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) 2016-2022 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 #define PCRE2_HASACCEPT 0x00800000 /* contains (*ACCEPT) */ 532 533 #define PCRE2_MODE_MASK (PCRE2_MODE8 | PCRE2_MODE16 | PCRE2_MODE32) 534 535 /* Values for the matchedby field in a match data block. */ 536 537 enum { PCRE2_MATCHEDBY_INTERPRETER, /* pcre2_match() */ 538 PCRE2_MATCHEDBY_DFA_INTERPRETER, /* pcre2_dfa_match() */ 539 PCRE2_MATCHEDBY_JIT }; /* pcre2_jit_match() */ 540 541 /* Values for the flags field in a match data block. */ 542 543 #define PCRE2_MD_COPIED_SUBJECT 0x01u 544 545 /* Magic number to provide a small check against being handed junk. */ 546 547 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ 548 549 /* The maximum remaining length of subject we are prepared to search for a 550 req_unit match from an anchored pattern. In 8-bit mode, memchr() is used and is 551 much faster than the search loop that has to be used in 16-bit and 32-bit 552 modes. */ 553 554 #if PCRE2_CODE_UNIT_WIDTH == 8 555 #define REQ_CU_MAX 5000 556 #else 557 #define REQ_CU_MAX 2000 558 #endif 559 560 /* Offsets for the bitmap tables in the cbits set of tables. Each table 561 contains a set of bits for a class map. Some classes are built by combining 562 these tables. */ 563 564 #define cbit_space 0 /* [:space:] or \s */ 565 #define cbit_xdigit 32 /* [:xdigit:] */ 566 #define cbit_digit 64 /* [:digit:] or \d */ 567 #define cbit_upper 96 /* [:upper:] */ 568 #define cbit_lower 128 /* [:lower:] */ 569 #define cbit_word 160 /* [:word:] or \w */ 570 #define cbit_graph 192 /* [:graph:] */ 571 #define cbit_print 224 /* [:print:] */ 572 #define cbit_punct 256 /* [:punct:] */ 573 #define cbit_cntrl 288 /* [:cntrl:] */ 574 #define cbit_length 320 /* Length of the cbits table */ 575 576 /* Bit definitions for entries in the ctypes table. Do not change these values 577 without checking pcre2_jit_compile.c, which has an assertion to ensure that 578 ctype_word has the value 16. */ 579 580 #define ctype_space 0x01 581 #define ctype_letter 0x02 582 #define ctype_lcletter 0x04 583 #define ctype_digit 0x08 584 #define ctype_word 0x10 /* alphanumeric or '_' */ 585 586 /* Offsets of the various tables from the base tables pointer, and 587 total length of the tables. */ 588 589 #define lcc_offset 0 /* Lower case */ 590 #define fcc_offset 256 /* Flip case */ 591 #define cbits_offset 512 /* Character classes */ 592 #define ctypes_offset (cbits_offset + cbit_length) /* Character types */ 593 #define TABLES_LENGTH (ctypes_offset + 256) 594 595 596 /* -------------------- Character and string names ------------------------ */ 597 598 /* If PCRE2 is to support UTF-8 on EBCDIC platforms, we cannot use normal 599 character constants like '*' because the compiler would emit their EBCDIC code, 600 which is different from their ASCII/UTF-8 code. Instead we define macros for 601 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support 602 is enabled. When UTF-8 support is not enabled, the definitions use character 603 literals. Both character and string versions of each character are needed, and 604 there are some longer strings as well. 605 606 This means that, on EBCDIC platforms, the PCRE2 library can handle either 607 EBCDIC, or UTF-8, but not both. To support both in the same compiled library 608 would need different lookups depending on whether PCRE2_UTF was set or not. 609 This would make it impossible to use characters in switch/case statements, 610 which would reduce performance. For a theoretical use (which nobody has asked 611 for) in a minority area (EBCDIC platforms), this is not sensible. Any 612 application that did need both could compile two versions of the library, using 613 macros to give the functions distinct names. */ 614 615 #ifndef SUPPORT_UNICODE 616 617 /* UTF-8 support is not enabled; use the platform-dependent character literals 618 so that PCRE2 works in both ASCII and EBCDIC environments, but only in non-UTF 619 mode. Newline characters are problematic in EBCDIC. Though it has CR and LF 620 characters, a common practice has been to use its NL (0x15) character as the 621 line terminator in C-like processing environments. However, sometimes the LF 622 (0x25) character is used instead, according to this Unicode document: 623 624 http://unicode.org/standard/reports/tr13/tr13-5.html 625 626 PCRE2 defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25 627 instead. Whichever is *not* chosen is defined as NEL. 628 629 In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the 630 same code point. */ 631 632 #ifdef EBCDIC 633 634 #ifndef EBCDIC_NL25 635 #define CHAR_NL '\x15' 636 #define CHAR_NEL '\x25' 637 #define STR_NL "\x15" 638 #define STR_NEL "\x25" 639 #else 640 #define CHAR_NL '\x25' 641 #define CHAR_NEL '\x15' 642 #define STR_NL "\x25" 643 #define STR_NEL "\x15" 644 #endif 645 646 #define CHAR_LF CHAR_NL 647 #define STR_LF STR_NL 648 649 #define CHAR_ESC '\047' 650 #define CHAR_DEL '\007' 651 #define CHAR_NBSP ((unsigned char)'\x41') 652 #define STR_ESC "\047" 653 #define STR_DEL "\007" 654 655 #else /* Not EBCDIC */ 656 657 /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for 658 compatibility. NEL is the Unicode newline character; make sure it is 659 a positive value. */ 660 661 #define CHAR_LF '\n' 662 #define CHAR_NL CHAR_LF 663 #define CHAR_NEL ((unsigned char)'\x85') 664 #define CHAR_ESC '\033' 665 #define CHAR_DEL '\177' 666 #define CHAR_NBSP ((unsigned char)'\xa0') 667 668 #define STR_LF "\n" 669 #define STR_NL STR_LF 670 #define STR_NEL "\x85" 671 #define STR_ESC "\033" 672 #define STR_DEL "\177" 673 674 #endif /* EBCDIC */ 675 676 /* The remaining definitions work in both environments. */ 677 678 #define CHAR_NUL '\0' 679 #define CHAR_HT '\t' 680 #define CHAR_VT '\v' 681 #define CHAR_FF '\f' 682 #define CHAR_CR '\r' 683 #define CHAR_BS '\b' 684 #define CHAR_BEL '\a' 685 686 #define CHAR_SPACE ' ' 687 #define CHAR_EXCLAMATION_MARK '!' 688 #define CHAR_QUOTATION_MARK '"' 689 #define CHAR_NUMBER_SIGN '#' 690 #define CHAR_DOLLAR_SIGN '$' 691 #define CHAR_PERCENT_SIGN '%' 692 #define CHAR_AMPERSAND '&' 693 #define CHAR_APOSTROPHE '\'' 694 #define CHAR_LEFT_PARENTHESIS '(' 695 #define CHAR_RIGHT_PARENTHESIS ')' 696 #define CHAR_ASTERISK '*' 697 #define CHAR_PLUS '+' 698 #define CHAR_COMMA ',' 699 #define CHAR_MINUS '-' 700 #define CHAR_DOT '.' 701 #define CHAR_SLASH '/' 702 #define CHAR_0 '0' 703 #define CHAR_1 '1' 704 #define CHAR_2 '2' 705 #define CHAR_3 '3' 706 #define CHAR_4 '4' 707 #define CHAR_5 '5' 708 #define CHAR_6 '6' 709 #define CHAR_7 '7' 710 #define CHAR_8 '8' 711 #define CHAR_9 '9' 712 #define CHAR_COLON ':' 713 #define CHAR_SEMICOLON ';' 714 #define CHAR_LESS_THAN_SIGN '<' 715 #define CHAR_EQUALS_SIGN '=' 716 #define CHAR_GREATER_THAN_SIGN '>' 717 #define CHAR_QUESTION_MARK '?' 718 #define CHAR_COMMERCIAL_AT '@' 719 #define CHAR_A 'A' 720 #define CHAR_B 'B' 721 #define CHAR_C 'C' 722 #define CHAR_D 'D' 723 #define CHAR_E 'E' 724 #define CHAR_F 'F' 725 #define CHAR_G 'G' 726 #define CHAR_H 'H' 727 #define CHAR_I 'I' 728 #define CHAR_J 'J' 729 #define CHAR_K 'K' 730 #define CHAR_L 'L' 731 #define CHAR_M 'M' 732 #define CHAR_N 'N' 733 #define CHAR_O 'O' 734 #define CHAR_P 'P' 735 #define CHAR_Q 'Q' 736 #define CHAR_R 'R' 737 #define CHAR_S 'S' 738 #define CHAR_T 'T' 739 #define CHAR_U 'U' 740 #define CHAR_V 'V' 741 #define CHAR_W 'W' 742 #define CHAR_X 'X' 743 #define CHAR_Y 'Y' 744 #define CHAR_Z 'Z' 745 #define CHAR_LEFT_SQUARE_BRACKET '[' 746 #define CHAR_BACKSLASH '\\' 747 #define CHAR_RIGHT_SQUARE_BRACKET ']' 748 #define CHAR_CIRCUMFLEX_ACCENT '^' 749 #define CHAR_UNDERSCORE '_' 750 #define CHAR_GRAVE_ACCENT '`' 751 #define CHAR_a 'a' 752 #define CHAR_b 'b' 753 #define CHAR_c 'c' 754 #define CHAR_d 'd' 755 #define CHAR_e 'e' 756 #define CHAR_f 'f' 757 #define CHAR_g 'g' 758 #define CHAR_h 'h' 759 #define CHAR_i 'i' 760 #define CHAR_j 'j' 761 #define CHAR_k 'k' 762 #define CHAR_l 'l' 763 #define CHAR_m 'm' 764 #define CHAR_n 'n' 765 #define CHAR_o 'o' 766 #define CHAR_p 'p' 767 #define CHAR_q 'q' 768 #define CHAR_r 'r' 769 #define CHAR_s 's' 770 #define CHAR_t 't' 771 #define CHAR_u 'u' 772 #define CHAR_v 'v' 773 #define CHAR_w 'w' 774 #define CHAR_x 'x' 775 #define CHAR_y 'y' 776 #define CHAR_z 'z' 777 #define CHAR_LEFT_CURLY_BRACKET '{' 778 #define CHAR_VERTICAL_LINE '|' 779 #define CHAR_RIGHT_CURLY_BRACKET '}' 780 #define CHAR_TILDE '~' 781 782 #define STR_HT "\t" 783 #define STR_VT "\v" 784 #define STR_FF "\f" 785 #define STR_CR "\r" 786 #define STR_BS "\b" 787 #define STR_BEL "\a" 788 789 #define STR_SPACE " " 790 #define STR_EXCLAMATION_MARK "!" 791 #define STR_QUOTATION_MARK "\"" 792 #define STR_NUMBER_SIGN "#" 793 #define STR_DOLLAR_SIGN "$" 794 #define STR_PERCENT_SIGN "%" 795 #define STR_AMPERSAND "&" 796 #define STR_APOSTROPHE "'" 797 #define STR_LEFT_PARENTHESIS "(" 798 #define STR_RIGHT_PARENTHESIS ")" 799 #define STR_ASTERISK "*" 800 #define STR_PLUS "+" 801 #define STR_COMMA "," 802 #define STR_MINUS "-" 803 #define STR_DOT "." 804 #define STR_SLASH "/" 805 #define STR_0 "0" 806 #define STR_1 "1" 807 #define STR_2 "2" 808 #define STR_3 "3" 809 #define STR_4 "4" 810 #define STR_5 "5" 811 #define STR_6 "6" 812 #define STR_7 "7" 813 #define STR_8 "8" 814 #define STR_9 "9" 815 #define STR_COLON ":" 816 #define STR_SEMICOLON ";" 817 #define STR_LESS_THAN_SIGN "<" 818 #define STR_EQUALS_SIGN "=" 819 #define STR_GREATER_THAN_SIGN ">" 820 #define STR_QUESTION_MARK "?" 821 #define STR_COMMERCIAL_AT "@" 822 #define STR_A "A" 823 #define STR_B "B" 824 #define STR_C "C" 825 #define STR_D "D" 826 #define STR_E "E" 827 #define STR_F "F" 828 #define STR_G "G" 829 #define STR_H "H" 830 #define STR_I "I" 831 #define STR_J "J" 832 #define STR_K "K" 833 #define STR_L "L" 834 #define STR_M "M" 835 #define STR_N "N" 836 #define STR_O "O" 837 #define STR_P "P" 838 #define STR_Q "Q" 839 #define STR_R "R" 840 #define STR_S "S" 841 #define STR_T "T" 842 #define STR_U "U" 843 #define STR_V "V" 844 #define STR_W "W" 845 #define STR_X "X" 846 #define STR_Y "Y" 847 #define STR_Z "Z" 848 #define STR_LEFT_SQUARE_BRACKET "[" 849 #define STR_BACKSLASH "\\" 850 #define STR_RIGHT_SQUARE_BRACKET "]" 851 #define STR_CIRCUMFLEX_ACCENT "^" 852 #define STR_UNDERSCORE "_" 853 #define STR_GRAVE_ACCENT "`" 854 #define STR_a "a" 855 #define STR_b "b" 856 #define STR_c "c" 857 #define STR_d "d" 858 #define STR_e "e" 859 #define STR_f "f" 860 #define STR_g "g" 861 #define STR_h "h" 862 #define STR_i "i" 863 #define STR_j "j" 864 #define STR_k "k" 865 #define STR_l "l" 866 #define STR_m "m" 867 #define STR_n "n" 868 #define STR_o "o" 869 #define STR_p "p" 870 #define STR_q "q" 871 #define STR_r "r" 872 #define STR_s "s" 873 #define STR_t "t" 874 #define STR_u "u" 875 #define STR_v "v" 876 #define STR_w "w" 877 #define STR_x "x" 878 #define STR_y "y" 879 #define STR_z "z" 880 #define STR_LEFT_CURLY_BRACKET "{" 881 #define STR_VERTICAL_LINE "|" 882 #define STR_RIGHT_CURLY_BRACKET "}" 883 #define STR_TILDE "~" 884 885 #define STRING_ACCEPT0 "ACCEPT\0" 886 #define STRING_COMMIT0 "COMMIT\0" 887 #define STRING_F0 "F\0" 888 #define STRING_FAIL0 "FAIL\0" 889 #define STRING_MARK0 "MARK\0" 890 #define STRING_PRUNE0 "PRUNE\0" 891 #define STRING_SKIP0 "SKIP\0" 892 #define STRING_THEN "THEN" 893 894 #define STRING_atomic0 "atomic\0" 895 #define STRING_pla0 "pla\0" 896 #define STRING_plb0 "plb\0" 897 #define STRING_napla0 "napla\0" 898 #define STRING_naplb0 "naplb\0" 899 #define STRING_nla0 "nla\0" 900 #define STRING_nlb0 "nlb\0" 901 #define STRING_sr0 "sr\0" 902 #define STRING_asr0 "asr\0" 903 #define STRING_positive_lookahead0 "positive_lookahead\0" 904 #define STRING_positive_lookbehind0 "positive_lookbehind\0" 905 #define STRING_non_atomic_positive_lookahead0 "non_atomic_positive_lookahead\0" 906 #define STRING_non_atomic_positive_lookbehind0 "non_atomic_positive_lookbehind\0" 907 #define STRING_negative_lookahead0 "negative_lookahead\0" 908 #define STRING_negative_lookbehind0 "negative_lookbehind\0" 909 #define STRING_script_run0 "script_run\0" 910 #define STRING_atomic_script_run "atomic_script_run" 911 912 #define STRING_alpha0 "alpha\0" 913 #define STRING_lower0 "lower\0" 914 #define STRING_upper0 "upper\0" 915 #define STRING_alnum0 "alnum\0" 916 #define STRING_ascii0 "ascii\0" 917 #define STRING_blank0 "blank\0" 918 #define STRING_cntrl0 "cntrl\0" 919 #define STRING_digit0 "digit\0" 920 #define STRING_graph0 "graph\0" 921 #define STRING_print0 "print\0" 922 #define STRING_punct0 "punct\0" 923 #define STRING_space0 "space\0" 924 #define STRING_word0 "word\0" 925 #define STRING_xdigit "xdigit" 926 927 #define STRING_DEFINE "DEFINE" 928 #define STRING_VERSION "VERSION" 929 #define STRING_WEIRD_STARTWORD "[:<:]]" 930 #define STRING_WEIRD_ENDWORD "[:>:]]" 931 932 #define STRING_CR_RIGHTPAR "CR)" 933 #define STRING_LF_RIGHTPAR "LF)" 934 #define STRING_CRLF_RIGHTPAR "CRLF)" 935 #define STRING_ANY_RIGHTPAR "ANY)" 936 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)" 937 #define STRING_NUL_RIGHTPAR "NUL)" 938 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)" 939 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)" 940 #define STRING_UTF8_RIGHTPAR "UTF8)" 941 #define STRING_UTF16_RIGHTPAR "UTF16)" 942 #define STRING_UTF32_RIGHTPAR "UTF32)" 943 #define STRING_UTF_RIGHTPAR "UTF)" 944 #define STRING_UCP_RIGHTPAR "UCP)" 945 #define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)" 946 #define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR "NO_DOTSTAR_ANCHOR)" 947 #define STRING_NO_JIT_RIGHTPAR "NO_JIT)" 948 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)" 949 #define STRING_NOTEMPTY_RIGHTPAR "NOTEMPTY)" 950 #define STRING_NOTEMPTY_ATSTART_RIGHTPAR "NOTEMPTY_ATSTART)" 951 #define STRING_LIMIT_HEAP_EQ "LIMIT_HEAP=" 952 #define STRING_LIMIT_MATCH_EQ "LIMIT_MATCH=" 953 #define STRING_LIMIT_DEPTH_EQ "LIMIT_DEPTH=" 954 #define STRING_LIMIT_RECURSION_EQ "LIMIT_RECURSION=" 955 #define STRING_MARK "MARK" 956 957 #define STRING_bc "bc" 958 #define STRING_bidiclass "bidiclass" 959 #define STRING_sc "sc" 960 #define STRING_script "script" 961 #define STRING_scriptextensions "scriptextensions" 962 #define STRING_scx "scx" 963 964 #else /* SUPPORT_UNICODE */ 965 966 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This 967 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode 968 only. */ 969 970 #define CHAR_HT '\011' 971 #define CHAR_VT '\013' 972 #define CHAR_FF '\014' 973 #define CHAR_CR '\015' 974 #define CHAR_LF '\012' 975 #define CHAR_NL CHAR_LF 976 #define CHAR_NEL ((unsigned char)'\x85') 977 #define CHAR_BS '\010' 978 #define CHAR_BEL '\007' 979 #define CHAR_ESC '\033' 980 #define CHAR_DEL '\177' 981 982 #define CHAR_NUL '\0' 983 #define CHAR_SPACE '\040' 984 #define CHAR_EXCLAMATION_MARK '\041' 985 #define CHAR_QUOTATION_MARK '\042' 986 #define CHAR_NUMBER_SIGN '\043' 987 #define CHAR_DOLLAR_SIGN '\044' 988 #define CHAR_PERCENT_SIGN '\045' 989 #define CHAR_AMPERSAND '\046' 990 #define CHAR_APOSTROPHE '\047' 991 #define CHAR_LEFT_PARENTHESIS '\050' 992 #define CHAR_RIGHT_PARENTHESIS '\051' 993 #define CHAR_ASTERISK '\052' 994 #define CHAR_PLUS '\053' 995 #define CHAR_COMMA '\054' 996 #define CHAR_MINUS '\055' 997 #define CHAR_DOT '\056' 998 #define CHAR_SLASH '\057' 999 #define CHAR_0 '\060' 1000 #define CHAR_1 '\061' 1001 #define CHAR_2 '\062' 1002 #define CHAR_3 '\063' 1003 #define CHAR_4 '\064' 1004 #define CHAR_5 '\065' 1005 #define CHAR_6 '\066' 1006 #define CHAR_7 '\067' 1007 #define CHAR_8 '\070' 1008 #define CHAR_9 '\071' 1009 #define CHAR_COLON '\072' 1010 #define CHAR_SEMICOLON '\073' 1011 #define CHAR_LESS_THAN_SIGN '\074' 1012 #define CHAR_EQUALS_SIGN '\075' 1013 #define CHAR_GREATER_THAN_SIGN '\076' 1014 #define CHAR_QUESTION_MARK '\077' 1015 #define CHAR_COMMERCIAL_AT '\100' 1016 #define CHAR_A '\101' 1017 #define CHAR_B '\102' 1018 #define CHAR_C '\103' 1019 #define CHAR_D '\104' 1020 #define CHAR_E '\105' 1021 #define CHAR_F '\106' 1022 #define CHAR_G '\107' 1023 #define CHAR_H '\110' 1024 #define CHAR_I '\111' 1025 #define CHAR_J '\112' 1026 #define CHAR_K '\113' 1027 #define CHAR_L '\114' 1028 #define CHAR_M '\115' 1029 #define CHAR_N '\116' 1030 #define CHAR_O '\117' 1031 #define CHAR_P '\120' 1032 #define CHAR_Q '\121' 1033 #define CHAR_R '\122' 1034 #define CHAR_S '\123' 1035 #define CHAR_T '\124' 1036 #define CHAR_U '\125' 1037 #define CHAR_V '\126' 1038 #define CHAR_W '\127' 1039 #define CHAR_X '\130' 1040 #define CHAR_Y '\131' 1041 #define CHAR_Z '\132' 1042 #define CHAR_LEFT_SQUARE_BRACKET '\133' 1043 #define CHAR_BACKSLASH '\134' 1044 #define CHAR_RIGHT_SQUARE_BRACKET '\135' 1045 #define CHAR_CIRCUMFLEX_ACCENT '\136' 1046 #define CHAR_UNDERSCORE '\137' 1047 #define CHAR_GRAVE_ACCENT '\140' 1048 #define CHAR_a '\141' 1049 #define CHAR_b '\142' 1050 #define CHAR_c '\143' 1051 #define CHAR_d '\144' 1052 #define CHAR_e '\145' 1053 #define CHAR_f '\146' 1054 #define CHAR_g '\147' 1055 #define CHAR_h '\150' 1056 #define CHAR_i '\151' 1057 #define CHAR_j '\152' 1058 #define CHAR_k '\153' 1059 #define CHAR_l '\154' 1060 #define CHAR_m '\155' 1061 #define CHAR_n '\156' 1062 #define CHAR_o '\157' 1063 #define CHAR_p '\160' 1064 #define CHAR_q '\161' 1065 #define CHAR_r '\162' 1066 #define CHAR_s '\163' 1067 #define CHAR_t '\164' 1068 #define CHAR_u '\165' 1069 #define CHAR_v '\166' 1070 #define CHAR_w '\167' 1071 #define CHAR_x '\170' 1072 #define CHAR_y '\171' 1073 #define CHAR_z '\172' 1074 #define CHAR_LEFT_CURLY_BRACKET '\173' 1075 #define CHAR_VERTICAL_LINE '\174' 1076 #define CHAR_RIGHT_CURLY_BRACKET '\175' 1077 #define CHAR_TILDE '\176' 1078 #define CHAR_NBSP ((unsigned char)'\xa0') 1079 1080 #define STR_HT "\011" 1081 #define STR_VT "\013" 1082 #define STR_FF "\014" 1083 #define STR_CR "\015" 1084 #define STR_NL "\012" 1085 #define STR_BS "\010" 1086 #define STR_BEL "\007" 1087 #define STR_ESC "\033" 1088 #define STR_DEL "\177" 1089 1090 #define STR_SPACE "\040" 1091 #define STR_EXCLAMATION_MARK "\041" 1092 #define STR_QUOTATION_MARK "\042" 1093 #define STR_NUMBER_SIGN "\043" 1094 #define STR_DOLLAR_SIGN "\044" 1095 #define STR_PERCENT_SIGN "\045" 1096 #define STR_AMPERSAND "\046" 1097 #define STR_APOSTROPHE "\047" 1098 #define STR_LEFT_PARENTHESIS "\050" 1099 #define STR_RIGHT_PARENTHESIS "\051" 1100 #define STR_ASTERISK "\052" 1101 #define STR_PLUS "\053" 1102 #define STR_COMMA "\054" 1103 #define STR_MINUS "\055" 1104 #define STR_DOT "\056" 1105 #define STR_SLASH "\057" 1106 #define STR_0 "\060" 1107 #define STR_1 "\061" 1108 #define STR_2 "\062" 1109 #define STR_3 "\063" 1110 #define STR_4 "\064" 1111 #define STR_5 "\065" 1112 #define STR_6 "\066" 1113 #define STR_7 "\067" 1114 #define STR_8 "\070" 1115 #define STR_9 "\071" 1116 #define STR_COLON "\072" 1117 #define STR_SEMICOLON "\073" 1118 #define STR_LESS_THAN_SIGN "\074" 1119 #define STR_EQUALS_SIGN "\075" 1120 #define STR_GREATER_THAN_SIGN "\076" 1121 #define STR_QUESTION_MARK "\077" 1122 #define STR_COMMERCIAL_AT "\100" 1123 #define STR_A "\101" 1124 #define STR_B "\102" 1125 #define STR_C "\103" 1126 #define STR_D "\104" 1127 #define STR_E "\105" 1128 #define STR_F "\106" 1129 #define STR_G "\107" 1130 #define STR_H "\110" 1131 #define STR_I "\111" 1132 #define STR_J "\112" 1133 #define STR_K "\113" 1134 #define STR_L "\114" 1135 #define STR_M "\115" 1136 #define STR_N "\116" 1137 #define STR_O "\117" 1138 #define STR_P "\120" 1139 #define STR_Q "\121" 1140 #define STR_R "\122" 1141 #define STR_S "\123" 1142 #define STR_T "\124" 1143 #define STR_U "\125" 1144 #define STR_V "\126" 1145 #define STR_W "\127" 1146 #define STR_X "\130" 1147 #define STR_Y "\131" 1148 #define STR_Z "\132" 1149 #define STR_LEFT_SQUARE_BRACKET "\133" 1150 #define STR_BACKSLASH "\134" 1151 #define STR_RIGHT_SQUARE_BRACKET "\135" 1152 #define STR_CIRCUMFLEX_ACCENT "\136" 1153 #define STR_UNDERSCORE "\137" 1154 #define STR_GRAVE_ACCENT "\140" 1155 #define STR_a "\141" 1156 #define STR_b "\142" 1157 #define STR_c "\143" 1158 #define STR_d "\144" 1159 #define STR_e "\145" 1160 #define STR_f "\146" 1161 #define STR_g "\147" 1162 #define STR_h "\150" 1163 #define STR_i "\151" 1164 #define STR_j "\152" 1165 #define STR_k "\153" 1166 #define STR_l "\154" 1167 #define STR_m "\155" 1168 #define STR_n "\156" 1169 #define STR_o "\157" 1170 #define STR_p "\160" 1171 #define STR_q "\161" 1172 #define STR_r "\162" 1173 #define STR_s "\163" 1174 #define STR_t "\164" 1175 #define STR_u "\165" 1176 #define STR_v "\166" 1177 #define STR_w "\167" 1178 #define STR_x "\170" 1179 #define STR_y "\171" 1180 #define STR_z "\172" 1181 #define STR_LEFT_CURLY_BRACKET "\173" 1182 #define STR_VERTICAL_LINE "\174" 1183 #define STR_RIGHT_CURLY_BRACKET "\175" 1184 #define STR_TILDE "\176" 1185 1186 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0" 1187 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0" 1188 #define STRING_F0 STR_F "\0" 1189 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0" 1190 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0" 1191 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0" 1192 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0" 1193 #define STRING_THEN STR_T STR_H STR_E STR_N 1194 1195 #define STRING_atomic0 STR_a STR_t STR_o STR_m STR_i STR_c "\0" 1196 #define STRING_pla0 STR_p STR_l STR_a "\0" 1197 #define STRING_plb0 STR_p STR_l STR_b "\0" 1198 #define STRING_napla0 STR_n STR_a STR_p STR_l STR_a "\0" 1199 #define STRING_naplb0 STR_n STR_a STR_p STR_l STR_b "\0" 1200 #define STRING_nla0 STR_n STR_l STR_a "\0" 1201 #define STRING_nlb0 STR_n STR_l STR_b "\0" 1202 #define STRING_sr0 STR_s STR_r "\0" 1203 #define STRING_asr0 STR_a STR_s STR_r "\0" 1204 #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" 1205 #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" 1206 #define STRING_non_atomic_positive_lookahead0 STR_n STR_o STR_n STR_UNDERSCORE STR_a STR_t STR_o STR_m STR_i STR_c STR_UNDERSCORE 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" 1207 #define STRING_non_atomic_positive_lookbehind0 STR_n STR_o STR_n STR_UNDERSCORE STR_a STR_t STR_o STR_m STR_i STR_c STR_UNDERSCORE 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" 1208 #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" 1209 #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" 1210 #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" 1211 #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 1212 1213 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0" 1214 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0" 1215 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0" 1216 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0" 1217 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0" 1218 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0" 1219 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0" 1220 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0" 1221 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0" 1222 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0" 1223 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0" 1224 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0" 1225 #define STRING_word0 STR_w STR_o STR_r STR_d "\0" 1226 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t 1227 1228 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E 1229 #define STRING_VERSION STR_V STR_E STR_R STR_S STR_I STR_O STR_N 1230 #define STRING_WEIRD_STARTWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1231 #define STRING_WEIRD_ENDWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1232 1233 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS 1234 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS 1235 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1236 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS 1237 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1238 #define STRING_NUL_RIGHTPAR STR_N STR_U STR_L STR_RIGHT_PARENTHESIS 1239 #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 1240 #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 1241 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS 1242 #define STRING_UTF16_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS 1243 #define STRING_UTF32_RIGHTPAR STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS 1244 #define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_RIGHT_PARENTHESIS 1245 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS 1246 #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 1247 #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 1248 #define STRING_NO_JIT_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_J STR_I STR_T STR_RIGHT_PARENTHESIS 1249 #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 1250 #define STRING_NOTEMPTY_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_RIGHT_PARENTHESIS 1251 #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 1252 #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 1253 #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 1254 #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 1255 #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 1256 #define STRING_MARK STR_M STR_A STR_R STR_K 1257 1258 #define STRING_bc STR_b STR_c 1259 #define STRING_bidiclass STR_b STR_i STR_d STR_i STR_c STR_l STR_a STR_s STR_s 1260 #define STRING_sc STR_s STR_c 1261 #define STRING_script STR_s STR_c STR_r STR_i STR_p STR_t 1262 #define STRING_scriptextensions STR_s STR_c STR_r STR_i STR_p STR_t STR_e STR_x STR_t STR_e STR_n STR_s STR_i STR_o STR_n STR_s 1263 #define STRING_scx STR_s STR_c STR_x 1264 1265 1266 #endif /* SUPPORT_UNICODE */ 1267 1268 /* -------------------- End of character and string names -------------------*/ 1269 1270 /* -------------------- Definitions for compiled patterns -------------------*/ 1271 1272 /* Codes for different types of Unicode property. If these definitions are 1273 changed, the autopossessifying table in pcre2_auto_possess.c must be updated to 1274 match. */ 1275 1276 #define PT_ANY 0 /* Any property - matches all chars */ 1277 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */ 1278 #define PT_GC 2 /* Specified general characteristic (e.g. L) */ 1279 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */ 1280 #define PT_SC 4 /* Script only (e.g. Han) */ 1281 #define PT_SCX 5 /* Script extensions (includes SC) */ 1282 #define PT_ALNUM 6 /* Alphanumeric - the union of L and N */ 1283 #define PT_SPACE 7 /* Perl space - general category Z plus 9,10,12,13 */ 1284 #define PT_PXSPACE 8 /* POSIX space - Z plus 9,10,11,12,13 */ 1285 #define PT_WORD 9 /* Word - L plus N plus underscore */ 1286 #define PT_CLIST 10 /* Pseudo-property: match character list */ 1287 #define PT_UCNC 11 /* Universal Character nameable character */ 1288 #define PT_BIDICL 12 /* Specified bidi class */ 1289 #define PT_BOOL 13 /* Boolean property */ 1290 #define PT_TABSIZE 14 /* Size of square table for autopossessify tests */ 1291 1292 /* The following special properties are used only in XCLASS items, when POSIX 1293 classes are specified and PCRE2_UCP is set - in other words, for Unicode 1294 handling of these classes. They are not available via the \p or \P escapes like 1295 those in the above list, and so they do not take part in the autopossessifying 1296 table. */ 1297 1298 #define PT_PXGRAPH 14 /* [:graph:] - characters that mark the paper */ 1299 #define PT_PXPRINT 15 /* [:print:] - [:graph:] plus non-control spaces */ 1300 #define PT_PXPUNCT 16 /* [:punct:] - punctuation characters */ 1301 1302 /* This value is used when parsing \p and \P escapes to indicate that neither 1303 \p{script:...} nor \p{scx:...} has been encountered. */ 1304 1305 #define PT_NOTSCRIPT 255 1306 1307 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that 1308 contain characters with values greater than 255. */ 1309 1310 #define XCL_NOT 0x01 /* Flag: this is a negative class */ 1311 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ 1312 #define XCL_HASPROP 0x04 /* Flag: property checks are present. */ 1313 1314 #define XCL_END 0 /* Marks end of individual items */ 1315 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */ 1316 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */ 1317 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */ 1318 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */ 1319 1320 /* These are escaped items that aren't just an encoding of a particular data 1321 value such as \n. They must have non-zero values, as check_escape() returns 0 1322 for a data character. In the escapes[] table in pcre2_compile.c their values 1323 are negated in order to distinguish them from data values. 1324 1325 They must appear here in the same order as in the opcode definitions below, up 1326 to ESC_z. There's a dummy for OP_ALLANY because it corresponds to "." in DOTALL 1327 mode rather than an escape sequence. It is also used for [^] in JavaScript 1328 compatibility mode, and for \C in non-utf mode. In non-DOTALL mode, "." behaves 1329 like \N. 1330 1331 Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in 1332 check_escape(). There are tests in the code for an escape greater than ESC_b 1333 and less than ESC_Z to detect the types that may be repeated. These are the 1334 types that consume characters. If any new escapes are put in between that don't 1335 consume a character, that code will have to change. */ 1336 1337 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, 1338 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, 1339 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, 1340 ESC_E, ESC_Q, ESC_g, ESC_k }; 1341 1342 1343 /********************** Opcode definitions ******************/ 1344 1345 /****** NOTE NOTE NOTE ****** 1346 1347 Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in 1348 order to the list of escapes immediately above. Furthermore, values up to 1349 OP_DOLLM must not be changed without adjusting the table called autoposstab in 1350 pcre2_auto_possess.c. 1351 1352 Whenever this list is updated, the two macro definitions that follow must be 1353 updated to match. The possessification table called "opcode_possessify" in 1354 pcre2_compile.c must also be updated, and also the tables called "coptable" 1355 and "poptable" in pcre2_dfa_match.c. 1356 1357 ****** NOTE NOTE NOTE ******/ 1358 1359 1360 /* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive, 1361 are used in a table for deciding whether a repeated character type can be 1362 auto-possessified. */ 1363 1364 #define FIRST_AUTOTAB_OP OP_NOT_DIGIT 1365 #define LAST_AUTOTAB_LEFT_OP OP_EXTUNI 1366 #define LAST_AUTOTAB_RIGHT_OP OP_DOLLM 1367 1368 enum { 1369 OP_END, /* 0 End of pattern */ 1370 1371 /* Values corresponding to backslashed metacharacters */ 1372 1373 OP_SOD, /* 1 Start of data: \A */ 1374 OP_SOM, /* 2 Start of match (subject + offset): \G */ 1375 OP_SET_SOM, /* 3 Set start of match (\K) */ 1376 OP_NOT_WORD_BOUNDARY, /* 4 \B */ 1377 OP_WORD_BOUNDARY, /* 5 \b */ 1378 OP_NOT_DIGIT, /* 6 \D */ 1379 OP_DIGIT, /* 7 \d */ 1380 OP_NOT_WHITESPACE, /* 8 \S */ 1381 OP_WHITESPACE, /* 9 \s */ 1382 OP_NOT_WORDCHAR, /* 10 \W */ 1383 OP_WORDCHAR, /* 11 \w */ 1384 1385 OP_ANY, /* 12 Match any character except newline (\N) */ 1386 OP_ALLANY, /* 13 Match any character */ 1387 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */ 1388 OP_NOTPROP, /* 15 \P (not Unicode property) */ 1389 OP_PROP, /* 16 \p (Unicode property) */ 1390 OP_ANYNL, /* 17 \R (any newline sequence) */ 1391 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */ 1392 OP_HSPACE, /* 19 \h (horizontal whitespace) */ 1393 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */ 1394 OP_VSPACE, /* 21 \v (vertical whitespace) */ 1395 OP_EXTUNI, /* 22 \X (extended Unicode sequence */ 1396 OP_EODN, /* 23 End of data or \n at end of data (\Z) */ 1397 OP_EOD, /* 24 End of data (\z) */ 1398 1399 /* Line end assertions */ 1400 1401 OP_DOLL, /* 25 End of line - not multiline */ 1402 OP_DOLLM, /* 26 End of line - multiline */ 1403 OP_CIRC, /* 27 Start of line - not multiline */ 1404 OP_CIRCM, /* 28 Start of line - multiline */ 1405 1406 /* Single characters; caseful must precede the caseless ones, and these 1407 must remain in this order, and adjacent. */ 1408 1409 OP_CHAR, /* 29 Match one character, casefully */ 1410 OP_CHARI, /* 30 Match one character, caselessly */ 1411 OP_NOT, /* 31 Match one character, not the given one, casefully */ 1412 OP_NOTI, /* 32 Match one character, not the given one, caselessly */ 1413 1414 /* The following sets of 13 opcodes must always be kept in step because 1415 the offset from the first one is used to generate the others. */ 1416 1417 /* Repeated characters; caseful must precede the caseless ones */ 1418 1419 OP_STAR, /* 33 The maximizing and minimizing versions of */ 1420 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */ 1421 OP_PLUS, /* 35 the minimizing one second. */ 1422 OP_MINPLUS, /* 36 */ 1423 OP_QUERY, /* 37 */ 1424 OP_MINQUERY, /* 38 */ 1425 1426 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/ 1427 OP_MINUPTO, /* 40 */ 1428 OP_EXACT, /* 41 Exactly n matches */ 1429 1430 OP_POSSTAR, /* 42 Possessified star, caseful */ 1431 OP_POSPLUS, /* 43 Possessified plus, caseful */ 1432 OP_POSQUERY, /* 44 Posesssified query, caseful */ 1433 OP_POSUPTO, /* 45 Possessified upto, caseful */ 1434 1435 /* Repeated characters; caseless must follow the caseful ones */ 1436 1437 OP_STARI, /* 46 */ 1438 OP_MINSTARI, /* 47 */ 1439 OP_PLUSI, /* 48 */ 1440 OP_MINPLUSI, /* 49 */ 1441 OP_QUERYI, /* 50 */ 1442 OP_MINQUERYI, /* 51 */ 1443 1444 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */ 1445 OP_MINUPTOI, /* 53 */ 1446 OP_EXACTI, /* 54 */ 1447 1448 OP_POSSTARI, /* 55 Possessified star, caseless */ 1449 OP_POSPLUSI, /* 56 Possessified plus, caseless */ 1450 OP_POSQUERYI, /* 57 Posesssified query, caseless */ 1451 OP_POSUPTOI, /* 58 Possessified upto, caseless */ 1452 1453 /* The negated ones must follow the non-negated ones, and match them */ 1454 /* Negated repeated character, caseful; must precede the caseless ones */ 1455 1456 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */ 1457 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */ 1458 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */ 1459 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */ 1460 OP_NOTQUERY, /* 63 */ 1461 OP_NOTMINQUERY, /* 64 */ 1462 1463 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */ 1464 OP_NOTMINUPTO, /* 66 */ 1465 OP_NOTEXACT, /* 67 Exactly n matches */ 1466 1467 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */ 1468 OP_NOTPOSPLUS, /* 69 */ 1469 OP_NOTPOSQUERY, /* 70 */ 1470 OP_NOTPOSUPTO, /* 71 */ 1471 1472 /* Negated repeated character, caseless; must follow the caseful ones */ 1473 1474 OP_NOTSTARI, /* 72 */ 1475 OP_NOTMINSTARI, /* 73 */ 1476 OP_NOTPLUSI, /* 74 */ 1477 OP_NOTMINPLUSI, /* 75 */ 1478 OP_NOTQUERYI, /* 76 */ 1479 OP_NOTMINQUERYI, /* 77 */ 1480 1481 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */ 1482 OP_NOTMINUPTOI, /* 79 */ 1483 OP_NOTEXACTI, /* 80 Exactly n matches */ 1484 1485 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */ 1486 OP_NOTPOSPLUSI, /* 82 */ 1487 OP_NOTPOSQUERYI, /* 83 */ 1488 OP_NOTPOSUPTOI, /* 84 */ 1489 1490 /* Character types */ 1491 1492 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */ 1493 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */ 1494 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */ 1495 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */ 1496 OP_TYPEQUERY, /* 89 */ 1497 OP_TYPEMINQUERY, /* 90 */ 1498 1499 OP_TYPEUPTO, /* 91 From 0 to n matches */ 1500 OP_TYPEMINUPTO, /* 92 */ 1501 OP_TYPEEXACT, /* 93 Exactly n matches */ 1502 1503 OP_TYPEPOSSTAR, /* 94 Possessified versions */ 1504 OP_TYPEPOSPLUS, /* 95 */ 1505 OP_TYPEPOSQUERY, /* 96 */ 1506 OP_TYPEPOSUPTO, /* 97 */ 1507 1508 /* These are used for character classes and back references; only the 1509 first six are the same as the sets above. */ 1510 1511 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */ 1512 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */ 1513 OP_CRPLUS, /* 100 the minimizing one second. These codes must */ 1514 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */ 1515 OP_CRQUERY, /* 102 */ 1516 OP_CRMINQUERY, /* 103 */ 1517 1518 OP_CRRANGE, /* 104 These are different to the three sets above. */ 1519 OP_CRMINRANGE, /* 105 */ 1520 1521 OP_CRPOSSTAR, /* 106 Possessified versions */ 1522 OP_CRPOSPLUS, /* 107 */ 1523 OP_CRPOSQUERY, /* 108 */ 1524 OP_CRPOSRANGE, /* 109 */ 1525 1526 /* End of quantifier opcodes */ 1527 1528 OP_CLASS, /* 110 Match a character class, chars < 256 only */ 1529 OP_NCLASS, /* 111 Same, but the bitmap was created from a negative 1530 class - the difference is relevant only when a 1531 character > 255 is encountered. */ 1532 OP_XCLASS, /* 112 Extended class for handling > 255 chars within the 1533 class. This does both positive and negative. */ 1534 OP_REF, /* 113 Match a back reference, casefully */ 1535 OP_REFI, /* 114 Match a back reference, caselessly */ 1536 OP_DNREF, /* 115 Match a duplicate name backref, casefully */ 1537 OP_DNREFI, /* 116 Match a duplicate name backref, caselessly */ 1538 OP_RECURSE, /* 117 Match a numbered subpattern (possibly recursive) */ 1539 OP_CALLOUT, /* 118 Call out to external function if provided */ 1540 OP_CALLOUT_STR, /* 119 Call out with string argument */ 1541 1542 OP_ALT, /* 120 Start of alternation */ 1543 OP_KET, /* 121 End of group that doesn't have an unbounded repeat */ 1544 OP_KETRMAX, /* 122 These two must remain together and in this */ 1545 OP_KETRMIN, /* 123 order. They are for groups the repeat for ever. */ 1546 OP_KETRPOS, /* 124 Possessive unlimited repeat. */ 1547 1548 /* The assertions must come before BRA, CBRA, ONCE, and COND. */ 1549 1550 OP_REVERSE, /* 125 Move pointer back - used in lookbehind assertions */ 1551 OP_ASSERT, /* 126 Positive lookahead */ 1552 OP_ASSERT_NOT, /* 127 Negative lookahead */ 1553 OP_ASSERTBACK, /* 128 Positive lookbehind */ 1554 OP_ASSERTBACK_NOT, /* 129 Negative lookbehind */ 1555 OP_ASSERT_NA, /* 130 Positive non-atomic lookahead */ 1556 OP_ASSERTBACK_NA, /* 131 Positive non-atomic lookbehind */ 1557 1558 /* ONCE, SCRIPT_RUN, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come 1559 immediately after the assertions, with ONCE first, as there's a test for >= 1560 ONCE for a subpattern that isn't an assertion. The POS versions must 1561 immediately follow the non-POS versions in each case. */ 1562 1563 OP_ONCE, /* 132 Atomic group, contains captures */ 1564 OP_SCRIPT_RUN, /* 133 Non-capture, but check characters' scripts */ 1565 OP_BRA, /* 134 Start of non-capturing bracket */ 1566 OP_BRAPOS, /* 135 Ditto, with unlimited, possessive repeat */ 1567 OP_CBRA, /* 136 Start of capturing bracket */ 1568 OP_CBRAPOS, /* 137 Ditto, with unlimited, possessive repeat */ 1569 OP_COND, /* 138 Conditional group */ 1570 1571 /* These five must follow the previous five, in the same order. There's a 1572 check for >= SBRA to distinguish the two sets. */ 1573 1574 OP_SBRA, /* 139 Start of non-capturing bracket, check empty */ 1575 OP_SBRAPOS, /* 149 Ditto, with unlimited, possessive repeat */ 1576 OP_SCBRA, /* 141 Start of capturing bracket, check empty */ 1577 OP_SCBRAPOS, /* 142 Ditto, with unlimited, possessive repeat */ 1578 OP_SCOND, /* 143 Conditional group, check empty */ 1579 1580 /* The next two pairs must (respectively) be kept together. */ 1581 1582 OP_CREF, /* 144 Used to hold a capture number as condition */ 1583 OP_DNCREF, /* 145 Used to point to duplicate names as a condition */ 1584 OP_RREF, /* 146 Used to hold a recursion number as condition */ 1585 OP_DNRREF, /* 147 Used to point to duplicate names as a condition */ 1586 OP_FALSE, /* 148 Always false (used by DEFINE and VERSION) */ 1587 OP_TRUE, /* 149 Always true (used by VERSION) */ 1588 1589 OP_BRAZERO, /* 150 These two must remain together and in this */ 1590 OP_BRAMINZERO, /* 151 order. */ 1591 OP_BRAPOSZERO, /* 152 */ 1592 1593 /* These are backtracking control verbs */ 1594 1595 OP_MARK, /* 153 always has an argument */ 1596 OP_PRUNE, /* 154 */ 1597 OP_PRUNE_ARG, /* 155 same, but with argument */ 1598 OP_SKIP, /* 156 */ 1599 OP_SKIP_ARG, /* 157 same, but with argument */ 1600 OP_THEN, /* 158 */ 1601 OP_THEN_ARG, /* 159 same, but with argument */ 1602 OP_COMMIT, /* 160 */ 1603 OP_COMMIT_ARG, /* 161 same, but with argument */ 1604 1605 /* These are forced failure and success verbs. FAIL and ACCEPT do accept an 1606 argument, but these cases can be compiled as, for example, (*MARK:X)(*FAIL) 1607 without the need for a special opcode. */ 1608 1609 OP_FAIL, /* 162 */ 1610 OP_ACCEPT, /* 163 */ 1611 OP_ASSERT_ACCEPT, /* 164 Used inside assertions */ 1612 OP_CLOSE, /* 165 Used before OP_ACCEPT to close open captures */ 1613 1614 /* This is used to skip a subpattern with a {0} quantifier */ 1615 1616 OP_SKIPZERO, /* 166 */ 1617 1618 /* This is used to identify a DEFINE group during compilation so that it can 1619 be checked for having only one branch. It is changed to OP_FALSE before 1620 compilation finishes. */ 1621 1622 OP_DEFINE, /* 167 */ 1623 1624 /* This is not an opcode, but is used to check that tables indexed by opcode 1625 are the correct length, in order to catch updating errors - there have been 1626 some in the past. */ 1627 1628 OP_TABLE_LENGTH 1629 1630 }; 1631 1632 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro 1633 definitions that follow must also be updated to match. There are also tables 1634 called "opcode_possessify" in pcre2_compile.c and "coptable" and "poptable" in 1635 pcre2_dfa_match.c that must be updated. */ 1636 1637 1638 /* This macro defines textual names for all the opcodes. These are used only 1639 for debugging, and some of them are only partial names. The macro is referenced 1640 only in pcre2_printint.c, which fills out the full names in many cases (and in 1641 some cases doesn't actually use these names at all). */ 1642 1643 #define OP_NAME_LIST \ 1644 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \ 1645 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \ 1646 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \ 1647 "extuni", "\\Z", "\\z", \ 1648 "$", "$", "^", "^", "char", "chari", "not", "noti", \ 1649 "*", "*?", "+", "+?", "?", "??", \ 1650 "{", "{", "{", \ 1651 "*+","++", "?+", "{", \ 1652 "*", "*?", "+", "+?", "?", "??", \ 1653 "{", "{", "{", \ 1654 "*+","++", "?+", "{", \ 1655 "*", "*?", "+", "+?", "?", "??", \ 1656 "{", "{", "{", \ 1657 "*+","++", "?+", "{", \ 1658 "*", "*?", "+", "+?", "?", "??", \ 1659 "{", "{", "{", \ 1660 "*+","++", "?+", "{", \ 1661 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \ 1662 "*+","++", "?+", "{", \ 1663 "*", "*?", "+", "+?", "?", "??", "{", "{", \ 1664 "*+","++", "?+", "{", \ 1665 "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi", \ 1666 "Recurse", "Callout", "CalloutStr", \ 1667 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \ 1668 "Reverse", "Assert", "Assert not", \ 1669 "Assert back", "Assert back not", \ 1670 "Non-atomic assert", "Non-atomic assert back", \ 1671 "Once", \ 1672 "Script run", \ 1673 "Bra", "BraPos", "CBra", "CBraPos", \ 1674 "Cond", \ 1675 "SBra", "SBraPos", "SCBra", "SCBraPos", \ 1676 "SCond", \ 1677 "Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", \ 1678 "Cond false", "Cond true", \ 1679 "Brazero", "Braminzero", "Braposzero", \ 1680 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \ 1681 "*THEN", "*THEN", "*COMMIT", "*COMMIT", "*FAIL", \ 1682 "*ACCEPT", "*ASSERT_ACCEPT", \ 1683 "Close", "Skip zero", "Define" 1684 1685 1686 /* This macro defines the length of fixed length operations in the compiled 1687 regex. The lengths are used when searching for specific things, and also in the 1688 debugging printing of a compiled regex. We use a macro so that it can be 1689 defined close to the definitions of the opcodes themselves. 1690 1691 As things have been extended, some of these are no longer fixed lenths, but are 1692 minima instead. For example, the length of a single-character repeat may vary 1693 in UTF-8 mode. The code that uses this table must know about such things. */ 1694 1695 #define OP_LENGTHS \ 1696 1, /* End */ \ 1697 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \ 1698 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \ 1699 1, 1, 1, /* Any, AllAny, Anybyte */ \ 1700 3, 3, /* \P, \p */ \ 1701 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \ 1702 1, /* \X */ \ 1703 1, 1, 1, 1, 1, 1, /* \Z, \z, $, $M ^, ^M */ \ 1704 2, /* Char - the minimum length */ \ 1705 2, /* Chari - the minimum length */ \ 1706 2, /* not */ \ 1707 2, /* noti */ \ 1708 /* Positive single-char repeats ** These are */ \ 1709 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ 1710 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \ 1711 2+IMM2_SIZE, /* exact */ \ 1712 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \ 1713 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \ 1714 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \ 1715 2+IMM2_SIZE, /* exact I */ \ 1716 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \ 1717 /* Negative single-char repeats - only for chars < 256 */ \ 1718 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ 1719 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \ 1720 2+IMM2_SIZE, /* NOT exact */ \ 1721 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \ 1722 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \ 1723 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \ 1724 2+IMM2_SIZE, /* NOT exact I */ \ 1725 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \ 1726 /* Positive type repeats */ \ 1727 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ 1728 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \ 1729 2+IMM2_SIZE, /* Type exact */ \ 1730 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \ 1731 /* Character class & ref repeats */ \ 1732 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ 1733 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \ 1734 1, 1, 1, 1+2*IMM2_SIZE, /* Possessive *+, ++, ?+, CRPOSRANGE */ \ 1735 1+(32/sizeof(PCRE2_UCHAR)), /* CLASS */ \ 1736 1+(32/sizeof(PCRE2_UCHAR)), /* NCLASS */ \ 1737 0, /* XCLASS - variable length */ \ 1738 1+IMM2_SIZE, /* REF */ \ 1739 1+IMM2_SIZE, /* REFI */ \ 1740 1+2*IMM2_SIZE, /* DNREF */ \ 1741 1+2*IMM2_SIZE, /* DNREFI */ \ 1742 1+LINK_SIZE, /* RECURSE */ \ 1743 1+2*LINK_SIZE+1, /* CALLOUT */ \ 1744 0, /* CALLOUT_STR - variable length */ \ 1745 1+LINK_SIZE, /* Alt */ \ 1746 1+LINK_SIZE, /* Ket */ \ 1747 1+LINK_SIZE, /* KetRmax */ \ 1748 1+LINK_SIZE, /* KetRmin */ \ 1749 1+LINK_SIZE, /* KetRpos */ \ 1750 1+LINK_SIZE, /* Reverse */ \ 1751 1+LINK_SIZE, /* Assert */ \ 1752 1+LINK_SIZE, /* Assert not */ \ 1753 1+LINK_SIZE, /* Assert behind */ \ 1754 1+LINK_SIZE, /* Assert behind not */ \ 1755 1+LINK_SIZE, /* NA Assert */ \ 1756 1+LINK_SIZE, /* NA Assert behind */ \ 1757 1+LINK_SIZE, /* ONCE */ \ 1758 1+LINK_SIZE, /* SCRIPT_RUN */ \ 1759 1+LINK_SIZE, /* BRA */ \ 1760 1+LINK_SIZE, /* BRAPOS */ \ 1761 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \ 1762 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \ 1763 1+LINK_SIZE, /* COND */ \ 1764 1+LINK_SIZE, /* SBRA */ \ 1765 1+LINK_SIZE, /* SBRAPOS */ \ 1766 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \ 1767 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \ 1768 1+LINK_SIZE, /* SCOND */ \ 1769 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* CREF, DNCREF */ \ 1770 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* RREF, DNRREF */ \ 1771 1, 1, /* FALSE, TRUE */ \ 1772 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ 1773 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ 1774 1, 3, /* SKIP, SKIP_ARG */ \ 1775 1, 3, /* THEN, THEN_ARG */ \ 1776 1, 3, /* COMMIT, COMMIT_ARG */ \ 1777 1, 1, 1, /* FAIL, ACCEPT, ASSERT_ACCEPT */ \ 1778 1+IMM2_SIZE, 1, /* CLOSE, SKIPZERO */ \ 1779 1 /* DEFINE */ 1780 1781 /* A magic value for OP_RREF to indicate the "any recursion" condition. */ 1782 1783 #define RREF_ANY 0xffff 1784 1785 1786 /* ---------- Private structures that are mode-independent. ---------- */ 1787 1788 /* Structure to hold data for custom memory management. */ 1789 1790 typedef struct pcre2_memctl { 1791 void * (*malloc)(size_t, void *); 1792 void (*free)(void *, void *); 1793 void *memory_data; 1794 } pcre2_memctl; 1795 1796 /* Structure for building a chain of open capturing subpatterns during 1797 compiling, so that instructions to close them can be compiled when (*ACCEPT) is 1798 encountered. */ 1799 1800 typedef struct open_capitem { 1801 struct open_capitem *next; /* Chain link */ 1802 uint16_t number; /* Capture number */ 1803 uint16_t assert_depth; /* Assertion depth when opened */ 1804 } open_capitem; 1805 1806 /* Layout of the UCP type table that translates property names into types and 1807 codes. Each entry used to point directly to a name, but to reduce the number of 1808 relocations in shared libraries, it now has an offset into a single string 1809 instead. */ 1810 1811 typedef struct { 1812 uint16_t name_offset; 1813 uint16_t type; 1814 uint16_t value; 1815 } ucp_type_table; 1816 1817 /* Unicode character database (UCD) record format */ 1818 1819 typedef struct { 1820 uint8_t script; /* ucp_Arabic, etc. */ 1821 uint8_t chartype; /* ucp_Cc, etc. (general categories) */ 1822 uint8_t gbprop; /* ucp_gbControl, etc. (grapheme break property) */ 1823 uint8_t caseset; /* offset to multichar other cases or zero */ 1824 int32_t other_case; /* offset to other case, or zero if none */ 1825 uint16_t scriptx_bidiclass; /* script extension (11 bit) and bidi class (5 bit) values */ 1826 uint16_t bprops; /* binary properties offset */ 1827 } ucd_record; 1828 1829 /* UCD access macros */ 1830 1831 #define UCD_BLOCK_SIZE 128 1832 #define REAL_GET_UCD(ch) (PRIV(ucd_records) + \ 1833 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \ 1834 UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE]) 1835 1836 #if PCRE2_CODE_UNIT_WIDTH == 32 1837 #define GET_UCD(ch) ((ch > MAX_UTF_CODE_POINT)? \ 1838 PRIV(dummy_ucd_record) : REAL_GET_UCD(ch)) 1839 #else 1840 #define GET_UCD(ch) REAL_GET_UCD(ch) 1841 #endif 1842 1843 #define UCD_SCRIPTX_MASK 0x3ff 1844 #define UCD_BIDICLASS_SHIFT 11 1845 #define UCD_BPROPS_MASK 0xfff 1846 1847 #define UCD_SCRIPTX_PROP(prop) ((prop)->scriptx_bidiclass & UCD_SCRIPTX_MASK) 1848 #define UCD_BIDICLASS_PROP(prop) ((prop)->scriptx_bidiclass >> UCD_BIDICLASS_SHIFT) 1849 #define UCD_BPROPS_PROP(prop) ((prop)->bprops & UCD_BPROPS_MASK) 1850 1851 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype 1852 #define UCD_SCRIPT(ch) GET_UCD(ch)->script 1853 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)] 1854 #define UCD_GRAPHBREAK(ch) GET_UCD(ch)->gbprop 1855 #define UCD_CASESET(ch) GET_UCD(ch)->caseset 1856 #define UCD_OTHERCASE(ch) ((uint32_t)((int)ch + (int)(GET_UCD(ch)->other_case))) 1857 #define UCD_SCRIPTX(ch) UCD_SCRIPTX_PROP(GET_UCD(ch)) 1858 #define UCD_BPROPS(ch) UCD_BPROPS_PROP(GET_UCD(ch)) 1859 #define UCD_BIDICLASS(ch) UCD_BIDICLASS_PROP(GET_UCD(ch)) 1860 1861 /* The "scriptx" and bprops fields contain offsets into vectors of 32-bit words 1862 that form a bitmap representing a list of scripts or boolean properties. These 1863 macros test or set a bit in the map by number. */ 1864 1865 #define MAPBIT(map,n) ((map)[(n)/32]&(1u<<((n)%32))) 1866 #define MAPSET(map,n) ((map)[(n)/32]|=(1u<<((n)%32))) 1867 1868 /* Header for serialized pcre2 codes. */ 1869 1870 typedef struct pcre2_serialized_data { 1871 uint32_t magic; 1872 uint32_t version; 1873 uint32_t config; 1874 int32_t number_of_codes; 1875 } pcre2_serialized_data; 1876 1877 1878 1879 /* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */ 1880 1881 /* When this file is included by pcre2test, PCRE2_CODE_UNIT_WIDTH is defined as 1882 0, so the following items are omitted. */ 1883 1884 #if defined PCRE2_CODE_UNIT_WIDTH && PCRE2_CODE_UNIT_WIDTH != 0 1885 1886 /* EBCDIC is supported only for the 8-bit library. */ 1887 1888 #if defined EBCDIC && PCRE2_CODE_UNIT_WIDTH != 8 1889 #error EBCDIC is not supported for the 16-bit or 32-bit libraries 1890 #endif 1891 1892 /* This is the largest non-UTF code point. */ 1893 1894 #define MAX_NON_UTF_CHAR (0xffffffffU >> (32 - PCRE2_CODE_UNIT_WIDTH)) 1895 1896 /* Internal shared data tables and variables. These are used by more than one 1897 of the exported public functions. They have to be "external" in the C sense, 1898 but are not part of the PCRE2 public API. Although the data for some of them is 1899 identical in all libraries, they must have different names so that multiple 1900 libraries can be simultaneously linked to a single application. However, UTF-8 1901 tables are needed only when compiling the 8-bit library. */ 1902 1903 #if PCRE2_CODE_UNIT_WIDTH == 8 1904 extern const int PRIV(utf8_table1)[]; 1905 extern const int PRIV(utf8_table1_size); 1906 extern const int PRIV(utf8_table2)[]; 1907 extern const int PRIV(utf8_table3)[]; 1908 extern const uint8_t PRIV(utf8_table4)[]; 1909 #endif 1910 1911 #define _pcre2_OP_lengths PCRE2_SUFFIX(_pcre2_OP_lengths_) 1912 #define _pcre2_callout_end_delims PCRE2_SUFFIX(_pcre2_callout_end_delims_) 1913 #define _pcre2_callout_start_delims PCRE2_SUFFIX(_pcre2_callout_start_delims_) 1914 #define _pcre2_default_compile_context PCRE2_SUFFIX(_pcre2_default_compile_context_) 1915 #define _pcre2_default_convert_context PCRE2_SUFFIX(_pcre2_default_convert_context_) 1916 #define _pcre2_default_match_context PCRE2_SUFFIX(_pcre2_default_match_context_) 1917 #define _pcre2_default_tables PCRE2_SUFFIX(_pcre2_default_tables_) 1918 #if PCRE2_CODE_UNIT_WIDTH == 32 1919 #define _pcre2_dummy_ucd_record PCRE2_SUFFIX(_pcre2_dummy_ucd_record_) 1920 #endif 1921 #define _pcre2_hspace_list PCRE2_SUFFIX(_pcre2_hspace_list_) 1922 #define _pcre2_vspace_list PCRE2_SUFFIX(_pcre2_vspace_list_) 1923 #define _pcre2_ucd_boolprop_sets PCRE2_SUFFIX(_pcre2_ucd_boolprop_sets_) 1924 #define _pcre2_ucd_caseless_sets PCRE2_SUFFIX(_pcre2_ucd_caseless_sets_) 1925 #define _pcre2_ucd_digit_sets PCRE2_SUFFIX(_pcre2_ucd_digit_sets_) 1926 #define _pcre2_ucd_script_sets PCRE2_SUFFIX(_pcre2_ucd_script_sets_) 1927 #define _pcre2_ucd_records PCRE2_SUFFIX(_pcre2_ucd_records_) 1928 #define _pcre2_ucd_stage1 PCRE2_SUFFIX(_pcre2_ucd_stage1_) 1929 #define _pcre2_ucd_stage2 PCRE2_SUFFIX(_pcre2_ucd_stage2_) 1930 #define _pcre2_ucp_gbtable PCRE2_SUFFIX(_pcre2_ucp_gbtable_) 1931 #define _pcre2_ucp_gentype PCRE2_SUFFIX(_pcre2_ucp_gentype_) 1932 #define _pcre2_ucp_typerange PCRE2_SUFFIX(_pcre2_ucp_typerange_) 1933 #define _pcre2_unicode_version PCRE2_SUFFIX(_pcre2_unicode_version_) 1934 #define _pcre2_utt PCRE2_SUFFIX(_pcre2_utt_) 1935 #define _pcre2_utt_names PCRE2_SUFFIX(_pcre2_utt_names_) 1936 #define _pcre2_utt_size PCRE2_SUFFIX(_pcre2_utt_size_) 1937 1938 extern const uint8_t PRIV(OP_lengths)[]; 1939 extern const uint32_t PRIV(callout_end_delims)[]; 1940 extern const uint32_t PRIV(callout_start_delims)[]; 1941 extern const pcre2_compile_context PRIV(default_compile_context); 1942 extern const pcre2_convert_context PRIV(default_convert_context); 1943 extern const pcre2_match_context PRIV(default_match_context); 1944 extern const uint8_t PRIV(default_tables)[]; 1945 extern const uint32_t PRIV(hspace_list)[]; 1946 extern const uint32_t PRIV(vspace_list)[]; 1947 extern const uint32_t PRIV(ucd_boolprop_sets)[]; 1948 extern const uint32_t PRIV(ucd_caseless_sets)[]; 1949 extern const uint32_t PRIV(ucd_digit_sets)[]; 1950 extern const uint32_t PRIV(ucd_script_sets)[]; 1951 extern const ucd_record PRIV(ucd_records)[]; 1952 #if PCRE2_CODE_UNIT_WIDTH == 32 1953 extern const ucd_record PRIV(dummy_ucd_record)[]; 1954 #endif 1955 extern const uint16_t PRIV(ucd_stage1)[]; 1956 extern const uint16_t PRIV(ucd_stage2)[]; 1957 extern const uint32_t PRIV(ucp_gbtable)[]; 1958 extern const uint32_t PRIV(ucp_gentype)[]; 1959 #ifdef SUPPORT_JIT 1960 extern const int PRIV(ucp_typerange)[]; 1961 #endif 1962 extern const char *PRIV(unicode_version); 1963 extern const ucp_type_table PRIV(utt)[]; 1964 extern const char PRIV(utt_names)[]; 1965 extern const size_t PRIV(utt_size); 1966 1967 /* Mode-dependent macros and hidden and private structures are defined in a 1968 separate file so that pcre2test can include them at all supported widths. When 1969 compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we can 1970 include them at the appropriate width, after setting up suffix macros for the 1971 private structures. */ 1972 1973 #define branch_chain PCRE2_SUFFIX(branch_chain_) 1974 #define compile_block PCRE2_SUFFIX(compile_block_) 1975 #define dfa_match_block PCRE2_SUFFIX(dfa_match_block_) 1976 #define match_block PCRE2_SUFFIX(match_block_) 1977 #define named_group PCRE2_SUFFIX(named_group_) 1978 1979 #include "pcre2_intmodedep.h" 1980 1981 /* Private "external" functions. These are internal functions that are called 1982 from modules other than the one in which they are defined. They have to be 1983 "external" in the C sense, but are not part of the PCRE2 public API. They are 1984 not referenced from pcre2test, and must not be defined when no code unit width 1985 is available. */ 1986 1987 #define _pcre2_auto_possessify PCRE2_SUFFIX(_pcre2_auto_possessify_) 1988 #define _pcre2_check_escape PCRE2_SUFFIX(_pcre2_check_escape_) 1989 #define _pcre2_extuni PCRE2_SUFFIX(_pcre2_extuni_) 1990 #define _pcre2_find_bracket PCRE2_SUFFIX(_pcre2_find_bracket_) 1991 #define _pcre2_is_newline PCRE2_SUFFIX(_pcre2_is_newline_) 1992 #define _pcre2_jit_free_rodata PCRE2_SUFFIX(_pcre2_jit_free_rodata_) 1993 #define _pcre2_jit_free PCRE2_SUFFIX(_pcre2_jit_free_) 1994 #define _pcre2_jit_get_size PCRE2_SUFFIX(_pcre2_jit_get_size_) 1995 #define _pcre2_jit_get_target PCRE2_SUFFIX(_pcre2_jit_get_target_) 1996 #define _pcre2_memctl_malloc PCRE2_SUFFIX(_pcre2_memctl_malloc_) 1997 #define _pcre2_ord2utf PCRE2_SUFFIX(_pcre2_ord2utf_) 1998 #define _pcre2_script_run PCRE2_SUFFIX(_pcre2_script_run_) 1999 #define _pcre2_strcmp PCRE2_SUFFIX(_pcre2_strcmp_) 2000 #define _pcre2_strcmp_c8 PCRE2_SUFFIX(_pcre2_strcmp_c8_) 2001 #define _pcre2_strcpy_c8 PCRE2_SUFFIX(_pcre2_strcpy_c8_) 2002 #define _pcre2_strlen PCRE2_SUFFIX(_pcre2_strlen_) 2003 #define _pcre2_strncmp PCRE2_SUFFIX(_pcre2_strncmp_) 2004 #define _pcre2_strncmp_c8 PCRE2_SUFFIX(_pcre2_strncmp_c8_) 2005 #define _pcre2_study PCRE2_SUFFIX(_pcre2_study_) 2006 #define _pcre2_valid_utf PCRE2_SUFFIX(_pcre2_valid_utf_) 2007 #define _pcre2_was_newline PCRE2_SUFFIX(_pcre2_was_newline_) 2008 #define _pcre2_xclass PCRE2_SUFFIX(_pcre2_xclass_) 2009 2010 extern int _pcre2_auto_possessify(PCRE2_UCHAR *, 2011 const compile_block *); 2012 extern int _pcre2_check_escape(PCRE2_SPTR *, PCRE2_SPTR, uint32_t *, 2013 int *, uint32_t, uint32_t, BOOL, compile_block *); 2014 extern PCRE2_SPTR _pcre2_extuni(uint32_t, PCRE2_SPTR, PCRE2_SPTR, PCRE2_SPTR, 2015 BOOL, int *); 2016 extern PCRE2_SPTR _pcre2_find_bracket(PCRE2_SPTR, BOOL, int); 2017 extern BOOL _pcre2_is_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 2018 uint32_t *, BOOL); 2019 extern void _pcre2_jit_free_rodata(void *, void *); 2020 extern void _pcre2_jit_free(void *, pcre2_memctl *); 2021 extern size_t _pcre2_jit_get_size(void *); 2022 const char * _pcre2_jit_get_target(void); 2023 extern void * _pcre2_memctl_malloc(size_t, pcre2_memctl *); 2024 extern unsigned int _pcre2_ord2utf(uint32_t, PCRE2_UCHAR *); 2025 extern BOOL _pcre2_script_run(PCRE2_SPTR, PCRE2_SPTR, BOOL); 2026 extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR); 2027 extern int _pcre2_strcmp_c8(PCRE2_SPTR, const char *); 2028 extern PCRE2_SIZE _pcre2_strcpy_c8(PCRE2_UCHAR *, const char *); 2029 extern PCRE2_SIZE _pcre2_strlen(PCRE2_SPTR); 2030 extern int _pcre2_strncmp(PCRE2_SPTR, PCRE2_SPTR, size_t); 2031 extern int _pcre2_strncmp_c8(PCRE2_SPTR, const char *, size_t); 2032 extern int _pcre2_study(pcre2_real_code *); 2033 extern int _pcre2_valid_utf(PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE *); 2034 extern BOOL _pcre2_was_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 2035 uint32_t *, BOOL); 2036 extern BOOL _pcre2_xclass(uint32_t, PCRE2_SPTR, BOOL); 2037 2038 /* This function is needed only when memmove() is not available. */ 2039 2040 #if !defined(VPCOMPAT) && !defined(HAVE_MEMMOVE) 2041 #define _pcre2_memmove PCRE2_SUFFIX(_pcre2_memmove) 2042 extern void * _pcre2_memmove(void *, const void *, size_t); 2043 #endif 2044 2045 #endif /* PCRE2_CODE_UNIT_WIDTH */ 2046 #endif /* PCRE2_INTERNAL_H_IDEMPOTENT_GUARD */ 2047 2048 /* End of pcre2_internal.h */ 2049
