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
|
static const char rcsid[] = "$Id: s_regex.c,v 1.10 2004/04/30 17:00:58 will Exp $";
/*
* regex - Regular expression pattern matching
* and replacement
*
*
* By: Ozan S. Yigit (oz)
* Dept. of Computer Science
* York University
*
*
* 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.
*
* ** NAMES CHANGED TO BE PREFIXED WITH s_, TO AVOID CONFLICTS **
* ** -Will Deich, Feb 1997 **
*
* Routines:
* s_re_comp: compile a regular expression into
* a DFA.
*
* char *s_re_comp(s)
* char *s;
*
* s_re_exec: execute the DFA to match a pattern.
*
* int s_re_exec(s)
* char *s;
*
* s_re_modw change s_re_exec's understanding of what
* a "word" looks like (for \< and \>)
* by adding into the hidden word-character
* table.
*
* void s_re_modw(s)
* char *s;
*
* s_re_subs: substitute the matched portions in
* a new string.
*
* int s_re_subs(src, dst)
* char *src;
* char *dst;
*
* s_re_fail: failure routine for s_re_exec.
*
* void s_re_fail(msg, op)
* char *msg;
* char op;
*
* Regular Expressions:
*
* [1] char matches itself, unless it is a special
* character (metachar): . \ [ ] * + ^ $
*
* [2] . matches any character.
*
* [3] \ matches the character following it, except
* when followed by a left or right round bracket,
* a digit 1 to 9 or a left or right angle bracket.
* (see [7], [8] and [9])
* It is used as an escape character for all
* other meta-characters, and itself. When used
* in a set ([4]), it is treated as an ordinary
* character.
*
* [4] [set] matches one of the characters in the set.
* If the first character in the set is "^",
* it matches a character NOT in the set. A
* shorthand S-E is used to specify a set of
* characters S upto E, inclusive. The special
* characters "]" and "-" have no special
* meaning if they appear as the first chars
* in the set.
* examples: match:
*
* [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], followed by
* closure char (*) matches zero or more matches of
* that form.
*
* [6] + same as [5], except it matches one or more.
*
* [7] a regular expression in the form [1] to [10], enclosed
* as \(form\) matches what form matches. The enclosure
* creates a set of tags, used for [8] and for
* pattern substution. The tagged forms are numbered
* starting from 1.
*
* [8] a \ followed by a digit 1 to 9 matches whatever a
* previously tagged regular expression ([7]) matched.
*
* [9] \< 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 _. It must also be preceded and/or
* followed by any character outside those mentioned.
*
* [10] a composite regular expression xy where x and y
* are in the form [1] to [10] matches the longest
* match of x followed by a match for y.
*
* [11] ^ 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 128-bit block. Thus, CCL or NCL always takes a
* constant 16 bytes in the internal dfa, and s_re_exec 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 2 o b 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 ...
*
*/
#define MAXDFA 1024
#define MAXTAG 10
#define OKP 1
#define NOP 0
#define CHR 1
#define ANY 2
#define CCL 3
#define NCL 4
#define BOL 5
#define EOL 6
#define BOT 7
#define EOT 8
#define BOW 9
#define EOW 10
#define REF 11
#define CLO 12
#define END 0
/*
* The following defines are not meant
* to be changeable. They are for readibility
* only.
*
*/
#define MAXCHR 128
#define CHRBIT 8
#define BITBLK MAXCHR/CHRBIT
#define BLKIND 0170
#define BITIND 07
#define ASCIIB 0177
typedef unsigned char CHAR;
static int tagstk[MAXTAG]; /* subpat tag stack..*/
static CHAR dfa[MAXDFA]; /* automaton.. */
static int sta = NOP; /* status of lastpat */
static CHAR bittab[BITBLK]; /* bit table for CCL */
static void
chset(c)
register CHAR c;
{
bittab[((int) ((c)&BLKIND))>>3] |= 1<<((c)& (unsigned) BITIND);
}
#define badpat(x) return(*dfa = END, x)
#define store(x) *mp++ = x
char *
s_re_comp(pat)
char *pat;
{
register char *p; /* pattern pointer */
register CHAR *mp=dfa; /* dfa pointer */
register CHAR *lp; /* saved pointer.. */
register CHAR *sp=dfa; /* another one.. */
register int tagi = 0; /* tag stack index */
register int tagc = 1; /* actual tag count */
register int n;
int c1, c2;
if (!pat || !*pat) {
if (sta)
return(0);
else
badpat("No previous regular expression");
}
sta = NOP;
for (p = pat; *p; p++) {
lp = mp;
switch(*p) {
case '.': /* match any char.. */
store(ANY);
break;
case '^': /* match beginning.. */
if (p == pat)
store(BOL);
else {
store(CHR);
store(*p);
}
break;
case '$': /* match endofline.. */
if (!*(p+1))
store(EOL);
else {
store(CHR);
store(*p);
}
break;
case '[': /* match char class..*/
if (*++p == '^') {
store(NCL);
p++;
}
else
store(CCL);
if (*p == '-') /* real dash */
chset(*p++);
if (*p == ']') /* real brac */
chset(*p++);
while (*p && *p != ']') {
if (*p == '-' && *(p+1) && *(p+1) != ']') {
p++;
c1 = *(p-2) + 1;
c2 = *p++;
while (c1 <= c2)
chset(c1++);
}
#ifdef EXTEND
else if (*p == '\\' && *(p+1)) {
p++;
chset(*p++);
}
#endif
else
chset(*p++);
}
if (!*p)
badpat("Missing ]");
for (n = 0; n < BITBLK; bittab[n++] = (char) 0)
store(bittab[n]);
break;
case '*': /* match 0 or more.. */
case '+': /* match 1 or more.. */
if (p == pat)
badpat("Empty closure");
lp = sp; /* previous opcode */
if (*lp == CLO) /* equivalence.. */
break;
switch(*lp) {
case BOL:
case BOT:
case EOT:
case BOW:
case EOW:
case REF:
badpat("Illegal closure");
default:
break;
}
if (*p == '+')
for (sp = mp; lp < sp; lp++)
store(*lp);
store(END);
store(END);
sp = mp;
while (--mp > lp)
*mp = mp[-1];
store(CLO);
mp = sp;
break;
case '\\': /* tags, backrefs .. */
switch(*++p) {
case '(':
if (tagc < MAXTAG) {
tagstk[++tagi] = tagc;
store(BOT);
store(tagc++);
}
else
badpat("Too many \\(\\) pairs");
break;
case ')':
if (*sp == BOT)
badpat("Null pattern inside \\(\\)");
if (tagi > 0) {
store(EOT);
store(tagstk[tagi--]);
}
else
badpat("Unmatched \\)");
break;
case '<':
store(BOW);
break;
case '>':
if (*sp == BOW)
badpat("Null pattern inside \\<\\>");
store(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)
badpat("Cyclical reference");
if (tagc > n) {
store(REF);
store(n);
}
else
badpat("Undetermined reference");
break;
#ifdef EXTEND
case 'b':
store(CHR);
store('\b');
break;
case 'n':
store(CHR);
store('\n');
break;
case 'f':
store(CHR);
store('\f');
break;
case 'r':
store(CHR);
store('\r');
break;
case 't':
store(CHR);
store('\t');
break;
#endif
default:
store(CHR);
store(*p);
}
break;
default : /* an ordinary char */
store(CHR);
store(*p);
break;
}
sp = lp;
}
if (tagi > 0)
badpat("Unmatched \\(");
store(END);
sta = OKP;
return(0);
}
static char *bol;
static char *bopat[MAXTAG];
static char *eopat[MAXTAG];
static char *pmatch();
/*
* s_re_exec:
* execute dfa to find a match.
*
* special cases: (dfa[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
* s_re_comp 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
s_re_exec(lp)
register char *lp;
{
register char c;
register char *ep = 0;
register CHAR *ap = dfa;
bol = lp;
bopat[0] = 0;
bopat[1] = 0;
bopat[2] = 0;
bopat[3] = 0;
bopat[4] = 0;
bopat[5] = 0;
bopat[6] = 0;
bopat[7] = 0;
bopat[8] = 0;
bopat[9] = 0;
switch(*ap) {
case BOL: /* anchored: match from BOL only */
ep = pmatch(lp,ap);
break;
case CHR: /* ordinary char: locate it fast */
c = *(ap+1);
while (*lp && *lp != c)
lp++;
if (!*lp) /* if EOS, fail, else fall thru. */
return(0);
default: /* regular matching all the way. */
while (*lp) {
if ((ep = pmatch(lp,ap)))
break;
lp++;
}
break;
case END: /* munged automaton. fail always */
return(0);
}
if (!ep)
return(0);
bopat[0] = lp;
eopat[0] = ep;
return(1);
}
/*
* pmatch:
* internal routine for the hard part
*
* This code is mostly snarfed from an early
* grep written by David Conroy. The backref and
* tag stuff, and various other mods are by oZ.
*
* special cases: (dfa[n], dfa[n+1])
* CLO ANY
* We KNOW ".*" will match ANYTHING
* upto the end of line. Thus, go to
* the end of line straight, without
* calling pmatch recursively. 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 without recursion,
* and at the point of failure, we execute
* the remaining dfa recursively, as
* described 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 s_re_fail();
/*
* character classification table for word boundary
* operators BOW and EOW. the reason for not using
* ctype macros is that we can let the user add into
* our own table. see s_re_modw. This table is not in
* the bitset form, since we may wish to extend it
* in the future for other character classifications.
*
* TRUE for 0-9 A-Z a-z _
*/
static char chrtyp[MAXCHR] = {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 0, 0,
0, 0, 0, 0, 0, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 0, 0, 0, 0, 1, 0, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 0, 0, 0, 0, 0
};
#define inascii(x) (0177&(x))
#define iswordc(x) chrtyp[inascii(x)]
#define isinset(x,y) ((x)[((y)&BLKIND)>>3] & (1<<((y)&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 18 /* CLO CCL 16bytes END ... */
static char *
pmatch(lp, ap)
register char *lp;
register CHAR *ap;
{
register char *e; /* extra pointer for CLO */
register char *bp; /* beginning of subpat.. */
register char *ep; /* ending of subpat.. */
register int op, c, n;
char *are; /* to save the line ptr. */
while ((op = *ap++) != END)
switch(op) {
case CHR:
if ((unsigned char) *lp++ != *ap++)
return(0);
break;
case ANY:
if (!*lp++)
return(0);
break;
case CCL:
c = *lp++;
if (!isinset(ap,c))
return(0);
ap += BITBLK;
break;
case NCL:
c = *lp++;
if (isinset(ap,c))
return(0);
ap += BITBLK;
break;
case BOL:
if (lp != bol)
return(0);
break;
case EOL:
if (*lp)
return(0);
break;
case BOT:
bopat[*ap++] = lp;
break;
case EOT:
eopat[*ap++] = lp;
break;
case BOW:
if (!(lp!=bol && iswordc(lp[-1])) && iswordc(*lp))
break;
return(0);
case EOW:
if ((lp!=bol && iswordc(lp[-1])) && !iswordc(*lp))
break;
return(0);
case REF:
n = *ap++;
bp = bopat[n];
ep = eopat[n];
while (bp < ep)
if (*bp++ != *lp++)
return(0);
break;
case CLO:
are = lp;
switch(*ap) {
case ANY:
while (*lp)
lp++;
n = ANYSKIP;
break;
case CHR:
c = *(ap+1);
while (*lp && c == *lp)
lp++;
n = CHRSKIP;
break;
case CCL:
case NCL:
while (*lp && (e = pmatch(lp, ap)))
lp = e;
n = CCLSKIP;
break;
default:
s_re_fail("closure: bad dfa.", *ap);
return(0);
}
ap += n;
while (lp >= are) {
if ((e = pmatch(lp, ap)))
return(e);
--lp;
}
return(0);
default:
s_re_fail("s_re_exec: bad dfa.", op);
return(0);
}
return(lp);
}
/*
* s_re_modw:
* add new characters into the word table to
* change the s_re_exec's understanding of what
* a word should look like. Note that we only
* accept additions into the word definition.
*
* If the string parameter is 0 or null string,
* the table is reset back to the default, which
* contains A-Z a-z 0-9 _. [We use the compact
* bitset representation for the default table]
*
*/
static unsigned char deftab[16] = {
0, 0, 0, 0, 0, 0, 0377, 003, 0376, 0377, 0377, 0207,
0376, 0377, 0377, 007
};
void
s_re_modw(s)
register char *s;
{
register int i;
if (!s || !*s) {
for (i = 0; i < MAXCHR; i++)
if (!isinset(deftab,i))
iswordc(i) = 0;
}
else
while(*s)
iswordc(*s++) = 1;
}
/*
* s_re_subs:
* substitute the matched portions of the src in
* dst.
*
* & substitute the entire matched pattern.
*
* \digit substitute a subpattern, with the given
* tag number. Tags are numbered from 1 to
* 9. If the particular tagged subpattern
* does not exist, null is substituted.
*
*/
int
s_re_subs(src, dst)
register char *src;
register char *dst;
{
register char c;
register int pin;
register char *bp;
register char *ep;
if (!*src || !bopat[0])
return(0);
while ((c = *src++)) {
switch(c) {
case '&':
pin = 0;
break;
case '\\':
c = *src++;
if (c >= '0' && c <= '9') {
pin = c - '0';
break;
}
default:
*dst++ = c;
continue;
}
if ((bp = bopat[pin]) && (ep = eopat[pin])) {
while (*bp && bp < ep)
*dst++ = *bp++;
if (bp < ep)
return(0);
}
}
*dst = (char) 0;
return(1);
}
#ifdef DEBUG
/*
* symbolic - produce a symbolic dump of the
* dfa
*/
symbolic(s)
char *s;
{
printf("pattern: %s\n", s);
printf("dfacode:\n");
dfadump(dfa);
}
static
dfadump(ap)
CHAR *ap;
{
register int n;
while (*ap != END)
switch(*ap++) {
case CLO:
printf("CLOSURE");
dfadump(ap);
switch(*ap) {
case CHR:
n = CHRSKIP;
break;
case ANY:
n = ANYSKIP;
break;
case CCL:
case NCL:
n = CCLSKIP;
break;
}
ap += n;
break;
case CHR:
printf("\tCHR %c\n",*ap++);
break;
case ANY:
printf("\tANY .\n");
break;
case BOL:
printf("\tBOL -\n");
break;
case EOL:
printf("\tEOL -\n");
break;
case BOT:
printf("BOT: %d\n",*ap++);
break;
case EOT:
printf("EOT: %d\n",*ap++);
break;
case BOW:
printf("BOW\n");
break;
case EOW:
printf("EOW\n");
break;
case REF:
printf("REF: %d\n",*ap++);
break;
case CCL:
printf("\tCCL [");
for (n = 0; n < MAXCHR; n++)
if (isinset(ap,(CHAR)n))
printf("%c",n);
printf("]\n");
ap += BITBLK;
break;
case NCL:
printf("\tNCL [");
for (n = 0; n < MAXCHR; n++)
if (isinset(ap,(CHAR)n))
printf("%c",n);
printf("]\n");
ap += BITBLK;
break;
default:
printf("bad dfa. opcode %o\n", ap[-1]);
exit(1);
break;
}
}
#endif
|