TGLexer.cpp 30.7 KB
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
//===- TGLexer.cpp - Lexer for TableGen -----------------------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// Implement the Lexer for TableGen.
//
//===----------------------------------------------------------------------===//

#include "TGLexer.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/ADT/Twine.h"
#include "llvm/Config/config.h" // for strtoull()/strtoll() define
#include "llvm/Support/Compiler.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/TableGen/Error.h"
#include <algorithm>
#include <cctype>
#include <cerrno>
#include <cstdint>
#include <cstdio>
#include <cstdlib>
#include <cstring>

using namespace llvm;

namespace {
// A list of supported preprocessing directives with their
// internal token kinds and names.
struct {
  tgtok::TokKind Kind;
  const char *Word;
} PreprocessorDirs[] = {
  { tgtok::Ifdef, "ifdef" },
  { tgtok::Ifndef, "ifndef" },
  { tgtok::Else, "else" },
  { tgtok::Endif, "endif" },
  { tgtok::Define, "define" }
};
} // end anonymous namespace

TGLexer::TGLexer(SourceMgr &SM, ArrayRef<std::string> Macros) : SrcMgr(SM) {
  CurBuffer = SrcMgr.getMainFileID();
  CurBuf = SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer();
  CurPtr = CurBuf.begin();
  TokStart = nullptr;

  // Pretend that we enter the "top-level" include file.
  PrepIncludeStack.push_back(
      std::make_unique<std::vector<PreprocessorControlDesc>>());

  // Put all macros defined in the command line into the DefinedMacros set.
  std::for_each(Macros.begin(), Macros.end(),
                [this](const std::string &MacroName) {
                  DefinedMacros.insert(MacroName);
                });
}

SMLoc TGLexer::getLoc() const {
  return SMLoc::getFromPointer(TokStart);
}

/// ReturnError - Set the error to the specified string at the specified
/// location.  This is defined to always return tgtok::Error.
tgtok::TokKind TGLexer::ReturnError(SMLoc Loc, const Twine &Msg) {
  PrintError(Loc, Msg);
  return tgtok::Error;
}

tgtok::TokKind TGLexer::ReturnError(const char *Loc, const Twine &Msg) {
  return ReturnError(SMLoc::getFromPointer(Loc), Msg);
}

bool TGLexer::processEOF() {
  SMLoc ParentIncludeLoc = SrcMgr.getParentIncludeLoc(CurBuffer);
  if (ParentIncludeLoc != SMLoc()) {
    // If prepExitInclude() detects a problem with the preprocessing
    // control stack, it will return false.  Pretend that we reached
    // the final EOF and stop lexing more tokens by returning false
    // to LexToken().
    if (!prepExitInclude(false))
      return false;

    CurBuffer = SrcMgr.FindBufferContainingLoc(ParentIncludeLoc);
    CurBuf = SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer();
    CurPtr = ParentIncludeLoc.getPointer();
    // Make sure TokStart points into the parent file's buffer.
    // LexToken() assigns to it before calling getNextChar(),
    // so it is pointing into the included file now.
    TokStart = CurPtr;
    return true;
  }

  // Pretend that we exit the "top-level" include file.
  // Note that in case of an error (e.g. control stack imbalance)
  // the routine will issue a fatal error.
  prepExitInclude(true);
  return false;
}

int TGLexer::getNextChar() {
  char CurChar = *CurPtr++;
  switch (CurChar) {
  default:
    return (unsigned char)CurChar;
  case 0: {
    // A nul character in the stream is either the end of the current buffer or
    // a random nul in the file.  Disambiguate that here.
    if (CurPtr-1 != CurBuf.end())
      return 0;  // Just whitespace.

    // Otherwise, return end of file.
    --CurPtr;  // Another call to lex will return EOF again.
    return EOF;
  }
  case '\n':
  case '\r':
    // Handle the newline character by ignoring it and incrementing the line
    // count.  However, be careful about 'dos style' files with \n\r in them.
    // Only treat a \n\r or \r\n as a single line.
    if ((*CurPtr == '\n' || (*CurPtr == '\r')) &&
        *CurPtr != CurChar)
      ++CurPtr;  // Eat the two char newline sequence.
    return '\n';
  }
}

int TGLexer::peekNextChar(int Index) const {
  return *(CurPtr + Index);
}

tgtok::TokKind TGLexer::LexToken(bool FileOrLineStart) {
  TokStart = CurPtr;
  // This always consumes at least one character.
  int CurChar = getNextChar();

  switch (CurChar) {
  default:
    // Handle letters: [a-zA-Z_]
    if (isalpha(CurChar) || CurChar == '_')
      return LexIdentifier();

    // Unknown character, emit an error.
    return ReturnError(TokStart, "Unexpected character");
  case EOF:
    // Lex next token, if we just left an include file.
    // Note that leaving an include file means that the next
    // symbol is located at the end of 'include "..."'
    // construct, so LexToken() is called with default
    // false parameter.
    if (processEOF())
      return LexToken();

    // Return EOF denoting the end of lexing.
    return tgtok::Eof;

  case ':': return tgtok::colon;
  case ';': return tgtok::semi;
  case ',': return tgtok::comma;
  case '<': return tgtok::less;
  case '>': return tgtok::greater;
  case ']': return tgtok::r_square;
  case '{': return tgtok::l_brace;
  case '}': return tgtok::r_brace;
  case '(': return tgtok::l_paren;
  case ')': return tgtok::r_paren;
  case '=': return tgtok::equal;
  case '?': return tgtok::question;
  case '#':
    if (FileOrLineStart) {
      tgtok::TokKind Kind = prepIsDirective();
      if (Kind != tgtok::Error)
        return lexPreprocessor(Kind);
    }

    return tgtok::paste;

  // The period is a separate case so we can recognize the "..."
  // range punctuator.
  case '.':
    if (peekNextChar(0) == '.') {
      ++CurPtr; // Eat second dot.
      if (peekNextChar(0) == '.') {
        ++CurPtr; // Eat third dot.
        return tgtok::dotdotdot;
      }
      return ReturnError(TokStart, "Invalid '..' punctuation");
    }
    return tgtok::dot;

  case '\r':
    PrintFatalError("getNextChar() must never return '\r'");
    return tgtok::Error;

  case 0:
  case ' ':
  case '\t':
    // Ignore whitespace.
    return LexToken(FileOrLineStart);
  case '\n':
    // Ignore whitespace, and identify the new line.
    return LexToken(true);
  case '/':
    // If this is the start of a // comment, skip until the end of the line or
    // the end of the buffer.
    if (*CurPtr == '/')
      SkipBCPLComment();
    else if (*CurPtr == '*') {
      if (SkipCComment())
        return tgtok::Error;
    } else // Otherwise, this is an error.
      return ReturnError(TokStart, "Unexpected character");
    return LexToken(FileOrLineStart);
  case '-': case '+':
  case '0': case '1': case '2': case '3': case '4': case '5': case '6':
  case '7': case '8': case '9': {
    int NextChar = 0;
    if (isdigit(CurChar)) {
      // Allow identifiers to start with a number if it is followed by
      // an identifier.  This can happen with paste operations like
      // foo#8i.
      int i = 0;
      do {
        NextChar = peekNextChar(i++);
      } while (isdigit(NextChar));

      if (NextChar == 'x' || NextChar == 'b') {
        // If this is [0-9]b[01] or [0-9]x[0-9A-fa-f] this is most
        // likely a number.
        int NextNextChar = peekNextChar(i);
        switch (NextNextChar) {
        default:
          break;
        case '0': case '1':
          if (NextChar == 'b')
            return LexNumber();
          LLVM_FALLTHROUGH;
        case '2': case '3': case '4': case '5':
        case '6': case '7': case '8': case '9':
        case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
        case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
          if (NextChar == 'x')
            return LexNumber();
          break;
        }
      }
    }

    if (isalpha(NextChar) || NextChar == '_')
      return LexIdentifier();

    return LexNumber();
  }
  case '"': return LexString();
  case '$': return LexVarName();
  case '[': return LexBracket();
  case '!': return LexExclaim();
  }
}

/// LexString - Lex "[^"]*"
tgtok::TokKind TGLexer::LexString() {
  const char *StrStart = CurPtr;

  CurStrVal = "";

  while (*CurPtr != '"') {
    // If we hit the end of the buffer, report an error.
    if (*CurPtr == 0 && CurPtr == CurBuf.end())
      return ReturnError(StrStart, "End of file in string literal");

    if (*CurPtr == '\n' || *CurPtr == '\r')
      return ReturnError(StrStart, "End of line in string literal");

    if (*CurPtr != '\\') {
      CurStrVal += *CurPtr++;
      continue;
    }

    ++CurPtr;

    switch (*CurPtr) {
    case '\\': case '\'': case '"':
      // These turn into their literal character.
      CurStrVal += *CurPtr++;
      break;
    case 't':
      CurStrVal += '\t';
      ++CurPtr;
      break;
    case 'n':
      CurStrVal += '\n';
      ++CurPtr;
      break;

    case '\n':
    case '\r':
      return ReturnError(CurPtr, "escaped newlines not supported in tblgen");

    // If we hit the end of the buffer, report an error.
    case '\0':
      if (CurPtr == CurBuf.end())
        return ReturnError(StrStart, "End of file in string literal");
      LLVM_FALLTHROUGH;
    default:
      return ReturnError(CurPtr, "invalid escape in string literal");
    }
  }

  ++CurPtr;
  return tgtok::StrVal;
}

tgtok::TokKind TGLexer::LexVarName() {
  if (!isalpha(CurPtr[0]) && CurPtr[0] != '_')
    return ReturnError(TokStart, "Invalid variable name");

  // Otherwise, we're ok, consume the rest of the characters.
  const char *VarNameStart = CurPtr++;

  while (isalpha(*CurPtr) || isdigit(*CurPtr) || *CurPtr == '_')
    ++CurPtr;

  CurStrVal.assign(VarNameStart, CurPtr);
  return tgtok::VarName;
}

tgtok::TokKind TGLexer::LexIdentifier() {
  // The first letter is [a-zA-Z_].
  const char *IdentStart = TokStart;

  // Match the rest of the identifier regex: [0-9a-zA-Z_]*
  while (isalpha(*CurPtr) || isdigit(*CurPtr) || *CurPtr == '_')
    ++CurPtr;

  // Check to see if this identifier is a keyword.
  StringRef Str(IdentStart, CurPtr-IdentStart);

  if (Str == "include") {
    if (LexInclude()) return tgtok::Error;
    return Lex();
  }

  tgtok::TokKind Kind = StringSwitch<tgtok::TokKind>(Str)
    .Case("int", tgtok::Int)
    .Case("bit", tgtok::Bit)
    .Case("bits", tgtok::Bits)
    .Case("string", tgtok::String)
    .Case("list", tgtok::List)
    .Case("code", tgtok::Code)
    .Case("dag", tgtok::Dag)
    .Case("class", tgtok::Class)
    .Case("def", tgtok::Def)
    .Case("foreach", tgtok::Foreach)
    .Case("defm", tgtok::Defm)
    .Case("defset", tgtok::Defset)
    .Case("multiclass", tgtok::MultiClass)
    .Case("field", tgtok::Field)
    .Case("let", tgtok::Let)
    .Case("in", tgtok::In)
    .Case("defvar", tgtok::Defvar)
    .Case("if", tgtok::If)
    .Case("then", tgtok::Then)
    .Case("else", tgtok::ElseKW)
    .Default(tgtok::Id);

  if (Kind == tgtok::Id)
    CurStrVal.assign(Str.begin(), Str.end());
  return Kind;
}

/// LexInclude - We just read the "include" token.  Get the string token that
/// comes next and enter the include.
bool TGLexer::LexInclude() {
  // The token after the include must be a string.
  tgtok::TokKind Tok = LexToken();
  if (Tok == tgtok::Error) return true;
  if (Tok != tgtok::StrVal) {
    PrintError(getLoc(), "Expected filename after include");
    return true;
  }

  // Get the string.
  std::string Filename = CurStrVal;
  std::string IncludedFile;

  CurBuffer = SrcMgr.AddIncludeFile(Filename, SMLoc::getFromPointer(CurPtr),
                                    IncludedFile);
  if (!CurBuffer) {
    PrintError(getLoc(), "Could not find include file '" + Filename + "'");
    return true;
  }

  Dependencies.insert(IncludedFile);
  // Save the line number and lex buffer of the includer.
  CurBuf = SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer();
  CurPtr = CurBuf.begin();

  PrepIncludeStack.push_back(
      std::make_unique<std::vector<PreprocessorControlDesc>>());
  return false;
}

void TGLexer::SkipBCPLComment() {
  ++CurPtr;  // skip the second slash.
  while (true) {
    switch (*CurPtr) {
    case '\n':
    case '\r':
      return;  // Newline is end of comment.
    case 0:
      // If this is the end of the buffer, end the comment.
      if (CurPtr == CurBuf.end())
        return;
      break;
    }
    // Otherwise, skip the character.
    ++CurPtr;
  }
}

/// SkipCComment - This skips C-style /**/ comments.  The only difference from C
/// is that we allow nesting.
bool TGLexer::SkipCComment() {
  ++CurPtr;  // skip the star.
  unsigned CommentDepth = 1;

  while (true) {
    int CurChar = getNextChar();
    switch (CurChar) {
    case EOF:
      PrintError(TokStart, "Unterminated comment!");
      return true;
    case '*':
      // End of the comment?
      if (CurPtr[0] != '/') break;

      ++CurPtr;   // End the */.
      if (--CommentDepth == 0)
        return false;
      break;
    case '/':
      // Start of a nested comment?
      if (CurPtr[0] != '*') break;
      ++CurPtr;
      ++CommentDepth;
      break;
    }
  }
}

/// LexNumber - Lex:
///    [-+]?[0-9]+
///    0x[0-9a-fA-F]+
///    0b[01]+
tgtok::TokKind TGLexer::LexNumber() {
  if (CurPtr[-1] == '0') {
    if (CurPtr[0] == 'x') {
      ++CurPtr;
      const char *NumStart = CurPtr;
      while (isxdigit(CurPtr[0]))
        ++CurPtr;

      // Requires at least one hex digit.
      if (CurPtr == NumStart)
        return ReturnError(TokStart, "Invalid hexadecimal number");

      errno = 0;
      CurIntVal = strtoll(NumStart, nullptr, 16);
      if (errno == EINVAL)
        return ReturnError(TokStart, "Invalid hexadecimal number");
      if (errno == ERANGE) {
        errno = 0;
        CurIntVal = (int64_t)strtoull(NumStart, nullptr, 16);
        if (errno == EINVAL)
          return ReturnError(TokStart, "Invalid hexadecimal number");
        if (errno == ERANGE)
          return ReturnError(TokStart, "Hexadecimal number out of range");
      }
      return tgtok::IntVal;
    } else if (CurPtr[0] == 'b') {
      ++CurPtr;
      const char *NumStart = CurPtr;
      while (CurPtr[0] == '0' || CurPtr[0] == '1')
        ++CurPtr;

      // Requires at least one binary digit.
      if (CurPtr == NumStart)
        return ReturnError(CurPtr-2, "Invalid binary number");
      CurIntVal = strtoll(NumStart, nullptr, 2);
      return tgtok::BinaryIntVal;
    }
  }

  // Check for a sign without a digit.
  if (!isdigit(CurPtr[0])) {
    if (CurPtr[-1] == '-')
      return tgtok::minus;
    else if (CurPtr[-1] == '+')
      return tgtok::plus;
  }

  while (isdigit(CurPtr[0]))
    ++CurPtr;
  CurIntVal = strtoll(TokStart, nullptr, 10);
  return tgtok::IntVal;
}

/// LexBracket - We just read '['.  If this is a code block, return it,
/// otherwise return the bracket.  Match: '[' and '[{ ( [^}]+ | }[^]] )* }]'
tgtok::TokKind TGLexer::LexBracket() {
  if (CurPtr[0] != '{')
    return tgtok::l_square;
  ++CurPtr;
  const char *CodeStart = CurPtr;
  while (true) {
    int Char = getNextChar();
    if (Char == EOF) break;

    if (Char != '}') continue;

    Char = getNextChar();
    if (Char == EOF) break;
    if (Char == ']') {
      CurStrVal.assign(CodeStart, CurPtr-2);
      return tgtok::CodeFragment;
    }
  }

  return ReturnError(CodeStart-2, "Unterminated Code Block");
}

/// LexExclaim - Lex '!' and '![a-zA-Z]+'.
tgtok::TokKind TGLexer::LexExclaim() {
  if (!isalpha(*CurPtr))
    return ReturnError(CurPtr - 1, "Invalid \"!operator\"");

  const char *Start = CurPtr++;
  while (isalpha(*CurPtr))
    ++CurPtr;

  // Check to see which operator this is.
  tgtok::TokKind Kind =
    StringSwitch<tgtok::TokKind>(StringRef(Start, CurPtr - Start))
    .Case("eq", tgtok::XEq)
    .Case("ne", tgtok::XNe)
    .Case("le", tgtok::XLe)
    .Case("lt", tgtok::XLt)
    .Case("ge", tgtok::XGe)
    .Case("gt", tgtok::XGt)
    .Case("if", tgtok::XIf)
    .Case("cond", tgtok::XCond)
    .Case("isa", tgtok::XIsA)
    .Case("head", tgtok::XHead)
    .Case("tail", tgtok::XTail)
    .Case("size", tgtok::XSize)
    .Case("con", tgtok::XConcat)
    .Case("dag", tgtok::XDag)
    .Case("add", tgtok::XADD)
    .Case("mul", tgtok::XMUL)
    .Case("and", tgtok::XAND)
    .Case("or", tgtok::XOR)
    .Case("shl", tgtok::XSHL)
    .Case("sra", tgtok::XSRA)
    .Case("srl", tgtok::XSRL)
    .Case("cast", tgtok::XCast)
    .Case("empty", tgtok::XEmpty)
    .Case("subst", tgtok::XSubst)
    .Case("foldl", tgtok::XFoldl)
    .Case("foreach", tgtok::XForEach)
    .Case("listconcat", tgtok::XListConcat)
    .Case("listsplat", tgtok::XListSplat)
    .Case("strconcat", tgtok::XStrConcat)
    .Case("setop", tgtok::XSetOp)
    .Case("getop", tgtok::XGetOp)
    .Default(tgtok::Error);

  return Kind != tgtok::Error ? Kind : ReturnError(Start-1, "Unknown operator");
}

bool TGLexer::prepExitInclude(bool IncludeStackMustBeEmpty) {
  // Report an error, if preprocessor control stack for the current
  // file is not empty.
  if (!PrepIncludeStack.back()->empty()) {
    prepReportPreprocessorStackError();

    return false;
  }

  // Pop the preprocessing controls from the include stack.
  if (PrepIncludeStack.empty()) {
    PrintFatalError("Preprocessor include stack is empty");
  }

  PrepIncludeStack.pop_back();

  if (IncludeStackMustBeEmpty) {
    if (!PrepIncludeStack.empty())
      PrintFatalError("Preprocessor include stack is not empty");
  } else {
    if (PrepIncludeStack.empty())
      PrintFatalError("Preprocessor include stack is empty");
  }

  return true;
}

tgtok::TokKind TGLexer::prepIsDirective() const {
  for (unsigned ID = 0; ID < llvm::array_lengthof(PreprocessorDirs); ++ID) {
    int NextChar = *CurPtr;
    bool Match = true;
    unsigned I = 0;
    for (; I < strlen(PreprocessorDirs[ID].Word); ++I) {
      if (NextChar != PreprocessorDirs[ID].Word[I]) {
        Match = false;
        break;
      }

      NextChar = peekNextChar(I + 1);
    }

    // Check for whitespace after the directive.  If there is no whitespace,
    // then we do not recognize it as a preprocessing directive.
    if (Match) {
      tgtok::TokKind Kind = PreprocessorDirs[ID].Kind;

      // New line and EOF may follow only #else/#endif.  It will be reported
      // as an error for #ifdef/#define after the call to prepLexMacroName().
      if (NextChar == ' ' || NextChar == '\t' || NextChar == EOF ||
          NextChar == '\n' ||
          // It looks like TableGen does not support '\r' as the actual
          // carriage return, e.g. getNextChar() treats a single '\r'
          // as '\n'.  So we do the same here.
          NextChar == '\r')
        return Kind;

      // Allow comments after some directives, e.g.:
      //     #else// OR #else/**/
      //     #endif// OR #endif/**/
      //
      // Note that we do allow comments after #ifdef/#define here, e.g.
      //     #ifdef/**/ AND #ifdef//
      //     #define/**/ AND #define//
      //
      // These cases will be reported as incorrect after calling
      // prepLexMacroName().  We could have supported C-style comments
      // after #ifdef/#define, but this would complicate the code
      // for little benefit.
      if (NextChar == '/') {
        NextChar = peekNextChar(I + 1);

        if (NextChar == '*' || NextChar == '/')
          return Kind;

        // Pretend that we do not recognize the directive.
      }
    }
  }

  return tgtok::Error;
}

bool TGLexer::prepEatPreprocessorDirective(tgtok::TokKind Kind) {
  TokStart = CurPtr;

  for (unsigned ID = 0; ID < llvm::array_lengthof(PreprocessorDirs); ++ID)
    if (PreprocessorDirs[ID].Kind == Kind) {
      // Advance CurPtr to the end of the preprocessing word.
      CurPtr += strlen(PreprocessorDirs[ID].Word);
      return true;
    }

  PrintFatalError("Unsupported preprocessing token in "
                  "prepEatPreprocessorDirective()");
  return false;
}

tgtok::TokKind TGLexer::lexPreprocessor(
    tgtok::TokKind Kind, bool ReturnNextLiveToken) {

  // We must be looking at a preprocessing directive.  Eat it!
  if (!prepEatPreprocessorDirective(Kind))
    PrintFatalError("lexPreprocessor() called for unknown "
                    "preprocessor directive");

  if (Kind == tgtok::Ifdef || Kind == tgtok::Ifndef) {
    StringRef MacroName = prepLexMacroName();
    StringRef IfTokName = Kind == tgtok::Ifdef ? "#ifdef" : "#ifndef";
    if (MacroName.empty())
      return ReturnError(TokStart, "Expected macro name after " + IfTokName);

    bool MacroIsDefined = DefinedMacros.count(MacroName) != 0;

    // Canonicalize ifndef to ifdef equivalent
    if (Kind == tgtok::Ifndef) {
      MacroIsDefined = !MacroIsDefined;
      Kind = tgtok::Ifdef;
    }

    // Regardless of whether we are processing tokens or not,
    // we put the #ifdef control on stack.
    PrepIncludeStack.back()->push_back(
        {Kind, MacroIsDefined, SMLoc::getFromPointer(TokStart)});

    if (!prepSkipDirectiveEnd())
      return ReturnError(CurPtr, "Only comments are supported after " +
                                     IfTokName + " NAME");

    // If we were not processing tokens before this #ifdef,
    // then just return back to the lines skipping code.
    if (!ReturnNextLiveToken)
      return Kind;

    // If we were processing tokens before this #ifdef,
    // and the macro is defined, then just return the next token.
    if (MacroIsDefined)
      return LexToken();

    // We were processing tokens before this #ifdef, and the macro
    // is not defined, so we have to start skipping the lines.
    // If the skipping is successful, it will return the token following
    // either #else or #endif corresponding to this #ifdef.
    if (prepSkipRegion(ReturnNextLiveToken))
      return LexToken();

    return tgtok::Error;
  } else if (Kind == tgtok::Else) {
    // Check if this #else is correct before calling prepSkipDirectiveEnd(),
    // which will move CurPtr away from the beginning of #else.
    if (PrepIncludeStack.back()->empty())
      return ReturnError(TokStart, "#else without #ifdef or #ifndef");

    PreprocessorControlDesc IfdefEntry = PrepIncludeStack.back()->back();

    if (IfdefEntry.Kind != tgtok::Ifdef) {
      PrintError(TokStart, "double #else");
      return ReturnError(IfdefEntry.SrcPos, "Previous #else is here");
    }

    // Replace the corresponding #ifdef's control with its negation
    // on the control stack.
    PrepIncludeStack.back()->pop_back();
    PrepIncludeStack.back()->push_back(
        {Kind, !IfdefEntry.IsDefined, SMLoc::getFromPointer(TokStart)});

    if (!prepSkipDirectiveEnd())
      return ReturnError(CurPtr, "Only comments are supported after #else");

    // If we were processing tokens before this #else,
    // we have to start skipping lines until the matching #endif.
    if (ReturnNextLiveToken) {
      if (prepSkipRegion(ReturnNextLiveToken))
        return LexToken();

      return tgtok::Error;
    }

    // Return to the lines skipping code.
    return Kind;
  } else if (Kind == tgtok::Endif) {
    // Check if this #endif is correct before calling prepSkipDirectiveEnd(),
    // which will move CurPtr away from the beginning of #endif.
    if (PrepIncludeStack.back()->empty())
      return ReturnError(TokStart, "#endif without #ifdef");

    auto &IfdefOrElseEntry = PrepIncludeStack.back()->back();

    if (IfdefOrElseEntry.Kind != tgtok::Ifdef &&
        IfdefOrElseEntry.Kind != tgtok::Else) {
      PrintFatalError("Invalid preprocessor control on the stack");
      return tgtok::Error;
    }

    if (!prepSkipDirectiveEnd())
      return ReturnError(CurPtr, "Only comments are supported after #endif");

    PrepIncludeStack.back()->pop_back();

    // If we were processing tokens before this #endif, then
    // we should continue it.
    if (ReturnNextLiveToken) {
      return LexToken();
    }

    // Return to the lines skipping code.
    return Kind;
  } else if (Kind == tgtok::Define) {
    StringRef MacroName = prepLexMacroName();
    if (MacroName.empty())
      return ReturnError(TokStart, "Expected macro name after #define");

    if (!DefinedMacros.insert(MacroName).second)
      PrintWarning(getLoc(),
                   "Duplicate definition of macro: " + Twine(MacroName));

    if (!prepSkipDirectiveEnd())
      return ReturnError(CurPtr,
                         "Only comments are supported after #define NAME");

    if (!ReturnNextLiveToken) {
      PrintFatalError("#define must be ignored during the lines skipping");
      return tgtok::Error;
    }

    return LexToken();
  }

  PrintFatalError("Preprocessing directive is not supported");
  return tgtok::Error;
}

bool TGLexer::prepSkipRegion(bool MustNeverBeFalse) {
  if (!MustNeverBeFalse)
    PrintFatalError("Invalid recursion.");

  do {
    // Skip all symbols to the line end.
    prepSkipToLineEnd();

    // Find the first non-whitespace symbol in the next line(s).
    if (!prepSkipLineBegin())
      return false;

    // If the first non-blank/comment symbol on the line is '#',
    // it may be a start of preprocessing directive.
    //
    // If it is not '#' just go to the next line.
    if (*CurPtr == '#')
      ++CurPtr;
    else
      continue;

    tgtok::TokKind Kind = prepIsDirective();

    // If we did not find a preprocessing directive or it is #define,
    // then just skip to the next line.  We do not have to do anything
    // for #define in the line-skipping mode.
    if (Kind == tgtok::Error || Kind == tgtok::Define)
      continue;

    tgtok::TokKind ProcessedKind = lexPreprocessor(Kind, false);

    // If lexPreprocessor() encountered an error during lexing this
    // preprocessor idiom, then return false to the calling lexPreprocessor().
    // This will force tgtok::Error to be returned to the tokens processing.
    if (ProcessedKind == tgtok::Error)
      return false;

    if (Kind != ProcessedKind)
      PrintFatalError("prepIsDirective() and lexPreprocessor() "
                      "returned different token kinds");

    // If this preprocessing directive enables tokens processing,
    // then return to the lexPreprocessor() and get to the next token.
    // We can move from line-skipping mode to processing tokens only
    // due to #else or #endif.
    if (prepIsProcessingEnabled()) {
      if (Kind != tgtok::Else && Kind != tgtok::Endif) {
        PrintFatalError("Tokens processing was enabled by an unexpected "
                        "preprocessing directive");
        return false;
      }

      return true;
    }
  } while (CurPtr != CurBuf.end());

  // We have reached the end of the file, but never left the lines-skipping
  // mode.  This means there is no matching #endif.
  prepReportPreprocessorStackError();
  return false;
}

StringRef TGLexer::prepLexMacroName() {
  // Skip whitespaces between the preprocessing directive and the macro name.
  while (*CurPtr == ' ' || *CurPtr == '\t')
    ++CurPtr;

  TokStart = CurPtr;
  // Macro names start with [a-zA-Z_].
  if (*CurPtr != '_' && !isalpha(*CurPtr))
    return "";

  // Match the rest of the identifier regex: [0-9a-zA-Z_]*
  while (isalpha(*CurPtr) || isdigit(*CurPtr) || *CurPtr == '_')
    ++CurPtr;

  return StringRef(TokStart, CurPtr - TokStart);
}

bool TGLexer::prepSkipLineBegin() {
  while (CurPtr != CurBuf.end()) {
    switch (*CurPtr) {
    case ' ':
    case '\t':
    case '\n':
    case '\r':
      break;

    case '/': {
      int NextChar = peekNextChar(1);
      if (NextChar == '*') {
        // Skip C-style comment.
        // Note that we do not care about skipping the C++-style comments.
        // If the line contains "//", it may not contain any processable
        // preprocessing directive.  Just return CurPtr pointing to
        // the first '/' in this case.  We also do not care about
        // incorrect symbols after the first '/' - we are in lines-skipping
        // mode, so incorrect code is allowed to some extent.

        // Set TokStart to the beginning of the comment to enable proper
        // diagnostic printing in case of error in SkipCComment().
        TokStart = CurPtr;

        // CurPtr must point to '*' before call to SkipCComment().
        ++CurPtr;
        if (SkipCComment())
          return false;
      } else {
        // CurPtr points to the non-whitespace '/'.
        return true;
      }

      // We must not increment CurPtr after the comment was lexed.
      continue;
    }

    default:
      return true;
    }

    ++CurPtr;
  }

  // We have reached the end of the file.  Return to the lines skipping
  // code, and allow it to handle the EOF as needed.
  return true;
}

bool TGLexer::prepSkipDirectiveEnd() {
  while (CurPtr != CurBuf.end()) {
    switch (*CurPtr) {
    case ' ':
    case '\t':
      break;

    case '\n':
    case '\r':
      return true;

    case '/': {
      int NextChar = peekNextChar(1);
      if (NextChar == '/') {
        // Skip C++-style comment.
        // We may just return true now, but let's skip to the line/buffer end
        // to simplify the method specification.
        ++CurPtr;
        SkipBCPLComment();
      } else if (NextChar == '*') {
        // When we are skipping C-style comment at the end of a preprocessing
        // directive, we can skip several lines.  If any meaningful TD token
        // follows the end of the C-style comment on the same line, it will
        // be considered as an invalid usage of TD token.
        // For example, we want to forbid usages like this one:
        //     #define MACRO class Class {}
        // But with C-style comments we also disallow the following:
        //     #define MACRO /* This macro is used
        //                      to ... */ class Class {}
        // One can argue that this should be allowed, but it does not seem
        // to be worth of the complication.  Moreover, this matches
        // the C preprocessor behavior.

        // Set TokStart to the beginning of the comment to enable proper
        // diagnostic printer in case of error in SkipCComment().
        TokStart = CurPtr;
        ++CurPtr;
        if (SkipCComment())
          return false;
      } else {
        TokStart = CurPtr;
        PrintError(CurPtr, "Unexpected character");
        return false;
      }

      // We must not increment CurPtr after the comment was lexed.
      continue;
    }

    default:
      // Do not allow any non-whitespaces after the directive.
      TokStart = CurPtr;
      return false;
    }

    ++CurPtr;
  }

  return true;
}

void TGLexer::prepSkipToLineEnd() {
  while (*CurPtr != '\n' && *CurPtr != '\r' && CurPtr != CurBuf.end())
    ++CurPtr;
}

bool TGLexer::prepIsProcessingEnabled() {
  for (auto I = PrepIncludeStack.back()->rbegin(),
            E = PrepIncludeStack.back()->rend();
       I != E; ++I) {
    if (!I->IsDefined)
      return false;
  }

  return true;
}

void TGLexer::prepReportPreprocessorStackError() {
  if (PrepIncludeStack.back()->empty())
    PrintFatalError("prepReportPreprocessorStackError() called with "
                    "empty control stack");

  auto &PrepControl = PrepIncludeStack.back()->back();
  PrintError(CurBuf.end(), "Reached EOF without matching #endif");
  PrintError(PrepControl.SrcPos, "The latest preprocessor control is here");

  TokStart = CurPtr;
}