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 pcre2_match() backtracking points 224 is allocated on the heap, of this size (bytes) or ten times the frame size if 225 larger, unless the heap limit is smaller. Typical frame sizes are a few hundred 226 bytes (it depends on the number of capturing parentheses) so 20KiB handles 227 quite a few frames. A larger vector on the heap is obtained for matches that 228 need more frames, subject to the heap limit. */ 229 230 #define START_FRAMES_SIZE 20480 231 232 /* For DFA matching, an initial internal workspace vector is allocated on the 233 stack. The heap is used only if this turns out to be too small. */ 234 235 #define DFA_START_RWS_SIZE 30720 236 237 /* Define the default BSR convention. */ 238 239 #ifdef BSR_ANYCRLF 240 #define BSR_DEFAULT PCRE2_BSR_ANYCRLF 241 #else 242 #define BSR_DEFAULT PCRE2_BSR_UNICODE 243 #endif 244 245 246 /* ---------------- Basic UTF-8 macros ---------------- */ 247 248 /* These UTF-8 macros are always defined because they are used in pcre2test for 249 handling wide characters in 16-bit and 32-bit modes, even if an 8-bit library 250 is not supported. */ 251 252 /* Tests whether a UTF-8 code point needs extra bytes to decode. */ 253 254 #define HASUTF8EXTRALEN(c) ((c) >= 0xc0) 255 256 /* The following macros were originally written in the form of loops that used 257 data from the tables whose names start with PRIV(utf8_table). They were 258 rewritten by a user so as not to use loops, because in some environments this 259 gives a significant performance advantage, and it seems never to do any harm. 260 */ 261 262 /* Base macro to pick up the remaining bytes of a UTF-8 character, not 263 advancing the pointer. */ 264 265 #define GETUTF8(c, eptr) \ 266 { \ 267 if ((c & 0x20u) == 0) \ 268 c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \ 269 else if ((c & 0x10u) == 0) \ 270 c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 271 else if ((c & 0x08u) == 0) \ 272 c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \ 273 ((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \ 274 else if ((c & 0x04u) == 0) \ 275 c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \ 276 ((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \ 277 (eptr[4] & 0x3fu); \ 278 else \ 279 c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \ 280 ((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \ 281 ((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \ 282 } 283 284 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing 285 the pointer. */ 286 287 #define GETUTF8INC(c, eptr) \ 288 { \ 289 if ((c & 0x20u) == 0) \ 290 c = ((c & 0x1fu) << 6) | (*eptr++ & 0x3fu); \ 291 else if ((c & 0x10u) == 0) \ 292 { \ 293 c = ((c & 0x0fu) << 12) | ((*eptr & 0x3fu) << 6) | (eptr[1] & 0x3fu); \ 294 eptr += 2; \ 295 } \ 296 else if ((c & 0x08u) == 0) \ 297 { \ 298 c = ((c & 0x07u) << 18) | ((*eptr & 0x3fu) << 12) | \ 299 ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 300 eptr += 3; \ 301 } \ 302 else if ((c & 0x04u) == 0) \ 303 { \ 304 c = ((c & 0x03u) << 24) | ((*eptr & 0x3fu) << 18) | \ 305 ((eptr[1] & 0x3fu) << 12) | ((eptr[2] & 0x3fu) << 6) | \ 306 (eptr[3] & 0x3fu); \ 307 eptr += 4; \ 308 } \ 309 else \ 310 { \ 311 c = ((c & 0x01u) << 30) | ((*eptr & 0x3fu) << 24) | \ 312 ((eptr[1] & 0x3fu) << 18) | ((eptr[2] & 0x3fu) << 12) | \ 313 ((eptr[3] & 0x3fu) << 6) | (eptr[4] & 0x3fu); \ 314 eptr += 5; \ 315 } \ 316 } 317 318 /* Base macro to pick up the remaining bytes of a UTF-8 character, not 319 advancing the pointer, incrementing the length. */ 320 321 #define GETUTF8LEN(c, eptr, len) \ 322 { \ 323 if ((c & 0x20u) == 0) \ 324 { \ 325 c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \ 326 len++; \ 327 } \ 328 else if ((c & 0x10u) == 0) \ 329 { \ 330 c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \ 331 len += 2; \ 332 } \ 333 else if ((c & 0x08u) == 0) \ 334 {\ 335 c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \ 336 ((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \ 337 len += 3; \ 338 } \ 339 else if ((c & 0x04u) == 0) \ 340 { \ 341 c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \ 342 ((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \ 343 (eptr[4] & 0x3fu); \ 344 len += 4; \ 345 } \ 346 else \ 347 {\ 348 c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \ 349 ((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \ 350 ((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \ 351 len += 5; \ 352 } \ 353 } 354 355 /* --------------- Whitespace macros ---------------- */ 356 357 /* Tests for Unicode horizontal and vertical whitespace characters must check a 358 number of different values. Using a switch statement for this generates the 359 fastest code (no loop, no memory access), and there are several places in the 360 interpreter code where this happens. In order to ensure that all the case lists 361 remain in step, we use macros so that there is only one place where the lists 362 are defined. 363 364 These values are also required as lists in pcre2_compile.c when processing \h, 365 \H, \v and \V in a character class. The lists are defined in pcre2_tables.c, 366 but macros that define the values are here so that all the definitions are 367 together. The lists must be in ascending character order, terminated by 368 NOTACHAR (which is 0xffffffff). 369 370 Any changes should ensure that the various macros are kept in step with each 371 other. NOTE: The values also appear in pcre2_jit_compile.c. */ 372 373 /* -------------- ASCII/Unicode environments -------------- */ 374 375 #ifndef EBCDIC 376 377 /* Character U+180E (Mongolian Vowel Separator) is not included in the list of 378 spaces in the Unicode file PropList.txt, and Perl does not recognize it as a 379 space. However, in many other sources it is listed as a space and has been in 380 PCRE (both APIs) for a long time. */ 381 382 #define HSPACE_LIST \ 383 CHAR_HT, CHAR_SPACE, CHAR_NBSP, \ 384 0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \ 385 0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \ 386 NOTACHAR 387 388 #define HSPACE_MULTIBYTE_CASES \ 389 case 0x1680: /* OGHAM SPACE MARK */ \ 390 case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ \ 391 case 0x2000: /* EN QUAD */ \ 392 case 0x2001: /* EM QUAD */ \ 393 case 0x2002: /* EN SPACE */ \ 394 case 0x2003: /* EM SPACE */ \ 395 case 0x2004: /* THREE-PER-EM SPACE */ \ 396 case 0x2005: /* FOUR-PER-EM SPACE */ \ 397 case 0x2006: /* SIX-PER-EM SPACE */ \ 398 case 0x2007: /* FIGURE SPACE */ \ 399 case 0x2008: /* PUNCTUATION SPACE */ \ 400 case 0x2009: /* THIN SPACE */ \ 401 case 0x200A: /* HAIR SPACE */ \ 402 case 0x202f: /* NARROW NO-BREAK SPACE */ \ 403 case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ \ 404 case 0x3000 /* IDEOGRAPHIC SPACE */ 405 406 #define HSPACE_BYTE_CASES \ 407 case CHAR_HT: \ 408 case CHAR_SPACE: \ 409 case CHAR_NBSP 410 411 #define HSPACE_CASES \ 412 HSPACE_BYTE_CASES: \ 413 HSPACE_MULTIBYTE_CASES 414 415 #define VSPACE_LIST \ 416 CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR 417 418 #define VSPACE_MULTIBYTE_CASES \ 419 case 0x2028: /* LINE SEPARATOR */ \ 420 case 0x2029 /* PARAGRAPH SEPARATOR */ 421 422 #define VSPACE_BYTE_CASES \ 423 case CHAR_LF: \ 424 case CHAR_VT: \ 425 case CHAR_FF: \ 426 case CHAR_CR: \ 427 case CHAR_NEL 428 429 #define VSPACE_CASES \ 430 VSPACE_BYTE_CASES: \ 431 VSPACE_MULTIBYTE_CASES 432 433 /* -------------- EBCDIC environments -------------- */ 434 435 #else 436 #define HSPACE_LIST CHAR_HT, CHAR_SPACE, CHAR_NBSP, NOTACHAR 437 438 #define HSPACE_BYTE_CASES \ 439 case CHAR_HT: \ 440 case CHAR_SPACE: \ 441 case CHAR_NBSP 442 443 #define HSPACE_CASES HSPACE_BYTE_CASES 444 445 #ifdef EBCDIC_NL25 446 #define VSPACE_LIST \ 447 CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR 448 #else 449 #define VSPACE_LIST \ 450 CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR 451 #endif 452 453 #define VSPACE_BYTE_CASES \ 454 case CHAR_LF: \ 455 case CHAR_VT: \ 456 case CHAR_FF: \ 457 case CHAR_CR: \ 458 case CHAR_NEL 459 460 #define VSPACE_CASES VSPACE_BYTE_CASES 461 #endif /* EBCDIC */ 462 463 /* -------------- End of whitespace macros -------------- */ 464 465 466 /* PCRE2 is able to support several different kinds of newline (CR, LF, CRLF, 467 "any" and "anycrlf" at present). The following macros are used to package up 468 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various 469 modules to indicate in which datablock the parameters exist, and what the 470 start/end of string field names are. */ 471 472 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */ 473 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */ 474 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */ 475 476 /* This macro checks for a newline at the given position */ 477 478 #define IS_NEWLINE(p) \ 479 ((NLBLOCK->nltype != NLTYPE_FIXED)? \ 480 ((p) < NLBLOCK->PSEND && \ 481 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \ 482 &(NLBLOCK->nllen), utf)) \ 483 : \ 484 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ 485 UCHAR21TEST(p) == NLBLOCK->nl[0] && \ 486 (NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1]) \ 487 ) \ 488 ) 489 490 /* This macro checks for a newline immediately preceding the given position */ 491 492 #define WAS_NEWLINE(p) \ 493 ((NLBLOCK->nltype != NLTYPE_FIXED)? \ 494 ((p) > NLBLOCK->PSSTART && \ 495 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ 496 &(NLBLOCK->nllen), utf)) \ 497 : \ 498 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ 499 UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \ 500 (NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \ 501 ) \ 502 ) 503 504 /* Private flags containing information about the compiled pattern. The first 505 three must not be changed, because whichever is set is actually the number of 506 bytes in a code unit in that mode. */ 507 508 #define PCRE2_MODE8 0x00000001 /* compiled in 8 bit mode */ 509 #define PCRE2_MODE16 0x00000002 /* compiled in 16 bit mode */ 510 #define PCRE2_MODE32 0x00000004 /* compiled in 32 bit mode */ 511 #define PCRE2_FIRSTSET 0x00000010 /* first_code unit is set */ 512 #define PCRE2_FIRSTCASELESS 0x00000020 /* caseless first code unit */ 513 #define PCRE2_FIRSTMAPSET 0x00000040 /* bitmap of first code units is set */ 514 #define PCRE2_LASTSET 0x00000080 /* last code unit is set */ 515 #define PCRE2_LASTCASELESS 0x00000100 /* caseless last code unit */ 516 #define PCRE2_STARTLINE 0x00000200 /* start after \n for multiline */ 517 #define PCRE2_JCHANGED 0x00000400 /* j option used in pattern */ 518 #define PCRE2_HASCRORLF 0x00000800 /* explicit \r or \n in pattern */ 519 #define PCRE2_HASTHEN 0x00001000 /* pattern contains (*THEN) */ 520 #define PCRE2_MATCH_EMPTY 0x00002000 /* pattern can match empty string */ 521 #define PCRE2_BSR_SET 0x00004000 /* BSR was set in the pattern */ 522 #define PCRE2_NL_SET 0x00008000 /* newline was set in the pattern */ 523 #define PCRE2_NOTEMPTY_SET 0x00010000 /* (*NOTEMPTY) used ) keep */ 524 #define PCRE2_NE_ATST_SET 0x00020000 /* (*NOTEMPTY_ATSTART) used) together */ 525 #define PCRE2_DEREF_TABLES 0x00040000 /* release character tables */ 526 #define PCRE2_NOJIT 0x00080000 /* (*NOJIT) used */ 527 #define PCRE2_HASBKPORX 0x00100000 /* contains \P, \p, or \X */ 528 #define PCRE2_DUPCAPUSED 0x00200000 /* contains (?| */ 529 #define PCRE2_HASBKC 0x00400000 /* contains \C */ 530 #define PCRE2_HASACCEPT 0x00800000 /* contains (*ACCEPT) */ 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 from an anchored pattern. In 8-bit mode, memchr() is used and is 550 much faster than the search loop that has to be used in 16-bit and 32-bit 551 modes. */ 552 553 #if PCRE2_CODE_UNIT_WIDTH == 8 554 #define REQ_CU_MAX 5000 555 #else 556 #define REQ_CU_MAX 2000 557 #endif 558 559 /* Offsets for the bitmap tables in the cbits set of tables. Each table 560 contains a set of bits for a class map. Some classes are built by combining 561 these tables. */ 562 563 #define cbit_space 0 /* [:space:] or \s */ 564 #define cbit_xdigit 32 /* [:xdigit:] */ 565 #define cbit_digit 64 /* [:digit:] or \d */ 566 #define cbit_upper 96 /* [:upper:] */ 567 #define cbit_lower 128 /* [:lower:] */ 568 #define cbit_word 160 /* [:word:] or \w */ 569 #define cbit_graph 192 /* [:graph:] */ 570 #define cbit_print 224 /* [:print:] */ 571 #define cbit_punct 256 /* [:punct:] */ 572 #define cbit_cntrl 288 /* [:cntrl:] */ 573 #define cbit_length 320 /* Length of the cbits table */ 574 575 /* Bit definitions for entries in the ctypes table. Do not change these values 576 without checking pcre2_jit_compile.c, which has an assertion to ensure that 577 ctype_word has the value 16. */ 578 579 #define ctype_space 0x01 580 #define ctype_letter 0x02 581 #define ctype_lcletter 0x04 582 #define ctype_digit 0x08 583 #define ctype_word 0x10 /* alphanumeric or '_' */ 584 585 /* Offsets of the various tables from the base tables pointer, and 586 total length of the tables. */ 587 588 #define lcc_offset 0 /* Lower case */ 589 #define fcc_offset 256 /* Flip case */ 590 #define cbits_offset 512 /* Character classes */ 591 #define ctypes_offset (cbits_offset + cbit_length) /* Character types */ 592 #define TABLES_LENGTH (ctypes_offset + 256) 593 594 595 /* -------------------- Character and string names ------------------------ */ 596 597 /* If PCRE2 is to support UTF-8 on EBCDIC platforms, we cannot use normal 598 character constants like '*' because the compiler would emit their EBCDIC code, 599 which is different from their ASCII/UTF-8 code. Instead we define macros for 600 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support 601 is enabled. When UTF-8 support is not enabled, the definitions use character 602 literals. Both character and string versions of each character are needed, and 603 there are some longer strings as well. 604 605 This means that, on EBCDIC platforms, the PCRE2 library can handle either 606 EBCDIC, or UTF-8, but not both. To support both in the same compiled library 607 would need different lookups depending on whether PCRE2_UTF was set or not. 608 This would make it impossible to use characters in switch/case statements, 609 which would reduce performance. For a theoretical use (which nobody has asked 610 for) in a minority area (EBCDIC platforms), this is not sensible. Any 611 application that did need both could compile two versions of the library, using 612 macros to give the functions distinct names. */ 613 614 #ifndef SUPPORT_UNICODE 615 616 /* UTF-8 support is not enabled; use the platform-dependent character literals 617 so that PCRE2 works in both ASCII and EBCDIC environments, but only in non-UTF 618 mode. Newline characters are problematic in EBCDIC. Though it has CR and LF 619 characters, a common practice has been to use its NL (0x15) character as the 620 line terminator in C-like processing environments. However, sometimes the LF 621 (0x25) character is used instead, according to this Unicode document: 622 623 http://unicode.org/standard/reports/tr13/tr13-5.html 624 625 PCRE2 defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25 626 instead. Whichever is *not* chosen is defined as NEL. 627 628 In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the 629 same code point. */ 630 631 #ifdef EBCDIC 632 633 #ifndef EBCDIC_NL25 634 #define CHAR_NL '\x15' 635 #define CHAR_NEL '\x25' 636 #define STR_NL "\x15" 637 #define STR_NEL "\x25" 638 #else 639 #define CHAR_NL '\x25' 640 #define CHAR_NEL '\x15' 641 #define STR_NL "\x25" 642 #define STR_NEL "\x15" 643 #endif 644 645 #define CHAR_LF CHAR_NL 646 #define STR_LF STR_NL 647 648 #define CHAR_ESC '\047' 649 #define CHAR_DEL '\007' 650 #define CHAR_NBSP ((unsigned char)'\x41') 651 #define STR_ESC "\047" 652 #define STR_DEL "\007" 653 654 #else /* Not EBCDIC */ 655 656 /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for 657 compatibility. NEL is the Unicode newline character; make sure it is 658 a positive value. */ 659 660 #define CHAR_LF '\n' 661 #define CHAR_NL CHAR_LF 662 #define CHAR_NEL ((unsigned char)'\x85') 663 #define CHAR_ESC '\033' 664 #define CHAR_DEL '\177' 665 #define CHAR_NBSP ((unsigned char)'\xa0') 666 667 #define STR_LF "\n" 668 #define STR_NL STR_LF 669 #define STR_NEL "\x85" 670 #define STR_ESC "\033" 671 #define STR_DEL "\177" 672 673 #endif /* EBCDIC */ 674 675 /* The remaining definitions work in both environments. */ 676 677 #define CHAR_NUL '\0' 678 #define CHAR_HT '\t' 679 #define CHAR_VT '\v' 680 #define CHAR_FF '\f' 681 #define CHAR_CR '\r' 682 #define CHAR_BS '\b' 683 #define CHAR_BEL '\a' 684 685 #define CHAR_SPACE ' ' 686 #define CHAR_EXCLAMATION_MARK '!' 687 #define CHAR_QUOTATION_MARK '"' 688 #define CHAR_NUMBER_SIGN '#' 689 #define CHAR_DOLLAR_SIGN '$' 690 #define CHAR_PERCENT_SIGN '%' 691 #define CHAR_AMPERSAND '&' 692 #define CHAR_APOSTROPHE '\'' 693 #define CHAR_LEFT_PARENTHESIS '(' 694 #define CHAR_RIGHT_PARENTHESIS ')' 695 #define CHAR_ASTERISK '*' 696 #define CHAR_PLUS '+' 697 #define CHAR_COMMA ',' 698 #define CHAR_MINUS '-' 699 #define CHAR_DOT '.' 700 #define CHAR_SLASH '/' 701 #define CHAR_0 '0' 702 #define CHAR_1 '1' 703 #define CHAR_2 '2' 704 #define CHAR_3 '3' 705 #define CHAR_4 '4' 706 #define CHAR_5 '5' 707 #define CHAR_6 '6' 708 #define CHAR_7 '7' 709 #define CHAR_8 '8' 710 #define CHAR_9 '9' 711 #define CHAR_COLON ':' 712 #define CHAR_SEMICOLON ';' 713 #define CHAR_LESS_THAN_SIGN '<' 714 #define CHAR_EQUALS_SIGN '=' 715 #define CHAR_GREATER_THAN_SIGN '>' 716 #define CHAR_QUESTION_MARK '?' 717 #define CHAR_COMMERCIAL_AT '@' 718 #define CHAR_A 'A' 719 #define CHAR_B 'B' 720 #define CHAR_C 'C' 721 #define CHAR_D 'D' 722 #define CHAR_E 'E' 723 #define CHAR_F 'F' 724 #define CHAR_G 'G' 725 #define CHAR_H 'H' 726 #define CHAR_I 'I' 727 #define CHAR_J 'J' 728 #define CHAR_K 'K' 729 #define CHAR_L 'L' 730 #define CHAR_M 'M' 731 #define CHAR_N 'N' 732 #define CHAR_O 'O' 733 #define CHAR_P 'P' 734 #define CHAR_Q 'Q' 735 #define CHAR_R 'R' 736 #define CHAR_S 'S' 737 #define CHAR_T 'T' 738 #define CHAR_U 'U' 739 #define CHAR_V 'V' 740 #define CHAR_W 'W' 741 #define CHAR_X 'X' 742 #define CHAR_Y 'Y' 743 #define CHAR_Z 'Z' 744 #define CHAR_LEFT_SQUARE_BRACKET '[' 745 #define CHAR_BACKSLASH '\\' 746 #define CHAR_RIGHT_SQUARE_BRACKET ']' 747 #define CHAR_CIRCUMFLEX_ACCENT '^' 748 #define CHAR_UNDERSCORE '_' 749 #define CHAR_GRAVE_ACCENT '`' 750 #define CHAR_a 'a' 751 #define CHAR_b 'b' 752 #define CHAR_c 'c' 753 #define CHAR_d 'd' 754 #define CHAR_e 'e' 755 #define CHAR_f 'f' 756 #define CHAR_g 'g' 757 #define CHAR_h 'h' 758 #define CHAR_i 'i' 759 #define CHAR_j 'j' 760 #define CHAR_k 'k' 761 #define CHAR_l 'l' 762 #define CHAR_m 'm' 763 #define CHAR_n 'n' 764 #define CHAR_o 'o' 765 #define CHAR_p 'p' 766 #define CHAR_q 'q' 767 #define CHAR_r 'r' 768 #define CHAR_s 's' 769 #define CHAR_t 't' 770 #define CHAR_u 'u' 771 #define CHAR_v 'v' 772 #define CHAR_w 'w' 773 #define CHAR_x 'x' 774 #define CHAR_y 'y' 775 #define CHAR_z 'z' 776 #define CHAR_LEFT_CURLY_BRACKET '{' 777 #define CHAR_VERTICAL_LINE '|' 778 #define CHAR_RIGHT_CURLY_BRACKET '}' 779 #define CHAR_TILDE '~' 780 781 #define STR_HT "\t" 782 #define STR_VT "\v" 783 #define STR_FF "\f" 784 #define STR_CR "\r" 785 #define STR_BS "\b" 786 #define STR_BEL "\a" 787 788 #define STR_SPACE " " 789 #define STR_EXCLAMATION_MARK "!" 790 #define STR_QUOTATION_MARK "\"" 791 #define STR_NUMBER_SIGN "#" 792 #define STR_DOLLAR_SIGN "$" 793 #define STR_PERCENT_SIGN "%" 794 #define STR_AMPERSAND "&" 795 #define STR_APOSTROPHE "'" 796 #define STR_LEFT_PARENTHESIS "(" 797 #define STR_RIGHT_PARENTHESIS ")" 798 #define STR_ASTERISK "*" 799 #define STR_PLUS "+" 800 #define STR_COMMA "," 801 #define STR_MINUS "-" 802 #define STR_DOT "." 803 #define STR_SLASH "/" 804 #define STR_0 "0" 805 #define STR_1 "1" 806 #define STR_2 "2" 807 #define STR_3 "3" 808 #define STR_4 "4" 809 #define STR_5 "5" 810 #define STR_6 "6" 811 #define STR_7 "7" 812 #define STR_8 "8" 813 #define STR_9 "9" 814 #define STR_COLON ":" 815 #define STR_SEMICOLON ";" 816 #define STR_LESS_THAN_SIGN "<" 817 #define STR_EQUALS_SIGN "=" 818 #define STR_GREATER_THAN_SIGN ">" 819 #define STR_QUESTION_MARK "?" 820 #define STR_COMMERCIAL_AT "@" 821 #define STR_A "A" 822 #define STR_B "B" 823 #define STR_C "C" 824 #define STR_D "D" 825 #define STR_E "E" 826 #define STR_F "F" 827 #define STR_G "G" 828 #define STR_H "H" 829 #define STR_I "I" 830 #define STR_J "J" 831 #define STR_K "K" 832 #define STR_L "L" 833 #define STR_M "M" 834 #define STR_N "N" 835 #define STR_O "O" 836 #define STR_P "P" 837 #define STR_Q "Q" 838 #define STR_R "R" 839 #define STR_S "S" 840 #define STR_T "T" 841 #define STR_U "U" 842 #define STR_V "V" 843 #define STR_W "W" 844 #define STR_X "X" 845 #define STR_Y "Y" 846 #define STR_Z "Z" 847 #define STR_LEFT_SQUARE_BRACKET "[" 848 #define STR_BACKSLASH "\\" 849 #define STR_RIGHT_SQUARE_BRACKET "]" 850 #define STR_CIRCUMFLEX_ACCENT "^" 851 #define STR_UNDERSCORE "_" 852 #define STR_GRAVE_ACCENT "`" 853 #define STR_a "a" 854 #define STR_b "b" 855 #define STR_c "c" 856 #define STR_d "d" 857 #define STR_e "e" 858 #define STR_f "f" 859 #define STR_g "g" 860 #define STR_h "h" 861 #define STR_i "i" 862 #define STR_j "j" 863 #define STR_k "k" 864 #define STR_l "l" 865 #define STR_m "m" 866 #define STR_n "n" 867 #define STR_o "o" 868 #define STR_p "p" 869 #define STR_q "q" 870 #define STR_r "r" 871 #define STR_s "s" 872 #define STR_t "t" 873 #define STR_u "u" 874 #define STR_v "v" 875 #define STR_w "w" 876 #define STR_x "x" 877 #define STR_y "y" 878 #define STR_z "z" 879 #define STR_LEFT_CURLY_BRACKET "{" 880 #define STR_VERTICAL_LINE "|" 881 #define STR_RIGHT_CURLY_BRACKET "}" 882 #define STR_TILDE "~" 883 884 #define STRING_ACCEPT0 "ACCEPT\0" 885 #define STRING_COMMIT0 "COMMIT\0" 886 #define STRING_F0 "F\0" 887 #define STRING_FAIL0 "FAIL\0" 888 #define STRING_MARK0 "MARK\0" 889 #define STRING_PRUNE0 "PRUNE\0" 890 #define STRING_SKIP0 "SKIP\0" 891 #define STRING_THEN "THEN" 892 893 #define STRING_atomic0 "atomic\0" 894 #define STRING_pla0 "pla\0" 895 #define STRING_plb0 "plb\0" 896 #define STRING_napla0 "napla\0" 897 #define STRING_naplb0 "naplb\0" 898 #define STRING_nla0 "nla\0" 899 #define STRING_nlb0 "nlb\0" 900 #define STRING_sr0 "sr\0" 901 #define STRING_asr0 "asr\0" 902 #define STRING_positive_lookahead0 "positive_lookahead\0" 903 #define STRING_positive_lookbehind0 "positive_lookbehind\0" 904 #define STRING_non_atomic_positive_lookahead0 "non_atomic_positive_lookahead\0" 905 #define STRING_non_atomic_positive_lookbehind0 "non_atomic_positive_lookbehind\0" 906 #define STRING_negative_lookahead0 "negative_lookahead\0" 907 #define STRING_negative_lookbehind0 "negative_lookbehind\0" 908 #define STRING_script_run0 "script_run\0" 909 #define STRING_atomic_script_run "atomic_script_run" 910 911 #define STRING_alpha0 "alpha\0" 912 #define STRING_lower0 "lower\0" 913 #define STRING_upper0 "upper\0" 914 #define STRING_alnum0 "alnum\0" 915 #define STRING_ascii0 "ascii\0" 916 #define STRING_blank0 "blank\0" 917 #define STRING_cntrl0 "cntrl\0" 918 #define STRING_digit0 "digit\0" 919 #define STRING_graph0 "graph\0" 920 #define STRING_print0 "print\0" 921 #define STRING_punct0 "punct\0" 922 #define STRING_space0 "space\0" 923 #define STRING_word0 "word\0" 924 #define STRING_xdigit "xdigit" 925 926 #define STRING_DEFINE "DEFINE" 927 #define STRING_VERSION "VERSION" 928 #define STRING_WEIRD_STARTWORD "[:<:]]" 929 #define STRING_WEIRD_ENDWORD "[:>:]]" 930 931 #define STRING_CR_RIGHTPAR "CR)" 932 #define STRING_LF_RIGHTPAR "LF)" 933 #define STRING_CRLF_RIGHTPAR "CRLF)" 934 #define STRING_ANY_RIGHTPAR "ANY)" 935 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)" 936 #define STRING_NUL_RIGHTPAR "NUL)" 937 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)" 938 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)" 939 #define STRING_UTF8_RIGHTPAR "UTF8)" 940 #define STRING_UTF16_RIGHTPAR "UTF16)" 941 #define STRING_UTF32_RIGHTPAR "UTF32)" 942 #define STRING_UTF_RIGHTPAR "UTF)" 943 #define STRING_UCP_RIGHTPAR "UCP)" 944 #define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)" 945 #define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR "NO_DOTSTAR_ANCHOR)" 946 #define STRING_NO_JIT_RIGHTPAR "NO_JIT)" 947 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)" 948 #define STRING_NOTEMPTY_RIGHTPAR "NOTEMPTY)" 949 #define STRING_NOTEMPTY_ATSTART_RIGHTPAR "NOTEMPTY_ATSTART)" 950 #define STRING_LIMIT_HEAP_EQ "LIMIT_HEAP=" 951 #define STRING_LIMIT_MATCH_EQ "LIMIT_MATCH=" 952 #define STRING_LIMIT_DEPTH_EQ "LIMIT_DEPTH=" 953 #define STRING_LIMIT_RECURSION_EQ "LIMIT_RECURSION=" 954 #define STRING_MARK "MARK" 955 956 #define STRING_bc "bc" 957 #define STRING_bidiclass "bidiclass" 958 #define STRING_sc "sc" 959 #define STRING_script "script" 960 #define STRING_scriptextensions "scriptextensions" 961 #define STRING_scx "scx" 962 963 #else /* SUPPORT_UNICODE */ 964 965 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This 966 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode 967 only. */ 968 969 #define CHAR_HT '\011' 970 #define CHAR_VT '\013' 971 #define CHAR_FF '\014' 972 #define CHAR_CR '\015' 973 #define CHAR_LF '\012' 974 #define CHAR_NL CHAR_LF 975 #define CHAR_NEL ((unsigned char)'\x85') 976 #define CHAR_BS '\010' 977 #define CHAR_BEL '\007' 978 #define CHAR_ESC '\033' 979 #define CHAR_DEL '\177' 980 981 #define CHAR_NUL '\0' 982 #define CHAR_SPACE '\040' 983 #define CHAR_EXCLAMATION_MARK '\041' 984 #define CHAR_QUOTATION_MARK '\042' 985 #define CHAR_NUMBER_SIGN '\043' 986 #define CHAR_DOLLAR_SIGN '\044' 987 #define CHAR_PERCENT_SIGN '\045' 988 #define CHAR_AMPERSAND '\046' 989 #define CHAR_APOSTROPHE '\047' 990 #define CHAR_LEFT_PARENTHESIS '\050' 991 #define CHAR_RIGHT_PARENTHESIS '\051' 992 #define CHAR_ASTERISK '\052' 993 #define CHAR_PLUS '\053' 994 #define CHAR_COMMA '\054' 995 #define CHAR_MINUS '\055' 996 #define CHAR_DOT '\056' 997 #define CHAR_SLASH '\057' 998 #define CHAR_0 '\060' 999 #define CHAR_1 '\061' 1000 #define CHAR_2 '\062' 1001 #define CHAR_3 '\063' 1002 #define CHAR_4 '\064' 1003 #define CHAR_5 '\065' 1004 #define CHAR_6 '\066' 1005 #define CHAR_7 '\067' 1006 #define CHAR_8 '\070' 1007 #define CHAR_9 '\071' 1008 #define CHAR_COLON '\072' 1009 #define CHAR_SEMICOLON '\073' 1010 #define CHAR_LESS_THAN_SIGN '\074' 1011 #define CHAR_EQUALS_SIGN '\075' 1012 #define CHAR_GREATER_THAN_SIGN '\076' 1013 #define CHAR_QUESTION_MARK '\077' 1014 #define CHAR_COMMERCIAL_AT '\100' 1015 #define CHAR_A '\101' 1016 #define CHAR_B '\102' 1017 #define CHAR_C '\103' 1018 #define CHAR_D '\104' 1019 #define CHAR_E '\105' 1020 #define CHAR_F '\106' 1021 #define CHAR_G '\107' 1022 #define CHAR_H '\110' 1023 #define CHAR_I '\111' 1024 #define CHAR_J '\112' 1025 #define CHAR_K '\113' 1026 #define CHAR_L '\114' 1027 #define CHAR_M '\115' 1028 #define CHAR_N '\116' 1029 #define CHAR_O '\117' 1030 #define CHAR_P '\120' 1031 #define CHAR_Q '\121' 1032 #define CHAR_R '\122' 1033 #define CHAR_S '\123' 1034 #define CHAR_T '\124' 1035 #define CHAR_U '\125' 1036 #define CHAR_V '\126' 1037 #define CHAR_W '\127' 1038 #define CHAR_X '\130' 1039 #define CHAR_Y '\131' 1040 #define CHAR_Z '\132' 1041 #define CHAR_LEFT_SQUARE_BRACKET '\133' 1042 #define CHAR_BACKSLASH '\134' 1043 #define CHAR_RIGHT_SQUARE_BRACKET '\135' 1044 #define CHAR_CIRCUMFLEX_ACCENT '\136' 1045 #define CHAR_UNDERSCORE '\137' 1046 #define CHAR_GRAVE_ACCENT '\140' 1047 #define CHAR_a '\141' 1048 #define CHAR_b '\142' 1049 #define CHAR_c '\143' 1050 #define CHAR_d '\144' 1051 #define CHAR_e '\145' 1052 #define CHAR_f '\146' 1053 #define CHAR_g '\147' 1054 #define CHAR_h '\150' 1055 #define CHAR_i '\151' 1056 #define CHAR_j '\152' 1057 #define CHAR_k '\153' 1058 #define CHAR_l '\154' 1059 #define CHAR_m '\155' 1060 #define CHAR_n '\156' 1061 #define CHAR_o '\157' 1062 #define CHAR_p '\160' 1063 #define CHAR_q '\161' 1064 #define CHAR_r '\162' 1065 #define CHAR_s '\163' 1066 #define CHAR_t '\164' 1067 #define CHAR_u '\165' 1068 #define CHAR_v '\166' 1069 #define CHAR_w '\167' 1070 #define CHAR_x '\170' 1071 #define CHAR_y '\171' 1072 #define CHAR_z '\172' 1073 #define CHAR_LEFT_CURLY_BRACKET '\173' 1074 #define CHAR_VERTICAL_LINE '\174' 1075 #define CHAR_RIGHT_CURLY_BRACKET '\175' 1076 #define CHAR_TILDE '\176' 1077 #define CHAR_NBSP ((unsigned char)'\xa0') 1078 1079 #define STR_HT "\011" 1080 #define STR_VT "\013" 1081 #define STR_FF "\014" 1082 #define STR_CR "\015" 1083 #define STR_NL "\012" 1084 #define STR_BS "\010" 1085 #define STR_BEL "\007" 1086 #define STR_ESC "\033" 1087 #define STR_DEL "\177" 1088 1089 #define STR_SPACE "\040" 1090 #define STR_EXCLAMATION_MARK "\041" 1091 #define STR_QUOTATION_MARK "\042" 1092 #define STR_NUMBER_SIGN "\043" 1093 #define STR_DOLLAR_SIGN "\044" 1094 #define STR_PERCENT_SIGN "\045" 1095 #define STR_AMPERSAND "\046" 1096 #define STR_APOSTROPHE "\047" 1097 #define STR_LEFT_PARENTHESIS "\050" 1098 #define STR_RIGHT_PARENTHESIS "\051" 1099 #define STR_ASTERISK "\052" 1100 #define STR_PLUS "\053" 1101 #define STR_COMMA "\054" 1102 #define STR_MINUS "\055" 1103 #define STR_DOT "\056" 1104 #define STR_SLASH "\057" 1105 #define STR_0 "\060" 1106 #define STR_1 "\061" 1107 #define STR_2 "\062" 1108 #define STR_3 "\063" 1109 #define STR_4 "\064" 1110 #define STR_5 "\065" 1111 #define STR_6 "\066" 1112 #define STR_7 "\067" 1113 #define STR_8 "\070" 1114 #define STR_9 "\071" 1115 #define STR_COLON "\072" 1116 #define STR_SEMICOLON "\073" 1117 #define STR_LESS_THAN_SIGN "\074" 1118 #define STR_EQUALS_SIGN "\075" 1119 #define STR_GREATER_THAN_SIGN "\076" 1120 #define STR_QUESTION_MARK "\077" 1121 #define STR_COMMERCIAL_AT "\100" 1122 #define STR_A "\101" 1123 #define STR_B "\102" 1124 #define STR_C "\103" 1125 #define STR_D "\104" 1126 #define STR_E "\105" 1127 #define STR_F "\106" 1128 #define STR_G "\107" 1129 #define STR_H "\110" 1130 #define STR_I "\111" 1131 #define STR_J "\112" 1132 #define STR_K "\113" 1133 #define STR_L "\114" 1134 #define STR_M "\115" 1135 #define STR_N "\116" 1136 #define STR_O "\117" 1137 #define STR_P "\120" 1138 #define STR_Q "\121" 1139 #define STR_R "\122" 1140 #define STR_S "\123" 1141 #define STR_T "\124" 1142 #define STR_U "\125" 1143 #define STR_V "\126" 1144 #define STR_W "\127" 1145 #define STR_X "\130" 1146 #define STR_Y "\131" 1147 #define STR_Z "\132" 1148 #define STR_LEFT_SQUARE_BRACKET "\133" 1149 #define STR_BACKSLASH "\134" 1150 #define STR_RIGHT_SQUARE_BRACKET "\135" 1151 #define STR_CIRCUMFLEX_ACCENT "\136" 1152 #define STR_UNDERSCORE "\137" 1153 #define STR_GRAVE_ACCENT "\140" 1154 #define STR_a "\141" 1155 #define STR_b "\142" 1156 #define STR_c "\143" 1157 #define STR_d "\144" 1158 #define STR_e "\145" 1159 #define STR_f "\146" 1160 #define STR_g "\147" 1161 #define STR_h "\150" 1162 #define STR_i "\151" 1163 #define STR_j "\152" 1164 #define STR_k "\153" 1165 #define STR_l "\154" 1166 #define STR_m "\155" 1167 #define STR_n "\156" 1168 #define STR_o "\157" 1169 #define STR_p "\160" 1170 #define STR_q "\161" 1171 #define STR_r "\162" 1172 #define STR_s "\163" 1173 #define STR_t "\164" 1174 #define STR_u "\165" 1175 #define STR_v "\166" 1176 #define STR_w "\167" 1177 #define STR_x "\170" 1178 #define STR_y "\171" 1179 #define STR_z "\172" 1180 #define STR_LEFT_CURLY_BRACKET "\173" 1181 #define STR_VERTICAL_LINE "\174" 1182 #define STR_RIGHT_CURLY_BRACKET "\175" 1183 #define STR_TILDE "\176" 1184 1185 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0" 1186 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0" 1187 #define STRING_F0 STR_F "\0" 1188 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0" 1189 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0" 1190 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0" 1191 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0" 1192 #define STRING_THEN STR_T STR_H STR_E STR_N 1193 1194 #define STRING_atomic0 STR_a STR_t STR_o STR_m STR_i STR_c "\0" 1195 #define STRING_pla0 STR_p STR_l STR_a "\0" 1196 #define STRING_plb0 STR_p STR_l STR_b "\0" 1197 #define STRING_napla0 STR_n STR_a STR_p STR_l STR_a "\0" 1198 #define STRING_naplb0 STR_n STR_a STR_p STR_l STR_b "\0" 1199 #define STRING_nla0 STR_n STR_l STR_a "\0" 1200 #define STRING_nlb0 STR_n STR_l STR_b "\0" 1201 #define STRING_sr0 STR_s STR_r "\0" 1202 #define STRING_asr0 STR_a STR_s STR_r "\0" 1203 #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" 1204 #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" 1205 #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" 1206 #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" 1207 #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" 1208 #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" 1209 #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" 1210 #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 1211 1212 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0" 1213 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0" 1214 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0" 1215 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0" 1216 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0" 1217 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0" 1218 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0" 1219 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0" 1220 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0" 1221 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0" 1222 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0" 1223 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0" 1224 #define STRING_word0 STR_w STR_o STR_r STR_d "\0" 1225 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t 1226 1227 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E 1228 #define STRING_VERSION STR_V STR_E STR_R STR_S STR_I STR_O STR_N 1229 #define STRING_WEIRD_STARTWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1230 #define STRING_WEIRD_ENDWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET 1231 1232 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS 1233 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS 1234 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1235 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS 1236 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS 1237 #define STRING_NUL_RIGHTPAR STR_N STR_U STR_L STR_RIGHT_PARENTHESIS 1238 #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 1239 #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 1240 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS 1241 #define STRING_UTF16_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS 1242 #define STRING_UTF32_RIGHTPAR STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS 1243 #define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_RIGHT_PARENTHESIS 1244 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS 1245 #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 1246 #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 1247 #define STRING_NO_JIT_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_J STR_I STR_T STR_RIGHT_PARENTHESIS 1248 #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 1249 #define STRING_NOTEMPTY_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_RIGHT_PARENTHESIS 1250 #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 1251 #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 1252 #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 1253 #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 1254 #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 1255 #define STRING_MARK STR_M STR_A STR_R STR_K 1256 1257 #define STRING_bc STR_b STR_c 1258 #define STRING_bidiclass STR_b STR_i STR_d STR_i STR_c STR_l STR_a STR_s STR_s 1259 #define STRING_sc STR_s STR_c 1260 #define STRING_script STR_s STR_c STR_r STR_i STR_p STR_t 1261 #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 1262 #define STRING_scx STR_s STR_c STR_x 1263 1264 1265 #endif /* SUPPORT_UNICODE */ 1266 1267 /* -------------------- End of character and string names -------------------*/ 1268 1269 /* -------------------- Definitions for compiled patterns -------------------*/ 1270 1271 /* Codes for different types of Unicode property. If these definitions are 1272 changed, the autopossessifying table in pcre2_auto_possess.c must be updated to 1273 match. */ 1274 1275 #define PT_ANY 0 /* Any property - matches all chars */ 1276 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */ 1277 #define PT_GC 2 /* Specified general characteristic (e.g. L) */ 1278 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */ 1279 #define PT_SC 4 /* Script only (e.g. Han) */ 1280 #define PT_SCX 5 /* Script extensions (includes SC) */ 1281 #define PT_ALNUM 6 /* Alphanumeric - the union of L and N */ 1282 #define PT_SPACE 7 /* Perl space - general category Z plus 9,10,12,13 */ 1283 #define PT_PXSPACE 8 /* POSIX space - Z plus 9,10,11,12,13 */ 1284 #define PT_WORD 9 /* Word - L plus N plus underscore */ 1285 #define PT_CLIST 10 /* Pseudo-property: match character list */ 1286 #define PT_UCNC 11 /* Universal Character nameable character */ 1287 #define PT_BIDICL 12 /* Specified bidi class */ 1288 #define PT_BOOL 13 /* Boolean property */ 1289 #define PT_TABSIZE 14 /* Size of square table for autopossessify tests */ 1290 1291 /* The following special properties are used only in XCLASS items, when POSIX 1292 classes are specified and PCRE2_UCP is set - in other words, for Unicode 1293 handling of these classes. They are not available via the \p or \P escapes like 1294 those in the above list, and so they do not take part in the autopossessifying 1295 table. */ 1296 1297 #define PT_PXGRAPH 14 /* [:graph:] - characters that mark the paper */ 1298 #define PT_PXPRINT 15 /* [:print:] - [:graph:] plus non-control spaces */ 1299 #define PT_PXPUNCT 16 /* [:punct:] - punctuation characters */ 1300 1301 /* This value is used when parsing \p and \P escapes to indicate that neither 1302 \p{script:...} nor \p{scx:...} has been encountered. */ 1303 1304 #define PT_NOTSCRIPT 255 1305 1306 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that 1307 contain characters with values greater than 255. */ 1308 1309 #define XCL_NOT 0x01 /* Flag: this is a negative class */ 1310 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ 1311 #define XCL_HASPROP 0x04 /* Flag: property checks are present. */ 1312 1313 #define XCL_END 0 /* Marks end of individual items */ 1314 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */ 1315 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */ 1316 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */ 1317 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */ 1318 1319 /* These are escaped items that aren't just an encoding of a particular data 1320 value such as \n. They must have non-zero values, as check_escape() returns 0 1321 for a data character. In the escapes[] table in pcre2_compile.c their values 1322 are negated in order to distinguish them from data values. 1323 1324 They must appear here in the same order as in the opcode definitions below, up 1325 to ESC_z. There's a dummy for OP_ALLANY because it corresponds to "." in DOTALL 1326 mode rather than an escape sequence. It is also used for [^] in JavaScript 1327 compatibility mode, and for \C in non-utf mode. In non-DOTALL mode, "." behaves 1328 like \N. 1329 1330 Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in 1331 check_escape(). There are tests in the code for an escape greater than ESC_b 1332 and less than ESC_Z to detect the types that may be repeated. These are the 1333 types that consume characters. If any new escapes are put in between that don't 1334 consume a character, that code will have to change. */ 1335 1336 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, 1337 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, 1338 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, 1339 ESC_E, ESC_Q, ESC_g, ESC_k }; 1340 1341 1342 /********************** Opcode definitions ******************/ 1343 1344 /****** NOTE NOTE NOTE ****** 1345 1346 Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in 1347 order to the list of escapes immediately above. Furthermore, values up to 1348 OP_DOLLM must not be changed without adjusting the table called autoposstab in 1349 pcre2_auto_possess.c. 1350 1351 Whenever this list is updated, the two macro definitions that follow must be 1352 updated to match. The possessification table called "opcode_possessify" in 1353 pcre2_compile.c must also be updated, and also the tables called "coptable" 1354 and "poptable" in pcre2_dfa_match.c. 1355 1356 ****** NOTE NOTE NOTE ******/ 1357 1358 1359 /* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive, 1360 are used in a table for deciding whether a repeated character type can be 1361 auto-possessified. */ 1362 1363 #define FIRST_AUTOTAB_OP OP_NOT_DIGIT 1364 #define LAST_AUTOTAB_LEFT_OP OP_EXTUNI 1365 #define LAST_AUTOTAB_RIGHT_OP OP_DOLLM 1366 1367 enum { 1368 OP_END, /* 0 End of pattern */ 1369 1370 /* Values corresponding to backslashed metacharacters */ 1371 1372 OP_SOD, /* 1 Start of data: \A */ 1373 OP_SOM, /* 2 Start of match (subject + offset): \G */ 1374 OP_SET_SOM, /* 3 Set start of match (\K) */ 1375 OP_NOT_WORD_BOUNDARY, /* 4 \B */ 1376 OP_WORD_BOUNDARY, /* 5 \b */ 1377 OP_NOT_DIGIT, /* 6 \D */ 1378 OP_DIGIT, /* 7 \d */ 1379 OP_NOT_WHITESPACE, /* 8 \S */ 1380 OP_WHITESPACE, /* 9 \s */ 1381 OP_NOT_WORDCHAR, /* 10 \W */ 1382 OP_WORDCHAR, /* 11 \w */ 1383 1384 OP_ANY, /* 12 Match any character except newline (\N) */ 1385 OP_ALLANY, /* 13 Match any character */ 1386 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */ 1387 OP_NOTPROP, /* 15 \P (not Unicode property) */ 1388 OP_PROP, /* 16 \p (Unicode property) */ 1389 OP_ANYNL, /* 17 \R (any newline sequence) */ 1390 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */ 1391 OP_HSPACE, /* 19 \h (horizontal whitespace) */ 1392 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */ 1393 OP_VSPACE, /* 21 \v (vertical whitespace) */ 1394 OP_EXTUNI, /* 22 \X (extended Unicode sequence */ 1395 OP_EODN, /* 23 End of data or \n at end of data (\Z) */ 1396 OP_EOD, /* 24 End of data (\z) */ 1397 1398 /* Line end assertions */ 1399 1400 OP_DOLL, /* 25 End of line - not multiline */ 1401 OP_DOLLM, /* 26 End of line - multiline */ 1402 OP_CIRC, /* 27 Start of line - not multiline */ 1403 OP_CIRCM, /* 28 Start of line - multiline */ 1404 1405 /* Single characters; caseful must precede the caseless ones, and these 1406 must remain in this order, and adjacent. */ 1407 1408 OP_CHAR, /* 29 Match one character, casefully */ 1409 OP_CHARI, /* 30 Match one character, caselessly */ 1410 OP_NOT, /* 31 Match one character, not the given one, casefully */ 1411 OP_NOTI, /* 32 Match one character, not the given one, caselessly */ 1412 1413 /* The following sets of 13 opcodes must always be kept in step because 1414 the offset from the first one is used to generate the others. */ 1415 1416 /* Repeated characters; caseful must precede the caseless ones */ 1417 1418 OP_STAR, /* 33 The maximizing and minimizing versions of */ 1419 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */ 1420 OP_PLUS, /* 35 the minimizing one second. */ 1421 OP_MINPLUS, /* 36 */ 1422 OP_QUERY, /* 37 */ 1423 OP_MINQUERY, /* 38 */ 1424 1425 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/ 1426 OP_MINUPTO, /* 40 */ 1427 OP_EXACT, /* 41 Exactly n matches */ 1428 1429 OP_POSSTAR, /* 42 Possessified star, caseful */ 1430 OP_POSPLUS, /* 43 Possessified plus, caseful */ 1431 OP_POSQUERY, /* 44 Posesssified query, caseful */ 1432 OP_POSUPTO, /* 45 Possessified upto, caseful */ 1433 1434 /* Repeated characters; caseless must follow the caseful ones */ 1435 1436 OP_STARI, /* 46 */ 1437 OP_MINSTARI, /* 47 */ 1438 OP_PLUSI, /* 48 */ 1439 OP_MINPLUSI, /* 49 */ 1440 OP_QUERYI, /* 50 */ 1441 OP_MINQUERYI, /* 51 */ 1442 1443 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */ 1444 OP_MINUPTOI, /* 53 */ 1445 OP_EXACTI, /* 54 */ 1446 1447 OP_POSSTARI, /* 55 Possessified star, caseless */ 1448 OP_POSPLUSI, /* 56 Possessified plus, caseless */ 1449 OP_POSQUERYI, /* 57 Posesssified query, caseless */ 1450 OP_POSUPTOI, /* 58 Possessified upto, caseless */ 1451 1452 /* The negated ones must follow the non-negated ones, and match them */ 1453 /* Negated repeated character, caseful; must precede the caseless ones */ 1454 1455 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */ 1456 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */ 1457 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */ 1458 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */ 1459 OP_NOTQUERY, /* 63 */ 1460 OP_NOTMINQUERY, /* 64 */ 1461 1462 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */ 1463 OP_NOTMINUPTO, /* 66 */ 1464 OP_NOTEXACT, /* 67 Exactly n matches */ 1465 1466 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */ 1467 OP_NOTPOSPLUS, /* 69 */ 1468 OP_NOTPOSQUERY, /* 70 */ 1469 OP_NOTPOSUPTO, /* 71 */ 1470 1471 /* Negated repeated character, caseless; must follow the caseful ones */ 1472 1473 OP_NOTSTARI, /* 72 */ 1474 OP_NOTMINSTARI, /* 73 */ 1475 OP_NOTPLUSI, /* 74 */ 1476 OP_NOTMINPLUSI, /* 75 */ 1477 OP_NOTQUERYI, /* 76 */ 1478 OP_NOTMINQUERYI, /* 77 */ 1479 1480 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */ 1481 OP_NOTMINUPTOI, /* 79 */ 1482 OP_NOTEXACTI, /* 80 Exactly n matches */ 1483 1484 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */ 1485 OP_NOTPOSPLUSI, /* 82 */ 1486 OP_NOTPOSQUERYI, /* 83 */ 1487 OP_NOTPOSUPTOI, /* 84 */ 1488 1489 /* Character types */ 1490 1491 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */ 1492 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */ 1493 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */ 1494 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */ 1495 OP_TYPEQUERY, /* 89 */ 1496 OP_TYPEMINQUERY, /* 90 */ 1497 1498 OP_TYPEUPTO, /* 91 From 0 to n matches */ 1499 OP_TYPEMINUPTO, /* 92 */ 1500 OP_TYPEEXACT, /* 93 Exactly n matches */ 1501 1502 OP_TYPEPOSSTAR, /* 94 Possessified versions */ 1503 OP_TYPEPOSPLUS, /* 95 */ 1504 OP_TYPEPOSQUERY, /* 96 */ 1505 OP_TYPEPOSUPTO, /* 97 */ 1506 1507 /* These are used for character classes and back references; only the 1508 first six are the same as the sets above. */ 1509 1510 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */ 1511 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */ 1512 OP_CRPLUS, /* 100 the minimizing one second. These codes must */ 1513 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */ 1514 OP_CRQUERY, /* 102 */ 1515 OP_CRMINQUERY, /* 103 */ 1516 1517 OP_CRRANGE, /* 104 These are different to the three sets above. */ 1518 OP_CRMINRANGE, /* 105 */ 1519 1520 OP_CRPOSSTAR, /* 106 Possessified versions */ 1521 OP_CRPOSPLUS, /* 107 */ 1522 OP_CRPOSQUERY, /* 108 */ 1523 OP_CRPOSRANGE, /* 109 */ 1524 1525 /* End of quantifier opcodes */ 1526 1527 OP_CLASS, /* 110 Match a character class, chars < 256 only */ 1528 OP_NCLASS, /* 111 Same, but the bitmap was created from a negative 1529 class - the difference is relevant only when a 1530 character > 255 is encountered. */ 1531 OP_XCLASS, /* 112 Extended class for handling > 255 chars within the 1532 class. This does both positive and negative. */ 1533 OP_REF, /* 113 Match a back reference, casefully */ 1534 OP_REFI, /* 114 Match a back reference, caselessly */ 1535 OP_DNREF, /* 115 Match a duplicate name backref, casefully */ 1536 OP_DNREFI, /* 116 Match a duplicate name backref, caselessly */ 1537 OP_RECURSE, /* 117 Match a numbered subpattern (possibly recursive) */ 1538 OP_CALLOUT, /* 118 Call out to external function if provided */ 1539 OP_CALLOUT_STR, /* 119 Call out with string argument */ 1540 1541 OP_ALT, /* 120 Start of alternation */ 1542 OP_KET, /* 121 End of group that doesn't have an unbounded repeat */ 1543 OP_KETRMAX, /* 122 These two must remain together and in this */ 1544 OP_KETRMIN, /* 123 order. They are for groups the repeat for ever. */ 1545 OP_KETRPOS, /* 124 Possessive unlimited repeat. */ 1546 1547 /* The assertions must come before BRA, CBRA, ONCE, and COND. */ 1548 1549 OP_REVERSE, /* 125 Move pointer back - used in lookbehind assertions */ 1550 OP_ASSERT, /* 126 Positive lookahead */ 1551 OP_ASSERT_NOT, /* 127 Negative lookahead */ 1552 OP_ASSERTBACK, /* 128 Positive lookbehind */ 1553 OP_ASSERTBACK_NOT, /* 129 Negative lookbehind */ 1554 OP_ASSERT_NA, /* 130 Positive non-atomic lookahead */ 1555 OP_ASSERTBACK_NA, /* 131 Positive non-atomic lookbehind */ 1556 1557 /* ONCE, SCRIPT_RUN, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come 1558 immediately after the assertions, with ONCE first, as there's a test for >= 1559 ONCE for a subpattern that isn't an assertion. The POS versions must 1560 immediately follow the non-POS versions in each case. */ 1561 1562 OP_ONCE, /* 132 Atomic group, contains captures */ 1563 OP_SCRIPT_RUN, /* 133 Non-capture, but check characters' scripts */ 1564 OP_BRA, /* 134 Start of non-capturing bracket */ 1565 OP_BRAPOS, /* 135 Ditto, with unlimited, possessive repeat */ 1566 OP_CBRA, /* 136 Start of capturing bracket */ 1567 OP_CBRAPOS, /* 137 Ditto, with unlimited, possessive repeat */ 1568 OP_COND, /* 138 Conditional group */ 1569 1570 /* These five must follow the previous five, in the same order. There's a 1571 check for >= SBRA to distinguish the two sets. */ 1572 1573 OP_SBRA, /* 139 Start of non-capturing bracket, check empty */ 1574 OP_SBRAPOS, /* 149 Ditto, with unlimited, possessive repeat */ 1575 OP_SCBRA, /* 141 Start of capturing bracket, check empty */ 1576 OP_SCBRAPOS, /* 142 Ditto, with unlimited, possessive repeat */ 1577 OP_SCOND, /* 143 Conditional group, check empty */ 1578 1579 /* The next two pairs must (respectively) be kept together. */ 1580 1581 OP_CREF, /* 144 Used to hold a capture number as condition */ 1582 OP_DNCREF, /* 145 Used to point to duplicate names as a condition */ 1583 OP_RREF, /* 146 Used to hold a recursion number as condition */ 1584 OP_DNRREF, /* 147 Used to point to duplicate names as a condition */ 1585 OP_FALSE, /* 148 Always false (used by DEFINE and VERSION) */ 1586 OP_TRUE, /* 149 Always true (used by VERSION) */ 1587 1588 OP_BRAZERO, /* 150 These two must remain together and in this */ 1589 OP_BRAMINZERO, /* 151 order. */ 1590 OP_BRAPOSZERO, /* 152 */ 1591 1592 /* These are backtracking control verbs */ 1593 1594 OP_MARK, /* 153 always has an argument */ 1595 OP_PRUNE, /* 154 */ 1596 OP_PRUNE_ARG, /* 155 same, but with argument */ 1597 OP_SKIP, /* 156 */ 1598 OP_SKIP_ARG, /* 157 same, but with argument */ 1599 OP_THEN, /* 158 */ 1600 OP_THEN_ARG, /* 159 same, but with argument */ 1601 OP_COMMIT, /* 160 */ 1602 OP_COMMIT_ARG, /* 161 same, but with argument */ 1603 1604 /* These are forced failure and success verbs. FAIL and ACCEPT do accept an 1605 argument, but these cases can be compiled as, for example, (*MARK:X)(*FAIL) 1606 without the need for a special opcode. */ 1607 1608 OP_FAIL, /* 162 */ 1609 OP_ACCEPT, /* 163 */ 1610 OP_ASSERT_ACCEPT, /* 164 Used inside assertions */ 1611 OP_CLOSE, /* 165 Used before OP_ACCEPT to close open captures */ 1612 1613 /* This is used to skip a subpattern with a {0} quantifier */ 1614 1615 OP_SKIPZERO, /* 166 */ 1616 1617 /* This is used to identify a DEFINE group during compilation so that it can 1618 be checked for having only one branch. It is changed to OP_FALSE before 1619 compilation finishes. */ 1620 1621 OP_DEFINE, /* 167 */ 1622 1623 /* This is not an opcode, but is used to check that tables indexed by opcode 1624 are the correct length, in order to catch updating errors - there have been 1625 some in the past. */ 1626 1627 OP_TABLE_LENGTH 1628 1629 }; 1630 1631 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro 1632 definitions that follow must also be updated to match. There are also tables 1633 called "opcode_possessify" in pcre2_compile.c and "coptable" and "poptable" in 1634 pcre2_dfa_match.c that must be updated. */ 1635 1636 1637 /* This macro defines textual names for all the opcodes. These are used only 1638 for debugging, and some of them are only partial names. The macro is referenced 1639 only in pcre2_printint.c, which fills out the full names in many cases (and in 1640 some cases doesn't actually use these names at all). */ 1641 1642 #define OP_NAME_LIST \ 1643 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \ 1644 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \ 1645 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \ 1646 "extuni", "\\Z", "\\z", \ 1647 "$", "$", "^", "^", "char", "chari", "not", "noti", \ 1648 "*", "*?", "+", "+?", "?", "??", \ 1649 "{", "{", "{", \ 1650 "*+","++", "?+", "{", \ 1651 "*", "*?", "+", "+?", "?", "??", \ 1652 "{", "{", "{", \ 1653 "*+","++", "?+", "{", \ 1654 "*", "*?", "+", "+?", "?", "??", \ 1655 "{", "{", "{", \ 1656 "*+","++", "?+", "{", \ 1657 "*", "*?", "+", "+?", "?", "??", \ 1658 "{", "{", "{", \ 1659 "*+","++", "?+", "{", \ 1660 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \ 1661 "*+","++", "?+", "{", \ 1662 "*", "*?", "+", "+?", "?", "??", "{", "{", \ 1663 "*+","++", "?+", "{", \ 1664 "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi", \ 1665 "Recurse", "Callout", "CalloutStr", \ 1666 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \ 1667 "Reverse", "Assert", "Assert not", \ 1668 "Assert back", "Assert back not", \ 1669 "Non-atomic assert", "Non-atomic assert back", \ 1670 "Once", \ 1671 "Script run", \ 1672 "Bra", "BraPos", "CBra", "CBraPos", \ 1673 "Cond", \ 1674 "SBra", "SBraPos", "SCBra", "SCBraPos", \ 1675 "SCond", \ 1676 "Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", \ 1677 "Cond false", "Cond true", \ 1678 "Brazero", "Braminzero", "Braposzero", \ 1679 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \ 1680 "*THEN", "*THEN", "*COMMIT", "*COMMIT", "*FAIL", \ 1681 "*ACCEPT", "*ASSERT_ACCEPT", \ 1682 "Close", "Skip zero", "Define" 1683 1684 1685 /* This macro defines the length of fixed length operations in the compiled 1686 regex. The lengths are used when searching for specific things, and also in the 1687 debugging printing of a compiled regex. We use a macro so that it can be 1688 defined close to the definitions of the opcodes themselves. 1689 1690 As things have been extended, some of these are no longer fixed lenths, but are 1691 minima instead. For example, the length of a single-character repeat may vary 1692 in UTF-8 mode. The code that uses this table must know about such things. */ 1693 1694 #define OP_LENGTHS \ 1695 1, /* End */ \ 1696 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \ 1697 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \ 1698 1, 1, 1, /* Any, AllAny, Anybyte */ \ 1699 3, 3, /* \P, \p */ \ 1700 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \ 1701 1, /* \X */ \ 1702 1, 1, 1, 1, 1, 1, /* \Z, \z, $, $M ^, ^M */ \ 1703 2, /* Char - the minimum length */ \ 1704 2, /* Chari - the minimum length */ \ 1705 2, /* not */ \ 1706 2, /* noti */ \ 1707 /* Positive single-char repeats ** These are */ \ 1708 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ 1709 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \ 1710 2+IMM2_SIZE, /* exact */ \ 1711 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \ 1712 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \ 1713 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \ 1714 2+IMM2_SIZE, /* exact I */ \ 1715 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \ 1716 /* Negative single-char repeats - only for chars < 256 */ \ 1717 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ 1718 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \ 1719 2+IMM2_SIZE, /* NOT exact */ \ 1720 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \ 1721 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \ 1722 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \ 1723 2+IMM2_SIZE, /* NOT exact I */ \ 1724 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \ 1725 /* Positive type repeats */ \ 1726 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ 1727 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \ 1728 2+IMM2_SIZE, /* Type exact */ \ 1729 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \ 1730 /* Character class & ref repeats */ \ 1731 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ 1732 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \ 1733 1, 1, 1, 1+2*IMM2_SIZE, /* Possessive *+, ++, ?+, CRPOSRANGE */ \ 1734 1+(32/sizeof(PCRE2_UCHAR)), /* CLASS */ \ 1735 1+(32/sizeof(PCRE2_UCHAR)), /* NCLASS */ \ 1736 0, /* XCLASS - variable length */ \ 1737 1+IMM2_SIZE, /* REF */ \ 1738 1+IMM2_SIZE, /* REFI */ \ 1739 1+2*IMM2_SIZE, /* DNREF */ \ 1740 1+2*IMM2_SIZE, /* DNREFI */ \ 1741 1+LINK_SIZE, /* RECURSE */ \ 1742 1+2*LINK_SIZE+1, /* CALLOUT */ \ 1743 0, /* CALLOUT_STR - variable length */ \ 1744 1+LINK_SIZE, /* Alt */ \ 1745 1+LINK_SIZE, /* Ket */ \ 1746 1+LINK_SIZE, /* KetRmax */ \ 1747 1+LINK_SIZE, /* KetRmin */ \ 1748 1+LINK_SIZE, /* KetRpos */ \ 1749 1+LINK_SIZE, /* Reverse */ \ 1750 1+LINK_SIZE, /* Assert */ \ 1751 1+LINK_SIZE, /* Assert not */ \ 1752 1+LINK_SIZE, /* Assert behind */ \ 1753 1+LINK_SIZE, /* Assert behind not */ \ 1754 1+LINK_SIZE, /* NA Assert */ \ 1755 1+LINK_SIZE, /* NA Assert behind */ \ 1756 1+LINK_SIZE, /* ONCE */ \ 1757 1+LINK_SIZE, /* SCRIPT_RUN */ \ 1758 1+LINK_SIZE, /* BRA */ \ 1759 1+LINK_SIZE, /* BRAPOS */ \ 1760 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \ 1761 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \ 1762 1+LINK_SIZE, /* COND */ \ 1763 1+LINK_SIZE, /* SBRA */ \ 1764 1+LINK_SIZE, /* SBRAPOS */ \ 1765 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \ 1766 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \ 1767 1+LINK_SIZE, /* SCOND */ \ 1768 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* CREF, DNCREF */ \ 1769 1+IMM2_SIZE, 1+2*IMM2_SIZE, /* RREF, DNRREF */ \ 1770 1, 1, /* FALSE, TRUE */ \ 1771 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ 1772 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ 1773 1, 3, /* SKIP, SKIP_ARG */ \ 1774 1, 3, /* THEN, THEN_ARG */ \ 1775 1, 3, /* COMMIT, COMMIT_ARG */ \ 1776 1, 1, 1, /* FAIL, ACCEPT, ASSERT_ACCEPT */ \ 1777 1+IMM2_SIZE, 1, /* CLOSE, SKIPZERO */ \ 1778 1 /* DEFINE */ 1779 1780 /* A magic value for OP_RREF to indicate the "any recursion" condition. */ 1781 1782 #define RREF_ANY 0xffff 1783 1784 1785 /* ---------- Private structures that are mode-independent. ---------- */ 1786 1787 /* Structure to hold data for custom memory management. */ 1788 1789 typedef struct pcre2_memctl { 1790 void * (*malloc)(size_t, void *); 1791 void (*free)(void *, void *); 1792 void *memory_data; 1793 } pcre2_memctl; 1794 1795 /* Structure for building a chain of open capturing subpatterns during 1796 compiling, so that instructions to close them can be compiled when (*ACCEPT) is 1797 encountered. */ 1798 1799 typedef struct open_capitem { 1800 struct open_capitem *next; /* Chain link */ 1801 uint16_t number; /* Capture number */ 1802 uint16_t assert_depth; /* Assertion depth when opened */ 1803 } open_capitem; 1804 1805 /* Layout of the UCP type table that translates property names into types and 1806 codes. Each entry used to point directly to a name, but to reduce the number of 1807 relocations in shared libraries, it now has an offset into a single string 1808 instead. */ 1809 1810 typedef struct { 1811 uint16_t name_offset; 1812 uint16_t type; 1813 uint16_t value; 1814 } ucp_type_table; 1815 1816 /* Unicode character database (UCD) record format */ 1817 1818 typedef struct { 1819 uint8_t script; /* ucp_Arabic, etc. */ 1820 uint8_t chartype; /* ucp_Cc, etc. (general categories) */ 1821 uint8_t gbprop; /* ucp_gbControl, etc. (grapheme break property) */ 1822 uint8_t caseset; /* offset to multichar other cases or zero */ 1823 int32_t other_case; /* offset to other case, or zero if none */ 1824 uint16_t scriptx_bidiclass; /* script extension (11 bit) and bidi class (5 bit) values */ 1825 uint16_t bprops; /* binary properties offset */ 1826 } ucd_record; 1827 1828 /* UCD access macros */ 1829 1830 #define UCD_BLOCK_SIZE 128 1831 #define REAL_GET_UCD(ch) (PRIV(ucd_records) + \ 1832 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \ 1833 UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE]) 1834 1835 #if PCRE2_CODE_UNIT_WIDTH == 32 1836 #define GET_UCD(ch) ((ch > MAX_UTF_CODE_POINT)? \ 1837 PRIV(dummy_ucd_record) : REAL_GET_UCD(ch)) 1838 #else 1839 #define GET_UCD(ch) REAL_GET_UCD(ch) 1840 #endif 1841 1842 #define UCD_SCRIPTX_MASK 0x3ff 1843 #define UCD_BIDICLASS_SHIFT 11 1844 #define UCD_BPROPS_MASK 0xfff 1845 1846 #define UCD_SCRIPTX_PROP(prop) ((prop)->scriptx_bidiclass & UCD_SCRIPTX_MASK) 1847 #define UCD_BIDICLASS_PROP(prop) ((prop)->scriptx_bidiclass >> UCD_BIDICLASS_SHIFT) 1848 #define UCD_BPROPS_PROP(prop) ((prop)->bprops & UCD_BPROPS_MASK) 1849 1850 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype 1851 #define UCD_SCRIPT(ch) GET_UCD(ch)->script 1852 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)] 1853 #define UCD_GRAPHBREAK(ch) GET_UCD(ch)->gbprop 1854 #define UCD_CASESET(ch) GET_UCD(ch)->caseset 1855 #define UCD_OTHERCASE(ch) ((uint32_t)((int)ch + (int)(GET_UCD(ch)->other_case))) 1856 #define UCD_SCRIPTX(ch) UCD_SCRIPTX_PROP(GET_UCD(ch)) 1857 #define UCD_BPROPS(ch) UCD_BPROPS_PROP(GET_UCD(ch)) 1858 #define UCD_BIDICLASS(ch) UCD_BIDICLASS_PROP(GET_UCD(ch)) 1859 1860 /* The "scriptx" and bprops fields contain offsets into vectors of 32-bit words 1861 that form a bitmap representing a list of scripts or boolean properties. These 1862 macros test or set a bit in the map by number. */ 1863 1864 #define MAPBIT(map,n) ((map)[(n)/32]&(1u<<((n)%32))) 1865 #define MAPSET(map,n) ((map)[(n)/32]|=(1u<<((n)%32))) 1866 1867 /* Header for serialized pcre2 codes. */ 1868 1869 typedef struct pcre2_serialized_data { 1870 uint32_t magic; 1871 uint32_t version; 1872 uint32_t config; 1873 int32_t number_of_codes; 1874 } pcre2_serialized_data; 1875 1876 1877 1878 /* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */ 1879 1880 /* When this file is included by pcre2test, PCRE2_CODE_UNIT_WIDTH is defined as 1881 0, so the following items are omitted. */ 1882 1883 #if defined PCRE2_CODE_UNIT_WIDTH && PCRE2_CODE_UNIT_WIDTH != 0 1884 1885 /* EBCDIC is supported only for the 8-bit library. */ 1886 1887 #if defined EBCDIC && PCRE2_CODE_UNIT_WIDTH != 8 1888 #error EBCDIC is not supported for the 16-bit or 32-bit libraries 1889 #endif 1890 1891 /* This is the largest non-UTF code point. */ 1892 1893 #define MAX_NON_UTF_CHAR (0xffffffffU >> (32 - PCRE2_CODE_UNIT_WIDTH)) 1894 1895 /* Internal shared data tables and variables. These are used by more than one 1896 of the exported public functions. They have to be "external" in the C sense, 1897 but are not part of the PCRE2 public API. Although the data for some of them is 1898 identical in all libraries, they must have different names so that multiple 1899 libraries can be simultaneously linked to a single application. However, UTF-8 1900 tables are needed only when compiling the 8-bit library. */ 1901 1902 #if PCRE2_CODE_UNIT_WIDTH == 8 1903 extern const int PRIV(utf8_table1)[]; 1904 extern const int PRIV(utf8_table1_size); 1905 extern const int PRIV(utf8_table2)[]; 1906 extern const int PRIV(utf8_table3)[]; 1907 extern const uint8_t PRIV(utf8_table4)[]; 1908 #endif 1909 1910 #define _pcre2_OP_lengths PCRE2_SUFFIX(_pcre2_OP_lengths_) 1911 #define _pcre2_callout_end_delims PCRE2_SUFFIX(_pcre2_callout_end_delims_) 1912 #define _pcre2_callout_start_delims PCRE2_SUFFIX(_pcre2_callout_start_delims_) 1913 #define _pcre2_default_compile_context PCRE2_SUFFIX(_pcre2_default_compile_context_) 1914 #define _pcre2_default_convert_context PCRE2_SUFFIX(_pcre2_default_convert_context_) 1915 #define _pcre2_default_match_context PCRE2_SUFFIX(_pcre2_default_match_context_) 1916 #define _pcre2_default_tables PCRE2_SUFFIX(_pcre2_default_tables_) 1917 #if PCRE2_CODE_UNIT_WIDTH == 32 1918 #define _pcre2_dummy_ucd_record PCRE2_SUFFIX(_pcre2_dummy_ucd_record_) 1919 #endif 1920 #define _pcre2_hspace_list PCRE2_SUFFIX(_pcre2_hspace_list_) 1921 #define _pcre2_vspace_list PCRE2_SUFFIX(_pcre2_vspace_list_) 1922 #define _pcre2_ucd_boolprop_sets PCRE2_SUFFIX(_pcre2_ucd_boolprop_sets_) 1923 #define _pcre2_ucd_caseless_sets PCRE2_SUFFIX(_pcre2_ucd_caseless_sets_) 1924 #define _pcre2_ucd_digit_sets PCRE2_SUFFIX(_pcre2_ucd_digit_sets_) 1925 #define _pcre2_ucd_script_sets PCRE2_SUFFIX(_pcre2_ucd_script_sets_) 1926 #define _pcre2_ucd_records PCRE2_SUFFIX(_pcre2_ucd_records_) 1927 #define _pcre2_ucd_stage1 PCRE2_SUFFIX(_pcre2_ucd_stage1_) 1928 #define _pcre2_ucd_stage2 PCRE2_SUFFIX(_pcre2_ucd_stage2_) 1929 #define _pcre2_ucp_gbtable PCRE2_SUFFIX(_pcre2_ucp_gbtable_) 1930 #define _pcre2_ucp_gentype PCRE2_SUFFIX(_pcre2_ucp_gentype_) 1931 #define _pcre2_ucp_typerange PCRE2_SUFFIX(_pcre2_ucp_typerange_) 1932 #define _pcre2_unicode_version PCRE2_SUFFIX(_pcre2_unicode_version_) 1933 #define _pcre2_utt PCRE2_SUFFIX(_pcre2_utt_) 1934 #define _pcre2_utt_names PCRE2_SUFFIX(_pcre2_utt_names_) 1935 #define _pcre2_utt_size PCRE2_SUFFIX(_pcre2_utt_size_) 1936 1937 extern const uint8_t PRIV(OP_lengths)[]; 1938 extern const uint32_t PRIV(callout_end_delims)[]; 1939 extern const uint32_t PRIV(callout_start_delims)[]; 1940 extern const pcre2_compile_context PRIV(default_compile_context); 1941 extern const pcre2_convert_context PRIV(default_convert_context); 1942 extern const pcre2_match_context PRIV(default_match_context); 1943 extern const uint8_t PRIV(default_tables)[]; 1944 extern const uint32_t PRIV(hspace_list)[]; 1945 extern const uint32_t PRIV(vspace_list)[]; 1946 extern const uint32_t PRIV(ucd_boolprop_sets)[]; 1947 extern const uint32_t PRIV(ucd_caseless_sets)[]; 1948 extern const uint32_t PRIV(ucd_digit_sets)[]; 1949 extern const uint32_t PRIV(ucd_script_sets)[]; 1950 extern const ucd_record PRIV(ucd_records)[]; 1951 #if PCRE2_CODE_UNIT_WIDTH == 32 1952 extern const ucd_record PRIV(dummy_ucd_record)[]; 1953 #endif 1954 extern const uint16_t PRIV(ucd_stage1)[]; 1955 extern const uint16_t PRIV(ucd_stage2)[]; 1956 extern const uint32_t PRIV(ucp_gbtable)[]; 1957 extern const uint32_t PRIV(ucp_gentype)[]; 1958 #ifdef SUPPORT_JIT 1959 extern const int PRIV(ucp_typerange)[]; 1960 #endif 1961 extern const char *PRIV(unicode_version); 1962 extern const ucp_type_table PRIV(utt)[]; 1963 extern const char PRIV(utt_names)[]; 1964 extern const size_t PRIV(utt_size); 1965 1966 /* Mode-dependent macros and hidden and private structures are defined in a 1967 separate file so that pcre2test can include them at all supported widths. When 1968 compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we can 1969 include them at the appropriate width, after setting up suffix macros for the 1970 private structures. */ 1971 1972 #define branch_chain PCRE2_SUFFIX(branch_chain_) 1973 #define compile_block PCRE2_SUFFIX(compile_block_) 1974 #define dfa_match_block PCRE2_SUFFIX(dfa_match_block_) 1975 #define match_block PCRE2_SUFFIX(match_block_) 1976 #define named_group PCRE2_SUFFIX(named_group_) 1977 1978 #include "pcre2_intmodedep.h" 1979 1980 /* Private "external" functions. These are internal functions that are called 1981 from modules other than the one in which they are defined. They have to be 1982 "external" in the C sense, but are not part of the PCRE2 public API. They are 1983 not referenced from pcre2test, and must not be defined when no code unit width 1984 is available. */ 1985 1986 #define _pcre2_auto_possessify PCRE2_SUFFIX(_pcre2_auto_possessify_) 1987 #define _pcre2_check_escape PCRE2_SUFFIX(_pcre2_check_escape_) 1988 #define _pcre2_extuni PCRE2_SUFFIX(_pcre2_extuni_) 1989 #define _pcre2_find_bracket PCRE2_SUFFIX(_pcre2_find_bracket_) 1990 #define _pcre2_is_newline PCRE2_SUFFIX(_pcre2_is_newline_) 1991 #define _pcre2_jit_free_rodata PCRE2_SUFFIX(_pcre2_jit_free_rodata_) 1992 #define _pcre2_jit_free PCRE2_SUFFIX(_pcre2_jit_free_) 1993 #define _pcre2_jit_get_size PCRE2_SUFFIX(_pcre2_jit_get_size_) 1994 #define _pcre2_jit_get_target PCRE2_SUFFIX(_pcre2_jit_get_target_) 1995 #define _pcre2_memctl_malloc PCRE2_SUFFIX(_pcre2_memctl_malloc_) 1996 #define _pcre2_ord2utf PCRE2_SUFFIX(_pcre2_ord2utf_) 1997 #define _pcre2_script_run PCRE2_SUFFIX(_pcre2_script_run_) 1998 #define _pcre2_strcmp PCRE2_SUFFIX(_pcre2_strcmp_) 1999 #define _pcre2_strcmp_c8 PCRE2_SUFFIX(_pcre2_strcmp_c8_) 2000 #define _pcre2_strcpy_c8 PCRE2_SUFFIX(_pcre2_strcpy_c8_) 2001 #define _pcre2_strlen PCRE2_SUFFIX(_pcre2_strlen_) 2002 #define _pcre2_strncmp PCRE2_SUFFIX(_pcre2_strncmp_) 2003 #define _pcre2_strncmp_c8 PCRE2_SUFFIX(_pcre2_strncmp_c8_) 2004 #define _pcre2_study PCRE2_SUFFIX(_pcre2_study_) 2005 #define _pcre2_valid_utf PCRE2_SUFFIX(_pcre2_valid_utf_) 2006 #define _pcre2_was_newline PCRE2_SUFFIX(_pcre2_was_newline_) 2007 #define _pcre2_xclass PCRE2_SUFFIX(_pcre2_xclass_) 2008 2009 extern int _pcre2_auto_possessify(PCRE2_UCHAR *, 2010 const compile_block *); 2011 extern int _pcre2_check_escape(PCRE2_SPTR *, PCRE2_SPTR, uint32_t *, 2012 int *, uint32_t, uint32_t, BOOL, compile_block *); 2013 extern PCRE2_SPTR _pcre2_extuni(uint32_t, PCRE2_SPTR, PCRE2_SPTR, PCRE2_SPTR, 2014 BOOL, int *); 2015 extern PCRE2_SPTR _pcre2_find_bracket(PCRE2_SPTR, BOOL, int); 2016 extern BOOL _pcre2_is_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 2017 uint32_t *, BOOL); 2018 extern void _pcre2_jit_free_rodata(void *, void *); 2019 extern void _pcre2_jit_free(void *, pcre2_memctl *); 2020 extern size_t _pcre2_jit_get_size(void *); 2021 const char * _pcre2_jit_get_target(void); 2022 extern void * _pcre2_memctl_malloc(size_t, pcre2_memctl *); 2023 extern unsigned int _pcre2_ord2utf(uint32_t, PCRE2_UCHAR *); 2024 extern BOOL _pcre2_script_run(PCRE2_SPTR, PCRE2_SPTR, BOOL); 2025 extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR); 2026 extern int _pcre2_strcmp_c8(PCRE2_SPTR, const char *); 2027 extern PCRE2_SIZE _pcre2_strcpy_c8(PCRE2_UCHAR *, const char *); 2028 extern PCRE2_SIZE _pcre2_strlen(PCRE2_SPTR); 2029 extern int _pcre2_strncmp(PCRE2_SPTR, PCRE2_SPTR, size_t); 2030 extern int _pcre2_strncmp_c8(PCRE2_SPTR, const char *, size_t); 2031 extern int _pcre2_study(pcre2_real_code *); 2032 extern int _pcre2_valid_utf(PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE *); 2033 extern BOOL _pcre2_was_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR, 2034 uint32_t *, BOOL); 2035 extern BOOL _pcre2_xclass(uint32_t, PCRE2_SPTR, BOOL); 2036 2037 /* This function is needed only when memmove() is not available. */ 2038 2039 #if !defined(VPCOMPAT) && !defined(HAVE_MEMMOVE) 2040 #define _pcre2_memmove PCRE2_SUFFIX(_pcre2_memmove) 2041 extern void * _pcre2_memmove(void *, const void *, size_t); 2042 #endif 2043 2044 #endif /* PCRE2_CODE_UNIT_WIDTH */ 2045 #endif /* PCRE2_INTERNAL_H_IDEMPOTENT_GUARD */ 2046 2047 /* End of pcre2_internal.h */ 2048
