| 7 |
and semantics are as close as possible to those of the Perl 5 language. |
and semantics are as close as possible to those of the Perl 5 language. |
| 8 |
|
|
| 9 |
Written by Philip Hazel |
Written by Philip Hazel |
| 10 |
Copyright (c) 1997-2010 University of Cambridge |
Copyright (c) 1997-2012 University of Cambridge |
| 11 |
|
|
| 12 |
----------------------------------------------------------------------------- |
----------------------------------------------------------------------------- |
| 13 |
Redistribution and use in source and binary forms, with or without |
Redistribution and use in source and binary forms, with or without |
| 40 |
|
|
| 41 |
/* This header contains definitions that are shared between the different |
/* This header contains definitions that are shared between the different |
| 42 |
modules, but which are not relevant to the exported API. This includes some |
modules, but which are not relevant to the exported API. This includes some |
| 43 |
functions whose names all begin with "_pcre_". */ |
functions whose names all begin with "_pcre_" or "_pcre16_" depending on |
| 44 |
|
the PRIV macro. */ |
| 45 |
|
|
| 46 |
#ifndef PCRE_INTERNAL_H |
#ifndef PCRE_INTERNAL_H |
| 47 |
#define PCRE_INTERNAL_H |
#define PCRE_INTERNAL_H |
| 52 |
#define PCRE_DEBUG |
#define PCRE_DEBUG |
| 53 |
#endif |
#endif |
| 54 |
|
|
| 55 |
/* We do not support both EBCDIC and UTF-8 at the same time. The "configure" |
/* PCRE is compiled as an 8 bit library if it is not requested otherwise. */ |
| 56 |
script prevents both being selected, but not everybody uses "configure". */ |
#ifndef COMPILE_PCRE16 |
| 57 |
|
#define COMPILE_PCRE8 |
|
#if defined EBCDIC && defined SUPPORT_UTF8 |
|
|
#error The use of both EBCDIC and SUPPORT_UTF8 is not supported. |
|
| 58 |
#endif |
#endif |
| 59 |
|
|
| 60 |
/* If SUPPORT_UCP is defined, SUPPORT_UTF8 must also be defined. The |
/* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The |
| 61 |
"configure" script ensures this, but not everybody uses "configure". */ |
"configure" script ensures this, but not everybody uses "configure". */ |
| 62 |
|
|
| 63 |
#if defined SUPPORT_UCP && !defined SUPPORT_UTF8 |
#if defined SUPPORT_UCP && !(defined SUPPORT_UTF) |
| 64 |
|
#define SUPPORT_UTF 1 |
| 65 |
|
#endif |
| 66 |
|
|
| 67 |
|
/* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility |
| 68 |
|
reasons with existing code. */ |
| 69 |
|
|
| 70 |
|
#if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF) |
| 71 |
|
#define SUPPORT_UTF 1 |
| 72 |
|
#endif |
| 73 |
|
|
| 74 |
|
/* Fixme: SUPPORT_UTF8 should be eventually disappear from the code. |
| 75 |
|
Until then we define it if SUPPORT_UTF is defined. */ |
| 76 |
|
|
| 77 |
|
#if defined SUPPORT_UTF && !(defined SUPPORT_UTF8) |
| 78 |
#define SUPPORT_UTF8 1 |
#define SUPPORT_UTF8 1 |
| 79 |
#endif |
#endif |
| 80 |
|
|
| 81 |
|
/* We do not support both EBCDIC and UTF-8/16 at the same time. The "configure" |
| 82 |
|
script prevents both being selected, but not everybody uses "configure". */ |
| 83 |
|
|
| 84 |
|
#if defined EBCDIC && defined SUPPORT_UTF |
| 85 |
|
#error The use of both EBCDIC and SUPPORT_UTF8/16 is not supported. |
| 86 |
|
#endif |
| 87 |
|
|
| 88 |
/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef |
/* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef |
| 89 |
inline, and there are *still* stupid compilers about that don't like indented |
inline, and there are *still* stupid compilers about that don't like indented |
| 90 |
pre-processor statements, or at least there were when I first wrote this. After |
pre-processor statements, or at least there were when I first wrote this. After |
| 178 |
#define PCRE_CALL_CONVENTION |
#define PCRE_CALL_CONVENTION |
| 179 |
#endif |
#endif |
| 180 |
|
|
| 181 |
/* We need to have types that specify unsigned 16-bit and 32-bit integers. We |
/* We need to have types that specify unsigned 8, 16 and 32-bit integers. We |
| 182 |
cannot determine these outside the compilation (e.g. by running a program as |
cannot determine these outside the compilation (e.g. by running a program as |
| 183 |
part of "configure") because PCRE is often cross-compiled for use on other |
part of "configure") because PCRE is often cross-compiled for use on other |
| 184 |
systems. Instead we make use of the maximum sizes that are available at |
systems. Instead we make use of the maximum sizes that are available at |
| 185 |
preprocessor time in standard C environments. */ |
preprocessor time in standard C environments. */ |
| 186 |
|
|
| 187 |
|
typedef unsigned char pcre_uint8; |
| 188 |
|
|
| 189 |
#if USHRT_MAX == 65535 |
#if USHRT_MAX == 65535 |
| 190 |
typedef unsigned short pcre_uint16; |
typedef unsigned short pcre_uint16; |
| 191 |
typedef short pcre_int16; |
typedef short pcre_int16; |
| 228 |
|
|
| 229 |
/* All character handling must be done as unsigned characters. Otherwise there |
/* All character handling must be done as unsigned characters. Otherwise there |
| 230 |
are problems with top-bit-set characters and functions such as isspace(). |
are problems with top-bit-set characters and functions such as isspace(). |
| 231 |
However, we leave the interface to the outside world as char *, because that |
However, we leave the interface to the outside world as char * or short *, |
| 232 |
should make things easier for callers. We define a short type for unsigned char |
because that should make things easier for callers. This character type is |
| 233 |
to save lots of typing. I tried "uchar", but it causes problems on Digital |
called pcre_uchar. |
| 234 |
Unix, where it is defined in sys/types, so use "uschar" instead. */ |
|
| 235 |
|
The IN_UCHARS macro multiply its argument with the byte size of the current |
| 236 |
|
pcre_uchar type. Useful for memcpy and such operations, whose require the |
| 237 |
|
byte size of their input/output buffers. |
| 238 |
|
|
| 239 |
|
The MAX_255 macro checks whether its pcre_uchar input is less than 256. |
| 240 |
|
|
| 241 |
|
The TABLE_GET macro is designed for accessing elements of tables whose contain |
| 242 |
|
exactly 256 items. When the character is able to contain more than 256 |
| 243 |
|
items, some check is needed before accessing these tables. |
| 244 |
|
*/ |
| 245 |
|
|
| 246 |
|
#ifdef COMPILE_PCRE8 |
| 247 |
|
|
| 248 |
|
typedef unsigned char pcre_uchar; |
| 249 |
|
#define IN_UCHARS(x) (x) |
| 250 |
|
#define MAX_255(c) 1 |
| 251 |
|
#define TABLE_GET(c, table, default) ((table)[c]) |
| 252 |
|
|
| 253 |
|
#else |
| 254 |
|
|
| 255 |
|
#ifdef COMPILE_PCRE16 |
| 256 |
|
#if USHRT_MAX != 65535 |
| 257 |
|
/* This is a warning message. Change PCRE_SCHAR16 to a 16 bit data type in |
| 258 |
|
pcre.h(.in) and disable (comment out) this message. */ |
| 259 |
|
#error Warning: PCRE_SCHAR16 is not a 16 bit data type. |
| 260 |
|
#endif |
| 261 |
|
|
| 262 |
|
typedef pcre_uint16 pcre_uchar; |
| 263 |
|
#define IN_UCHARS(x) ((x) << 1) |
| 264 |
|
#define MAX_255(c) ((c) <= 255u) |
| 265 |
|
#define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default)) |
| 266 |
|
|
| 267 |
typedef unsigned char uschar; |
#else |
| 268 |
|
#error Unsupported compiling mode |
| 269 |
|
#endif /* COMPILE_PCRE16 */ |
| 270 |
|
|
| 271 |
|
#endif /* COMPILE_PCRE8 */ |
| 272 |
|
|
| 273 |
/* This is an unsigned int value that no character can ever have. UTF-8 |
/* This is an unsigned int value that no character can ever have. UTF-8 |
| 274 |
characters only go up to 0x7fffffff (though Unicode doesn't go beyond |
characters only go up to 0x7fffffff (though Unicode doesn't go beyond |
| 291 |
#define IS_NEWLINE(p) \ |
#define IS_NEWLINE(p) \ |
| 292 |
((NLBLOCK->nltype != NLTYPE_FIXED)? \ |
((NLBLOCK->nltype != NLTYPE_FIXED)? \ |
| 293 |
((p) < NLBLOCK->PSEND && \ |
((p) < NLBLOCK->PSEND && \ |
| 294 |
_pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\ |
PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \ |
| 295 |
utf8)) \ |
&(NLBLOCK->nllen), utf)) \ |
| 296 |
: \ |
: \ |
| 297 |
((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ |
((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ |
| 298 |
(p)[0] == NLBLOCK->nl[0] && \ |
(p)[0] == NLBLOCK->nl[0] && \ |
| 305 |
#define WAS_NEWLINE(p) \ |
#define WAS_NEWLINE(p) \ |
| 306 |
((NLBLOCK->nltype != NLTYPE_FIXED)? \ |
((NLBLOCK->nltype != NLTYPE_FIXED)? \ |
| 307 |
((p) > NLBLOCK->PSSTART && \ |
((p) > NLBLOCK->PSSTART && \ |
| 308 |
_pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ |
PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ |
| 309 |
&(NLBLOCK->nllen), utf8)) \ |
&(NLBLOCK->nllen), utf)) \ |
| 310 |
: \ |
: \ |
| 311 |
((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ |
((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ |
| 312 |
(p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \ |
(p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \ |
| 324 |
must begin with PCRE_. */ |
must begin with PCRE_. */ |
| 325 |
|
|
| 326 |
#ifdef CUSTOM_SUBJECT_PTR |
#ifdef CUSTOM_SUBJECT_PTR |
| 327 |
#define PCRE_SPTR CUSTOM_SUBJECT_PTR |
#define PCRE_PUCHAR CUSTOM_SUBJECT_PTR |
|
#define USPTR CUSTOM_SUBJECT_PTR |
|
| 328 |
#else |
#else |
| 329 |
#define PCRE_SPTR const char * |
#define PCRE_PUCHAR const pcre_uchar * |
|
#define USPTR const unsigned char * |
|
| 330 |
#endif |
#endif |
| 331 |
|
|
|
|
|
|
|
|
| 332 |
/* Include the public PCRE header and the definitions of UCP character property |
/* Include the public PCRE header and the definitions of UCP character property |
| 333 |
values. */ |
values. */ |
| 334 |
|
|
| 396 |
the config.h file, but can be overridden by using -D on the command line. This |
the config.h file, but can be overridden by using -D on the command line. This |
| 397 |
is automated on Unix systems via the "configure" command. */ |
is automated on Unix systems via the "configure" command. */ |
| 398 |
|
|
| 399 |
|
#ifdef COMPILE_PCRE8 |
| 400 |
|
|
| 401 |
#if LINK_SIZE == 2 |
#if LINK_SIZE == 2 |
| 402 |
|
|
| 403 |
#define PUT(a,n,d) \ |
#define PUT(a,n,d) \ |
| 434 |
#define GET(a,n) \ |
#define GET(a,n) \ |
| 435 |
(((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3]) |
(((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3]) |
| 436 |
|
|
| 437 |
#define MAX_PATTERN_SIZE (1 << 30) /* Keep it positive */ |
/* Keep it positive */ |
| 438 |
|
#define MAX_PATTERN_SIZE (1 << 30) |
| 439 |
|
|
| 440 |
|
#else |
| 441 |
|
#error LINK_SIZE must be either 2, 3, or 4 |
| 442 |
|
#endif |
| 443 |
|
|
| 444 |
|
#else /* COMPILE_PCRE8 */ |
| 445 |
|
|
| 446 |
|
#ifdef COMPILE_PCRE16 |
| 447 |
|
|
| 448 |
|
#if LINK_SIZE == 2 |
| 449 |
|
|
| 450 |
|
#undef LINK_SIZE |
| 451 |
|
#define LINK_SIZE 1 |
| 452 |
|
|
| 453 |
|
#define PUT(a,n,d) \ |
| 454 |
|
(a[n] = (d)) |
| 455 |
|
|
| 456 |
|
#define GET(a,n) \ |
| 457 |
|
(a[n]) |
| 458 |
|
|
| 459 |
|
#define MAX_PATTERN_SIZE (1 << 16) |
| 460 |
|
|
| 461 |
|
#elif LINK_SIZE == 3 || LINK_SIZE == 4 |
| 462 |
|
|
| 463 |
|
#undef LINK_SIZE |
| 464 |
|
#define LINK_SIZE 2 |
| 465 |
|
|
| 466 |
|
#define PUT(a,n,d) \ |
| 467 |
|
(a[n] = (d) >> 16), \ |
| 468 |
|
(a[(n)+1] = (d) & 65535) |
| 469 |
|
|
| 470 |
|
#define GET(a,n) \ |
| 471 |
|
(((a)[n] << 16) | (a)[(n)+1]) |
| 472 |
|
|
| 473 |
|
/* Keep it positive */ |
| 474 |
|
#define MAX_PATTERN_SIZE (1 << 30) |
| 475 |
|
|
| 476 |
#else |
#else |
| 477 |
#error LINK_SIZE must be either 2, 3, or 4 |
#error LINK_SIZE must be either 2, 3, or 4 |
| 478 |
#endif |
#endif |
| 479 |
|
|
| 480 |
|
#else |
| 481 |
|
#error Unsupported compiling mode |
| 482 |
|
#endif /* COMPILE_PCRE16 */ |
| 483 |
|
|
| 484 |
|
#endif /* COMPILE_PCRE8 */ |
| 485 |
|
|
| 486 |
/* Convenience macro defined in terms of the others */ |
/* Convenience macro defined in terms of the others */ |
| 487 |
|
|
| 492 |
offsets changes. There are used for repeat counts and for other things such as |
offsets changes. There are used for repeat counts and for other things such as |
| 493 |
capturing parenthesis numbers in back references. */ |
capturing parenthesis numbers in back references. */ |
| 494 |
|
|
| 495 |
|
#ifdef COMPILE_PCRE8 |
| 496 |
|
|
| 497 |
|
#define IMM2_SIZE 2 |
| 498 |
|
|
| 499 |
#define PUT2(a,n,d) \ |
#define PUT2(a,n,d) \ |
| 500 |
a[n] = (d) >> 8; \ |
a[n] = (d) >> 8; \ |
| 501 |
a[(n)+1] = (d) & 255 |
a[(n)+1] = (d) & 255 |
| 503 |
#define GET2(a,n) \ |
#define GET2(a,n) \ |
| 504 |
(((a)[n] << 8) | (a)[(n)+1]) |
(((a)[n] << 8) | (a)[(n)+1]) |
| 505 |
|
|
| 506 |
#define PUT2INC(a,n,d) PUT2(a,n,d), a += 2 |
#else /* COMPILE_PCRE8 */ |
| 507 |
|
|
| 508 |
|
#ifdef COMPILE_PCRE16 |
| 509 |
|
|
| 510 |
|
#define IMM2_SIZE 1 |
| 511 |
|
|
| 512 |
|
#define PUT2(a,n,d) \ |
| 513 |
|
a[n] = d |
| 514 |
|
|
| 515 |
|
#define GET2(a,n) \ |
| 516 |
|
a[n] |
| 517 |
|
|
| 518 |
|
#else |
| 519 |
|
#error Unsupported compiling mode |
| 520 |
|
#endif /* COMPILE_PCRE16 */ |
| 521 |
|
|
| 522 |
|
#endif /* COMPILE_PCRE8 */ |
| 523 |
|
|
| 524 |
/* When UTF-8 encoding is being used, a character is no longer just a single |
#define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE |
|
byte. The macros for character handling generate simple sequences when used in |
|
|
byte-mode, and more complicated ones for UTF-8 characters. GETCHARLENTEST is |
|
|
not used when UTF-8 is not supported, so it is not defined, and BACKCHAR should |
|
|
never be called in byte mode. To make sure they can never even appear when |
|
|
UTF-8 support is omitted, we don't even define them. */ |
|
| 525 |
|
|
| 526 |
#ifndef SUPPORT_UTF8 |
/* When UTF encoding is being used, a character is no longer just a single |
| 527 |
|
character. The macros for character handling generate simple sequences when |
| 528 |
|
used in character-mode, and more complicated ones for UTF characters. |
| 529 |
|
GETCHARLENTEST and other macros are not used when UTF is not supported, |
| 530 |
|
so they are not defined. To make sure they can never even appear when |
| 531 |
|
UTF support is omitted, we don't even define them. */ |
| 532 |
|
|
| 533 |
|
#ifndef SUPPORT_UTF |
| 534 |
|
|
| 535 |
|
/* #define MAX_VALUE_FOR_SINGLE_CHAR */ |
| 536 |
|
/* #define HAS_EXTRALEN(c) */ |
| 537 |
|
/* #define GET_EXTRALEN(c) */ |
| 538 |
|
/* #define NOT_FIRSTCHAR(c) */ |
| 539 |
#define GETCHAR(c, eptr) c = *eptr; |
#define GETCHAR(c, eptr) c = *eptr; |
| 540 |
#define GETCHARTEST(c, eptr) c = *eptr; |
#define GETCHARTEST(c, eptr) c = *eptr; |
| 541 |
#define GETCHARINC(c, eptr) c = *eptr++; |
#define GETCHARINC(c, eptr) c = *eptr++; |
| 543 |
#define GETCHARLEN(c, eptr, len) c = *eptr; |
#define GETCHARLEN(c, eptr, len) c = *eptr; |
| 544 |
/* #define GETCHARLENTEST(c, eptr, len) */ |
/* #define GETCHARLENTEST(c, eptr, len) */ |
| 545 |
/* #define BACKCHAR(eptr) */ |
/* #define BACKCHAR(eptr) */ |
| 546 |
|
/* #define FORWARDCHAR(eptr) */ |
| 547 |
|
/* #define ACROSSCHAR(condition, eptr, action) */ |
| 548 |
|
|
| 549 |
|
#else /* SUPPORT_UTF */ |
| 550 |
|
|
| 551 |
#else /* SUPPORT_UTF8 */ |
#ifdef COMPILE_PCRE8 |
| 552 |
|
|
| 553 |
/* These macros were originally written in the form of loops that used data |
/* These macros were originally written in the form of loops that used data |
| 554 |
from the tables whose names start with _pcre_utf8_table. They were rewritten by |
from the tables whose names start with PRIV(utf8_table). They were rewritten by |
| 555 |
a user so as not to use loops, because in some environments this gives a |
a user so as not to use loops, because in some environments this gives a |
| 556 |
significant performance advantage, and it seems never to do any harm. */ |
significant performance advantage, and it seems never to do any harm. */ |
| 557 |
|
|
| 558 |
|
/* Tells the biggest code point which can be encoded as a single character. */ |
| 559 |
|
|
| 560 |
|
#define MAX_VALUE_FOR_SINGLE_CHAR 127 |
| 561 |
|
|
| 562 |
|
/* Tests whether the code point needs extra characters to decode. */ |
| 563 |
|
|
| 564 |
|
#define HAS_EXTRALEN(c) ((c) >= 0xc0) |
| 565 |
|
|
| 566 |
|
/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE. |
| 567 |
|
Otherwise it has an undefined behaviour. */ |
| 568 |
|
|
| 569 |
|
#define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f]) |
| 570 |
|
|
| 571 |
|
/* Returns TRUE, if the given character is not the first character |
| 572 |
|
of a UTF sequence. */ |
| 573 |
|
|
| 574 |
|
#define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80) |
| 575 |
|
|
| 576 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not |
| 577 |
advancing the pointer. */ |
advancing the pointer. */ |
| 578 |
|
|
| 607 |
|
|
| 608 |
#define GETCHARTEST(c, eptr) \ |
#define GETCHARTEST(c, eptr) \ |
| 609 |
c = *eptr; \ |
c = *eptr; \ |
| 610 |
if (utf8 && c >= 0xc0) GETUTF8(c, eptr); |
if (utf && c >= 0xc0) GETUTF8(c, eptr); |
| 611 |
|
|
| 612 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, advancing |
/* Base macro to pick up the remaining bytes of a UTF-8 character, advancing |
| 613 |
the pointer. */ |
the pointer. */ |
| 655 |
|
|
| 656 |
#define GETCHARINCTEST(c, eptr) \ |
#define GETCHARINCTEST(c, eptr) \ |
| 657 |
c = *eptr++; \ |
c = *eptr++; \ |
| 658 |
if (utf8 && c >= 0xc0) GETUTF8INC(c, eptr); |
if (utf && c >= 0xc0) GETUTF8INC(c, eptr); |
| 659 |
|
|
| 660 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not |
| 661 |
advancing the pointer, incrementing the length. */ |
advancing the pointer, incrementing the length. */ |
| 707 |
|
|
| 708 |
#define GETCHARLENTEST(c, eptr, len) \ |
#define GETCHARLENTEST(c, eptr, len) \ |
| 709 |
c = *eptr; \ |
c = *eptr; \ |
| 710 |
if (utf8 && c >= 0xc0) GETUTF8LEN(c, eptr, len); |
if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len); |
| 711 |
|
|
| 712 |
/* If the pointer is not at the start of a character, move it back until |
/* If the pointer is not at the start of a character, move it back until |
| 713 |
it is. This is called only in UTF-8 mode - we don't put a test within the macro |
it is. This is called only in UTF-8 mode - we don't put a test within the macro |
| 715 |
|
|
| 716 |
#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr-- |
#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr-- |
| 717 |
|
|
| 718 |
#endif /* SUPPORT_UTF8 */ |
/* Same as above, just in the other direction. */ |
| 719 |
|
#define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++ |
| 720 |
|
|
| 721 |
|
/* Same as above, but it allows a fully customizable form. */ |
| 722 |
|
#define ACROSSCHAR(condition, eptr, action) \ |
| 723 |
|
while((condition) && ((eptr) & 0xc0) == 0x80) action |
| 724 |
|
|
| 725 |
|
#else /* COMPILE_PCRE8 */ |
| 726 |
|
|
| 727 |
|
#ifdef COMPILE_PCRE16 |
| 728 |
|
|
| 729 |
|
/* Tells the biggest code point which can be encoded as a single character. */ |
| 730 |
|
|
| 731 |
|
#define MAX_VALUE_FOR_SINGLE_CHAR 65535 |
| 732 |
|
|
| 733 |
|
/* Tests whether the code point needs extra characters to decode. */ |
| 734 |
|
|
| 735 |
|
#define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800) |
| 736 |
|
|
| 737 |
|
/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE. |
| 738 |
|
Otherwise it has an undefined behaviour. */ |
| 739 |
|
|
| 740 |
|
#define GET_EXTRALEN(c) 1 |
| 741 |
|
|
| 742 |
|
/* Returns TRUE, if the given character is not the first character |
| 743 |
|
of a UTF sequence. */ |
| 744 |
|
|
| 745 |
|
#define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00) |
| 746 |
|
|
| 747 |
|
/* Base macro to pick up the low surrogate of a UTF-16 character, not |
| 748 |
|
advancing the pointer. */ |
| 749 |
|
|
| 750 |
|
#define GETUTF16(c, eptr) \ |
| 751 |
|
{ c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; } |
| 752 |
|
|
| 753 |
|
/* Get the next UTF-16 character, not advancing the pointer. This is called when |
| 754 |
|
we know we are in UTF-16 mode. */ |
| 755 |
|
|
| 756 |
|
#define GETCHAR(c, eptr) \ |
| 757 |
|
c = *eptr; \ |
| 758 |
|
if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr); |
| 759 |
|
|
| 760 |
|
/* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the |
| 761 |
|
pointer. */ |
| 762 |
|
|
| 763 |
|
#define GETCHARTEST(c, eptr) \ |
| 764 |
|
c = *eptr; \ |
| 765 |
|
if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr); |
| 766 |
|
|
| 767 |
|
/* Base macro to pick up the low surrogate of a UTF-16 character, advancing |
| 768 |
|
the pointer. */ |
| 769 |
|
|
| 770 |
|
#define GETUTF16INC(c, eptr) \ |
| 771 |
|
{ c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; } |
| 772 |
|
|
| 773 |
|
/* Get the next UTF-16 character, advancing the pointer. This is called when we |
| 774 |
|
know we are in UTF-16 mode. */ |
| 775 |
|
|
| 776 |
|
#define GETCHARINC(c, eptr) \ |
| 777 |
|
c = *eptr++; \ |
| 778 |
|
if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr); |
| 779 |
|
|
| 780 |
|
/* Get the next character, testing for UTF-16 mode, and advancing the pointer. |
| 781 |
|
This is called when we don't know if we are in UTF-16 mode. */ |
| 782 |
|
|
| 783 |
|
#define GETCHARINCTEST(c, eptr) \ |
| 784 |
|
c = *eptr++; \ |
| 785 |
|
if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr); |
| 786 |
|
|
| 787 |
|
/* Base macro to pick up the low surrogate of a UTF-16 character, not |
| 788 |
|
advancing the pointer, incrementing the length. */ |
| 789 |
|
|
| 790 |
|
#define GETUTF16LEN(c, eptr, len) \ |
| 791 |
|
{ c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; } |
| 792 |
|
|
| 793 |
|
/* Get the next UTF-16 character, not advancing the pointer, incrementing |
| 794 |
|
length if there is a low surrogate. This is called when we know we are in |
| 795 |
|
UTF-16 mode. */ |
| 796 |
|
|
| 797 |
|
#define GETCHARLEN(c, eptr, len) \ |
| 798 |
|
c = *eptr; \ |
| 799 |
|
if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len); |
| 800 |
|
|
| 801 |
|
/* Get the next UTF-816character, testing for UTF-16 mode, not advancing the |
| 802 |
|
pointer, incrementing length if there is a low surrogate. This is called when |
| 803 |
|
we do not know if we are in UTF-16 mode. */ |
| 804 |
|
|
| 805 |
|
#define GETCHARLENTEST(c, eptr, len) \ |
| 806 |
|
c = *eptr; \ |
| 807 |
|
if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len); |
| 808 |
|
|
| 809 |
|
/* If the pointer is not at the start of a character, move it back until |
| 810 |
|
it is. This is called only in UTF-16 mode - we don't put a test within the |
| 811 |
|
macro because almost all calls are already within a block of UTF-16 only |
| 812 |
|
code. */ |
| 813 |
|
|
| 814 |
|
#define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr-- |
| 815 |
|
|
| 816 |
|
/* Same as above, just in the other direction. */ |
| 817 |
|
#define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++ |
| 818 |
|
|
| 819 |
|
/* Same as above, but it allows a fully customizable form. */ |
| 820 |
|
#define ACROSSCHAR(condition, eptr, action) \ |
| 821 |
|
if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action |
| 822 |
|
|
| 823 |
|
#endif |
| 824 |
|
|
| 825 |
|
#endif /* COMPILE_PCRE8 */ |
| 826 |
|
|
| 827 |
|
#endif /* SUPPORT_UTF */ |
| 828 |
|
|
| 829 |
|
|
| 830 |
/* In case there is no definition of offsetof() provided - though any proper |
/* In case there is no definition of offsetof() provided - though any proper |
| 841 |
the restrictions on partial matching have been lifted. It remains for backwards |
the restrictions on partial matching have been lifted. It remains for backwards |
| 842 |
compatibility. */ |
compatibility. */ |
| 843 |
|
|
| 844 |
#define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */ |
#ifdef COMPILE_PCRE8 |
| 845 |
#define PCRE_FIRSTSET 0x0002 /* first_byte is set */ |
#define PCRE_MODE 0x0001 /* compiled in 8 bit mode */ |
| 846 |
#define PCRE_REQCHSET 0x0004 /* req_byte is set */ |
#endif |
| 847 |
#define PCRE_STARTLINE 0x0008 /* start after \n for multiline */ |
#ifdef COMPILE_PCRE16 |
| 848 |
#define PCRE_JCHANGED 0x0010 /* j option used in regex */ |
#define PCRE_MODE 0x0002 /* compiled in 16 bit mode */ |
| 849 |
#define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */ |
#endif |
| 850 |
|
#define PCRE_FIRSTSET 0x0010 /* first_char is set */ |
| 851 |
|
#define PCRE_FCH_CASELESS 0x0020 /* caseless first char */ |
| 852 |
|
#define PCRE_REQCHSET 0x0040 /* req_byte is set */ |
| 853 |
|
#define PCRE_RCH_CASELESS 0x0080 /* caseless requested char */ |
| 854 |
|
#define PCRE_STARTLINE 0x0100 /* start after \n for multiline */ |
| 855 |
|
#define PCRE_NOPARTIAL 0x0200 /* can't use partial with this regex */ |
| 856 |
|
#define PCRE_JCHANGED 0x0400 /* j option used in regex */ |
| 857 |
|
#define PCRE_HASCRORLF 0x0800 /* explicit \r or \n in pattern */ |
| 858 |
|
#define PCRE_HASTHEN 0x1000 /* pattern contains (*THEN) */ |
| 859 |
|
|
| 860 |
/* Flags for the "extra" block produced by pcre_study(). */ |
/* Flags for the "extra" block produced by pcre_study(). */ |
| 861 |
|
|
| 886 |
PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \ |
PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \ |
| 887 |
PCRE_NO_START_OPTIMIZE) |
PCRE_NO_START_OPTIMIZE) |
| 888 |
|
|
| 889 |
#define PUBLIC_STUDY_OPTIONS 0 /* None defined */ |
#define PUBLIC_STUDY_OPTIONS \ |
| 890 |
|
PCRE_STUDY_JIT_COMPILE |
| 891 |
|
|
| 892 |
/* Magic number to provide a small check against being handed junk. Also used |
/* Magic number to provide a small check against being handed junk. */ |
|
to detect whether a pattern was compiled on a host of different endianness. */ |
|
| 893 |
|
|
| 894 |
#define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ |
#define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ |
| 895 |
|
|
| 896 |
|
/* This variable is used to detect a loaded regular expression |
| 897 |
|
in different endianness. */ |
| 898 |
|
|
| 899 |
|
#define REVERSED_MAGIC_NUMBER 0x45524350UL /* 'ERCP' */ |
| 900 |
|
|
| 901 |
/* Negative values for the firstchar and reqchar variables */ |
/* Negative values for the firstchar and reqchar variables */ |
| 902 |
|
|
| 903 |
#define REQ_UNSET (-2) |
#define REQ_UNSET (-2) |
| 908 |
|
|
| 909 |
#define REQ_BYTE_MAX 1000 |
#define REQ_BYTE_MAX 1000 |
| 910 |
|
|
|
/* Flags added to firstbyte or reqbyte; a "non-literal" item is either a |
|
|
variable-length repeat, or a anything other than literal characters. */ |
|
|
|
|
|
#define REQ_CASELESS 0x0100 /* indicates caselessness */ |
|
|
#define REQ_VARY 0x0200 /* reqbyte followed non-literal item */ |
|
|
|
|
| 911 |
/* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in |
/* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in |
| 912 |
environments where these macros are defined elsewhere. Unfortunately, there |
environments where these macros are defined elsewhere. Unfortunately, there |
| 913 |
is no way to do the same for the typedef. */ |
is no way to do the same for the typedef. */ |
| 936 |
application that did need both could compile two versions of the library, using |
application that did need both could compile two versions of the library, using |
| 937 |
macros to give the functions distinct names. */ |
macros to give the functions distinct names. */ |
| 938 |
|
|
| 939 |
#ifndef SUPPORT_UTF8 |
#ifndef SUPPORT_UTF |
| 940 |
|
|
| 941 |
/* UTF-8 support is not enabled; use the platform-dependent character literals |
/* UTF-8 support is not enabled; use the platform-dependent character literals |
| 942 |
so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */ |
so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */ |
| 1196 |
#define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)" |
#define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)" |
| 1197 |
#define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)" |
#define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)" |
| 1198 |
#define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)" |
#define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)" |
| 1199 |
#define STRING_UTF8_RIGHTPAR "UTF8)" |
#ifdef COMPILE_PCRE8 |
| 1200 |
|
#define STRING_UTF_RIGHTPAR "UTF8)" |
| 1201 |
|
#endif |
| 1202 |
|
#ifdef COMPILE_PCRE16 |
| 1203 |
|
#define STRING_UTF_RIGHTPAR "UTF16)" |
| 1204 |
|
#endif |
| 1205 |
#define STRING_UCP_RIGHTPAR "UCP)" |
#define STRING_UCP_RIGHTPAR "UCP)" |
| 1206 |
#define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)" |
#define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)" |
| 1207 |
|
|
| 1208 |
#else /* SUPPORT_UTF8 */ |
#else /* SUPPORT_UTF */ |
| 1209 |
|
|
| 1210 |
/* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This |
/* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This |
| 1211 |
works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode |
works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode |
| 1456 |
#define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS |
#define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS |
| 1457 |
#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 |
#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 |
| 1458 |
#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 |
#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 |
| 1459 |
#define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS |
#ifdef COMPILE_PCRE8 |
| 1460 |
|
#define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS |
| 1461 |
|
#endif |
| 1462 |
|
#ifdef COMPILE_PCRE16 |
| 1463 |
|
#define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS |
| 1464 |
|
#endif |
| 1465 |
#define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS |
#define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS |
| 1466 |
#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 |
#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 |
| 1467 |
|
|
| 1468 |
#endif /* SUPPORT_UTF8 */ |
#endif /* SUPPORT_UTF */ |
| 1469 |
|
|
| 1470 |
/* Escape items that are just an encoding of a particular data value. */ |
/* Escape items that are just an encoding of a particular data value. */ |
| 1471 |
|
|
| 1505 |
#define PT_WORD 8 /* Word - L plus N plus underscore */ |
#define PT_WORD 8 /* Word - L plus N plus underscore */ |
| 1506 |
|
|
| 1507 |
/* Flag bits and data types for the extended class (OP_XCLASS) for classes that |
/* Flag bits and data types for the extended class (OP_XCLASS) for classes that |
| 1508 |
contain UTF-8 characters with values greater than 255. */ |
contain characters with values greater than 255. */ |
| 1509 |
|
|
| 1510 |
#define XCL_NOT 0x01 /* Flag: this is a negative class */ |
#define XCL_NOT 0x01 /* Flag: this is a negative class */ |
| 1511 |
#define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ |
#define XCL_MAP 0x02 /* Flag: a 32-byte map is present */ |
| 1521 |
their negation. Also, they must appear in the same order as in the opcode |
their negation. Also, they must appear in the same order as in the opcode |
| 1522 |
definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it |
definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it |
| 1523 |
corresponds to "." in DOTALL mode rather than an escape sequence. It is also |
corresponds to "." in DOTALL mode rather than an escape sequence. It is also |
| 1524 |
used for [^] in JavaScript compatibility mode. In non-DOTALL mode, "." behaves |
used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In |
| 1525 |
like \N. |
non-DOTALL mode, "." behaves like \N. |
| 1526 |
|
|
| 1527 |
The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc. |
The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc. |
| 1528 |
when PCRE_UCP is set, when replacement of \d etc by \p sequences is required. |
when PCRE_UCP is set, when replacement of \d etc by \p sequences is required. |
| 1702 |
OP_CLASS, /* 106 Match a character class, chars < 256 only */ |
OP_CLASS, /* 106 Match a character class, chars < 256 only */ |
| 1703 |
OP_NCLASS, /* 107 Same, but the bitmap was created from a negative |
OP_NCLASS, /* 107 Same, but the bitmap was created from a negative |
| 1704 |
class - the difference is relevant only when a |
class - the difference is relevant only when a |
| 1705 |
UTF-8 character > 255 is encountered. */ |
character > 255 is encountered. */ |
| 1706 |
OP_XCLASS, /* 108 Extended class for handling UTF-8 chars within the |
OP_XCLASS, /* 108 Extended class for handling > 255 chars within the |
| 1707 |
class. This does both positive and negative. */ |
class. This does both positive and negative. */ |
| 1708 |
OP_REF, /* 109 Match a back reference, casefully */ |
OP_REF, /* 109 Match a back reference, casefully */ |
| 1709 |
OP_REFI, /* 110 Match a back reference, caselessly */ |
OP_REFI, /* 110 Match a back reference, caselessly */ |
| 1716 |
OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */ |
OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */ |
| 1717 |
OP_KETRPOS, /* 117 Possessive unlimited repeat. */ |
OP_KETRPOS, /* 117 Possessive unlimited repeat. */ |
| 1718 |
|
|
| 1719 |
/* The assertions must come before BRA, CBRA, ONCE, and COND.*/ |
/* The assertions must come before BRA, CBRA, ONCE, and COND, and the four |
| 1720 |
|
asserts must remain in order. */ |
| 1721 |
|
|
| 1722 |
OP_ASSERT, /* 118 Positive lookahead */ |
OP_REVERSE, /* 118 Move pointer back - used in lookbehind assertions */ |
| 1723 |
OP_ASSERT_NOT, /* 119 Negative lookahead */ |
OP_ASSERT, /* 119 Positive lookahead */ |
| 1724 |
OP_ASSERTBACK, /* 120 Positive lookbehind */ |
OP_ASSERT_NOT, /* 120 Negative lookahead */ |
| 1725 |
OP_ASSERTBACK_NOT, /* 121 Negative lookbehind */ |
OP_ASSERTBACK, /* 121 Positive lookbehind */ |
| 1726 |
OP_REVERSE, /* 122 Move pointer back - used in lookbehind assertions */ |
OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */ |
| 1727 |
|
|
| 1728 |
/* ONCE, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come after the assertions, |
/* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately |
| 1729 |
with ONCE first, as there's a test for >= ONCE for a subpattern that isn't an |
after the assertions, with ONCE first, as there's a test for >= ONCE for a |
| 1730 |
assertion. The POS versions must immediately follow the non-POS versions in |
subpattern that isn't an assertion. The POS versions must immediately follow |
| 1731 |
each case. */ |
the non-POS versions in each case. */ |
| 1732 |
|
|
| 1733 |
OP_ONCE, /* 123 Atomic group */ |
OP_ONCE, /* 123 Atomic group, contains captures */ |
| 1734 |
OP_BRA, /* 124 Start of non-capturing bracket */ |
OP_ONCE_NC, /* 124 Atomic group containing no captures */ |
| 1735 |
OP_BRAPOS, /* 125 Ditto, with unlimited, possessive repeat */ |
OP_BRA, /* 125 Start of non-capturing bracket */ |
| 1736 |
OP_CBRA, /* 126 Start of capturing bracket */ |
OP_BRAPOS, /* 126 Ditto, with unlimited, possessive repeat */ |
| 1737 |
OP_CBRAPOS, /* 127 Ditto, with unlimited, possessive repeat */ |
OP_CBRA, /* 127 Start of capturing bracket */ |
| 1738 |
OP_COND, /* 128 Conditional group */ |
OP_CBRAPOS, /* 128 Ditto, with unlimited, possessive repeat */ |
| 1739 |
|
OP_COND, /* 129 Conditional group */ |
| 1740 |
|
|
| 1741 |
/* These five must follow the previous five, in the same order. There's a |
/* These five must follow the previous five, in the same order. There's a |
| 1742 |
check for >= SBRA to distinguish the two sets. */ |
check for >= SBRA to distinguish the two sets. */ |
| 1743 |
|
|
| 1744 |
OP_SBRA, /* 129 Start of non-capturing bracket, check empty */ |
OP_SBRA, /* 130 Start of non-capturing bracket, check empty */ |
| 1745 |
OP_SBRAPOS, /* 130 Ditto, with unlimited, possessive repeat */ |
OP_SBRAPOS, /* 131 Ditto, with unlimited, possessive repeat */ |
| 1746 |
OP_SCBRA, /* 131 Start of capturing bracket, check empty */ |
OP_SCBRA, /* 132 Start of capturing bracket, check empty */ |
| 1747 |
OP_SCBRAPOS, /* 132 Ditto, with unlimited, possessive repeat */ |
OP_SCBRAPOS, /* 133 Ditto, with unlimited, possessive repeat */ |
| 1748 |
OP_SCOND, /* 133 Conditional group, check empty */ |
OP_SCOND, /* 134 Conditional group, check empty */ |
| 1749 |
|
|
| 1750 |
/* The next two pairs must (respectively) be kept together. */ |
/* The next two pairs must (respectively) be kept together. */ |
| 1751 |
|
|
| 1752 |
OP_CREF, /* 134 Used to hold a capture number as condition */ |
OP_CREF, /* 135 Used to hold a capture number as condition */ |
| 1753 |
OP_NCREF, /* 135 Same, but generated by a name reference*/ |
OP_NCREF, /* 136 Same, but generated by a name reference*/ |
| 1754 |
OP_RREF, /* 136 Used to hold a recursion number as condition */ |
OP_RREF, /* 137 Used to hold a recursion number as condition */ |
| 1755 |
OP_NRREF, /* 137 Same, but generated by a name reference*/ |
OP_NRREF, /* 138 Same, but generated by a name reference*/ |
| 1756 |
OP_DEF, /* 138 The DEFINE condition */ |
OP_DEF, /* 139 The DEFINE condition */ |
| 1757 |
|
|
| 1758 |
OP_BRAZERO, /* 139 These two must remain together and in this */ |
OP_BRAZERO, /* 140 These two must remain together and in this */ |
| 1759 |
OP_BRAMINZERO, /* 140 order. */ |
OP_BRAMINZERO, /* 141 order. */ |
| 1760 |
OP_BRAPOSZERO, /* 141 */ |
OP_BRAPOSZERO, /* 142 */ |
| 1761 |
|
|
| 1762 |
/* These are backtracking control verbs */ |
/* These are backtracking control verbs */ |
| 1763 |
|
|
| 1764 |
OP_MARK, /* 142 always has an argument */ |
OP_MARK, /* 143 always has an argument */ |
| 1765 |
OP_PRUNE, /* 143 */ |
OP_PRUNE, /* 144 */ |
| 1766 |
OP_PRUNE_ARG, /* 144 same, but with argument */ |
OP_PRUNE_ARG, /* 145 same, but with argument */ |
| 1767 |
OP_SKIP, /* 145 */ |
OP_SKIP, /* 146 */ |
| 1768 |
OP_SKIP_ARG, /* 146 same, but with argument */ |
OP_SKIP_ARG, /* 147 same, but with argument */ |
| 1769 |
OP_THEN, /* 147 */ |
OP_THEN, /* 148 */ |
| 1770 |
OP_THEN_ARG, /* 148 same, but with argument */ |
OP_THEN_ARG, /* 149 same, but with argument */ |
| 1771 |
OP_COMMIT, /* 149 */ |
OP_COMMIT, /* 150 */ |
| 1772 |
|
|
| 1773 |
/* These are forced failure and success verbs */ |
/* These are forced failure and success verbs */ |
| 1774 |
|
|
| 1775 |
OP_FAIL, /* 150 */ |
OP_FAIL, /* 151 */ |
| 1776 |
OP_ACCEPT, /* 151 */ |
OP_ACCEPT, /* 152 */ |
| 1777 |
OP_CLOSE, /* 152 Used before OP_ACCEPT to close open captures */ |
OP_ASSERT_ACCEPT, /* 153 Used inside assertions */ |
| 1778 |
|
OP_CLOSE, /* 154 Used before OP_ACCEPT to close open captures */ |
| 1779 |
|
|
| 1780 |
/* This is used to skip a subpattern with a {0} quantifier */ |
/* This is used to skip a subpattern with a {0} quantifier */ |
| 1781 |
|
|
| 1782 |
OP_SKIPZERO, /* 153 */ |
OP_SKIPZERO, /* 155 */ |
| 1783 |
|
|
| 1784 |
/* This is not an opcode, but is used to check that tables indexed by opcode |
/* This is not an opcode, but is used to check that tables indexed by opcode |
| 1785 |
are the correct length, in order to catch updating errors - there have been |
are the correct length, in order to catch updating errors - there have been |
| 1822 |
"class", "nclass", "xclass", "Ref", "Refi", \ |
"class", "nclass", "xclass", "Ref", "Refi", \ |
| 1823 |
"Recurse", "Callout", \ |
"Recurse", "Callout", \ |
| 1824 |
"Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \ |
"Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \ |
| 1825 |
"Assert", "Assert not", "AssertB", "AssertB not", "Reverse", \ |
"Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \ |
| 1826 |
"Once", \ |
"Once", "Once_NC", \ |
| 1827 |
"Bra", "BraPos", "CBra", "CBraPos", \ |
"Bra", "BraPos", "CBra", "CBraPos", \ |
| 1828 |
"Cond", \ |
"Cond", \ |
| 1829 |
"SBra", "SBraPos", "SCBra", "SCBraPos", \ |
"SBra", "SBraPos", "SCBra", "SCBraPos", \ |
| 1831 |
"Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \ |
"Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \ |
| 1832 |
"Brazero", "Braminzero", "Braposzero", \ |
"Brazero", "Braminzero", "Braposzero", \ |
| 1833 |
"*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \ |
"*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \ |
| 1834 |
"*THEN", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT", \ |
"*THEN", "*THEN", "*COMMIT", "*FAIL", \ |
| 1835 |
|
"*ACCEPT", "*ASSERT_ACCEPT", \ |
| 1836 |
"Close", "Skip zero" |
"Close", "Skip zero" |
| 1837 |
|
|
| 1838 |
|
|
| 1860 |
2, /* noti */ \ |
2, /* noti */ \ |
| 1861 |
/* Positive single-char repeats ** These are */ \ |
/* Positive single-char repeats ** These are */ \ |
| 1862 |
2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ |
2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \ |
| 1863 |
4, 4, 4, /* upto, minupto, exact ** mode */ \ |
2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \ |
| 1864 |
2, 2, 2, 4, /* *+, ++, ?+, upto+ */ \ |
2+IMM2_SIZE, /* exact */ \ |
| 1865 |
|
2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \ |
| 1866 |
2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \ |
2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \ |
| 1867 |
4, 4, 4, /* upto I, minupto I, exact I */ \ |
2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \ |
| 1868 |
2, 2, 2, 4, /* *+I, ++I, ?+I, upto+I */ \ |
2+IMM2_SIZE, /* exact I */ \ |
| 1869 |
|
2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \ |
| 1870 |
/* Negative single-char repeats - only for chars < 256 */ \ |
/* Negative single-char repeats - only for chars < 256 */ \ |
| 1871 |
2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ |
2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \ |
| 1872 |
4, 4, 4, /* NOT upto, minupto, exact */ \ |
2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \ |
| 1873 |
2, 2, 2, 4, /* Possessive NOT *, +, ?, upto */ \ |
2+IMM2_SIZE, /* NOT exact */ \ |
| 1874 |
|
2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \ |
| 1875 |
2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \ |
2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \ |
| 1876 |
4, 4, 4, /* NOT upto I, minupto I, exact I */ \ |
2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \ |
| 1877 |
2, 2, 2, 4, /* Possessive NOT *I, +I, ?I, upto I */ \ |
2+IMM2_SIZE, /* NOT exact I */ \ |
| 1878 |
|
2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \ |
| 1879 |
/* Positive type repeats */ \ |
/* Positive type repeats */ \ |
| 1880 |
2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ |
2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \ |
| 1881 |
4, 4, 4, /* Type upto, minupto, exact */ \ |
2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \ |
| 1882 |
2, 2, 2, 4, /* Possessive *+, ++, ?+, upto+ */ \ |
2+IMM2_SIZE, /* Type exact */ \ |
| 1883 |
|
2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \ |
| 1884 |
/* Character class & ref repeats */ \ |
/* Character class & ref repeats */ \ |
| 1885 |
1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ |
1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \ |
| 1886 |
5, 5, /* CRRANGE, CRMINRANGE */ \ |
1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \ |
| 1887 |
33, /* CLASS */ \ |
1+(32/sizeof(pcre_uchar)), /* CLASS */ \ |
| 1888 |
33, /* NCLASS */ \ |
1+(32/sizeof(pcre_uchar)), /* NCLASS */ \ |
| 1889 |
0, /* XCLASS - variable length */ \ |
0, /* XCLASS - variable length */ \ |
| 1890 |
3, /* REF */ \ |
1+IMM2_SIZE, /* REF */ \ |
| 1891 |
3, /* REFI */ \ |
1+IMM2_SIZE, /* REFI */ \ |
| 1892 |
1+LINK_SIZE, /* RECURSE */ \ |
1+LINK_SIZE, /* RECURSE */ \ |
| 1893 |
2+2*LINK_SIZE, /* CALLOUT */ \ |
2+2*LINK_SIZE, /* CALLOUT */ \ |
| 1894 |
1+LINK_SIZE, /* Alt */ \ |
1+LINK_SIZE, /* Alt */ \ |
| 1896 |
1+LINK_SIZE, /* KetRmax */ \ |
1+LINK_SIZE, /* KetRmax */ \ |
| 1897 |
1+LINK_SIZE, /* KetRmin */ \ |
1+LINK_SIZE, /* KetRmin */ \ |
| 1898 |
1+LINK_SIZE, /* KetRpos */ \ |
1+LINK_SIZE, /* KetRpos */ \ |
| 1899 |
|
1+LINK_SIZE, /* Reverse */ \ |
| 1900 |
1+LINK_SIZE, /* Assert */ \ |
1+LINK_SIZE, /* Assert */ \ |
| 1901 |
1+LINK_SIZE, /* Assert not */ \ |
1+LINK_SIZE, /* Assert not */ \ |
| 1902 |
1+LINK_SIZE, /* Assert behind */ \ |
1+LINK_SIZE, /* Assert behind */ \ |
| 1903 |
1+LINK_SIZE, /* Assert behind not */ \ |
1+LINK_SIZE, /* Assert behind not */ \ |
|
1+LINK_SIZE, /* Reverse */ \ |
|
| 1904 |
1+LINK_SIZE, /* ONCE */ \ |
1+LINK_SIZE, /* ONCE */ \ |
| 1905 |
|
1+LINK_SIZE, /* ONCE_NC */ \ |
| 1906 |
1+LINK_SIZE, /* BRA */ \ |
1+LINK_SIZE, /* BRA */ \ |
| 1907 |
1+LINK_SIZE, /* BRAPOS */ \ |
1+LINK_SIZE, /* BRAPOS */ \ |
| 1908 |
3+LINK_SIZE, /* CBRA */ \ |
1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \ |
| 1909 |
3+LINK_SIZE, /* CBRAPOS */ \ |
1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \ |
| 1910 |
1+LINK_SIZE, /* COND */ \ |
1+LINK_SIZE, /* COND */ \ |
| 1911 |
1+LINK_SIZE, /* SBRA */ \ |
1+LINK_SIZE, /* SBRA */ \ |
| 1912 |
1+LINK_SIZE, /* SBRAPOS */ \ |
1+LINK_SIZE, /* SBRAPOS */ \ |
| 1913 |
3+LINK_SIZE, /* SCBRA */ \ |
1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \ |
| 1914 |
3+LINK_SIZE, /* SCBRAPOS */ \ |
1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \ |
| 1915 |
1+LINK_SIZE, /* SCOND */ \ |
1+LINK_SIZE, /* SCOND */ \ |
| 1916 |
3, 3, /* CREF, NCREF */ \ |
1+IMM2_SIZE, 1+IMM2_SIZE, /* CREF, NCREF */ \ |
| 1917 |
3, 3, /* RREF, NRREF */ \ |
1+IMM2_SIZE, 1+IMM2_SIZE, /* RREF, NRREF */ \ |
| 1918 |
1, /* DEF */ \ |
1, /* DEF */ \ |
| 1919 |
1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ |
1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \ |
| 1920 |
3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ |
3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \ |
| 1921 |
1, 3, /* SKIP, SKIP_ARG */ \ |
1, 3, /* SKIP, SKIP_ARG */ \ |
| 1922 |
1+LINK_SIZE, 3+LINK_SIZE, /* THEN, THEN_ARG */ \ |
1, 3, /* THEN, THEN_ARG */ \ |
| 1923 |
1, 1, 1, 3, 1 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */ \ |
1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \ |
| 1924 |
|
1+IMM2_SIZE, 1 /* CLOSE, SKIPZERO */ |
| 1925 |
|
|
| 1926 |
/* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion" |
/* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion" |
| 1927 |
condition. */ |
condition. */ |
| 1938 |
ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, |
ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, |
| 1939 |
ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, |
ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, |
| 1940 |
ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, |
ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, |
| 1941 |
ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, |
ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69, |
| 1942 |
ERRCOUNT }; |
ERR70, ERR71, ERR72, ERR73, ERR74, ERRCOUNT }; |
| 1943 |
|
|
| 1944 |
/* The real format of the start of the pcre block; the index of names and the |
/* The real format of the start of the pcre block; the index of names and the |
| 1945 |
code vector run on as long as necessary after the end. We store an explicit |
code vector run on as long as necessary after the end. We store an explicit |
| 1958 |
NOTE NOTE NOTE |
NOTE NOTE NOTE |
| 1959 |
*/ |
*/ |
| 1960 |
|
|
| 1961 |
typedef struct real_pcre { |
#ifdef COMPILE_PCRE8 |
| 1962 |
|
#define REAL_PCRE real_pcre |
| 1963 |
|
#else |
| 1964 |
|
#define REAL_PCRE real_pcre16 |
| 1965 |
|
#endif |
| 1966 |
|
|
| 1967 |
|
typedef struct REAL_PCRE { |
| 1968 |
pcre_uint32 magic_number; |
pcre_uint32 magic_number; |
| 1969 |
pcre_uint32 size; /* Total that was malloced */ |
pcre_uint32 size; /* Total that was malloced */ |
| 1970 |
pcre_uint32 options; /* Public options */ |
pcre_uint32 options; /* Public options */ |
| 1972 |
pcre_uint16 dummy1; /* For future use */ |
pcre_uint16 dummy1; /* For future use */ |
| 1973 |
pcre_uint16 top_bracket; |
pcre_uint16 top_bracket; |
| 1974 |
pcre_uint16 top_backref; |
pcre_uint16 top_backref; |
| 1975 |
pcre_uint16 first_byte; |
pcre_uint16 first_char; /* Starting character */ |
| 1976 |
pcre_uint16 req_byte; |
pcre_uint16 req_char; /* This character must be seen */ |
| 1977 |
pcre_uint16 name_table_offset; /* Offset to name table that follows */ |
pcre_uint16 name_table_offset; /* Offset to name table that follows */ |
| 1978 |
pcre_uint16 name_entry_size; /* Size of any name items */ |
pcre_uint16 name_entry_size; /* Size of any name items */ |
| 1979 |
pcre_uint16 name_count; /* Number of name items */ |
pcre_uint16 name_count; /* Number of name items */ |
| 1980 |
pcre_uint16 ref_count; /* Reference count */ |
pcre_uint16 ref_count; /* Reference count */ |
| 1981 |
|
|
| 1982 |
const unsigned char *tables; /* Pointer to tables or NULL for std */ |
const pcre_uint8 *tables; /* Pointer to tables or NULL for std */ |
| 1983 |
const unsigned char *nullpad; /* NULL padding */ |
const pcre_uint8 *nullpad; /* NULL padding */ |
| 1984 |
} real_pcre; |
} REAL_PCRE; |
| 1985 |
|
|
| 1986 |
/* The format of the block used to store data from pcre_study(). The same |
/* The format of the block used to store data from pcre_study(). The same |
| 1987 |
remark (see NOTE above) about extending this structure applies. */ |
remark (see NOTE above) about extending this structure applies. */ |
| 1989 |
typedef struct pcre_study_data { |
typedef struct pcre_study_data { |
| 1990 |
pcre_uint32 size; /* Total that was malloced */ |
pcre_uint32 size; /* Total that was malloced */ |
| 1991 |
pcre_uint32 flags; /* Private flags */ |
pcre_uint32 flags; /* Private flags */ |
| 1992 |
uschar start_bits[32]; /* Starting char bits */ |
pcre_uint8 start_bits[32]; /* Starting char bits */ |
| 1993 |
pcre_uint32 minlength; /* Minimum subject length */ |
pcre_uint32 minlength; /* Minimum subject length */ |
| 1994 |
} pcre_study_data; |
} pcre_study_data; |
| 1995 |
|
|
| 2008 |
doing the compiling, so that they are thread-safe. */ |
doing the compiling, so that they are thread-safe. */ |
| 2009 |
|
|
| 2010 |
typedef struct compile_data { |
typedef struct compile_data { |
| 2011 |
const uschar *lcc; /* Points to lower casing table */ |
const pcre_uint8 *lcc; /* Points to lower casing table */ |
| 2012 |
const uschar *fcc; /* Points to case-flipping table */ |
const pcre_uint8 *fcc; /* Points to case-flipping table */ |
| 2013 |
const uschar *cbits; /* Points to character type table */ |
const pcre_uint8 *cbits; /* Points to character type table */ |
| 2014 |
const uschar *ctypes; /* Points to table of type maps */ |
const pcre_uint8 *ctypes; /* Points to table of type maps */ |
| 2015 |
const uschar *start_workspace;/* The start of working space */ |
const pcre_uchar *start_workspace;/* The start of working space */ |
| 2016 |
const uschar *start_code; /* The start of the compiled code */ |
const pcre_uchar *start_code; /* The start of the compiled code */ |
| 2017 |
const uschar *start_pattern; /* The start of the pattern */ |
const pcre_uchar *start_pattern; /* The start of the pattern */ |
| 2018 |
const uschar *end_pattern; /* The end of the pattern */ |
const pcre_uchar *end_pattern; /* The end of the pattern */ |
| 2019 |
open_capitem *open_caps; /* Chain of open capture items */ |
open_capitem *open_caps; /* Chain of open capture items */ |
| 2020 |
uschar *hwm; /* High watermark of workspace */ |
pcre_uchar *hwm; /* High watermark of workspace */ |
| 2021 |
uschar *name_table; /* The name/number table */ |
pcre_uchar *name_table; /* The name/number table */ |
| 2022 |
int names_found; /* Number of entries so far */ |
int names_found; /* Number of entries so far */ |
| 2023 |
int name_entry_size; /* Size of each entry */ |
int name_entry_size; /* Size of each entry */ |
| 2024 |
int bracount; /* Count of capturing parens as we compile */ |
int workspace_size; /* Size of workspace */ |
| 2025 |
int final_bracount; /* Saved value after first pass */ |
int bracount; /* Count of capturing parens as we compile */ |
| 2026 |
int top_backref; /* Maximum back reference */ |
int final_bracount; /* Saved value after first pass */ |
| 2027 |
unsigned int backref_map; /* Bitmap of low back refs */ |
int top_backref; /* Maximum back reference */ |
| 2028 |
int external_options; /* External (initial) options */ |
unsigned int backref_map; /* Bitmap of low back refs */ |
| 2029 |
int external_flags; /* External flag bits to be set */ |
int assert_depth; /* Depth of nested assertions */ |
| 2030 |
int req_varyopt; /* "After variable item" flag for reqbyte */ |
int external_options; /* External (initial) options */ |
| 2031 |
BOOL had_accept; /* (*ACCEPT) encountered */ |
int external_flags; /* External flag bits to be set */ |
| 2032 |
BOOL check_lookbehind; /* Lookbehinds need later checking */ |
int req_varyopt; /* "After variable item" flag for reqbyte */ |
| 2033 |
int nltype; /* Newline type */ |
BOOL had_accept; /* (*ACCEPT) encountered */ |
| 2034 |
int nllen; /* Newline string length */ |
BOOL check_lookbehind; /* Lookbehinds need later checking */ |
| 2035 |
uschar nl[4]; /* Newline string when fixed length */ |
int nltype; /* Newline type */ |
| 2036 |
|
int nllen; /* Newline string length */ |
| 2037 |
|
pcre_uchar nl[4]; /* Newline string when fixed length */ |
| 2038 |
} compile_data; |
} compile_data; |
| 2039 |
|
|
| 2040 |
/* Structure for maintaining a chain of pointers to the currently incomplete |
/* Structure for maintaining a chain of pointers to the currently incomplete |
| 2041 |
branches, for testing for left recursion. */ |
branches, for testing for left recursion while compiling. */ |
| 2042 |
|
|
| 2043 |
typedef struct branch_chain { |
typedef struct branch_chain { |
| 2044 |
struct branch_chain *outer; |
struct branch_chain *outer; |
| 2045 |
uschar *current_branch; |
pcre_uchar *current_branch; |
| 2046 |
} branch_chain; |
} branch_chain; |
| 2047 |
|
|
| 2048 |
/* Structure for items in a linked list that represents an explicit recursive |
/* Structure for items in a linked list that represents an explicit recursive |
| 2049 |
call within the pattern. */ |
call within the pattern; used by pcre_exec(). */ |
| 2050 |
|
|
| 2051 |
typedef struct recursion_info { |
typedef struct recursion_info { |
| 2052 |
struct recursion_info *prevrec; /* Previous recursion record (or NULL) */ |
struct recursion_info *prevrec; /* Previous recursion record (or NULL) */ |
| 2053 |
int group_num; /* Number of group that was called */ |
int group_num; /* Number of group that was called */ |
| 2054 |
const uschar *after_call; /* "Return value": points after the call in the expr */ |
int *offset_save; /* Pointer to start of saved offsets */ |
| 2055 |
int *offset_save; /* Pointer to start of saved offsets */ |
int saved_max; /* Number of saved offsets */ |
| 2056 |
int saved_max; /* Number of saved offsets */ |
PCRE_PUCHAR subject_position; /* Position at start of recursion */ |
|
int save_offset_top; /* Current value of offset_top */ |
|
| 2057 |
} recursion_info; |
} recursion_info; |
| 2058 |
|
|
| 2059 |
|
/* A similar structure for pcre_dfa_exec(). */ |
| 2060 |
|
|
| 2061 |
|
typedef struct dfa_recursion_info { |
| 2062 |
|
struct dfa_recursion_info *prevrec; |
| 2063 |
|
int group_num; |
| 2064 |
|
PCRE_PUCHAR subject_position; |
| 2065 |
|
} dfa_recursion_info; |
| 2066 |
|
|
| 2067 |
/* Structure for building a chain of data for holding the values of the subject |
/* Structure for building a chain of data for holding the values of the subject |
| 2068 |
pointer at the start of each subpattern, so as to detect when an empty string |
pointer at the start of each subpattern, so as to detect when an empty string |
| 2069 |
has been matched by a subpattern - to break infinite loops. */ |
has been matched by a subpattern - to break infinite loops; used by |
| 2070 |
|
pcre_exec(). */ |
| 2071 |
|
|
| 2072 |
typedef struct eptrblock { |
typedef struct eptrblock { |
| 2073 |
struct eptrblock *epb_prev; |
struct eptrblock *epb_prev; |
| 2074 |
USPTR epb_saved_eptr; |
PCRE_PUCHAR epb_saved_eptr; |
| 2075 |
} eptrblock; |
} eptrblock; |
| 2076 |
|
|
| 2077 |
|
|
| 2082 |
unsigned long int match_call_count; /* As it says */ |
unsigned long int match_call_count; /* As it says */ |
| 2083 |
unsigned long int match_limit; /* As it says */ |
unsigned long int match_limit; /* As it says */ |
| 2084 |
unsigned long int match_limit_recursion; /* As it says */ |
unsigned long int match_limit_recursion; /* As it says */ |
| 2085 |
int *offset_vector; /* Offset vector */ |
int *offset_vector; /* Offset vector */ |
| 2086 |
int offset_end; /* One past the end */ |
int offset_end; /* One past the end */ |
| 2087 |
int offset_max; /* The maximum usable for return data */ |
int offset_max; /* The maximum usable for return data */ |
| 2088 |
int nltype; /* Newline type */ |
int nltype; /* Newline type */ |
| 2089 |
int nllen; /* Newline string length */ |
int nllen; /* Newline string length */ |
| 2090 |
int name_count; /* Number of names in name table */ |
int name_count; /* Number of names in name table */ |
| 2091 |
int name_entry_size; /* Size of entry in names table */ |
int name_entry_size; /* Size of entry in names table */ |
| 2092 |
uschar *name_table; /* Table of names */ |
pcre_uchar *name_table; /* Table of names */ |
| 2093 |
uschar nl[4]; /* Newline string when fixed */ |
pcre_uchar nl[4]; /* Newline string when fixed */ |
| 2094 |
const uschar *lcc; /* Points to lower casing table */ |
const pcre_uint8 *lcc; /* Points to lower casing table */ |
| 2095 |
const uschar *ctypes; /* Points to table of type maps */ |
const pcre_uint8 *fcc; /* Points to case-flipping table */ |
| 2096 |
BOOL offset_overflow; /* Set if too many extractions */ |
const pcre_uint8 *ctypes; /* Points to table of type maps */ |
| 2097 |
BOOL notbol; /* NOTBOL flag */ |
BOOL offset_overflow; /* Set if too many extractions */ |
| 2098 |
BOOL noteol; /* NOTEOL flag */ |
BOOL notbol; /* NOTBOL flag */ |
| 2099 |
BOOL utf8; /* UTF8 flag */ |
BOOL noteol; /* NOTEOL flag */ |
| 2100 |
BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */ |
BOOL utf; /* UTF-8 / UTF-16 flag */ |
| 2101 |
BOOL use_ucp; /* PCRE_UCP flag */ |
BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */ |
| 2102 |
BOOL endonly; /* Dollar not before final \n */ |
BOOL use_ucp; /* PCRE_UCP flag */ |
| 2103 |
BOOL notempty; /* Empty string match not wanted */ |
BOOL endonly; /* Dollar not before final \n */ |
| 2104 |
BOOL notempty_atstart; /* Empty string match at start not wanted */ |
BOOL notempty; /* Empty string match not wanted */ |
| 2105 |
BOOL hitend; /* Hit the end of the subject at some point */ |
BOOL notempty_atstart; /* Empty string match at start not wanted */ |
| 2106 |
BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */ |
BOOL hitend; /* Hit the end of the subject at some point */ |
| 2107 |
const uschar *start_code; /* For use when recursing */ |
BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */ |
| 2108 |
USPTR start_subject; /* Start of the subject string */ |
BOOL hasthen; /* Pattern contains (*THEN) */ |
| 2109 |
USPTR end_subject; /* End of the subject string */ |
BOOL ignore_skip_arg; /* For re-run when SKIP name not found */ |
| 2110 |
USPTR start_match_ptr; /* Start of matched string */ |
const pcre_uchar *start_code; /* For use when recursing */ |
| 2111 |
USPTR end_match_ptr; /* Subject position at end match */ |
PCRE_PUCHAR start_subject; /* Start of the subject string */ |
| 2112 |
USPTR start_used_ptr; /* Earliest consulted character */ |
PCRE_PUCHAR end_subject; /* End of the subject string */ |
| 2113 |
int partial; /* PARTIAL options */ |
PCRE_PUCHAR start_match_ptr; /* Start of matched string */ |
| 2114 |
int end_offset_top; /* Highwater mark at end of match */ |
PCRE_PUCHAR end_match_ptr; /* Subject position at end match */ |
| 2115 |
int capture_last; /* Most recent capture number */ |
PCRE_PUCHAR start_used_ptr; /* Earliest consulted character */ |
| 2116 |
int start_offset; /* The start offset value */ |
int partial; /* PARTIAL options */ |
| 2117 |
int match_function_type; /* Set for certain special calls of MATCH() */ |
int end_offset_top; /* Highwater mark at end of match */ |
| 2118 |
eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */ |
int capture_last; /* Most recent capture number */ |
| 2119 |
int eptrn; /* Next free eptrblock */ |
int start_offset; /* The start offset value */ |
| 2120 |
recursion_info *recursive; /* Linked list of recursion data */ |
int match_function_type; /* Set for certain special calls of MATCH() */ |
| 2121 |
void *callout_data; /* To pass back to callouts */ |
eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */ |
| 2122 |
const uschar *mark; /* Mark pointer to pass back */ |
int eptrn; /* Next free eptrblock */ |
| 2123 |
|
recursion_info *recursive; /* Linked list of recursion data */ |
| 2124 |
|
void *callout_data; /* To pass back to callouts */ |
| 2125 |
|
const pcre_uchar *mark; /* Mark pointer to pass back on success */ |
| 2126 |
|
const pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */ |
| 2127 |
|
const pcre_uchar *once_target; /* Where to back up to for atomic groups */ |
| 2128 |
} match_data; |
} match_data; |
| 2129 |
|
|
| 2130 |
/* A similar structure is used for the same purpose by the DFA matching |
/* A similar structure is used for the same purpose by the DFA matching |
| 2131 |
functions. */ |
functions. */ |
| 2132 |
|
|
| 2133 |
typedef struct dfa_match_data { |
typedef struct dfa_match_data { |
| 2134 |
const uschar *start_code; /* Start of the compiled pattern */ |
const pcre_uchar *start_code; /* Start of the compiled pattern */ |
| 2135 |
const uschar *start_subject; /* Start of the subject string */ |
const pcre_uchar *start_subject ; /* Start of the subject string */ |
| 2136 |
const uschar *end_subject; /* End of subject string */ |
const pcre_uchar *end_subject; /* End of subject string */ |
| 2137 |
const uschar *start_used_ptr; /* Earliest consulted character */ |
const pcre_uchar *start_used_ptr; /* Earliest consulted character */ |
| 2138 |
const uschar *tables; /* Character tables */ |
const pcre_uint8 *tables; /* Character tables */ |
| 2139 |
int start_offset; /* The start offset value */ |
int start_offset; /* The start offset value */ |
| 2140 |
int moptions; /* Match options */ |
int moptions; /* Match options */ |
| 2141 |
int poptions; /* Pattern options */ |
int poptions; /* Pattern options */ |
| 2142 |
int nltype; /* Newline type */ |
int nltype; /* Newline type */ |
| 2143 |
int nllen; /* Newline string length */ |
int nllen; /* Newline string length */ |
| 2144 |
uschar nl[4]; /* Newline string when fixed */ |
pcre_uchar nl[4]; /* Newline string when fixed */ |
| 2145 |
void *callout_data; /* To pass back to callouts */ |
void *callout_data; /* To pass back to callouts */ |
| 2146 |
|
dfa_recursion_info *recursive; /* Linked list of recursion data */ |
| 2147 |
} dfa_match_data; |
} dfa_match_data; |
| 2148 |
|
|
| 2149 |
/* Bit definitions for entries in the pcre_ctypes table. */ |
/* Bit definitions for entries in the pcre_ctypes table. */ |
| 2179 |
#define ctypes_offset (cbits_offset + cbit_length) |
#define ctypes_offset (cbits_offset + cbit_length) |
| 2180 |
#define tables_length (ctypes_offset + 256) |
#define tables_length (ctypes_offset + 256) |
| 2181 |
|
|
| 2182 |
|
/* Internal function prefix */ |
| 2183 |
|
|
| 2184 |
|
#ifdef COMPILE_PCRE8 |
| 2185 |
|
#define PUBL(name) pcre_##name |
| 2186 |
|
#define PRIV(name) _pcre_##name |
| 2187 |
|
#else |
| 2188 |
|
#ifdef COMPILE_PCRE16 |
| 2189 |
|
#define PUBL(name) pcre16_##name |
| 2190 |
|
#define PRIV(name) _pcre16_##name |
| 2191 |
|
#else |
| 2192 |
|
#error Unsupported compiling mode |
| 2193 |
|
#endif /* COMPILE_PCRE16 */ |
| 2194 |
|
#endif /* COMPILE_PCRE8 */ |
| 2195 |
|
|
| 2196 |
/* Layout of the UCP type table that translates property names into types and |
/* Layout of the UCP type table that translates property names into types and |
| 2197 |
codes. Each entry used to point directly to a name, but to reduce the number of |
codes. Each entry used to point directly to a name, but to reduce the number of |
| 2198 |
relocations in shared libraries, it now has an offset into a single string |
relocations in shared libraries, it now has an offset into a single string |
| 2210 |
but are not part of the PCRE public API. The data for these tables is in the |
but are not part of the PCRE public API. The data for these tables is in the |
| 2211 |
pcre_tables.c module. */ |
pcre_tables.c module. */ |
| 2212 |
|
|
| 2213 |
extern const int _pcre_utf8_table1[]; |
#ifdef COMPILE_PCRE8 |
| 2214 |
extern const int _pcre_utf8_table2[]; |
|
| 2215 |
extern const int _pcre_utf8_table3[]; |
extern const int PRIV(utf8_table1)[]; |
| 2216 |
extern const uschar _pcre_utf8_table4[]; |
extern const int PRIV(utf8_table1_size); |
| 2217 |
|
extern const int PRIV(utf8_table2)[]; |
| 2218 |
|
extern const int PRIV(utf8_table3)[]; |
| 2219 |
|
extern const pcre_uint8 PRIV(utf8_table4)[]; |
| 2220 |
|
|
| 2221 |
extern const int _pcre_utf8_table1_size; |
#endif /* COMPILE_PCRE8 */ |
| 2222 |
|
|
| 2223 |
extern const char _pcre_utt_names[]; |
extern const char PRIV(utt_names)[]; |
| 2224 |
extern const ucp_type_table _pcre_utt[]; |
extern const ucp_type_table PRIV(utt)[]; |
| 2225 |
extern const int _pcre_utt_size; |
extern const int PRIV(utt_size); |
| 2226 |
|
|
| 2227 |
extern const uschar _pcre_default_tables[]; |
extern const pcre_uint8 PRIV(default_tables)[]; |
| 2228 |
|
|
| 2229 |
extern const uschar _pcre_OP_lengths[]; |
extern const pcre_uint8 PRIV(OP_lengths)[]; |
| 2230 |
|
|
| 2231 |
|
|
| 2232 |
/* Internal shared functions. These are functions that are used by more than |
/* Internal shared functions. These are functions that are used by more than |
| 2233 |
one of the exported public functions. They have to be "external" in the C |
one of the exported public functions. They have to be "external" in the C |
| 2234 |
sense, but are not part of the PCRE public API. */ |
sense, but are not part of the PCRE public API. */ |
| 2235 |
|
|
| 2236 |
extern const uschar *_pcre_find_bracket(const uschar *, BOOL, int); |
/* String comparison functions. */ |
| 2237 |
extern BOOL _pcre_is_newline(USPTR, int, USPTR, int *, BOOL); |
#ifdef COMPILE_PCRE8 |
|
extern int _pcre_ord2utf8(int, uschar *); |
|
|
extern real_pcre *_pcre_try_flipped(const real_pcre *, real_pcre *, |
|
|
const pcre_study_data *, pcre_study_data *); |
|
|
extern int _pcre_valid_utf8(USPTR, int, int *); |
|
|
extern BOOL _pcre_was_newline(USPTR, int, USPTR, int *, BOOL); |
|
|
extern BOOL _pcre_xclass(int, const uschar *); |
|
| 2238 |
|
|
| 2239 |
|
#define STRCMP_UC_UC(str1, str2) \ |
| 2240 |
|
strcmp((char *)(str1), (char *)(str2)) |
| 2241 |
|
#define STRCMP_UC_C8(str1, str2) \ |
| 2242 |
|
strcmp((char *)(str1), (str2)) |
| 2243 |
|
#define STRNCMP_UC_UC(str1, str2, num) \ |
| 2244 |
|
strncmp((char *)(str1), (char *)(str2), (num)) |
| 2245 |
|
#define STRNCMP_UC_C8(str1, str2, num) \ |
| 2246 |
|
strncmp((char *)(str1), (str2), (num)) |
| 2247 |
|
#define STRLEN_UC(str) strlen((const char *)str) |
| 2248 |
|
|
| 2249 |
|
#else |
| 2250 |
|
|
| 2251 |
|
extern int PRIV(strcmp_uc_uc)(const pcre_uchar *, |
| 2252 |
|
const pcre_uchar *); |
| 2253 |
|
extern int PRIV(strcmp_uc_c8)(const pcre_uchar *, |
| 2254 |
|
const char *); |
| 2255 |
|
extern int PRIV(strncmp_uc_uc)(const pcre_uchar *, |
| 2256 |
|
const pcre_uchar *, unsigned int num); |
| 2257 |
|
extern int PRIV(strncmp_uc_c8)(const pcre_uchar *, |
| 2258 |
|
const char *, unsigned int num); |
| 2259 |
|
extern unsigned int PRIV(strlen_uc)(const pcre_uchar *str); |
| 2260 |
|
|
| 2261 |
|
#define STRCMP_UC_UC(str1, str2) \ |
| 2262 |
|
PRIV(strcmp_uc_uc)((str1), (str2)) |
| 2263 |
|
#define STRCMP_UC_C8(str1, str2) \ |
| 2264 |
|
PRIV(strcmp_uc_c8)((str1), (str2)) |
| 2265 |
|
#define STRNCMP_UC_UC(str1, str2, num) \ |
| 2266 |
|
PRIV(strncmp_uc_uc)((str1), (str2), (num)) |
| 2267 |
|
#define STRNCMP_UC_C8(str1, str2, num) \ |
| 2268 |
|
PRIV(strncmp_uc_c8)((str1), (str2), (num)) |
| 2269 |
|
#define STRLEN_UC(str) PRIV(strlen_uc)(str) |
| 2270 |
|
|
| 2271 |
|
#endif /* COMPILE_PCRE8 */ |
| 2272 |
|
|
| 2273 |
|
extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int); |
| 2274 |
|
extern BOOL PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR, |
| 2275 |
|
int *, BOOL); |
| 2276 |
|
extern int PRIV(ord2utf)(pcre_uint32, pcre_uchar *); |
| 2277 |
|
extern int PRIV(valid_utf)(PCRE_PUCHAR, int, int *); |
| 2278 |
|
extern BOOL PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR, |
| 2279 |
|
int *, BOOL); |
| 2280 |
|
extern BOOL PRIV(xclass)(int, const pcre_uchar *, BOOL); |
| 2281 |
|
|
| 2282 |
|
#ifdef SUPPORT_JIT |
| 2283 |
|
extern void PRIV(jit_compile)(const REAL_PCRE *, PUBL(extra) *); |
| 2284 |
|
extern int PRIV(jit_exec)(const REAL_PCRE *, void *, |
| 2285 |
|
const pcre_uchar *, int, int, int, int, int *, int); |
| 2286 |
|
extern void PRIV(jit_free)(void *); |
| 2287 |
|
extern int PRIV(jit_get_size)(void *); |
| 2288 |
|
#endif |
| 2289 |
|
|
| 2290 |
/* Unicode character database (UCD) */ |
/* Unicode character database (UCD) */ |
| 2291 |
|
|
| 2292 |
typedef struct { |
typedef struct { |
| 2293 |
uschar script; |
pcre_uint8 script; |
| 2294 |
uschar chartype; |
pcre_uint8 chartype; |
| 2295 |
pcre_int32 other_case; |
pcre_int32 other_case; |
| 2296 |
} ucd_record; |
} ucd_record; |
| 2297 |
|
|
| 2298 |
extern const ucd_record _pcre_ucd_records[]; |
extern const ucd_record PRIV(ucd_records)[]; |
| 2299 |
extern const uschar _pcre_ucd_stage1[]; |
extern const pcre_uint8 PRIV(ucd_stage1)[]; |
| 2300 |
extern const pcre_uint16 _pcre_ucd_stage2[]; |
extern const pcre_uint16 PRIV(ucd_stage2)[]; |
| 2301 |
extern const int _pcre_ucp_gentype[]; |
extern const int PRIV(ucp_gentype)[]; |
| 2302 |
|
#ifdef SUPPORT_JIT |
| 2303 |
|
extern const int PRIV(ucp_typerange)[]; |
| 2304 |
|
#endif |
| 2305 |
|
|
| 2306 |
|
#ifdef SUPPORT_UCP |
| 2307 |
/* UCD access macros */ |
/* UCD access macros */ |
| 2308 |
|
|
| 2309 |
#define UCD_BLOCK_SIZE 128 |
#define UCD_BLOCK_SIZE 128 |
| 2310 |
#define GET_UCD(ch) (_pcre_ucd_records + \ |
#define GET_UCD(ch) (PRIV(ucd_records) + \ |
| 2311 |
_pcre_ucd_stage2[_pcre_ucd_stage1[(ch) / UCD_BLOCK_SIZE] * \ |
PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \ |
| 2312 |
UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE]) |
UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE]) |
| 2313 |
|
|
| 2314 |
#define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype |
#define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype |
| 2315 |
#define UCD_SCRIPT(ch) GET_UCD(ch)->script |
#define UCD_SCRIPT(ch) GET_UCD(ch)->script |
| 2316 |
#define UCD_CATEGORY(ch) _pcre_ucp_gentype[UCD_CHARTYPE(ch)] |
#define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)] |
| 2317 |
#define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case) |
#define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case) |
| 2318 |
|
|
| 2319 |
|
#endif /* SUPPORT_UCP */ |
| 2320 |
|
|
| 2321 |
#endif |
#endif |
| 2322 |
|
|
| 2323 |
/* End of pcre_internal.h */ |
/* End of pcre_internal.h */ |