summaryrefslogtreecommitdiff
path: root/clang/lib/Rewrite/HTMLRewrite.cpp
blob: dc39dde7ff1f9b7ab33e54ac6fde27fcd23cb012 (about) (plain)
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
//== HTMLRewrite.cpp - Translate source code into prettified HTML --*- C++ -*-//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
//  This file defines the HTMLRewriter clas, which is used to translate the
//  text of a source file into prettified HTML.
//
//===----------------------------------------------------------------------===//

#include "clang/Lex/Preprocessor.h"
#include "clang/Rewrite/Rewriter.h"
#include "clang/Rewrite/HTMLRewrite.h"
#include "clang/Lex/TokenConcatenation.h"
#include "clang/Lex/Preprocessor.h"
#include "clang/Basic/SourceManager.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/OwningPtr.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/raw_ostream.h"
using namespace clang;


/// HighlightRange - Highlight a range in the source code with the specified
/// start/end tags.  B/E must be in the same file.  This ensures that
/// start/end tags are placed at the start/end of each line if the range is
/// multiline.
void html::HighlightRange(Rewriter &R, SourceLocation B, SourceLocation E,
                          const char *StartTag, const char *EndTag) {
  SourceManager &SM = R.getSourceMgr();
  B = SM.getExpansionLoc(B);
  E = SM.getExpansionLoc(E);
  FileID FID = SM.getFileID(B);
  assert(SM.getFileID(E) == FID && "B/E not in the same file!");

  unsigned BOffset = SM.getFileOffset(B);
  unsigned EOffset = SM.getFileOffset(E);

  // Include the whole end token in the range.
  EOffset += Lexer::MeasureTokenLength(E, R.getSourceMgr(), R.getLangOpts());

  bool Invalid = false;
  const char *BufferStart = SM.getBufferData(FID, &Invalid).data();
  if (Invalid)
    return;
  
  HighlightRange(R.getEditBuffer(FID), BOffset, EOffset,
                 BufferStart, StartTag, EndTag);
}

/// HighlightRange - This is the same as the above method, but takes
/// decomposed file locations.
void html::HighlightRange(RewriteBuffer &RB, unsigned B, unsigned E,
                          const char *BufferStart,
                          const char *StartTag, const char *EndTag) {
  // Insert the tag at the absolute start/end of the range.
  RB.InsertTextAfter(B, StartTag);
  RB.InsertTextBefore(E, EndTag);

  // Scan the range to see if there is a \r or \n.  If so, and if the line is
  // not blank, insert tags on that line as well.
  bool HadOpenTag = true;

  unsigned LastNonWhiteSpace = B;
  for (unsigned i = B; i != E; ++i) {
    switch (BufferStart[i]) {
    case '\r':
    case '\n':
      // Okay, we found a newline in the range.  If we have an open tag, we need
      // to insert a close tag at the first non-whitespace before the newline.
      if (HadOpenTag)
        RB.InsertTextBefore(LastNonWhiteSpace+1, EndTag);

      // Instead of inserting an open tag immediately after the newline, we
      // wait until we see a non-whitespace character.  This prevents us from
      // inserting tags around blank lines, and also allows the open tag to
      // be put *after* whitespace on a non-blank line.
      HadOpenTag = false;
      break;
    case '\0':
    case ' ':
    case '\t':
    case '\f':
    case '\v':
      // Ignore whitespace.
      break;

    default:
      // If there is no tag open, do it now.
      if (!HadOpenTag) {
        RB.InsertTextAfter(i, StartTag);
        HadOpenTag = true;
      }

      // Remember this character.
      LastNonWhiteSpace = i;
      break;
    }
  }
}

void html::EscapeText(Rewriter &R, FileID FID,
                      bool EscapeSpaces, bool ReplaceTabs) {

  const llvm::MemoryBuffer *Buf = R.getSourceMgr().getBuffer(FID);
  const char* C = Buf->getBufferStart();
  const char* FileEnd = Buf->getBufferEnd();

  assert (C <= FileEnd);

  RewriteBuffer &RB = R.getEditBuffer(FID);

  unsigned ColNo = 0;
  for (unsigned FilePos = 0; C != FileEnd ; ++C, ++FilePos) {
    switch (*C) {
    default: ++ColNo; break;
    case '\n':
    case '\r':
      ColNo = 0;
      break;

    case ' ':
      if (EscapeSpaces)
        RB.ReplaceText(FilePos, 1, "&nbsp;");
      ++ColNo;
      break;
    case '\f':
      RB.ReplaceText(FilePos, 1, "<hr>");
      ColNo = 0;
      break;

    case '\t': {
      if (!ReplaceTabs)
        break;
      unsigned NumSpaces = 8-(ColNo&7);
      if (EscapeSpaces)
        RB.ReplaceText(FilePos, 1,
                       StringRef("&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"
                                       "&nbsp;&nbsp;&nbsp;", 6*NumSpaces));
      else
        RB.ReplaceText(FilePos, 1, StringRef("        ", NumSpaces));
      ColNo += NumSpaces;
      break;
    }
    case '<':
      RB.ReplaceText(FilePos, 1, "&lt;");
      ++ColNo;
      break;

    case '>':
      RB.ReplaceText(FilePos, 1, "&gt;");
      ++ColNo;
      break;

    case '&':
      RB.ReplaceText(FilePos, 1, "&amp;");
      ++ColNo;
      break;
    }
  }
}

std::string html::EscapeText(const std::string& s, bool EscapeSpaces,
                             bool ReplaceTabs) {

  unsigned len = s.size();
  std::string Str;
  llvm::raw_string_ostream os(Str);

  for (unsigned i = 0 ; i < len; ++i) {

    char c = s[i];
    switch (c) {
    default:
      os << c; break;

    case ' ':
      if (EscapeSpaces) os << "&nbsp;";
      else os << ' ';
      break;

    case '\t':
      if (ReplaceTabs) {
        if (EscapeSpaces)
          for (unsigned i = 0; i < 4; ++i)
            os << "&nbsp;";
        else
          for (unsigned i = 0; i < 4; ++i)
            os << " ";
      }
      else
        os << c;

      break;

    case '<': os << "&lt;"; break;
    case '>': os << "&gt;"; break;
    case '&': os << "&amp;"; break;
    }
  }

  return os.str();
}

static void AddLineNumber(RewriteBuffer &RB, unsigned LineNo,
                          unsigned B, unsigned E) {
  SmallString<256> Str;
  llvm::raw_svector_ostream OS(Str);

  OS << "<tr><td class=\"num\" id=\"LN"
     << LineNo << "\">"
     << LineNo << "</td><td class=\"line\">";

  if (B == E) { // Handle empty lines.
    OS << " </td></tr>";
    RB.InsertTextBefore(B, OS.str());
  } else {
    RB.InsertTextBefore(B, OS.str());
    RB.InsertTextBefore(E, "</td></tr>");
  }
}

void html::AddLineNumbers(Rewriter& R, FileID FID) {

  const llvm::MemoryBuffer *Buf = R.getSourceMgr().getBuffer(FID);
  const char* FileBeg = Buf->getBufferStart();
  const char* FileEnd = Buf->getBufferEnd();
  const char* C = FileBeg;
  RewriteBuffer &RB = R.getEditBuffer(FID);

  assert (C <= FileEnd);

  unsigned LineNo = 0;
  unsigned FilePos = 0;

  while (C != FileEnd) {

    ++LineNo;
    unsigned LineStartPos = FilePos;
    unsigned LineEndPos = FileEnd - FileBeg;

    assert (FilePos <= LineEndPos);
    assert (C < FileEnd);

    // Scan until the newline (or end-of-file).

    while (C != FileEnd) {
      char c = *C;
      ++C;

      if (c == '\n') {
        LineEndPos = FilePos++;
        break;
      }

      ++FilePos;
    }

    AddLineNumber(RB, LineNo, LineStartPos, LineEndPos);
  }

  // Add one big table tag that surrounds all of the code.
  RB.InsertTextBefore(0, "<table class=\"code\">\n");
  RB.InsertTextAfter(FileEnd - FileBeg, "</table>");
}

void html::AddHeaderFooterInternalBuiltinCSS(Rewriter& R, FileID FID,
                                             const char *title) {

  const llvm::MemoryBuffer *Buf = R.getSourceMgr().getBuffer(FID);
  const char* FileStart = Buf->getBufferStart();
  const char* FileEnd = Buf->getBufferEnd();

  SourceLocation StartLoc = R.getSourceMgr().getLocForStartOfFile(FID);
  SourceLocation EndLoc = StartLoc.getLocWithOffset(FileEnd-FileStart);

  std::string s;
  llvm::raw_string_ostream os(s);
  os << "<!doctype html>\n" // Use HTML 5 doctype
        "<html>\n<head>\n";

  if (title)
    os << "<title>" << html::EscapeText(title) << "</title>\n";

  os << "<style type=\"text/css\">\n"
      " body { color:#000000; background-color:#ffffff }\n"
      " body { font-family:Helvetica, sans-serif; font-size:10pt }\n"
      " h1 { font-size:14pt }\n"
      " .code { border-collapse:collapse; width:100%; }\n"
      " .code { font-family: \"Monospace\", monospace; font-size:10pt }\n"
      " .code { line-height: 1.2em }\n"
      " .comment { color: green; font-style: oblique }\n"
      " .keyword { color: blue }\n"
      " .string_literal { color: red }\n"
      " .directive { color: darkmagenta }\n"
      // Macro expansions.
      " .expansion { display: none; }\n"
      " .macro:hover .expansion { display: block; border: 2px solid #FF0000; "
          "padding: 2px; background-color:#FFF0F0; font-weight: normal; "
          "  -webkit-border-radius:5px;  -webkit-box-shadow:1px 1px 7px #000; "
          "position: absolute; top: -1em; left:10em; z-index: 1 } \n"
      " .macro { color: darkmagenta; background-color:LemonChiffon;"
             // Macros are position: relative to provide base for expansions.
             " position: relative }\n"
      " .num { width:2.5em; padding-right:2ex; background-color:#eeeeee }\n"
      " .num { text-align:right; font-size:8pt }\n"
      " .num { color:#444444 }\n"
      " .line { padding-left: 1ex; border-left: 3px solid #ccc }\n"
      " .line { white-space: pre }\n"
      " .msg { -webkit-box-shadow:1px 1px 7px #000 }\n"
      " .msg { -webkit-border-radius:5px }\n"
      " .msg { font-family:Helvetica, sans-serif; font-size:8pt }\n"
      " .msg { float:left }\n"
      " .msg { padding:0.25em 1ex 0.25em 1ex }\n"
      " .msg { margin-top:10px; margin-bottom:10px }\n"
      " .msg { font-weight:bold }\n"
      " .msg { max-width:60em; word-wrap: break-word; white-space: pre-wrap }\n"
      " .msgT { padding:0x; spacing:0x }\n"
      " .msgEvent { background-color:#fff8b4; color:#000000 }\n"
      " .msgControl { background-color:#bbbbbb; color:#000000 }\n"
      " .mrange { background-color:#dfddf3 }\n"
      " .mrange { border-bottom:1px solid #6F9DBE }\n"
      " .PathIndex { font-weight: bold; padding:0px 5px 0px 5px; "
        "margin-right:5px; }\n"
      " .PathIndex { -webkit-border-radius:8px }\n"
      " .PathIndexEvent { background-color:#bfba87 }\n"
      " .PathIndexControl { background-color:#8c8c8c }\n"
      " .CodeInsertionHint { font-weight: bold; background-color: #10dd10 }\n"
      " .CodeRemovalHint { background-color:#de1010 }\n"
      " .CodeRemovalHint { border-bottom:1px solid #6F9DBE }\n"
      " table.simpletable {\n"
      "   padding: 5px;\n"
      "   font-size:12pt;\n"
      "   margin:20px;\n"
      "   border-collapse: collapse; border-spacing: 0px;\n"
      " }\n"
      " td.rowname {\n"
      "   text-align:right; font-weight:bold; color:#444444;\n"
      "   padding-right:2ex; }\n"
      "</style>\n</head>\n<body>";

  // Generate header
  R.InsertTextBefore(StartLoc, os.str());
  // Generate footer

  R.InsertTextAfter(EndLoc, "</body></html>\n");
}

/// SyntaxHighlight - Relex the specified FileID and annotate the HTML with
/// information about keywords, macro expansions etc.  This uses the macro
/// table state from the end of the file, so it won't be perfectly perfect,
/// but it will be reasonably close.
void html::SyntaxHighlight(Rewriter &R, FileID FID, const Preprocessor &PP) {
  RewriteBuffer &RB = R.getEditBuffer(FID);

  const SourceManager &SM = PP.getSourceManager();
  const llvm::MemoryBuffer *FromFile = SM.getBuffer(FID);
  Lexer L(FID, FromFile, SM, PP.getLangOpts());
  const char *BufferStart = L.getBufferStart();

  // Inform the preprocessor that we want to retain comments as tokens, so we
  // can highlight them.
  L.SetCommentRetentionState(true);

  // Lex all the tokens in raw mode, to avoid entering #includes or expanding
  // macros.
  Token Tok;
  L.LexFromRawLexer(Tok);

  while (Tok.isNot(tok::eof)) {
    // Since we are lexing unexpanded tokens, all tokens are from the main
    // FileID.
    unsigned TokOffs = SM.getFileOffset(Tok.getLocation());
    unsigned TokLen = Tok.getLength();
    switch (Tok.getKind()) {
    default: break;
    case tok::identifier:
      llvm_unreachable("tok::identifier in raw lexing mode!");
    case tok::raw_identifier: {
      // Fill in Result.IdentifierInfo and update the token kind,
      // looking up the identifier in the identifier table.
      PP.LookUpIdentifierInfo(Tok);

      // If this is a pp-identifier, for a keyword, highlight it as such.
      if (Tok.isNot(tok::identifier))
        HighlightRange(RB, TokOffs, TokOffs+TokLen, BufferStart,
                       "<span class='keyword'>", "</span>");
      break;
    }
    case tok::comment:
      HighlightRange(RB, TokOffs, TokOffs+TokLen, BufferStart,
                     "<span class='comment'>", "</span>");
      break;
    case tok::utf8_string_literal:
      // Chop off the u part of u8 prefix
      ++TokOffs;
      --TokLen;
      // FALL THROUGH to chop the 8
    case tok::wide_string_literal:
    case tok::utf16_string_literal:
    case tok::utf32_string_literal:
      // Chop off the L, u, U or 8 prefix
      ++TokOffs;
      --TokLen;
      // FALL THROUGH.
    case tok::string_literal:
      // FIXME: Exclude the optional ud-suffix from the highlighted range.
      HighlightRange(RB, TokOffs, TokOffs+TokLen, BufferStart,
                     "<span class='string_literal'>", "</span>");
      break;
    case tok::hash: {
      // If this is a preprocessor directive, all tokens to end of line are too.
      if (!Tok.isAtStartOfLine())
        break;

      // Eat all of the tokens until we get to the next one at the start of
      // line.
      unsigned TokEnd = TokOffs+TokLen;
      L.LexFromRawLexer(Tok);
      while (!Tok.isAtStartOfLine() && Tok.isNot(tok::eof)) {
        TokEnd = SM.getFileOffset(Tok.getLocation())+Tok.getLength();
        L.LexFromRawLexer(Tok);
      }

      // Find end of line.  This is a hack.
      HighlightRange(RB, TokOffs, TokEnd, BufferStart,
                     "<span class='directive'>", "</span>");

      // Don't skip the next token.
      continue;
    }
    }

    L.LexFromRawLexer(Tok);
  }
}

/// HighlightMacros - This uses the macro table state from the end of the
/// file, to re-expand macros and insert (into the HTML) information about the
/// macro expansions.  This won't be perfectly perfect, but it will be
/// reasonably close.
void html::HighlightMacros(Rewriter &R, FileID FID, const Preprocessor& PP) {
  // Re-lex the raw token stream into a token buffer.
  const SourceManager &SM = PP.getSourceManager();
  std::vector<Token> TokenStream;

  const llvm::MemoryBuffer *FromFile = SM.getBuffer(FID);
  Lexer L(FID, FromFile, SM, PP.getLangOpts());

  // Lex all the tokens in raw mode, to avoid entering #includes or expanding
  // macros.
  while (1) {
    Token Tok;
    L.LexFromRawLexer(Tok);

    // If this is a # at the start of a line, discard it from the token stream.
    // We don't want the re-preprocess step to see #defines, #includes or other
    // preprocessor directives.
    if (Tok.is(tok::hash) && Tok.isAtStartOfLine())
      continue;

    // If this is a ## token, change its kind to unknown so that repreprocessing
    // it will not produce an error.
    if (Tok.is(tok::hashhash))
      Tok.setKind(tok::unknown);

    // If this raw token is an identifier, the raw lexer won't have looked up
    // the corresponding identifier info for it.  Do this now so that it will be
    // macro expanded when we re-preprocess it.
    if (Tok.is(tok::raw_identifier))
      PP.LookUpIdentifierInfo(Tok);

    TokenStream.push_back(Tok);

    if (Tok.is(tok::eof)) break;
  }

  // Temporarily change the diagnostics object so that we ignore any generated
  // diagnostics from this pass.
  DiagnosticsEngine TmpDiags(PP.getDiagnostics().getDiagnosticIDs(),
                      new IgnoringDiagConsumer);

  // FIXME: This is a huge hack; we reuse the input preprocessor because we want
  // its state, but we aren't actually changing it (we hope). This should really
  // construct a copy of the preprocessor.
  Preprocessor &TmpPP = const_cast<Preprocessor&>(PP);
  DiagnosticsEngine *OldDiags = &TmpPP.getDiagnostics();
  TmpPP.setDiagnostics(TmpDiags);

  // Inform the preprocessor that we don't want comments.
  TmpPP.SetCommentRetentionState(false, false);

  // Enter the tokens we just lexed.  This will cause them to be macro expanded
  // but won't enter sub-files (because we removed #'s).
  TmpPP.EnterTokenStream(&TokenStream[0], TokenStream.size(), false, false);

  TokenConcatenation ConcatInfo(TmpPP);

  // Lex all the tokens.
  Token Tok;
  TmpPP.Lex(Tok);
  while (Tok.isNot(tok::eof)) {
    // Ignore non-macro tokens.
    if (!Tok.getLocation().isMacroID()) {
      TmpPP.Lex(Tok);
      continue;
    }

    // Okay, we have the first token of a macro expansion: highlight the
    // expansion by inserting a start tag before the macro expansion and
    // end tag after it.
    std::pair<SourceLocation, SourceLocation> LLoc =
      SM.getExpansionRange(Tok.getLocation());

    // Ignore tokens whose instantiation location was not the main file.
    if (SM.getFileID(LLoc.first) != FID) {
      TmpPP.Lex(Tok);
      continue;
    }

    assert(SM.getFileID(LLoc.second) == FID &&
           "Start and end of expansion must be in the same ultimate file!");

    std::string Expansion = EscapeText(TmpPP.getSpelling(Tok));
    unsigned LineLen = Expansion.size();

    Token PrevPrevTok;
    Token PrevTok = Tok;
    // Okay, eat this token, getting the next one.
    TmpPP.Lex(Tok);

    // Skip all the rest of the tokens that are part of this macro
    // instantiation.  It would be really nice to pop up a window with all the
    // spelling of the tokens or something.
    while (!Tok.is(tok::eof) &&
           SM.getExpansionLoc(Tok.getLocation()) == LLoc.first) {
      // Insert a newline if the macro expansion is getting large.
      if (LineLen > 60) {
        Expansion += "<br>";
        LineLen = 0;
      }

      LineLen -= Expansion.size();

      // If the tokens were already space separated, or if they must be to avoid
      // them being implicitly pasted, add a space between them.
      if (Tok.hasLeadingSpace() ||
          ConcatInfo.AvoidConcat(PrevPrevTok, PrevTok, Tok))
        Expansion += ' ';

      // Escape any special characters in the token text.
      Expansion += EscapeText(TmpPP.getSpelling(Tok));
      LineLen += Expansion.size();

      PrevPrevTok = PrevTok;
      PrevTok = Tok;
      TmpPP.Lex(Tok);
    }


    // Insert the expansion as the end tag, so that multi-line macros all get
    // highlighted.
    Expansion = "<span class='expansion'>" + Expansion + "</span></span>";

    HighlightRange(R, LLoc.first, LLoc.second,
                   "<span class='macro'>", Expansion.c_str());
  }

  // Restore diagnostics object back to its own thing.
  TmpPP.setDiagnostics(*OldDiags);
}