 1271 1272 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that 1273 contain characters with values greater than 255. */ 1274 1275 #define XCL_NOT 0x01 /* Flag: this is a negative class */ 1276 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ 1277 #define XCL_HASPROP 0x04 /* Flag: property checks are present. */ 1278 1279 #define XCL_END 0 /* Marks end of individual items */ 1280 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */ 1281 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */ 1282 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */ 1283 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */ 1284 1285 /* These are escaped items that aren't just an encoding of a particular data 1286 value such as \n. They must have non-zero values, as check_escape() returns 0 1287 for a data character. In the escapes[] table in pcre2_compile.c their values 1288 are negated in order to distinguish them from data values. 1289 1290 They must appear here in the same order as in the opcode definitions below, up 1291 to ESC_z. There's a dummy for OP_ALLANY because it corresponds to "." in DOTALL 1292 mode rather than an escape sequence. It is also used for [^] in JavaScript 1293 compatibility mode, and for \C in non-utf mode. In non-DOTALL mode, "." behaves 1294 like \N. 1295 1296 Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in 1297 check_escape(). There are tests in the code for an escape greater than ESC_b 1298 and less than ESC_Z to detect the types that may be repeated. These are the 1299 types that consume characters. If any new escapes are put in between that don't 1300 consume a character, that code will have to change. */ 1301 1302 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, 1303 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, 1304 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, 1305 ESC_E, ESC_Q, ESC_g, ESC_k }; 1306 1307 1308 /********************** Opcode definitions ******************/ 1309 1310 /****** NOTE NOTE NOTE ****** 1311 1312 Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in 1313 order to the list of escapes immediately above. Furthermore, values up to 1314 OP_DOLLM must not be changed without adjusting the table called autoposstab in 1315 pcre2_auto_possess.c 1316 1317 Whenever this list is updated, the two macro definitions that follow must be 1318 updated to match. The possessification table called "opcode_possessify" in 1319 pcre2_compile.c must also be updated, and also the tables called "coptable" 1320 and "poptable" in pcre2_dfa_match.c. 1321 1322 ****** NOTE NOTE NOTE ******/ 1323 1324 1325 /* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive, 1326 are used in a table for deciding whether a repeated character type can be 1327 auto-possessified. */ 1328 1329 #define FIRST_AUTOTAB_OP OP_NOT_DIGIT 1330 #define LAST_AUTOTAB_LEFT_OP OP_EXTUNI 1331 #define LAST_AUTOTAB_RIGHT_OP OP_DOLLM 1332 1333 enum { 1334 OP_END, /* 0 End of pattern */ 1335 1336 /* Values corresponding to backslashed metacharacters */ 1337 1338 OP_SOD, /* 1 Start of data: \A */ 1339 OP_SOM, /* 2 Start of match (subject + offset): \G */ 1340 OP_SET_SOM, /* 3 Set start of match (\K) */ 1341 OP_NOT_WORD_BOUNDARY, /* 4 \B */ 1342 OP_WORD_BOUNDARY, /* 5 \b */ 1343 OP_NOT_DIGIT, /* 6 \D */ 1344 OP_DIGIT, /* 7 \d */ 1345 OP_NOT_WHITESPACE, /* 8 \S */ 1346 OP_WHITESPACE, /* 9 \s */ 1347 OP_NOT_WORDCHAR, /* 10 \W */ 1348 OP_WORDCHAR, /* 11 \w */ 1349 1350 OP_ANY, /* 12 Match any character except newline (\N) */ 1351 OP_ALLANY, /* 13 Match any character */ 1352 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */ 1353 OP_NOTPROP, /* 15 \P (not Unicode property) */ 1354 OP_PROP, /* 16 \p (Unicode property) */ 1355 OP_ANYNL, /* 17 \R (any newline sequence) */ 1356 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */ 1357 OP_HSPACE, /* 19 \h (horizontal whitespace) */ 1358 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */ 1359 OP_VSPACE, /* 21 \v (vertical whitespace) */ 1360 OP_EXTUNI, /* 22 \X (extended Unicode sequence */ 1361 OP_EODN, /* 23 End of data or \n at end of data (\Z) */ 1362 OP_EOD, /* 24 End of data (\z) */ 1363 1364 /* Line end assertions */ 1365 1366 OP_DOLL, /* 25 End of line - not multiline */ 1367 OP_DOLLM, /* 26 End of line - multiline */ 1368 OP_CIRC, /* 27 Start of line - not multiline */ 1369 OP_CIRCM, /* 28 Start of line - multiline */ 1370 1371 /* Single characters; caseful must precede the caseless ones, and these 1372 must remain in this order, and adjacent. */ 1373 1374 OP_CHAR, /* 29 Match one character, casefully */ 1375 OP_CHARI, /* 30 Match one character, caselessly */ 1376 OP_NOT, /* 31 Match one character, not the given one, casefully */ 1377 OP_NOTI, /* 32 Match one character, not the given one, caselessly */ 1378 1379 /* The following sets of 13 opcodes must always be kept in step because 1380 the offset from the first one is used to generate the others. */ 1381 1382 /* Repeated characters; caseful must precede the caseless ones */ 1383 1384 OP_STAR, /* 33 The maximizing and minimizing versions of */ 1385 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */ 1386 OP_PLUS, /* 35 the minimizing one second. */ 1387 OP_MINPLUS, /* 36 */ 1388 OP_QUERY, /* 37 */ 1389 OP_MINQUERY, /* 38 */ 1390 1391 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/ 1392 OP_MINUPTO, /* 40 */ 1393 OP_EXACT, /* 41 Exactly n matches */ 1394 1395 OP_POSSTAR, /* 42 Possessified star, caseful */ 1396 OP_POSPLUS, /* 43 Possessified plus, caseful */ 1397 OP_POSQUERY, /* 44 Posesssified query, caseful */ 1398 OP_POSUPTO, /* 45 Possessified upto, caseful */ 1399 1400 /* Repeated characters; caseless must follow the caseful ones */ 1401 1402 OP_STARI, /* 46 */ 1403 OP_MINSTARI, /* 47 */ 1404 OP_PLUSI, /* 48 */ 1405 OP_MINPLUSI, /* 49 */ 1406 OP_QUERYI, /* 50 */ 1407 OP_MINQUERYI, /* 51 */ 1408 1409 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */ 1410 OP_MINUPTOI, /* 53 */ 1411 OP_EXACTI, /* 54 */ 1412 1413 OP_POSSTARI, /* 55 Possessified star, caseless */ 1414 OP_POSPLUSI, /* 56 Possessified plus, caseless */ 1415 OP_POSQUERYI, /* 57 Posesssified query, caseless */ 1416 OP_POSUPTOI, /* 58 Possessified upto, caseless */ 1417 1418 /* The negated ones must follow the non-negated ones, and match them */ 1419 /* Negated repeated character, caseful; must precede the caseless ones */ 1420 1421 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */ 1422 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */ 1423 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */ 1424 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */ 1425 OP_NOTQUERY, /* 63 */ 1426 OP_NOTMINQUERY, /* 64 */ 1427 1428 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */ 1429 OP_NOTMINUPTO, /* 66 */ 1430 OP_NOTEXACT, /* 67 Exactly n matches */ 1431 1432 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */ 1433 OP_NOTPOSPLUS, /* 69 */ 1434 OP_NOTPOSQUERY, /* 70 */ 1435 OP_NOTPOSUPTO, /* 71 */ 1436 1437 /* Negated repeated character, caseless; must follow the caseful ones */ 1438 1439 OP_NOTSTARI, /* 72 */ 1440 OP_NOTMINSTARI, /* 73 */ 1441 OP_NOTPLUSI, /* 74 */ 1442 OP_NOTMINPLUSI, /* 75 */ 1443 OP_NOTQUERYI, /* 76 */ 1444 OP_NOTMINQUERYI, /* 77 */ 1445 1446 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */ 1447 OP_NOTMINUPTOI, /* 79 */ 1448 OP_NOTEXACTI, /* 80 Exactly n matches */ 1449 1450 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */ 1451 OP_NOTPOSPLUSI, /* 82 */ 1452 OP_NOTPOSQUERYI, /* 83 */ 1453 OP_NOTPOSUPTOI, /* 84 */ 1454 1455 /* Character types */ 1456 1457 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */ 1458 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */ 1459 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */ 1460 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */ 1461 OP_TYPEQUERY, /* 89 */ 1462 OP_TYPEMINQUERY, /* 90 */ 1463 1464 OP_TYPEUPTO, /* 91 From 0 to n matches */ 1465 OP_TYPEMINUPTO, /* 92 */ 1466 OP_TYPEEXACT, /* 93 Exactly n matches */ 1467 1468 OP_TYPEPOSSTAR, /* 94 Possessified versions */ 1469 OP_TYPEPOSPLUS, /* 95 */ 1470 OP_TYPEPOSQUERY, /* 96 */ 1471 OP_TYPEPOSUPTO, /* 97 */ 1472 1473 /* These are used for character classes and back references; only the 1474 first six are the same as the sets above. */ 1475 1476 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */ 1477 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */ 1478 OP_CRPLUS, /* 100 the minimizing one second. These codes must */ 1479 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */ 1480 OP_CRQUERY, /* 102 */ 1481 OP_CRMINQUERY, /* 103 */ 1482 1483 OP_CRRANGE, /* 104 These are different to the three sets above. */ 1484 OP_CRMINRANGE, /* 105 */ 1485 1486 OP_CRPOSSTAR, /* 106 Possessified versions */ 1487 OP_CRPOSPLUS, /* 107 */ 1488 OP_CRPOSQUERY, /* 108 */ 1489 OP_CRPOSRANGE, /* 109 */ 1490 1491 /* End of quantifier opcodes */ 1492 1493 OP_CLASS, /* 110 Match a character class, chars < 256 only */ 1494 OP_NCLASS, /* 111 Same, but the bitmap was created from a negative 1495 class - the difference is relevant only when a 1496 character > 255 is encountered. */ 1497 OP_XCLASS, /* 112 Extended class for handling > 255 chars within the 1498 class. This does both positive and negative. */ 1499 OP_REF, /* 113 Match a back reference, casefully */ 1500 OP_REFI, /* 114 Match a back reference, caselessly */ 1501 OP_DNREF, /* 115 Match a duplicate name backref, casefully */ 1502 OP_DNREFI, /* 116 Match a duplicate name backref, caselessly */ 1503 OP_RECURSE, /* 117 Match a numbered subpattern (possibly recursive) */ 1504 OP_CALLOUT, /* 118 Call out to external function if provided */ 1505 OP_CALLOUT_STR, /* 119 Call out with string argument */ 1506 1507 OP_ALT, /* 120 Start of alternation */ 1508 OP_KET, /* 121 End of group that doesn't have an unbounded repeat */ 1509 OP_KETRMAX, /* 122 These two must remain together and in this */ 1510 OP_KETRMIN, /* 123 order. They are for groups the repeat for ever. */ 1511 OP_KETRPOS, /* 124 Possessive unlimited repeat. */ 1512 1513 /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four 1514 asserts must remain in order. */ 1515 1516 OP_REVERSE, /* 125 Move pointer back - used in lookbehind assertions */ 1517 OP_ASSERT, /* 126 Positive lookahead */ 1518 OP_ASSERT_NOT, /* 127 Negative lookahead */ 1519 OP_ASSERTBACK, /* 128 Positive lookbehind */ 1520 OP_ASSERTBACK_NOT, /* 129 Negative lookbehind */ 1521 1522 /* ONCE, SCRIPT_RUN, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come 1523 immediately after the assertions, with ONCE first, as there's a test for >= 1524 ONCE for a subpattern that isn't an assertion. The POS versions must 1525 immediately follow the non-POS versions in each case. */ 1526 1527 OP_ONCE, /* 130 Atomic group, contains captures */ 1528 OP_SCRIPT_RUN, /* 131 Non-capture, but check characters' scripts */ 1529 OP_BRA, /* 132 Start of non-capturing bracket */ 1530 OP_BRAPOS, /* 133 Ditto, with unlimited, possessive repeat */ 1531 OP_CBRA, /* 134 Start of capturing bracket */ 1532 OP_CBRAPOS, /* 135 Ditto, with unlimited, possessive repeat */ 1533 OP_COND, /* 136 Conditional group */ 1534 1535 /* These five must follow the previous five, in the same order. There's a 1536 check for >= SBRA to distinguish the two sets. */ 1537 1538 OP_SBRA, /* 137 Start of non-capturing bracket, check empty */ 1539 OP_SBRAPOS, /* 138 Ditto, with unlimited, possessive repeat */ 1540 OP_SCBRA, /* 139 Start of capturing bracket, check empty */ 1541 OP_SCBRAPOS, /* 140 Ditto, with unlimited, possessive repeat */ 1542 OP_SCOND, /* 141 Conditional group, check empty */ 1543 1544 /* The next two pairs must (respectively) be kept together. */ 1545 1546 OP_CREF, /* 142 Used to hold a capture number as condition */ 1547 OP_DNCREF, /* 143 Used to point to duplicate names as a condition */ 1548 OP_RREF, /* 144 Used to hold a recursion number as condition */ 1549 OP_DNRREF, /* 145 Used to point to duplicate names as a condition */ 1550 OP_FALSE, /* 146 Always false (used by DEFINE and VERSION) */ 1551 OP_TRUE, /* 147 Always true (used by VERSION) */ 1552 1553 OP_BRAZERO, /* 148 These two must remain together and in this */ 1554 OP_BRAMINZERO, /* 149 order. */ 1555 OP_BRAPOSZERO, /* 150 */ 1556 1557 /* These are backtracking control verbs */ 1558 1559 OP_MARK, /* 151 always has an argument */ 1560 OP_PRUNE, /* 152 */ 1561 OP_PRUNE_ARG, /* 153 same, but with argument */ 1562 OP_SKIP, /* 154 */ 1563 OP_SKIP_ARG, /* 155 same, but with argument */ 1564 OP_THEN, /* 156 */ 1565 OP_THEN_ARG, /* 157 same, but with argument */ 1566 OP_COMMIT, /* 158 */ 1567 OP_COMMIT_ARG, /* 159 same, but with argument */ 1568 1569 /* These are forced failure and success verbs. FAIL and ACCEPT do accept an 1570 argument, but these cases can be compiled as, for example, (*MARK:X)(*FAIL) 1571 without the need for a special opcode. */ 1572 1573 OP_FAIL, /* 160 */ 1574 OP_ACCEPT, /* 161 */ 1575 OP_ASSERT_ACCEPT, /* 162 Used inside assertions */ 1576 OP_CLOSE, /* 163 Used before OP_ACCEPT to close open captures */ 1577 1578 /* This is used to skip a subpattern with a {0} quantifier */ 1579 1580 OP_SKIPZERO, /* 164 */ 1581 1582 /* This is used to identify a DEFINE group during compilation so that it can 1583 be checked for having only one branch. It is changed to OP_FALSE before 1584 compilation finishes. */ 1585 1586 OP_DEFINE, /* 165 */ 1587 1588 /* This is not an opcode, but is used to check that tables indexed by opcode 1589 are the correct length, in order to catch updating errors - there have been 1590 some in the past. */ 1591 1592 OP_TABLE_LENGTH 1593 1594 }; 1595 1596 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro 1597 definitions that follow must also be updated to match. There are also tables 1598 called "opcode_possessify" in pcre2_compile.c and "coptable" and "poptable" in 1599 pcre2_dfa_exec.c that must be updated. */ 1600 1601 1602 /* This macro defines textual names for all the opcodes. These are used only 1603 for debugging, and some of them are only partial names. The macro is referenced 1604 only in pcre2_printint.c, which fills out the full names in many cases (and in 1605 some cases doesn't actually use these names at all). */ 1606 1607 #define OP_NAME_LIST \ 1608 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \ 1609 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \ 1610 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \ 1611 "extuni", "\\Z", "\\z", \ 1612 "$", "$", "^", "^", "char", "chari", "not", "noti", \ 1613 "*", "*?", "+", "+?", "?", "??", \ 1614 "{", "{", "{", \ 1615 "*+","++", "?+", "{", \ 1616 "*", "*?", "+", "+?", "?", "??", \ 1617 "{", "{", "{", \ 1618 "*+","++", "?+", "{", \ 1619 "*", "*?", "+", "+?", "?", "??", \ 1620 "{", "{", "{", \ 1621 "*+","++", "?+", "{", \ 1622 "*", "*?", "+", "+?", "?", "??", \ 1623 "{", "{", "{", \ 1624 "*+","++", "?+", "{", \ 1625 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \ 1626 "*+","++", "?+", "{", \ 1627 "*", "*?", "+", "+?", "?", "??", "{", "{", \ 1628 "*+","++", "?+", "{", \ 1629 "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi", \ 1630 "Recurse", "Callout", "CalloutStr", \ 1631 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \ 1632 "Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \ 1633 "Once", \ 1634 "Script run", \ 1635 "Bra", "BraPos", "CBra", "CBraPos", \ 1636 "Cond", \ 1637 "SBra", "SBraPos", "SCBra", "SCBraPos", \ 1638 "SCond", \ 1639 "Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", \ 1640 "Cond false", "Cond true", \ 1641 "Brazero", "Braminzero", "Braposzero", \ 1642 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \ 1643 "*THEN", "*THEN", "*COMMIT", "*COMMIT", "*FAIL", \ 1644 "*ACCEPT", "*ASSERT_ACCEPT", \ 1645 "Close", "Skip zero", "Define" 1646 1647 1648 /* This macro defines the length of fixed length operations in the compiled 1649 regex. The lengths are used when searching for specific things, and also in the 1650 debugging printing of a compiled regex. We use a macro so that it can be 1651 defined close to the definitions of the opcodes themselves. 1652 1653 As things have been extended, some of these are no longer fixed lenths, but are 1654 minima instead. For example, the length of a single-character repeat may vary 1655 in UTF-8 mode. The code that uses this table must know about such things. */ 1656 1657 #define OP_LENGTHS \ 1658 1, /* End */ \ 1659 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \ 1660 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \ 1661 1, 1, 1, /* Any, AllAny, Anybyte */ \ 1662 3, 3, /* \P, \p */ \ 1663 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \ 1664 1, /* \X */ \ 1665 1, 1, 1, 1, 1, 1, /* \Z, \z, $, $M ^, ^M */ \ 1666 2, /* Char - the minimum length */ \ 1667 2, /* Chari - the minimum length */ \ 1668 2, /* not */ \ 1669 2, /* noti */ \ 1670 /* Positive single-char repeats ** These are */ \ 1671 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ 1672 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \ 1673 2+IMM2_SIZE, /* exact */ \ 1674 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \ 1675 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \ 1676 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \ 1677 2+IMM2_SIZE, /* exact I */ \ 1678 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \ 1679 /* Negative single-char repeats - only for chars < 256 */ \ 1680 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ 1681 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \ 1682 2+IMM2_SIZE, /* NOT exact */ \ 1683 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \ 1684 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \ 1685 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \ 1686 2+IMM2_SIZE, /* NOT exact I */ \ 1687 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \ 1688 /* Positive type repeats */ \ 1689 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ 1690 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \ 1691 2+IMM2_SIZE, /* Type exact */ \ 1692 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \ 1693 /* Character class & ref repeats */ \ 1694 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ 1695 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \ 1696 1, 1, 1, 1+2*IMM2_SIZE, /* Possessive *+, ++, ?+, CRPOSRANGE */ \ 1697 1+(32/sizeof(PCRE2_UCHAR)), /* CLASS */ \ 1698 1+(32/sizeof(PCRE2_UCHAR)), /* NCLASS */ \ 1699 0, /* XCLASS - variable length */ \ 1700 1+IMM2_SIZE, /* REF */ \ 1701 1+IMM2_SIZE, /* REFI */ \ 1702 1+2*IMM2_SIZE, /* DNREF */ \ 1703 1+2*IMM2_SIZE, /* DNREFI */ \ 1704 1+LINK_SIZE, /* RECURSE */ \ 1705 1+2*LINK_SIZE+1, /* CALLOUT */ \ 1706 0, /* CALLOUT_STR - variable length */ \ 1707 1+LINK_SIZE, /* Alt */ \ 1708 1+LINK_SIZE, /* Ket */ \ 1709 1+LINK_SIZE, /* KetRmax */ \ 1710 1+LINK_SIZE, /* KetRmin */ \ 1711 1+LINK_SIZE, /* KetRpos */ \ 1712 1+LINK_SIZE, /* Reverse */ \ 1713 1+LINK_SIZE, /* Assert */ \ 1714 1+LINK_SIZE, /* Assert not */ \ 1715 1+LINK_SIZE, /* Assert behind */ \ 1716 1+LINK_SIZE, /* Assert behind not */ \ 1717 1+LINK_SIZE, /* ONCE */ \ 1718 1+LINK_SIZE, /* SCRIPT_RUN */ \ 1719 1+LINK_SIZE, /* BRA */ \ 1720 1+LINK_SIZE, /* BRAPOS */ \ 1721 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \ 1722 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \ 1723 1+LINK_SIZE, /* COND */ \ 1724 1+LINK_SIZE, /* SBRA */ \ 1725 1+LINK_SIZE, /* SBRAPOS */ \ 1726 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \ 1727 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \ 1728 1+LINK_SIZE, /* SCOND */ \ 1729 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* CREF, DNCREF */ \ 1730 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* RREF, DNRREF */ \ 1731 1, 1, /* FALSE, TRUE */ \ 1732 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ 1733 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ 1734 1, 3, /* SKIP, SKIP_ARG */ \ 1735 1, 3, /* THEN, THEN_ARG */ \ 1736 1, 3, /* COMMIT, COMMIT_ARG */ \ 1737 1, 1, 1, /* FAIL, ACCEPT, ASSERT_ACCEPT */ \ 1738 1+IMM2_SIZE, 1, /* CLOSE, SKIPZERO */ \ 1739 1 /* DEFINE */ 1740 1741 /* A magic value for OP_RREF to indicate the "any recursion" condition. */ 1742 1743 #define RREF_ANY 0xffff 1744 1745 1746 /* ---------- Private structures that are mode-independent. ---------- */ 1747 1748 /* Structure to hold data for custom memory management. */ 1749 1750 typedef struct pcre2_memctl { 1751 void * (*malloc)(size_t, void *); 1752 void (*free)(void *, void *); 1753 void *memory_data; 1754 } pcre2_memctl; 1755 1756 /* Structure for building a chain of open capturing subpatterns during 1757 compiling, so that instructions to close them can be compiled when (*ACCEPT) is 1758 encountered. This is also used to identify subpatterns that contain recursive 1759 back references to themselves, so that they can be made atomic. */ 1760 1761 typedef struct open_capitem { 1762 struct open_capitem *next; /* Chain link */ 1763 uint16_t number; /* Capture number */ 1764 uint16_t flag; /* Set TRUE if recursive back ref */ 1765 uint16_t assert_depth; /* Assertion depth when opened */ 1766 } open_capitem; 1767 1768 /* Layout of the UCP type table that translates property names into types and 1769 codes. Each entry used to point directly to a name, but to reduce the number of 1770 relocations in shared libraries, it now has an offset into a single string 1771 instead. */ 1772 1773 typedef struct { 1774 uint16_t name_offset; 1775 uint16_t type; 1776 uint16_t value; 1777 } ucp_type_table; 1778 1779 /* Unicode character database (UCD) record format */ 1780 1781 typedef struct { 1782 uint8_t script; /* ucp_Arabic, etc. */ 1783 uint8_t chartype; /* ucp_Cc, etc. (general categories) */ 1784 uint8_t gbprop; /* ucp_gbControl, etc. (grapheme break property) */ 1785 uint8_t caseset; /* offset to multichar other cases or zero */ 1786 int32_t other_case; /* offset to other case, or zero if none */ 1787 int16_t scriptx; /* script extension value */ 1788 int16_t dummy; /* spare - to round to multiple of 4 bytes */ 1789 } ucd_record; 1790 1791 /* UCD access macros */ 1792 1793 #define UCD_BLOCK_SIZE 128 1794 #define REAL_GET_UCD(ch) (PRIV(ucd_records) + \ 1795 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \ 1796 UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE]) 1797 1798 #if PCRE2_CODE_UNIT_WIDTH == 32 1799 #define GET_UCD(ch) ((ch > MAX_UTF_CODE_POINT)? \ 1800 PRIV(dummy_ucd_record) : REAL_GET_UCD(ch)) 1801 #else 1802 #define GET_UCD(ch) REAL_GET_UCD(ch) 1803 #endif 1804 1805 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype 1806 #define UCD_SCRIPT(ch) GET_UCD(ch)->script 1807 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)] 1808 #define UCD_GRAPHBREAK(ch) GET_UCD(ch)->gbprop 1809 #define UCD_CASESET(ch) GET_UCD(ch)->caseset 1810 #define UCD_OTHERCASE(ch) ((uint32_t)((int)ch + (int)(GET_UCD(ch)->other_case))) 1811 #define UCD_SCRIPTX(ch) GET_UCD(ch)->scriptx 1812 1813 /* Header for serialized pcre2 codes. */ 1814 1815 typedef struct pcre2_serialized_data { 1816 uint32_t magic; 1817 uint32_t version; 1818 uint32_t config; 1819 int32_t number_of_codes; 1820 } pcre2_serialized_data; 1821 1822 1823 1824 /* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */ 1825 1826 /* When this file is included by pcre2test, PCRE2_CODE_UNIT_WIDTH is defined as 1827 0, so the following items are omitted. */ 1828 1829 #if defined PCRE2_CODE_UNIT_WIDTH && PCRE2_CODE_UNIT_WIDTH != 0 1830 1831 /* EBCDIC is supported only for the 8-bit library. */ 1832 1833 #if defined EBCDIC && PCRE2_CODE_UNIT_WIDTH != 8 1834 #error EBCDIC is not supported for the 16-bit or 32-bit libraries 1835 #endif 1836 1837 /* This is the largest non-UTF code point. */ 1838 1839 #define MAX_NON_UTF_CHAR (0xffffffffU >> (32 - PCRE2_CODE_UNIT_WIDTH)) 1840 1841 /* Internal shared data tables and variables. These are used by more than one 1842 of the exported public functions. They have to be "external" in the C sense, 1843 but are not part of the PCRE2 public API. Although the data for some of them is 1844 identical in all libraries, they must have different names so that multiple 1845 libraries can be simultaneously linked to a single application. However, UTF-8 1846 tables are needed only when compiling the 8-bit library. */ 1847 1848 #if PCRE2_CODE_UNIT_WIDTH == 8 1849 extern const int PRIV(utf8_table1)[]; 1850 extern const int PRIV(utf8_table1_size); 1851 extern const int PRIV(utf8_table2)[]; 1852 extern const int PRIV(utf8_table3)[]; 1853 extern const uint8_t PRIV(utf8_table4)[]; 1854 #endif 1855 1856 #define _pcre2_OP_lengths PCRE2_SUFFIX(_pcre2_OP_lengths_) 1857 #define _pcre2_callout_end_delims PCRE2_SUFFIX(_pcre2_callout_end_delims_) 1858 #define _pcre2_callout_start_delims PCRE2_SUFFIX(_pcre2_callout_start_delims_) 1859 #define _pcre2_default_compile_context PCRE2_SUFFIX(_pcre2_default_compile_context_) 1860 #define _pcre2_default_convert_context PCRE2_SUFFIX(_pcre2_default_convert_context_) 1861 #define _pcre2_default_match_context PCRE2_SUFFIX(_pcre2_default_match_context_) 1862 #define _pcre2_default_tables PCRE2_SUFFIX(_pcre2_default_tables_) 1863 #if PCRE2_CODE_UNIT_WIDTH == 32 1864 #define _pcre2_dummy_ucd_record PCRE2_SUFFIX(_pcre2_dummy_ucd_record_) 1865 #endif 1866 #define _pcre2_hspace_list PCRE2_SUFFIX(_pcre2_hspace_list_) 1867 #define _pcre2_vspace_list PCRE2_SUFFIX(_pcre2_vspace_list_) 1868 #define _pcre2_ucd_caseless_sets PCRE2_SUFFIX(_pcre2_ucd_caseless_sets_) 1869 #define _pcre2_ucd_digit_sets PCRE2_SUFFIX(_pcre2_ucd_digit_sets_) 1870 #define _pcre2_ucd_script_sets PCRE2_SUFFIX(_pcre2_ucd_script_sets_) 1871 #define _pcre2_ucd_records PCRE2_SUFFIX(_pcre2_ucd_records_) 1872 #define _pcre2_ucd_stage1 PCRE2_SUFFIX(_pcre2_ucd_stage1_) 1873 #define _pcre2_ucd_stage2 PCRE2_SUFFIX(_pcre2_ucd_stage2_) 1874 #define _pcre2_ucp_gbtable PCRE2_SUFFIX(_pcre2_ucp_gbtable_) 1875 #define _pcre2_ucp_gentype PCRE2_SUFFIX(_pcre2_ucp_gentype_) 1876 #define _pcre2_ucp_typerange PCRE2_SUFFIX(_pcre2_ucp_typerange_) 1877 #define _pcre2_unicode_version PCRE2_SUFFIX(_pcre2_unicode_version_) 1878 #define _pcre2_utt PCRE2_SUFFIX(_pcre2_utt_) 1879 #define _pcre2_utt_names PCRE2_SUFFIX(_pcre2_utt_names_) 1880 #define _pcre2_utt_size PCRE2_SUFFIX(_pcre2_utt_size_) 1881 1882 extern const uint8_t PRIV(OP_lengths)[]; 1883 extern const uint32_t PRIV(callout_end_delims)[]; 1884 extern const uint32_t PRIV(callout_start_delims)[]; 1885 extern const pcre2_compile_context PRIV(default_compile_context); 1886 extern const pcre2_convert_context PRIV(default_convert_context); 1887 extern const pcre2_match_context PRIV(default_match_context); 1888 extern const uint8_t PRIV(default_tables)[]; 1889 extern const uint32_t PRIV(hspace_list)[]; 1890 extern const uint32_t PRIV(vspace_list)[]; 1891 extern const uint32_t PRIV(ucd_caseless_sets)[]; 1892 extern const uint32_t PRIV(ucd_digit_sets)[]; 1893 extern const uint8_t PRIV(ucd_script_sets)[]; 1894 extern const ucd_record PRIV(ucd_records)[]; 1895 #if PCRE2_CODE_UNIT_WIDTH == 32 1896 extern const ucd_record PRIV(dummy_ucd_record)[]; 1897 #endif 1898 extern const uint16_t PRIV(ucd_stage1)[]; 1899 extern const uint16_t PRIV(ucd_stage2)[]; 1900 extern const uint32_t PRIV(ucp_gbtable)[]; 1901 extern const uint32_t PRIV(ucp_gentype)[]; 1902 #ifdef SUPPORT_JIT 1903 extern const int PRIV(ucp_typerange)[]; 1904 #endif 1905 extern const char *PRIV(unicode_version); 1906 extern const ucp_type_table PRIV(utt)[]; 1907 extern const char PRIV(utt_names)[]; 1908 extern const size_t PRIV(utt_size); 1909 1910 /* Mode-dependent macros and hidden and private structures are defined in a 1911 separate file so that pcre2test can include them at all supported widths. When 1912 compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we can 1913 include them at the appropriate width, after setting up suffix macros for the 1914 private structures. */ 1915 1916 #define branch_chain PCRE2_SUFFIX(branch_chain_) 1917 #define compile_block PCRE2_SUFFIX(compile_block_) 1918 #define dfa_match_block PCRE2_SUFFIX(dfa_match_block_) 1919 #define match_block PCRE2_SUFFIX(match_block_) 1920 #define named_group PCRE2_SUFFIX(named_group_) 1921 1922 #include "pcre2_intmodedep.h" 1923 1924 /* Private "external" functions. These are internal functions that are called 1925 from modules other than the one in which they are defined. They have to be 1926 "external" in the C sense, but are not part of the PCRE2 public API. They are 1927 not referenced from pcre2test, and must not be defined when no code unit width 1928 is available. */ 1929 1930 #define _pcre2_auto_possessify PCRE2_SUFFIX(_pcre2_auto_possessify_) 1931 #define _pcre2_check_escape PCRE2_SUFFIX(_pcre2_check_escape_) 1932 #define _pcre2_extuni PCRE2_SUFFIX(_pcre2_extuni_) 1933 #define _pcre2_find_bracket PCRE2_SUFFIX(_pcre2_find_bracket_) 1934 #define _pcre2_is_newline PCRE2_SUFFIX(_pcre2_is_newline_) 1935 #define _pcre2_jit_free_rodata PCRE2_SUFFIX(_pcre2_jit_free_rodata_) 1936 #define _pcre2_jit_free PCRE2_SUFFIX(_pcre2_jit_free_) 1937 #define _pcre2_jit_get_size PCRE2_SUFFIX(_pcre2_jit_get_size_) 1938 #define _pcre2_jit_get_target PCRE2_SUFFIX(_pcre2_jit_get_target_) 1939 #define _pcre2_memctl_malloc PCRE2_SUFFIX(_pcre2_memctl_malloc_) 1940 #define _pcre2_ord2utf PCRE2_SUFFIX(_pcre2_ord2utf_) 1941 #define _pcre2_script_run PCRE2_SUFFIX(_pcre2_script_run_) 1942 #define _pcre2_strcmp PCRE2_SUFFIX(_pcre2_strcmp_) 1943 #define _pcre2_strcmp_c8 PCRE2_SUFFIX(_pcre2_strcmp_c8_) 1944 #define _pcre2_strcpy_c8 PCRE2_SUFFIX(_pcre2_strcpy_c8_) 1945 #define _pcre2_strlen PCRE2_SUFFIX(_pcre2_strlen_) 1946 #define _pcre2_strncmp PCRE2_SUFFIX(_pcre2_strncmp_) 1947 #define _pcre2_strncmp_c8 PCRE2_SUFFIX(_pcre2_strncmp_c8_) 1948 #define _pcre2_study PCRE2_SUFFIX(_pcre2_study_) 1949 #define _pcre2_valid_utf PCRE2_SUFFIX(_pcre2_valid_utf_) 1950 #define _pcre2_was_newline PCRE2_SUFFIX(_pcre2_was_newline_) 1951 #define _pcre2_xclass PCRE2_SUFFIX(_pcre2_xclass_) 1952 1953 extern int _pcre2_auto_possessify(PCRE2_UCHAR *, BOOL, 1954 const compile_block *); 1955 extern int _pcre2_check_escape(PCRE2_SPTR *, PCRE2_SPTR, uint32_t *, 1956 int *, uint32_t, uint32_t, BOOL, compile_block *); 1957 extern PCRE2_SPTR _pcre2_extuni(uint32_t, PCRE2_SPTR, PCRE2_SPTR, PCRE2_SPTR, 1958 BOOL, int *); 1959 extern PCRE2_SPTR _pcre2_find_bracket(PCRE2_SPTR, BOOL, int); 1960 extern BOOL _pcre2_is_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 1961 uint32_t *, BOOL); 1962 extern void _pcre2_jit_free_rodata(void *, void *); 1963 extern void _pcre2_jit_free(void *, pcre2_memctl *); 1964 extern size_t _pcre2_jit_get_size(void *); 1965 const char * _pcre2_jit_get_target(void); 1966 extern void * _pcre2_memctl_malloc(size_t, pcre2_memctl *); 1967 extern unsigned int _pcre2_ord2utf(uint32_t, PCRE2_UCHAR *); 1968 extern BOOL _pcre2_script_run(PCRE2_SPTR, PCRE2_SPTR, BOOL); 1969 extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR); 1970 extern int _pcre2_strcmp_c8(PCRE2_SPTR, const char *); 1971 extern PCRE2_SIZE _pcre2_strcpy_c8(PCRE2_UCHAR *, const char *); 1972 extern PCRE2_SIZE _pcre2_strlen(PCRE2_SPTR); 1973 extern int _pcre2_strncmp(PCRE2_SPTR, PCRE2_SPTR, size_t); 1974 extern int _pcre2_strncmp_c8(PCRE2_SPTR, const char *, size_t); 1975 extern int _pcre2_study(pcre2_real_code *); 1976 extern int _pcre2_valid_utf(PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE *); 1977 extern BOOL _pcre2_was_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 1978 uint32_t *, BOOL); 1979 extern BOOL _pcre2_xclass(uint32_t, PCRE2_SPTR, BOOL); 1980 1981 /* This function is needed only when memmove() is not available. */ 1982 1983 #if !defined(VPCOMPAT) && !defined(HAVE_MEMMOVE) 1984 #define _pcre2_memmove PCRE2_SUFFIX(_pcre2_memmove) 1985 extern void * _pcre2_memmove(void *, const void *, size_t); 1986 #endif 1987 1988 #endif /* PCRE2_CODE_UNIT_WIDTH */ 1989 #endif /* PCRE2_INTERNAL_H_IDEMPOTENT_GUARD */ 1990 1991 /* End of pcre2_internal.h */ 1992