DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Mercurial (fbecf6c8a86f)

VCS Links

Line Code
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
// Copyright (c) 2005, Google Inc.
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
//     * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
//     * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
//     * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Author: Sanjay Ghemawat

#ifdef HAVE_CONFIG_H
#include <config.h>
#endif

#include <stdlib.h>
#include <stdio.h>
#include <ctype.h>
#include <limits.h>      /* for SHRT_MIN, USHRT_MAX, etc */
#include <assert.h>
#include <errno.h>
#include <string>
#include <algorithm>

#include "pcrecpp_internal.h"
#include <pcre.h>
#include "pcrecpp.h"
#include "pcre_stringpiece.h"


namespace pcrecpp {

// Maximum number of args we can set
static const int kMaxArgs = 16;
static const int kVecSize = (1 + kMaxArgs) * 3;  // results + PCRE workspace

// Special object that stands-in for no argument
PCRECPP_EXP_DEFN Arg no_arg((void*)NULL);

// If a regular expression has no error, its error_ field points here
static const string empty_string;

// If the user doesn't ask for any options, we just use this one
static RE_Options default_options;

void RE::Init(const string& pat, const RE_Options* options) {
  pattern_ = pat;
  if (options == NULL) {
    options_ = default_options;
  } else {
    options_ = *options;
  }
  error_ = &empty_string;
  re_full_ = NULL;
  re_partial_ = NULL;

  re_partial_ = Compile(UNANCHORED);
  if (re_partial_ != NULL) {
    re_full_ = Compile(ANCHOR_BOTH);
  }
}

void RE::Cleanup() {
  if (re_full_ != NULL)         (*pcre_free)(re_full_);
  if (re_partial_ != NULL)      (*pcre_free)(re_partial_);
  if (error_ != &empty_string)  delete error_;
}


RE::~RE() {
  Cleanup();
}


pcre* RE::Compile(Anchor anchor) {
  // First, convert RE_Options into pcre options
  int pcre_options = 0;
  pcre_options = options_.all_options();

  // Special treatment for anchoring.  This is needed because at
  // runtime pcre only provides an option for anchoring at the
  // beginning of a string (unless you use offset).
  //
  // There are three types of anchoring we want:
  //    UNANCHORED      Compile the original pattern, and use
  //                    a pcre unanchored match.
  //    ANCHOR_START    Compile the original pattern, and use
  //                    a pcre anchored match.
  //    ANCHOR_BOTH     Tack a "\z" to the end of the original pattern
  //                    and use a pcre anchored match.

  const char* compile_error;
  int eoffset;
  pcre* re;
  if (anchor != ANCHOR_BOTH) {
    re = pcre_compile(pattern_.c_str(), pcre_options,
                      &compile_error, &eoffset, NULL);
  } else {
    // Tack a '\z' at the end of RE.  Parenthesize it first so that
    // the '\z' applies to all top-level alternatives in the regexp.
    string wrapped = "(?:";  // A non-counting grouping operator
    wrapped += pattern_;
    wrapped += ")\\z";
    re = pcre_compile(wrapped.c_str(), pcre_options,
                      &compile_error, &eoffset, NULL);
  }
  if (re == NULL) {
    if (error_ == &empty_string) error_ = new string(compile_error);
  }
  return re;
}

/***** Matching interfaces *****/

bool RE::FullMatch(const StringPiece& text,
                   const Arg& ptr1,
                   const Arg& ptr2,
                   const Arg& ptr3,
                   const Arg& ptr4,
                   const Arg& ptr5,
                   const Arg& ptr6,
                   const Arg& ptr7,
                   const Arg& ptr8,
                   const Arg& ptr9,
                   const Arg& ptr10,
                   const Arg& ptr11,
                   const Arg& ptr12,
                   const Arg& ptr13,
                   const Arg& ptr14,
                   const Arg& ptr15,
                   const Arg& ptr16) const {
  const Arg* args[kMaxArgs];
  int n = 0;
  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
 done:

  int consumed;
  int vec[kVecSize];
  return DoMatchImpl(text, ANCHOR_BOTH, &consumed, args, n, vec, kVecSize);
}

bool RE::PartialMatch(const StringPiece& text,
                      const Arg& ptr1,
                      const Arg& ptr2,
                      const Arg& ptr3,
                      const Arg& ptr4,
                      const Arg& ptr5,
                      const Arg& ptr6,
                      const Arg& ptr7,
                      const Arg& ptr8,
                      const Arg& ptr9,
                      const Arg& ptr10,
                      const Arg& ptr11,
                      const Arg& ptr12,
                      const Arg& ptr13,
                      const Arg& ptr14,
                      const Arg& ptr15,
                      const Arg& ptr16) const {
  const Arg* args[kMaxArgs];
  int n = 0;
  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
 done:

  int consumed;
  int vec[kVecSize];
  return DoMatchImpl(text, UNANCHORED, &consumed, args, n, vec, kVecSize);
}

bool RE::Consume(StringPiece* input,
                 const Arg& ptr1,
                 const Arg& ptr2,
                 const Arg& ptr3,
                 const Arg& ptr4,
                 const Arg& ptr5,
                 const Arg& ptr6,
                 const Arg& ptr7,
                 const Arg& ptr8,
                 const Arg& ptr9,
                 const Arg& ptr10,
                 const Arg& ptr11,
                 const Arg& ptr12,
                 const Arg& ptr13,
                 const Arg& ptr14,
                 const Arg& ptr15,
                 const Arg& ptr16) const {
  const Arg* args[kMaxArgs];
  int n = 0;
  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
 done:

  int consumed;
  int vec[kVecSize];
  if (DoMatchImpl(*input, ANCHOR_START, &consumed,
                  args, n, vec, kVecSize)) {
    input->remove_prefix(consumed);
    return true;
  } else {
    return false;
  }
}

bool RE::FindAndConsume(StringPiece* input,
                        const Arg& ptr1,
                        const Arg& ptr2,
                        const Arg& ptr3,
                        const Arg& ptr4,
                        const Arg& ptr5,
                        const Arg& ptr6,
                        const Arg& ptr7,
                        const Arg& ptr8,
                        const Arg& ptr9,
                        const Arg& ptr10,
                        const Arg& ptr11,
                        const Arg& ptr12,
                        const Arg& ptr13,
                        const Arg& ptr14,
                        const Arg& ptr15,
                        const Arg& ptr16) const {
  const Arg* args[kMaxArgs];
  int n = 0;
  if (&ptr1  == &no_arg) goto done; args[n++] = &ptr1;
  if (&ptr2  == &no_arg) goto done; args[n++] = &ptr2;
  if (&ptr3  == &no_arg) goto done; args[n++] = &ptr3;
  if (&ptr4  == &no_arg) goto done; args[n++] = &ptr4;
  if (&ptr5  == &no_arg) goto done; args[n++] = &ptr5;
  if (&ptr6  == &no_arg) goto done; args[n++] = &ptr6;
  if (&ptr7  == &no_arg) goto done; args[n++] = &ptr7;
  if (&ptr8  == &no_arg) goto done; args[n++] = &ptr8;
  if (&ptr9  == &no_arg) goto done; args[n++] = &ptr9;
  if (&ptr10 == &no_arg) goto done; args[n++] = &ptr10;
  if (&ptr11 == &no_arg) goto done; args[n++] = &ptr11;
  if (&ptr12 == &no_arg) goto done; args[n++] = &ptr12;
  if (&ptr13 == &no_arg) goto done; args[n++] = &ptr13;
  if (&ptr14 == &no_arg) goto done; args[n++] = &ptr14;
  if (&ptr15 == &no_arg) goto done; args[n++] = &ptr15;
  if (&ptr16 == &no_arg) goto done; args[n++] = &ptr16;
 done:

  int consumed;
  int vec[kVecSize];
  if (DoMatchImpl(*input, UNANCHORED, &consumed,
                  args, n, vec, kVecSize)) {
    input->remove_prefix(consumed);
    return true;
  } else {
    return false;
  }
}

bool RE::Replace(const StringPiece& rewrite,
                 string *str) const {
  int vec[kVecSize];
  int matches = TryMatch(*str, 0, UNANCHORED, vec, kVecSize);
  if (matches == 0)
    return false;

  string s;
  if (!Rewrite(&s, rewrite, *str, vec, matches))
    return false;

  assert(vec[0] >= 0);
  assert(vec[1] >= 0);
  str->replace(vec[0], vec[1] - vec[0], s);
  return true;
}

// Returns PCRE_NEWLINE_CRLF, PCRE_NEWLINE_CR, or PCRE_NEWLINE_LF.
// Note that PCRE_NEWLINE_CRLF is defined to be P_N_CR | P_N_LF.
static int NewlineMode(int pcre_options) {
  // TODO: if we can make it threadsafe, cache this var
  int newline_mode = 0;
  /* if (newline_mode) return newline_mode; */  // do this once it's cached
  if (pcre_options & (PCRE_NEWLINE_CRLF|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF)) {
    newline_mode = (pcre_options &
                    (PCRE_NEWLINE_CRLF|PCRE_NEWLINE_CR|PCRE_NEWLINE_LF));
  } else {
    int newline;
    pcre_config(PCRE_CONFIG_NEWLINE, &newline);
    if (newline == 10)
      newline_mode = PCRE_NEWLINE_LF;
    else if (newline == 13)
      newline_mode = PCRE_NEWLINE_CR;
    else if (newline == 3338)
      newline_mode = PCRE_NEWLINE_CRLF;
    else
      assert("" == "Unexpected return value from pcre_config(NEWLINE)");
  }
  return newline_mode;
}

int RE::GlobalReplace(const StringPiece& rewrite,
                      string *str) const {
  int count = 0;
  int vec[kVecSize];
  string out;
  int start = 0;
  int lastend = -1;

  for (; start <= static_cast<int>(str->length()); count++) {
    int matches = TryMatch(*str, start, UNANCHORED, vec, kVecSize);
    if (matches <= 0)
      break;
    int matchstart = vec[0], matchend = vec[1];
    assert(matchstart >= start);
    assert(matchend >= matchstart);
    if (matchstart == matchend && matchstart == lastend) {
      // advance one character if we matched an empty string at the same
      // place as the last match occurred
      matchend = start + 1;
      // If the current char is CR and we're in CRLF mode, skip LF too.
      // Note it's better to call pcre_fullinfo() than to examine
      // all_options(), since options_ could have changed bewteen
      // compile-time and now, but this is simpler and safe enough.
      if (start+1 < static_cast<int>(str->length()) &&
          (*str)[start] == '\r' && (*str)[start+1] == '\n' &&
          NewlineMode(options_.all_options()) == PCRE_NEWLINE_CRLF) {
        matchend++;
      }
      // We also need to advance more than one char if we're in utf8 mode.
#ifdef SUPPORT_UTF8
      if (options_.utf8()) {
        while (matchend < static_cast<int>(str->length()) &&
               ((*str)[matchend] & 0xc0) == 0x80)
          matchend++;
      }
#endif
      if (matchend <= static_cast<int>(str->length()))
        out.append(*str, start, matchend - start);
      start = matchend;
    } else {
      out.append(*str, start, matchstart - start);
      Rewrite(&out, rewrite, *str, vec, matches);
      start = matchend;
      lastend = matchend;
      count++;
    }
  }

  if (count == 0)
    return 0;

  if (start < static_cast<int>(str->length()))
    out.append(*str, start, str->length() - start);
  swap(out, *str);
  return count;
}

bool RE::Extract(const StringPiece& rewrite,
                 const StringPiece& text,
                 string *out) const {
  int vec[kVecSize];
  int matches = TryMatch(text, 0, UNANCHORED, vec, kVecSize);
  if (matches == 0)
    return false;
  out->erase();
  return Rewrite(out, rewrite, text, vec, matches);
}

/*static*/ string RE::QuoteMeta(const StringPiece& unquoted) {
  string result;

  // Escape any ascii character not in [A-Za-z_0-9].
  //
  // Note that it's legal to escape a character even if it has no
  // special meaning in a regular expression -- so this function does
  // that.  (This also makes it identical to the perl function of the
  // same name; see `perldoc -f quotemeta`.)
  for (int ii = 0; ii < unquoted.size(); ++ii) {
    // Note that using 'isalnum' here raises the benchmark time from
    // 32ns to 58ns:
    if ((unquoted[ii] < 'a' || unquoted[ii] > 'z') &&
        (unquoted[ii] < 'A' || unquoted[ii] > 'Z') &&
        (unquoted[ii] < '0' || unquoted[ii] > '9') &&
        unquoted[ii] != '_' &&
        // If this is the part of a UTF8 or Latin1 character, we need
        // to copy this byte without escaping.  Experimentally this is
        // what works correctly with the regexp library.
        !(unquoted[ii] & 128)) {
      result += '\\';
    }
    result += unquoted[ii];
  }

  return result;
}

/***** Actual matching and rewriting code *****/

int RE::TryMatch(const StringPiece& text,
                 int startpos,
                 Anchor anchor,
                 int *vec,
                 int vecsize) const {
  pcre* re = (anchor == ANCHOR_BOTH) ? re_full_ : re_partial_;
  if (re == NULL) {
    //fprintf(stderr, "Matching against invalid re: %s\n", error_->c_str());
    return 0;
  }

  pcre_extra extra = { 0, 0, 0, 0, 0, 0 };
  if (options_.match_limit() > 0) {
    extra.flags |= PCRE_EXTRA_MATCH_LIMIT;
    extra.match_limit = options_.match_limit();
  }
  if (options_.match_limit_recursion() > 0) {
    extra.flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
    extra.match_limit_recursion = options_.match_limit_recursion();
  }
  int rc = pcre_exec(re,              // The regular expression object
                     &extra,
                     (text.data() == NULL) ? "" : text.data(),
                     text.size(),
                     startpos,
                     (anchor == UNANCHORED) ? 0 : PCRE_ANCHORED,
                     vec,
                     vecsize);

  // Handle errors
  if (rc == PCRE_ERROR_NOMATCH) {
    return 0;
  } else if (rc < 0) {
    //fprintf(stderr, "Unexpected return code: %d when matching '%s'\n",
    //        re, pattern_.c_str());
    return 0;
  } else if (rc == 0) {
    // pcre_exec() returns 0 as a special case when the number of
    // capturing subpatterns exceeds the size of the vector.
    // When this happens, there is a match and the output vector
    // is filled, but we miss out on the positions of the extra subpatterns.
    rc = vecsize / 2;
  }

  return rc;
}

bool RE::DoMatchImpl(const StringPiece& text,
                     Anchor anchor,
                     int* consumed,
                     const Arg* const* args,
                     int n,
                     int* vec,
                     int vecsize) const {
  assert((1 + n) * 3 <= vecsize);  // results + PCRE workspace
  int matches = TryMatch(text, 0, anchor, vec, vecsize);
  assert(matches >= 0);  // TryMatch never returns negatives
  if (matches == 0)
    return false;

  *consumed = vec[1];

  if (n == 0 || args == NULL) {
    // We are not interested in results
    return true;
  }

  if (NumberOfCapturingGroups() < n) {
    // RE has fewer capturing groups than number of arg pointers passed in
    return false;
  }

  // If we got here, we must have matched the whole pattern.
  // We do not need (can not do) any more checks on the value of 'matches' here
  // -- see the comment for TryMatch.
  for (int i = 0; i < n; i++) {
    const int start = vec[2*(i+1)];
    const int limit = vec[2*(i+1)+1];
    if (!args[i]->Parse(text.data() + start, limit-start)) {
      // TODO: Should we indicate what the error was?
      return false;
    }
  }

  return true;
}

bool RE::DoMatch(const StringPiece& text,
                 Anchor anchor,
                 int* consumed,
                 const Arg* const args[],
                 int n) const {
  assert(n >= 0);
  size_t const vecsize = (1 + n) * 3;  // results + PCRE workspace
                                       // (as for kVecSize)
  int space[21];   // use stack allocation for small vecsize (common case)
  int* vec = vecsize <= 21 ? space : new int[vecsize];
  bool retval = DoMatchImpl(text, anchor, consumed, args, n, vec, vecsize);
  if (vec != space) delete [] vec;
  return retval;
}

bool RE::Rewrite(string *out, const StringPiece &rewrite,
                 const StringPiece &text, int *vec, int veclen) const {
  for (const char *s = rewrite.data(), *end = s + rewrite.size();
       s < end; s++) {
    int c = *s;
    if (c == '\\') {
      c = *++s;
      if (isdigit(c)) {
        int n = (c - '0');
        if (n >= veclen) {
          //fprintf(stderr, requested group %d in regexp %.*s\n",
          //        n, rewrite.size(), rewrite.data());
          return false;
        }
        int start = vec[2 * n];
        if (start >= 0)
          out->append(text.data() + start, vec[2 * n + 1] - start);
      } else if (c == '\\') {
        out->push_back('\\');
      } else {
        //fprintf(stderr, "invalid rewrite pattern: %.*s\n",
        //        rewrite.size(), rewrite.data());
        return false;
      }
    } else {
      out->push_back(c);
    }
  }
  return true;
}

// Return the number of capturing subpatterns, or -1 if the
// regexp wasn't valid on construction.
int RE::NumberOfCapturingGroups() const {
  if (re_partial_ == NULL) return -1;

  int result;
  int pcre_retval = pcre_fullinfo(re_partial_,  // The regular expression object
                                  NULL,         // We did not study the pattern
                                  PCRE_INFO_CAPTURECOUNT,
                                  &result);
  assert(pcre_retval == 0);
  return result;
}

/***** Parsers for various types *****/

bool Arg::parse_null(const char* str, int n, void* dest) {
  // We fail if somebody asked us to store into a non-NULL void* pointer
  return (dest == NULL);
}

bool Arg::parse_string(const char* str, int n, void* dest) {
  reinterpret_cast<string*>(dest)->assign(str, n);
  return true;
}

bool Arg::parse_stringpiece(const char* str, int n, void* dest) {
  reinterpret_cast<StringPiece*>(dest)->set(str, n);
  return true;
}

bool Arg::parse_char(const char* str, int n, void* dest) {
  if (n != 1) return false;
  *(reinterpret_cast<char*>(dest)) = str[0];
  return true;
}

bool Arg::parse_uchar(const char* str, int n, void* dest) {
  if (n != 1) return false;
  *(reinterpret_cast<unsigned char*>(dest)) = str[0];
  return true;
}

// Largest number spec that we are willing to parse
static const int kMaxNumberLength = 32;

// REQUIRES "buf" must have length at least kMaxNumberLength+1
// REQUIRES "n > 0"
// Copies "str" into "buf" and null-terminates if necessary.
// Returns one of:
//      a. "str" if no termination is needed
//      b. "buf" if the string was copied and null-terminated
//      c. "" if the input was invalid and has no hope of being parsed
static const char* TerminateNumber(char* buf, const char* str, int n) {
  if ((n > 0) && isspace(*str)) {
    // We are less forgiving than the strtoxxx() routines and do not
    // allow leading spaces.
    return "";
  }

  // See if the character right after the input text may potentially
  // look like a digit.
  if (isdigit(str[n]) ||
      ((str[n] >= 'a') && (str[n] <= 'f')) ||
      ((str[n] >= 'A') && (str[n] <= 'F'))) {
    if (n > kMaxNumberLength) return ""; // Input too big to be a valid number
    memcpy(buf, str, n);
    buf[n] = '\0';
    return buf;
  } else {
    // We can parse right out of the supplied string, so return it.
    return str;
  }
}

bool Arg::parse_long_radix(const char* str,
                           int n,
                           void* dest,
                           int radix) {
  if (n == 0) return false;
  char buf[kMaxNumberLength+1];
  str = TerminateNumber(buf, str, n);
  char* end;
  errno = 0;
  long r = strtol(str, &end, radix);
  if (end != str + n) return false;   // Leftover junk
  if (errno) return false;
  *(reinterpret_cast<long*>(dest)) = r;
  return true;
}

bool Arg::parse_ulong_radix(const char* str,
                            int n,
                            void* dest,
                            int radix) {
  if (n == 0) return false;
  char buf[kMaxNumberLength+1];
  str = TerminateNumber(buf, str, n);
  if (str[0] == '-') return false;    // strtoul() on a negative number?!
  char* end;
  errno = 0;
  unsigned long r = strtoul(str, &end, radix);
  if (end != str + n) return false;   // Leftover junk
  if (errno) return false;
  *(reinterpret_cast<unsigned long*>(dest)) = r;
  return true;
}

bool Arg::parse_short_radix(const char* str,
                            int n,
                            void* dest,
                            int radix) {
  long r;
  if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
  if (r < SHRT_MIN || r > SHRT_MAX) return false;       // Out of range
  *(reinterpret_cast<short*>(dest)) = r;
  return true;
}

bool Arg::parse_ushort_radix(const char* str,
                             int n,
                             void* dest,
                             int radix) {
  unsigned long r;
  if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
  if (r > USHRT_MAX) return false;                      // Out of range
  *(reinterpret_cast<unsigned short*>(dest)) = r;
  return true;
}

bool Arg::parse_int_radix(const char* str,
                          int n,
                          void* dest,
                          int radix) {
  long r;
  if (!parse_long_radix(str, n, &r, radix)) return false; // Could not parse
  if (r < INT_MIN || r > INT_MAX) return false;         // Out of range
  *(reinterpret_cast<int*>(dest)) = r;
  return true;
}

bool Arg::parse_uint_radix(const char* str,
                           int n,
                           void* dest,
                           int radix) {
  unsigned long r;
  if (!parse_ulong_radix(str, n, &r, radix)) return false; // Could not parse
  if (r > UINT_MAX) return false;                       // Out of range
  *(reinterpret_cast<unsigned int*>(dest)) = r;
  return true;
}

bool Arg::parse_longlong_radix(const char* str,
                               int n,
                               void* dest,
                               int radix) {
#ifndef HAVE_LONG_LONG
  return false;
#else
  if (n == 0) return false;
  char buf[kMaxNumberLength+1];
  str = TerminateNumber(buf, str, n);
  char* end;
  errno = 0;
#if defined HAVE_STRTOQ
  long long r = strtoq(str, &end, radix);
#elif defined HAVE_STRTOLL
  long long r = strtoll(str, &end, radix);
#else
#error parse_longlong_radix: cannot convert input to a long-long
#endif
  if (end != str + n) return false;   // Leftover junk
  if (errno) return false;
  *(reinterpret_cast<long long*>(dest)) = r;
  return true;
#endif   /* HAVE_LONG_LONG */
}

bool Arg::parse_ulonglong_radix(const char* str,
                                int n,
                                void* dest,
                                int radix) {
#ifndef HAVE_UNSIGNED_LONG_LONG
  return false;
#else
  if (n == 0) return false;
  char buf[kMaxNumberLength+1];
  str = TerminateNumber(buf, str, n);
  if (str[0] == '-') return false;    // strtoull() on a negative number?!
  char* end;
  errno = 0;
#if defined HAVE_STRTOQ
  unsigned long long r = strtouq(str, &end, radix);
#elif defined HAVE_STRTOLL
  unsigned long long r = strtoull(str, &end, radix);
#else
#error parse_ulonglong_radix: cannot convert input to a long-long
#endif
  if (end != str + n) return false;   // Leftover junk
  if (errno) return false;
  *(reinterpret_cast<unsigned long long*>(dest)) = r;
  return true;
#endif   /* HAVE_UNSIGNED_LONG_LONG */
}

bool Arg::parse_double(const char* str, int n, void* dest) {
  if (n == 0) return false;
  static const int kMaxLength = 200;
  char buf[kMaxLength];
  if (n >= kMaxLength) return false;
  memcpy(buf, str, n);
  buf[n] = '\0';
  errno = 0;
  char* end;
  double r = strtod(buf, &end);
  if (end != buf + n) return false;   // Leftover junk
  if (errno) return false;
  *(reinterpret_cast<double*>(dest)) = r;
  return true;
}

bool Arg::parse_float(const char* str, int n, void* dest) {
  double r;
  if (!parse_double(str, n, &r)) return false;
  *(reinterpret_cast<float*>(dest)) = static_cast<float>(r);
  return true;
}


#define DEFINE_INTEGER_PARSERS(name)                                    \
  bool Arg::parse_##name(const char* str, int n, void* dest) {          \
    return parse_##name##_radix(str, n, dest, 10);                      \
  }                                                                     \
  bool Arg::parse_##name##_hex(const char* str, int n, void* dest) {    \
    return parse_##name##_radix(str, n, dest, 16);                      \
  }                                                                     \
  bool Arg::parse_##name##_octal(const char* str, int n, void* dest) {  \
    return parse_##name##_radix(str, n, dest, 8);                       \
  }                                                                     \
  bool Arg::parse_##name##_cradix(const char* str, int n, void* dest) { \
    return parse_##name##_radix(str, n, dest, 0);                       \
  }

DEFINE_INTEGER_PARSERS(short)      /*                                   */
DEFINE_INTEGER_PARSERS(ushort)     /*                                   */
DEFINE_INTEGER_PARSERS(int)        /* Don't use semicolons after these  */
DEFINE_INTEGER_PARSERS(uint)       /* statements because they can cause */
DEFINE_INTEGER_PARSERS(long)       /* compiler warnings if the checking */
DEFINE_INTEGER_PARSERS(ulong)      /* level is turned up high enough.   */
DEFINE_INTEGER_PARSERS(longlong)   /*                                   */
DEFINE_INTEGER_PARSERS(ulonglong)  /*                                   */

#undef DEFINE_INTEGER_PARSERS

}   // namespace pcrecpp