aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/clang14/lib/AST/CommentLexer.cpp
blob: 61ce8979f13f5af8aeb5f073fe8c33d9592ee8cd (plain) (blame)
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
//===--- CommentLexer.cpp -------------------------------------------------===//
//
// 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
//
//===----------------------------------------------------------------------===//

#include "clang/AST/CommentLexer.h"
#include "clang/AST/CommentCommandTraits.h"
#include "clang/AST/CommentDiagnostic.h"
#include "clang/Basic/CharInfo.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/Support/ConvertUTF.h"
#include "llvm/Support/ErrorHandling.h"

namespace clang {
namespace comments {

void Token::dump(const Lexer &L, const SourceManager &SM) const {
  llvm::errs() << "comments::Token Kind=" << Kind << " ";
  Loc.print(llvm::errs(), SM);
  llvm::errs() << " " << Length << " \"" << L.getSpelling(*this, SM) << "\"\n";
}

static inline bool isHTMLNamedCharacterReferenceCharacter(char C) {
  return isLetter(C);
}

static inline bool isHTMLDecimalCharacterReferenceCharacter(char C) {
  return isDigit(C);
}

static inline bool isHTMLHexCharacterReferenceCharacter(char C) {
  return isHexDigit(C);
}

static inline StringRef convertCodePointToUTF8(
                                      llvm::BumpPtrAllocator &Allocator,
                                      unsigned CodePoint) {
  char *Resolved = Allocator.Allocate<char>(UNI_MAX_UTF8_BYTES_PER_CODE_POINT);
  char *ResolvedPtr = Resolved;
  if (llvm::ConvertCodePointToUTF8(CodePoint, ResolvedPtr))
    return StringRef(Resolved, ResolvedPtr - Resolved);
  else
    return StringRef();
}

namespace {

#include "clang/AST/CommentHTMLTags.inc"
#include "clang/AST/CommentHTMLNamedCharacterReferences.inc"

} // end anonymous namespace

StringRef Lexer::resolveHTMLNamedCharacterReference(StringRef Name) const {
  // Fast path, first check a few most widely used named character references.
  return llvm::StringSwitch<StringRef>(Name)
      .Case("amp", "&")
      .Case("lt", "<")
      .Case("gt", ">")
      .Case("quot", "\"")
      .Case("apos", "\'")
      // Slow path.
      .Default(translateHTMLNamedCharacterReferenceToUTF8(Name));
}

StringRef Lexer::resolveHTMLDecimalCharacterReference(StringRef Name) const {
  unsigned CodePoint = 0;
  for (unsigned i = 0, e = Name.size(); i != e; ++i) {
    assert(isHTMLDecimalCharacterReferenceCharacter(Name[i]));
    CodePoint *= 10;
    CodePoint += Name[i] - '0';
  }
  return convertCodePointToUTF8(Allocator, CodePoint);
}

StringRef Lexer::resolveHTMLHexCharacterReference(StringRef Name) const {
  unsigned CodePoint = 0;
  for (unsigned i = 0, e = Name.size(); i != e; ++i) {
    CodePoint *= 16;
    const char C = Name[i];
    assert(isHTMLHexCharacterReferenceCharacter(C));
    CodePoint += llvm::hexDigitValue(C);
  }
  return convertCodePointToUTF8(Allocator, CodePoint);
}

void Lexer::skipLineStartingDecorations() {
  // This function should be called only for C comments
  assert(CommentState == LCS_InsideCComment);

  if (BufferPtr == CommentEnd)
    return;

  const char *NewBufferPtr = BufferPtr;
  while (isHorizontalWhitespace(*NewBufferPtr))
    if (++NewBufferPtr == CommentEnd)
      return;
  if (*NewBufferPtr == '*')
    BufferPtr = NewBufferPtr + 1;
}

namespace {
/// Returns pointer to the first newline character in the string.
const char *findNewline(const char *BufferPtr, const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (isVerticalWhitespace(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

const char *skipNewline(const char *BufferPtr, const char *BufferEnd) {
  if (BufferPtr == BufferEnd)
    return BufferPtr;

  if (*BufferPtr == '\n')
    BufferPtr++;
  else {
    assert(*BufferPtr == '\r');
    BufferPtr++;
    if (BufferPtr != BufferEnd && *BufferPtr == '\n')
      BufferPtr++;
  }
  return BufferPtr;
}

const char *skipNamedCharacterReference(const char *BufferPtr,
                                        const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isHTMLNamedCharacterReferenceCharacter(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

const char *skipDecimalCharacterReference(const char *BufferPtr,
                                          const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isHTMLDecimalCharacterReferenceCharacter(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

const char *skipHexCharacterReference(const char *BufferPtr,
                                      const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isHTMLHexCharacterReferenceCharacter(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

bool isHTMLIdentifierStartingCharacter(char C) {
  return isLetter(C);
}

bool isHTMLIdentifierCharacter(char C) {
  return isAlphanumeric(C);
}

const char *skipHTMLIdentifier(const char *BufferPtr, const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isHTMLIdentifierCharacter(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

/// Skip HTML string quoted in single or double quotes.  Escaping quotes inside
/// string allowed.
///
/// Returns pointer to closing quote.
const char *skipHTMLQuotedString(const char *BufferPtr, const char *BufferEnd)
{
  const char Quote = *BufferPtr;
  assert(Quote == '\"' || Quote == '\'');

  BufferPtr++;
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    const char C = *BufferPtr;
    if (C == Quote && BufferPtr[-1] != '\\')
      return BufferPtr;
  }
  return BufferEnd;
}

const char *skipWhitespace(const char *BufferPtr, const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isWhitespace(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

bool isWhitespace(const char *BufferPtr, const char *BufferEnd) {
  return skipWhitespace(BufferPtr, BufferEnd) == BufferEnd;
}

bool isCommandNameStartCharacter(char C) {
  return isLetter(C);
}

bool isCommandNameCharacter(char C) {
  return isAlphanumeric(C);
}

const char *skipCommandName(const char *BufferPtr, const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (!isCommandNameCharacter(*BufferPtr))
      return BufferPtr;
  }
  return BufferEnd;
}

/// Return the one past end pointer for BCPL comments.
/// Handles newlines escaped with backslash or trigraph for backslahs.
const char *findBCPLCommentEnd(const char *BufferPtr, const char *BufferEnd) {
  const char *CurPtr = BufferPtr;
  while (CurPtr != BufferEnd) {
    while (!isVerticalWhitespace(*CurPtr)) {
      CurPtr++;
      if (CurPtr == BufferEnd)
        return BufferEnd;
    }
    // We found a newline, check if it is escaped.
    const char *EscapePtr = CurPtr - 1;
    while(isHorizontalWhitespace(*EscapePtr))
      EscapePtr--;

    if (*EscapePtr == '\\' ||
        (EscapePtr - 2 >= BufferPtr && EscapePtr[0] == '/' &&
         EscapePtr[-1] == '?' && EscapePtr[-2] == '?')) {
      // We found an escaped newline.
      CurPtr = skipNewline(CurPtr, BufferEnd);
    } else
      return CurPtr; // Not an escaped newline.
  }
  return BufferEnd;
}

/// Return the one past end pointer for C comments.
/// Very dumb, does not handle escaped newlines or trigraphs.
const char *findCCommentEnd(const char *BufferPtr, const char *BufferEnd) {
  for ( ; BufferPtr != BufferEnd; ++BufferPtr) {
    if (*BufferPtr == '*') {
      assert(BufferPtr + 1 != BufferEnd);
      if (*(BufferPtr + 1) == '/')
        return BufferPtr;
    }
  }
  llvm_unreachable("buffer end hit before '*/' was seen");
}

} // end anonymous namespace

void Lexer::formTokenWithChars(Token &Result, const char *TokEnd,
                               tok::TokenKind Kind) {
  const unsigned TokLen = TokEnd - BufferPtr;
  Result.setLocation(getSourceLocation(BufferPtr));
  Result.setKind(Kind);
  Result.setLength(TokLen);
#ifndef NDEBUG
  Result.TextPtr = "<UNSET>";
  Result.IntVal = 7;
#endif
  BufferPtr = TokEnd;
}

const char *Lexer::skipTextToken() {
  const char *TokenPtr = BufferPtr;
  assert(TokenPtr < CommentEnd);
  StringRef TokStartSymbols = ParseCommands ? "\n\r\\@\"&<" : "\n\r";

again:
  size_t End =
      StringRef(TokenPtr, CommentEnd - TokenPtr).find_first_of(TokStartSymbols);
  if (End == StringRef::npos)
    return CommentEnd;

  // Doxygen doesn't recognize any commands in a one-line double quotation.
  // If we don't find an ending quotation mark, we pretend it never began.
  if (*(TokenPtr + End) == '\"') {
    TokenPtr += End + 1;
    End = StringRef(TokenPtr, CommentEnd - TokenPtr).find_first_of("\n\r\"");
    if (End != StringRef::npos && *(TokenPtr + End) == '\"')
      TokenPtr += End + 1;
    goto again;
  }
  return TokenPtr + End;
}

void Lexer::lexCommentText(Token &T) {
  assert(CommentState == LCS_InsideBCPLComment ||
         CommentState == LCS_InsideCComment);

  // Handles lexing non-command text, i.e. text and newline.
  auto HandleNonCommandToken = [&]() -> void {
    assert(State == LS_Normal);

    const char *TokenPtr = BufferPtr;
    assert(TokenPtr < CommentEnd);
    switch (*TokenPtr) {
      case '\n':
      case '\r':
          TokenPtr = skipNewline(TokenPtr, CommentEnd);
          formTokenWithChars(T, TokenPtr, tok::newline);

          if (CommentState == LCS_InsideCComment)
            skipLineStartingDecorations();
          return;

      default:
        return formTextToken(T, skipTextToken());
    }
  };

  if (!ParseCommands)
    return HandleNonCommandToken();

  switch (State) {
  case LS_Normal:
    break;
  case LS_VerbatimBlockFirstLine:
    lexVerbatimBlockFirstLine(T);
    return;
  case LS_VerbatimBlockBody:
    lexVerbatimBlockBody(T);
    return;
  case LS_VerbatimLineText:
    lexVerbatimLineText(T);
    return;
  case LS_HTMLStartTag:
    lexHTMLStartTag(T);
    return;
  case LS_HTMLEndTag:
    lexHTMLEndTag(T);
    return;
  }

  assert(State == LS_Normal);
  const char *TokenPtr = BufferPtr;
  assert(TokenPtr < CommentEnd);
  switch(*TokenPtr) {
    case '\\':
    case '@': {
      // Commands that start with a backslash and commands that start with
      // 'at' have equivalent semantics.  But we keep information about the
      // exact syntax in AST for comments.
      tok::TokenKind CommandKind =
          (*TokenPtr == '@') ? tok::at_command : tok::backslash_command;
      TokenPtr++;
      if (TokenPtr == CommentEnd) {
        formTextToken(T, TokenPtr);
        return;
      }
      char C = *TokenPtr;
      switch (C) {
      default:
        break;

      case '\\': case '@': case '&': case '$':
      case '#':  case '<': case '>': case '%':
      case '\"': case '.': case ':':
        // This is one of \\ \@ \& \$ etc escape sequences.
        TokenPtr++;
        if (C == ':' && TokenPtr != CommentEnd && *TokenPtr == ':') {
          // This is the \:: escape sequence.
          TokenPtr++;
        }
        StringRef UnescapedText(BufferPtr + 1, TokenPtr - (BufferPtr + 1));
        formTokenWithChars(T, TokenPtr, tok::text);
        T.setText(UnescapedText);
        return;
      }

      // Don't make zero-length commands.
      if (!isCommandNameStartCharacter(*TokenPtr)) {
        formTextToken(T, TokenPtr);
        return;
      }

      TokenPtr = skipCommandName(TokenPtr, CommentEnd);
      unsigned Length = TokenPtr - (BufferPtr + 1);

      // Hardcoded support for lexing LaTeX formula commands
      // \f$ \f( \f) \f[ \f] \f{ \f} as a single command.
      if (Length == 1 && TokenPtr[-1] == 'f' && TokenPtr != CommentEnd) {
        C = *TokenPtr;
        if (C == '$' || C == '(' || C == ')' || C == '[' || C == ']' ||
            C == '{' || C == '}') {
          TokenPtr++;
          Length++;
        }
      }

      StringRef CommandName(BufferPtr + 1, Length);

      const CommandInfo *Info = Traits.getCommandInfoOrNULL(CommandName);
      if (!Info) {
        if ((Info = Traits.getTypoCorrectCommandInfo(CommandName))) {
          StringRef CorrectedName = Info->Name;
          SourceLocation Loc = getSourceLocation(BufferPtr);
          SourceLocation EndLoc = getSourceLocation(TokenPtr);
          SourceRange FullRange = SourceRange(Loc, EndLoc);
          SourceRange CommandRange(Loc.getLocWithOffset(1), EndLoc);
          Diag(Loc, diag::warn_correct_comment_command_name)
            << FullRange << CommandName << CorrectedName
            << FixItHint::CreateReplacement(CommandRange, CorrectedName);
        } else {
          formTokenWithChars(T, TokenPtr, tok::unknown_command);
          T.setUnknownCommandName(CommandName);
          Diag(T.getLocation(), diag::warn_unknown_comment_command_name)
              << SourceRange(T.getLocation(), T.getEndLocation());
          return;
        }
      }
      if (Info->IsVerbatimBlockCommand) {
        setupAndLexVerbatimBlock(T, TokenPtr, *BufferPtr, Info);
        return;
      }
      if (Info->IsVerbatimLineCommand) {
        setupAndLexVerbatimLine(T, TokenPtr, Info);
        return;
      }
      formTokenWithChars(T, TokenPtr, CommandKind);
      T.setCommandID(Info->getID());
      return;
    }

    case '&':
      lexHTMLCharacterReference(T);
      return;

    case '<': {
      TokenPtr++;
      if (TokenPtr == CommentEnd) {
        formTextToken(T, TokenPtr);
        return;
      }
      const char C = *TokenPtr;
      if (isHTMLIdentifierStartingCharacter(C))
        setupAndLexHTMLStartTag(T);
      else if (C == '/')
        setupAndLexHTMLEndTag(T);
      else
        formTextToken(T, TokenPtr);
      return;
    }

    default:
      return HandleNonCommandToken();
  }
}

void Lexer::setupAndLexVerbatimBlock(Token &T,
                                     const char *TextBegin,
                                     char Marker, const CommandInfo *Info) {
  assert(Info->IsVerbatimBlockCommand);

  VerbatimBlockEndCommandName.clear();
  VerbatimBlockEndCommandName.append(Marker == '\\' ? "\\" : "@");
  VerbatimBlockEndCommandName.append(Info->EndCommandName);

  formTokenWithChars(T, TextBegin, tok::verbatim_block_begin);
  T.setVerbatimBlockID(Info->getID());

  // If there is a newline following the verbatim opening command, skip the
  // newline so that we don't create an tok::verbatim_block_line with empty
  // text content.
  if (BufferPtr != CommentEnd &&
      isVerticalWhitespace(*BufferPtr)) {
    BufferPtr = skipNewline(BufferPtr, CommentEnd);
    State = LS_VerbatimBlockBody;
    return;
  }

  State = LS_VerbatimBlockFirstLine;
}

void Lexer::lexVerbatimBlockFirstLine(Token &T) {
again:
  assert(BufferPtr < CommentEnd);

  // FIXME: It would be better to scan the text once, finding either the block
  // end command or newline.
  //
  // Extract current line.
  const char *Newline = findNewline(BufferPtr, CommentEnd);
  StringRef Line(BufferPtr, Newline - BufferPtr);

  // Look for end command in current line.
  size_t Pos = Line.find(VerbatimBlockEndCommandName);
  const char *TextEnd;
  const char *NextLine;
  if (Pos == StringRef::npos) {
    // Current line is completely verbatim.
    TextEnd = Newline;
    NextLine = skipNewline(Newline, CommentEnd);
  } else if (Pos == 0) {
    // Current line contains just an end command.
    const char *End = BufferPtr + VerbatimBlockEndCommandName.size();
    StringRef Name(BufferPtr + 1, End - (BufferPtr + 1));
    formTokenWithChars(T, End, tok::verbatim_block_end);
    T.setVerbatimBlockID(Traits.getCommandInfo(Name)->getID());
    State = LS_Normal;
    return;
  } else {
    // There is some text, followed by end command.  Extract text first.
    TextEnd = BufferPtr + Pos;
    NextLine = TextEnd;
    // If there is only whitespace before end command, skip whitespace.
    if (isWhitespace(BufferPtr, TextEnd)) {
      BufferPtr = TextEnd;
      goto again;
    }
  }

  StringRef Text(BufferPtr, TextEnd - BufferPtr);
  formTokenWithChars(T, NextLine, tok::verbatim_block_line);
  T.setVerbatimBlockText(Text);

  State = LS_VerbatimBlockBody;
}

void Lexer::lexVerbatimBlockBody(Token &T) {
  assert(State == LS_VerbatimBlockBody);

  if (CommentState == LCS_InsideCComment)
    skipLineStartingDecorations();

  if (BufferPtr == CommentEnd) {
    formTokenWithChars(T, BufferPtr, tok::verbatim_block_line);
    T.setVerbatimBlockText("");
    return;
  }

  lexVerbatimBlockFirstLine(T);
}

void Lexer::setupAndLexVerbatimLine(Token &T, const char *TextBegin,
                                    const CommandInfo *Info) {
  assert(Info->IsVerbatimLineCommand);
  formTokenWithChars(T, TextBegin, tok::verbatim_line_name);
  T.setVerbatimLineID(Info->getID());

  State = LS_VerbatimLineText;
}

void Lexer::lexVerbatimLineText(Token &T) {
  assert(State == LS_VerbatimLineText);

  // Extract current line.
  const char *Newline = findNewline(BufferPtr, CommentEnd);
  StringRef Text(BufferPtr, Newline - BufferPtr);
  formTokenWithChars(T, Newline, tok::verbatim_line_text);
  T.setVerbatimLineText(Text);

  State = LS_Normal;
}

void Lexer::lexHTMLCharacterReference(Token &T) {
  const char *TokenPtr = BufferPtr;
  assert(*TokenPtr == '&');
  TokenPtr++;
  if (TokenPtr == CommentEnd) {
    formTextToken(T, TokenPtr);
    return;
  }
  const char *NamePtr;
  bool isNamed = false;
  bool isDecimal = false;
  char C = *TokenPtr;
  if (isHTMLNamedCharacterReferenceCharacter(C)) {
    NamePtr = TokenPtr;
    TokenPtr = skipNamedCharacterReference(TokenPtr, CommentEnd);
    isNamed = true;
  } else if (C == '#') {
    TokenPtr++;
    if (TokenPtr == CommentEnd) {
      formTextToken(T, TokenPtr);
      return;
    }
    C = *TokenPtr;
    if (isHTMLDecimalCharacterReferenceCharacter(C)) {
      NamePtr = TokenPtr;
      TokenPtr = skipDecimalCharacterReference(TokenPtr, CommentEnd);
      isDecimal = true;
    } else if (C == 'x' || C == 'X') {
      TokenPtr++;
      NamePtr = TokenPtr;
      TokenPtr = skipHexCharacterReference(TokenPtr, CommentEnd);
    } else {
      formTextToken(T, TokenPtr);
      return;
    }
  } else {
    formTextToken(T, TokenPtr);
    return;
  }
  if (NamePtr == TokenPtr || TokenPtr == CommentEnd ||
      *TokenPtr != ';') {
    formTextToken(T, TokenPtr);
    return;
  }
  StringRef Name(NamePtr, TokenPtr - NamePtr);
  TokenPtr++; // Skip semicolon.
  StringRef Resolved;
  if (isNamed)
    Resolved = resolveHTMLNamedCharacterReference(Name);
  else if (isDecimal)
    Resolved = resolveHTMLDecimalCharacterReference(Name);
  else
    Resolved = resolveHTMLHexCharacterReference(Name);

  if (Resolved.empty()) {
    formTextToken(T, TokenPtr);
    return;
  }
  formTokenWithChars(T, TokenPtr, tok::text);
  T.setText(Resolved);
}

void Lexer::setupAndLexHTMLStartTag(Token &T) {
  assert(BufferPtr[0] == '<' &&
         isHTMLIdentifierStartingCharacter(BufferPtr[1]));
  const char *TagNameEnd = skipHTMLIdentifier(BufferPtr + 2, CommentEnd);
  StringRef Name(BufferPtr + 1, TagNameEnd - (BufferPtr + 1));
  if (!isHTMLTagName(Name)) {
    formTextToken(T, TagNameEnd);
    return;
  }

  formTokenWithChars(T, TagNameEnd, tok::html_start_tag);
  T.setHTMLTagStartName(Name);

  BufferPtr = skipWhitespace(BufferPtr, CommentEnd);

  const char C = *BufferPtr;
  if (BufferPtr != CommentEnd &&
      (C == '>' || C == '/' || isHTMLIdentifierStartingCharacter(C)))
    State = LS_HTMLStartTag;
}

void Lexer::lexHTMLStartTag(Token &T) {
  assert(State == LS_HTMLStartTag);

  const char *TokenPtr = BufferPtr;
  char C = *TokenPtr;
  if (isHTMLIdentifierCharacter(C)) {
    TokenPtr = skipHTMLIdentifier(TokenPtr, CommentEnd);
    StringRef Ident(BufferPtr, TokenPtr - BufferPtr);
    formTokenWithChars(T, TokenPtr, tok::html_ident);
    T.setHTMLIdent(Ident);
  } else {
    switch (C) {
    case '=':
      TokenPtr++;
      formTokenWithChars(T, TokenPtr, tok::html_equals);
      break;
    case '\"':
    case '\'': {
      const char *OpenQuote = TokenPtr;
      TokenPtr = skipHTMLQuotedString(TokenPtr, CommentEnd);
      const char *ClosingQuote = TokenPtr;
      if (TokenPtr != CommentEnd) // Skip closing quote.
        TokenPtr++;
      formTokenWithChars(T, TokenPtr, tok::html_quoted_string);
      T.setHTMLQuotedString(StringRef(OpenQuote + 1,
                                      ClosingQuote - (OpenQuote + 1)));
      break;
    }
    case '>':
      TokenPtr++;
      formTokenWithChars(T, TokenPtr, tok::html_greater);
      State = LS_Normal;
      return;
    case '/':
      TokenPtr++;
      if (TokenPtr != CommentEnd && *TokenPtr == '>') {
        TokenPtr++;
        formTokenWithChars(T, TokenPtr, tok::html_slash_greater);
      } else
        formTextToken(T, TokenPtr);

      State = LS_Normal;
      return;
    }
  }

  // Now look ahead and return to normal state if we don't see any HTML tokens
  // ahead.
  BufferPtr = skipWhitespace(BufferPtr, CommentEnd);
  if (BufferPtr == CommentEnd) {
    State = LS_Normal;
    return;
  }

  C = *BufferPtr;
  if (!isHTMLIdentifierStartingCharacter(C) &&
      C != '=' && C != '\"' && C != '\'' && C != '>') {
    State = LS_Normal;
    return;
  }
}

void Lexer::setupAndLexHTMLEndTag(Token &T) {
  assert(BufferPtr[0] == '<' && BufferPtr[1] == '/');

  const char *TagNameBegin = skipWhitespace(BufferPtr + 2, CommentEnd);
  const char *TagNameEnd = skipHTMLIdentifier(TagNameBegin, CommentEnd);
  StringRef Name(TagNameBegin, TagNameEnd - TagNameBegin);
  if (!isHTMLTagName(Name)) {
    formTextToken(T, TagNameEnd);
    return;
  }

  const char *End = skipWhitespace(TagNameEnd, CommentEnd);

  formTokenWithChars(T, End, tok::html_end_tag);
  T.setHTMLTagEndName(Name);

  if (BufferPtr != CommentEnd && *BufferPtr == '>')
    State = LS_HTMLEndTag;
}

void Lexer::lexHTMLEndTag(Token &T) {
  assert(BufferPtr != CommentEnd && *BufferPtr == '>');

  formTokenWithChars(T, BufferPtr + 1, tok::html_greater);
  State = LS_Normal;
}

Lexer::Lexer(llvm::BumpPtrAllocator &Allocator, DiagnosticsEngine &Diags,
             const CommandTraits &Traits, SourceLocation FileLoc,
             const char *BufferStart, const char *BufferEnd, bool ParseCommands)
    : Allocator(Allocator), Diags(Diags), Traits(Traits),
      BufferStart(BufferStart), BufferEnd(BufferEnd), BufferPtr(BufferStart),
      FileLoc(FileLoc), ParseCommands(ParseCommands),
      CommentState(LCS_BeforeComment), State(LS_Normal) {}

void Lexer::lex(Token &T) {
again:
  switch (CommentState) {
  case LCS_BeforeComment:
    if (BufferPtr == BufferEnd) {
      formTokenWithChars(T, BufferPtr, tok::eof);
      return;
    }

    assert(*BufferPtr == '/');
    BufferPtr++; // Skip first slash.
    switch(*BufferPtr) {
    case '/': { // BCPL comment.
      BufferPtr++; // Skip second slash.

      if (BufferPtr != BufferEnd) {
        // Skip Doxygen magic marker, if it is present.
        // It might be missing because of a typo //< or /*<, or because we
        // merged this non-Doxygen comment into a bunch of Doxygen comments
        // around it: /** ... */ /* ... */ /** ... */
        const char C = *BufferPtr;
        if (C == '/' || C == '!')
          BufferPtr++;
      }

      // Skip less-than symbol that marks trailing comments.
      // Skip it even if the comment is not a Doxygen one, because //< and /*<
      // are frequent typos.
      if (BufferPtr != BufferEnd && *BufferPtr == '<')
        BufferPtr++;

      CommentState = LCS_InsideBCPLComment;
      if (State != LS_VerbatimBlockBody && State != LS_VerbatimBlockFirstLine)
        State = LS_Normal;
      CommentEnd = findBCPLCommentEnd(BufferPtr, BufferEnd);
      goto again;
    }
    case '*': { // C comment.
      BufferPtr++; // Skip star.

      // Skip Doxygen magic marker.
      const char C = *BufferPtr;
      if ((C == '*' && *(BufferPtr + 1) != '/') || C == '!')
        BufferPtr++;

      // Skip less-than symbol that marks trailing comments.
      if (BufferPtr != BufferEnd && *BufferPtr == '<')
        BufferPtr++;

      CommentState = LCS_InsideCComment;
      State = LS_Normal;
      CommentEnd = findCCommentEnd(BufferPtr, BufferEnd);
      goto again;
    }
    default:
      llvm_unreachable("second character of comment should be '/' or '*'");
    }

  case LCS_BetweenComments: {
    // Consecutive comments are extracted only if there is only whitespace
    // between them.  So we can search for the start of the next comment.
    const char *EndWhitespace = BufferPtr;
    while(EndWhitespace != BufferEnd && *EndWhitespace != '/')
      EndWhitespace++;

    // Turn any whitespace between comments (and there is only whitespace
    // between them -- guaranteed by comment extraction) into a newline.  We
    // have two newlines between C comments in total (first one was synthesized
    // after a comment).
    formTokenWithChars(T, EndWhitespace, tok::newline);

    CommentState = LCS_BeforeComment;
    break;
  }

  case LCS_InsideBCPLComment:
  case LCS_InsideCComment:
    if (BufferPtr != CommentEnd) {
      lexCommentText(T);
      break;
    } else {
      // Skip C comment closing sequence.
      if (CommentState == LCS_InsideCComment) {
        assert(BufferPtr[0] == '*' && BufferPtr[1] == '/');
        BufferPtr += 2;
        assert(BufferPtr <= BufferEnd);

        // Synthenize newline just after the C comment, regardless if there is
        // actually a newline.
        formTokenWithChars(T, BufferPtr, tok::newline);

        CommentState = LCS_BetweenComments;
        break;
      } else {
        // Don't synthesized a newline after BCPL comment.
        CommentState = LCS_BetweenComments;
        goto again;
      }
    }
  }
}

StringRef Lexer::getSpelling(const Token &Tok,
                             const SourceManager &SourceMgr) const {
  SourceLocation Loc = Tok.getLocation();
  std::pair<FileID, unsigned> LocInfo = SourceMgr.getDecomposedLoc(Loc);

  bool InvalidTemp = false;
  StringRef File = SourceMgr.getBufferData(LocInfo.first, &InvalidTemp);
  if (InvalidTemp)
    return StringRef();

  const char *Begin = File.data() + LocInfo.second;
  return StringRef(Begin, Tok.getLength());
}

} // end namespace comments
} // end namespace clang