1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960
|
// Scintilla source code edit control
/** @file RESearch.cxx
** Regular expression search library.
**/
/*
* regex - Regular expression pattern matching and replacement
*
* By: Ozan S. Yigit (oz)
* Dept. of Computer Science
* York University
*
* Original code available from http://www.cs.yorku.ca/~oz/
* Translation to C++ by Neil Hodgson neilh@scintilla.org
* Removed all use of register.
* Converted to modern function prototypes.
* Put all global/static variables into an object so this code can be
* used from multiple threads, etc.
* Some extensions by Philippe Lhoste PhiLho(a)GMX.net
* '?' extensions by Michael Mullin masmullin@gmail.com
*
* These routines are the PUBLIC DOMAIN equivalents of regex
* routines as found in 4.nBSD UN*X, with minor extensions.
*
* These routines are derived from various implementations found
* in software tools books, and Conroy's grep. They are NOT derived
* from licensed/restricted software.
* For more interesting/academic/complicated implementations,
* see Henry Spencer's regexp routines, or GNU Emacs pattern
* matching module.
*
* Modification history removed.
*
* Interfaces:
* RESearch::Compile: compile a regular expression into a NFA.
*
* const char *RESearch::Compile(const char *pattern, int length,
* bool caseSensitive, bool posix)
*
* Returns a short error string if they fail.
*
* RESearch::Execute: execute the NFA to match a pattern.
*
* int RESearch::Execute(characterIndexer &ci, int lp, int endp)
*
* re_fail: failure routine for RESearch::Execute. (no longer used)
*
* void re_fail(char *msg, char op)
*
* Regular Expressions:
*
* [1] char matches itself, unless it is a special
* character (metachar): . \ [ ] * + ? ^ $
* and ( ) if posix option.
*
* [2] . matches any character.
*
* [3] \ matches the character following it, except:
* - \a, \b, \f, \n, \r, \t, \v match the corresponding C
* escape char, respectively BEL, BS, FF, LF, CR, TAB and VT;
* Note that \r and \n are never matched because Scintilla
* regex searches are made line per line
* (stripped of end-of-line chars).
* - if not in posix mode, when followed by a
* left or right round bracket (see [8]);
* - when followed by a digit 1 to 9 (see [9]);
* - when followed by a left or right angle bracket
* (see [10]);
* - when followed by d, D, s, S, w or W (see [11]);
* - when followed by x and two hexa digits (see [12].
* Backslash is used as an escape character for all
* other meta-characters, and itself.
*
* [4] [set] matches one of the characters in the set.
* If the first character in the set is "^",
* it matches the characters NOT in the set, i.e.
* complements the set. A shorthand S-E (start dash end)
* is used to specify a set of characters S up to
* E, inclusive. S and E must be characters, otherwise
* the dash is taken literally (eg. in expression [\d-a]).
* The special characters "]" and "-" have no special
* meaning if they appear as the first chars in the set.
* To include both, put - first: [-]A-Z]
* (or just backslash them).
* examples: match:
*
* [-]|] matches these 3 chars,
*
* []-|] matches from ] to | chars
*
* [a-z] any lowercase alpha
*
* [^-]] any char except - and ]
*
* [^A-Z] any char except uppercase
* alpha
*
* [a-zA-Z] any alpha
*
* [5] * any regular expression form [1] to [4]
* (except [8], [9] and [10] forms of [3]),
* followed by closure char (*)
* matches zero or more matches of that form.
*
* [6] + same as [5], except it matches one or more.
*
* [5-6] Both [5] and [6] are greedy (they match as much as possible).
* Unless they are followed by the 'lazy' quantifier (?)
* In which case both [5] and [6] try to match as little as possible
*
* [7] ? same as [5] except it matches zero or one.
*
* [8] a regular expression in the form [1] to [13], enclosed
* as \(form\) (or (form) with posix flag) matches what
* form matches. The enclosure creates a set of tags,
* used for [9] and for pattern substitution.
* The tagged forms are numbered starting from 1.
*
* [9] a \ followed by a digit 1 to 9 matches whatever a
* previously tagged regular expression ([8]) matched.
*
* [10] \< a regular expression starting with a \< construct
* \> and/or ending with a \> construct, restricts the
* pattern matching to the beginning of a word, and/or
* the end of a word. A word is defined to be a character
* string beginning and/or ending with the characters
* A-Z a-z 0-9 and _. Scintilla extends this definition
* by user setting. The word must also be preceded and/or
* followed by any character outside those mentioned.
*
* [11] \l a backslash followed by d, D, s, S, w or W,
* becomes a character class (both inside and
* outside sets []).
* d: decimal digits
* D: any char except decimal digits
* s: whitespace (space, \t \n \r \f \v)
* S: any char except whitespace (see above)
* w: alphanumeric & underscore (changed by user setting)
* W: any char except alphanumeric & underscore (see above)
*
* [12] \xHH a backslash followed by x and two hexa digits,
* becomes the character whose ASCII code is equal
* to these digits. If not followed by two digits,
* it is 'x' char itself.
*
* [13] a composite regular expression xy where x and y
* are in the form [1] to [12] matches the longest
* match of x followed by a match for y.
*
* [14] ^ a regular expression starting with a ^ character
* $ and/or ending with a $ character, restricts the
* pattern matching to the beginning of the line,
* or the end of line. [anchors] Elsewhere in the
* pattern, ^ and $ are treated as ordinary characters.
*
*
* Acknowledgements:
*
* HCR's Hugh Redelmeier has been most helpful in various
* stages of development. He convinced me to include BOW
* and EOW constructs, originally invented by Rob Pike at
* the University of Toronto.
*
* References:
* Software tools Kernighan & Plauger
* Software tools in Pascal Kernighan & Plauger
* Grep [rsx-11 C dist] David Conroy
* ed - text editor Un*x Programmer's Manual
* Advanced editing on Un*x B. W. Kernighan
* RegExp routines Henry Spencer
*
* Notes:
*
* This implementation uses a bit-set representation for character
* classes for speed and compactness. Each character is represented
* by one bit in a 256-bit block. Thus, CCL always takes a
* constant 32 bytes in the internal nfa, and RESearch::Execute does a single
* bit comparison to locate the character in the set.
*
* Examples:
*
* pattern: foo*.*
* compile: CHR f CHR o CLO CHR o END CLO ANY END END
* matches: fo foo fooo foobar fobar foxx ...
*
* pattern: fo[ob]a[rz]
* compile: CHR f CHR o CCL bitset CHR a CCL bitset END
* matches: fobar fooar fobaz fooaz
*
* pattern: foo\\+
* compile: CHR f CHR o CHR o CHR \ CLO CHR \ END END
* matches: foo\ foo\\ foo\\\ ...
*
* pattern: \(foo\)[1-3]\1 (same as foo[1-3]foo)
* compile: BOT 1 CHR f CHR o CHR o EOT 1 CCL bitset REF 1 END
* matches: foo1foo foo2foo foo3foo
*
* pattern: \(fo.*\)-\1
* compile: BOT 1 CHR f CHR o CLO ANY END EOT 1 CHR - REF 1 END
* matches: foo-foo fo-fo fob-fob foobar-foobar ...
*/
#include <cstddef>
#include <cstdlib>
#include <stdexcept>
#include <string>
#include <algorithm>
#include <iterator>
#include "Position.h"
#include "CharClassify.h"
#include "RESearch.h"
using namespace Scintilla;
#define OKP 1
#define NOP 0
#define CHR 1
#define ANY 2
#define CCL 3
#define BOL 4
#define EOL 5
#define BOT 6
#define EOT 7
#define BOW 8
#define EOW 9
#define REF 10
#define CLO 11
#define CLQ 12 /* 0 to 1 closure */
#define LCLO 13 /* lazy closure */
#define END 0
/*
* The following defines are not meant to be changeable.
* They are for readability only.
*/
#define BLKIND 0370
#define BITIND 07
static const char bitarr[] = { 1, 2, 4, 8, 16, 32, 64, '\200' };
#define badpat(x) (*nfa = END, x)
/*
* Character classification table for word boundary operators BOW
* and EOW is passed in by the creator of this object (Scintilla
* Document). The Document default state is that word chars are:
* 0-9, a-z, A-Z and _
*/
RESearch::RESearch(CharClassify *charClassTable) {
failure = 0;
charClass = charClassTable;
sta = NOP; /* status of lastpat */
bol = 0;
constexpr unsigned char nul = 0;
std::fill(bittab, std::end(bittab), nul);
std::fill(tagstk, std::end(tagstk), 0);
std::fill(nfa, std::end(nfa), '\0');
Clear();
}
RESearch::~RESearch() {
Clear();
}
void RESearch::Clear() noexcept {
for (int i = 0; i < MAXTAG; i++) {
pat[i].clear();
bopat[i] = NOTFOUND;
eopat[i] = NOTFOUND;
}
}
void RESearch::GrabMatches(const CharacterIndexer &ci) {
for (unsigned int i = 0; i < MAXTAG; i++) {
if ((bopat[i] != NOTFOUND) && (eopat[i] != NOTFOUND)) {
const Sci::Position len = eopat[i] - bopat[i];
pat[i].resize(len);
for (Sci::Position j = 0; j < len; j++)
pat[i][j] = ci.CharAt(bopat[i] + j);
}
}
}
void RESearch::ChSet(unsigned char c) noexcept {
bittab[((c) & BLKIND) >> 3] |= bitarr[(c) & BITIND];
}
void RESearch::ChSetWithCase(unsigned char c, bool caseSensitive) noexcept {
ChSet(c);
if (!caseSensitive) {
if ((c >= 'a') && (c <= 'z')) {
ChSet(c - 'a' + 'A');
} else if ((c >= 'A') && (c <= 'Z')) {
ChSet(c - 'A' + 'a');
}
}
}
static unsigned char escapeValue(unsigned char ch) noexcept {
switch (ch) {
case 'a': return '\a';
case 'b': return '\b';
case 'f': return '\f';
case 'n': return '\n';
case 'r': return '\r';
case 't': return '\t';
case 'v': return '\v';
}
return 0;
}
static int GetHexaChar(unsigned char hd1, unsigned char hd2) noexcept {
int hexValue = 0;
if (hd1 >= '0' && hd1 <= '9') {
hexValue += 16 * (hd1 - '0');
} else if (hd1 >= 'A' && hd1 <= 'F') {
hexValue += 16 * (hd1 - 'A' + 10);
} else if (hd1 >= 'a' && hd1 <= 'f') {
hexValue += 16 * (hd1 - 'a' + 10);
} else {
return -1;
}
if (hd2 >= '0' && hd2 <= '9') {
hexValue += hd2 - '0';
} else if (hd2 >= 'A' && hd2 <= 'F') {
hexValue += hd2 - 'A' + 10;
} else if (hd2 >= 'a' && hd2 <= 'f') {
hexValue += hd2 - 'a' + 10;
} else {
return -1;
}
return hexValue;
}
/**
* Called when the parser finds a backslash not followed
* by a valid expression (like \( in non-Posix mode).
* @param pattern : pointer on the char after the backslash.
* @param incr : (out) number of chars to skip after expression evaluation.
* @return the char if it resolves to a simple char,
* or -1 for a char class. In this case, bittab is changed.
*/
int RESearch::GetBackslashExpression(
const char *pattern,
int &incr) noexcept {
// Since error reporting is primitive and messages are not used anyway,
// I choose to interpret unexpected syntax in a logical way instead
// of reporting errors. Otherwise, we can stick on, eg., PCRE behaviour.
incr = 0; // Most of the time, will skip the char "naturally".
int c;
int result = -1;
const unsigned char bsc = *pattern;
if (!bsc) {
// Avoid overrun
result = '\\'; // \ at end of pattern, take it literally
return result;
}
switch (bsc) {
case 'a':
case 'b':
case 'n':
case 'f':
case 'r':
case 't':
case 'v':
result = escapeValue(bsc);
break;
case 'x': {
const unsigned char hd1 = *(pattern + 1);
const unsigned char hd2 = *(pattern + 2);
const int hexValue = GetHexaChar(hd1, hd2);
if (hexValue >= 0) {
result = hexValue;
incr = 2; // Must skip the digits
} else {
result = 'x'; // \x without 2 digits: see it as 'x'
}
}
break;
case 'd':
for (c = '0'; c <= '9'; c++) {
ChSet(static_cast<unsigned char>(c));
}
break;
case 'D':
for (c = 0; c < MAXCHR; c++) {
if (c < '0' || c > '9') {
ChSet(static_cast<unsigned char>(c));
}
}
break;
case 's':
ChSet(' ');
ChSet('\t');
ChSet('\n');
ChSet('\r');
ChSet('\f');
ChSet('\v');
break;
case 'S':
for (c = 0; c < MAXCHR; c++) {
if (c != ' ' && !(c >= 0x09 && c <= 0x0D)) {
ChSet(static_cast<unsigned char>(c));
}
}
break;
case 'w':
for (c = 0; c < MAXCHR; c++) {
if (iswordc(static_cast<unsigned char>(c))) {
ChSet(static_cast<unsigned char>(c));
}
}
break;
case 'W':
for (c = 0; c < MAXCHR; c++) {
if (!iswordc(static_cast<unsigned char>(c))) {
ChSet(static_cast<unsigned char>(c));
}
}
break;
default:
result = bsc;
}
return result;
}
const char *RESearch::Compile(const char *pattern, Sci::Position length, bool caseSensitive, bool posix) noexcept {
char *mp=nfa; /* nfa pointer */
char *lp; /* saved pointer */
char *sp=nfa; /* another one */
char *mpMax = mp + MAXNFA - BITBLK - 10;
int tagi = 0; /* tag stack index */
int tagc = 1; /* actual tag count */
int n;
char mask; /* xor mask -CCL/NCL */
int c1, c2, prevChar;
if (!pattern || !length) {
if (sta)
return nullptr;
else
return badpat("No previous regular expression");
}
sta = NOP;
const char *p=pattern; /* pattern pointer */
for (int i=0; i<length; i++, p++) {
if (mp > mpMax)
return badpat("Pattern too long");
lp = mp;
switch (*p) {
case '.': /* match any char */
*mp++ = ANY;
break;
case '^': /* match beginning */
if (p == pattern) {
*mp++ = BOL;
} else {
*mp++ = CHR;
*mp++ = *p;
}
break;
case '$': /* match endofline */
if (!*(p+1)) {
*mp++ = EOL;
} else {
*mp++ = CHR;
*mp++ = *p;
}
break;
case '[': /* match char class */
*mp++ = CCL;
prevChar = 0;
i++;
if (*++p == '^') {
mask = '\377';
i++;
p++;
} else {
mask = 0;
}
if (*p == '-') { /* real dash */
i++;
prevChar = *p;
ChSet(*p++);
}
if (*p == ']') { /* real brace */
i++;
prevChar = *p;
ChSet(*p++);
}
while (*p && *p != ']') {
if (*p == '-') {
if (prevChar < 0) {
// Previous def. was a char class like \d, take dash literally
prevChar = *p;
ChSet(*p);
} else if (*(p+1)) {
if (*(p+1) != ']') {
c1 = prevChar + 1;
i++;
c2 = static_cast<unsigned char>(*++p);
if (c2 == '\\') {
if (!*(p+1)) { // End of RE
return badpat("Missing ]");
} else {
i++;
p++;
int incr;
c2 = GetBackslashExpression(p, incr);
i += incr;
p += incr;
if (c2 >= 0) {
// Convention: \c (c is any char) is case sensitive, whatever the option
ChSet(static_cast<unsigned char>(c2));
prevChar = c2;
} else {
// bittab is already changed
prevChar = -1;
}
}
}
if (prevChar < 0) {
// Char after dash is char class like \d, take dash literally
prevChar = '-';
ChSet('-');
} else {
// Put all chars between c1 and c2 included in the char set
while (c1 <= c2) {
ChSetWithCase(static_cast<unsigned char>(c1++), caseSensitive);
}
}
} else {
// Dash before the ], take it literally
prevChar = *p;
ChSet(*p);
}
} else {
return badpat("Missing ]");
}
} else if (*p == '\\' && *(p+1)) {
i++;
p++;
int incr;
const int c = GetBackslashExpression(p, incr);
i += incr;
p += incr;
if (c >= 0) {
// Convention: \c (c is any char) is case sensitive, whatever the option
ChSet(static_cast<unsigned char>(c));
prevChar = c;
} else {
// bittab is already changed
prevChar = -1;
}
} else {
prevChar = static_cast<unsigned char>(*p);
ChSetWithCase(*p, caseSensitive);
}
i++;
p++;
}
if (!*p)
return badpat("Missing ]");
for (n = 0; n < BITBLK; bittab[n++] = 0)
*mp++ = static_cast<char>(mask ^ bittab[n]);
break;
case '*': /* match 0 or more... */
case '+': /* match 1 or more... */
case '?':
if (p == pattern)
return badpat("Empty closure");
lp = sp; /* previous opcode */
if (*lp == CLO || *lp == LCLO) /* equivalence... */
break;
switch (*lp) {
case BOL:
case BOT:
case EOT:
case BOW:
case EOW:
case REF:
return badpat("Illegal closure");
default:
break;
}
if (*p == '+')
for (sp = mp; lp < sp; lp++)
*mp++ = *lp;
*mp++ = END;
*mp++ = END;
sp = mp;
while (--mp > lp)
*mp = mp[-1];
if (*p == '?') *mp = CLQ;
else if (*(p+1) == '?') *mp = LCLO;
else *mp = CLO;
mp = sp;
break;
case '\\': /* tags, backrefs... */
i++;
switch (*++p) {
case '<':
*mp++ = BOW;
break;
case '>':
if (*sp == BOW)
return badpat("Null pattern inside \\<\\>");
*mp++ = EOW;
break;
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
n = *p-'0';
if (tagi > 0 && tagstk[tagi] == n)
return badpat("Cyclical reference");
if (tagc > n) {
*mp++ = REF;
*mp++ = static_cast<char>(n);
} else {
return badpat("Undetermined reference");
}
break;
default:
if (!posix && *p == '(') {
if (tagc < MAXTAG) {
tagstk[++tagi] = tagc;
*mp++ = BOT;
*mp++ = static_cast<char>(tagc++);
} else {
return badpat("Too many \\(\\) pairs");
}
} else if (!posix && *p == ')') {
if (*sp == BOT)
return badpat("Null pattern inside \\(\\)");
if (tagi > 0) {
*mp++ = EOT;
*mp++ = static_cast<char>(tagstk[tagi--]);
} else {
return badpat("Unmatched \\)");
}
} else {
int incr;
int c = GetBackslashExpression(p, incr);
i += incr;
p += incr;
if (c >= 0) {
*mp++ = CHR;
*mp++ = static_cast<unsigned char>(c);
} else {
*mp++ = CCL;
mask = 0;
for (n = 0; n < BITBLK; bittab[n++] = 0)
*mp++ = static_cast<char>(mask ^ bittab[n]);
}
}
}
break;
default : /* an ordinary char */
if (posix && *p == '(') {
if (tagc < MAXTAG) {
tagstk[++tagi] = tagc;
*mp++ = BOT;
*mp++ = static_cast<char>(tagc++);
} else {
return badpat("Too many () pairs");
}
} else if (posix && *p == ')') {
if (*sp == BOT)
return badpat("Null pattern inside ()");
if (tagi > 0) {
*mp++ = EOT;
*mp++ = static_cast<char>(tagstk[tagi--]);
} else {
return badpat("Unmatched )");
}
} else {
unsigned char c = *p;
if (!c) // End of RE
c = '\\'; // We take it as raw backslash
if (caseSensitive || !iswordc(c)) {
*mp++ = CHR;
*mp++ = c;
} else {
*mp++ = CCL;
mask = 0;
ChSetWithCase(c, false);
for (n = 0; n < BITBLK; bittab[n++] = 0)
*mp++ = static_cast<char>(mask ^ bittab[n]);
}
}
break;
}
sp = lp;
}
if (tagi > 0)
return badpat((posix ? "Unmatched (" : "Unmatched \\("));
*mp = END;
sta = OKP;
return nullptr;
}
/*
* RESearch::Execute:
* execute nfa to find a match.
*
* special cases: (nfa[0])
* BOL
* Match only once, starting from the
* beginning.
* CHR
* First locate the character without
* calling PMatch, and if found, call
* PMatch for the remaining string.
* END
* RESearch::Compile failed, poor luser did not
* check for it. Fail fast.
*
* If a match is found, bopat[0] and eopat[0] are set
* to the beginning and the end of the matched fragment,
* respectively.
*
*/
int RESearch::Execute(const CharacterIndexer &ci, Sci::Position lp, Sci::Position endp) {
unsigned char c;
Sci::Position ep = NOTFOUND;
char *ap = nfa;
bol = lp;
failure = 0;
Clear();
switch (*ap) {
case BOL: /* anchored: match from BOL only */
ep = PMatch(ci, lp, endp, ap);
break;
case EOL: /* just searching for end of line normal path doesn't work */
if (*(ap+1) == END) {
lp = endp;
ep = lp;
break;
} else {
return 0;
}
case CHR: /* ordinary char: locate it fast */
c = *(ap+1);
while ((lp < endp) && (static_cast<unsigned char>(ci.CharAt(lp)) != c))
lp++;
if (lp >= endp) /* if EOS, fail, else fall through. */
return 0;
[[fallthrough]];
default: /* regular matching all the way. */
while (lp < endp) {
ep = PMatch(ci, lp, endp, ap);
if (ep != NOTFOUND)
break;
lp++;
}
break;
case END: /* munged automaton. fail always */
return 0;
}
if (ep == NOTFOUND)
return 0;
bopat[0] = lp;
eopat[0] = ep;
return 1;
}
/*
* PMatch: internal routine for the hard part
*
* This code is partly snarfed from an early grep written by
* David Conroy. The backref and tag stuff, and various other
* innovations are by oz.
*
* special case optimizations: (nfa[n], nfa[n+1])
* CLO ANY
* We KNOW .* will match everything up to the
* end of line. Thus, directly go to the end of
* line, without recursive PMatch calls. As in
* the other closure cases, the remaining pattern
* must be matched by moving backwards on the
* string recursively, to find a match for xy
* (x is ".*" and y is the remaining pattern)
* where the match satisfies the LONGEST match for
* x followed by a match for y.
* CLO CHR
* We can again scan the string forward for the
* single char and at the point of failure, we
* execute the remaining nfa recursively, same as
* above.
*
* At the end of a successful match, bopat[n] and eopat[n]
* are set to the beginning and end of subpatterns matched
* by tagged expressions (n = 1 to 9).
*/
extern void re_fail(char *,char);
static inline int isinset(const char *ap, unsigned char c) noexcept {
return ap[(c & BLKIND) >> 3] & bitarr[c & BITIND];
}
/*
* skip values for CLO XXX to skip past the closure
*/
#define ANYSKIP 2 /* [CLO] ANY END */
#define CHRSKIP 3 /* [CLO] CHR chr END */
#define CCLSKIP 34 /* [CLO] CCL 32 bytes END */
Sci::Position RESearch::PMatch(const CharacterIndexer &ci, Sci::Position lp, Sci::Position endp, char *ap) {
int op, c, n;
Sci::Position e; /* extra pointer for CLO */
Sci::Position bp; /* beginning of subpat... */
Sci::Position ep; /* ending of subpat... */
Sci::Position are; /* to save the line ptr. */
Sci::Position llp; /* lazy lp for LCLO */
while ((op = *ap++) != END)
switch (op) {
case CHR:
if (ci.CharAt(lp++) != *ap++)
return NOTFOUND;
break;
case ANY:
if (lp++ >= endp)
return NOTFOUND;
break;
case CCL:
if (lp >= endp)
return NOTFOUND;
if (!isinset(ap, ci.CharAt(lp++)))
return NOTFOUND;
ap += BITBLK;
break;
case BOL:
if (lp != bol)
return NOTFOUND;
break;
case EOL:
if (lp < endp)
return NOTFOUND;
break;
case BOT:
bopat[static_cast<int>(*ap++)] = lp;
break;
case EOT:
eopat[static_cast<int>(*ap++)] = lp;
break;
case BOW:
if ((lp!=bol && iswordc(ci.CharAt(lp-1))) || !iswordc(ci.CharAt(lp)))
return NOTFOUND;
break;
case EOW:
if (lp==bol || !iswordc(ci.CharAt(lp-1)) || iswordc(ci.CharAt(lp)))
return NOTFOUND;
break;
case REF:
n = *ap++;
bp = bopat[n];
ep = eopat[n];
while (bp < ep)
if (ci.CharAt(bp++) != ci.CharAt(lp++))
return NOTFOUND;
break;
case LCLO:
case CLQ:
case CLO:
are = lp;
switch (*ap) {
case ANY:
if (op == CLO || op == LCLO)
while (lp < endp)
lp++;
else if (lp < endp)
lp++;
n = ANYSKIP;
break;
case CHR:
c = *(ap+1);
if (op == CLO || op == LCLO)
while ((lp < endp) && (c == ci.CharAt(lp)))
lp++;
else if ((lp < endp) && (c == ci.CharAt(lp)))
lp++;
n = CHRSKIP;
break;
case CCL:
while ((lp < endp) && isinset(ap+1, ci.CharAt(lp)))
lp++;
n = CCLSKIP;
break;
default:
failure = true;
//re_fail("closure: bad nfa.", *ap);
return NOTFOUND;
}
ap += n;
llp = lp;
e = NOTFOUND;
while (llp >= are) {
Sci::Position q;
if ((q = PMatch(ci, llp, endp, ap)) != NOTFOUND) {
e = q;
lp = llp;
if (op != LCLO) return e;
}
if (*ap == END) return e;
--llp;
}
if (*ap == EOT)
PMatch(ci, lp, endp, ap);
return e;
default:
//re_fail("RESearch::Execute: bad nfa.", static_cast<char>(op));
return NOTFOUND;
}
return lp;
}
|