C++ SourceManager::getDecomposedLoc方法代码示例

StringRef RawComment::getRawTextSlow(const SourceManager &SourceMgr) const {
  FileID BeginFileID;
  FileID EndFileID;
  unsigned BeginOffset;
  unsigned EndOffset;

  std::tie(BeginFileID, BeginOffset) =
      SourceMgr.getDecomposedLoc(Range.getBegin());
  std::tie(EndFileID, EndOffset) = SourceMgr.getDecomposedLoc(Range.getEnd());

  const unsigned Length = EndOffset - BeginOffset;
  if (Length < 2)
    return StringRef();

  // The comment can't begin in one file and end in another.
  assert(BeginFileID == EndFileID);

  bool Invalid = false;
  const char *BufferStart = SourceMgr.getBufferData(BeginFileID,
                                                    &Invalid).data();
  if (Invalid)
    return StringRef();

  return StringRef(BufferStart + BeginOffset, Length);
}

static std::string getText(const SourceManager &SourceManager, const T &Node) {
  SourceLocation StartSpellingLocatino =
      SourceManager.getSpellingLoc(Node.getLocStart());
  SourceLocation EndSpellingLocation =
      SourceManager.getSpellingLoc(Node.getLocEnd());
  if (!StartSpellingLocatino.isValid() || !EndSpellingLocation.isValid()) {
    return std::string();
  }
  bool Invalid = true;
  const char *Text =
      SourceManager.getCharacterData(StartSpellingLocatino, &Invalid);
  if (Invalid) {
    return std::string();
  }
  std::pair<FileID, unsigned> Start =
      SourceManager.getDecomposedLoc(StartSpellingLocatino);
  std::pair<FileID, unsigned> End =
      SourceManager.getDecomposedLoc(Lexer::getLocForEndOfToken(
          EndSpellingLocation, 0, SourceManager, LangOptions()));
  if (Start.first != End.first) {
    // Start and end are in different files.
    return std::string();
  }
  if (End.second < Start.second) {
    // Shuffling text with macros may cause this.
    return std::string();
  }
  return std::string(Text, End.second - Start.second);
}

// FIXME: This should go into the Lexer, but we need to figure out how
// to handle ranges for refactoring in general first - there is no obvious
// good way how to integrate this into the Lexer yet.
static int getRangeSize(SourceManager &Sources, const CharSourceRange &Range) {
  SourceLocation SpellingBegin = Sources.getSpellingLoc(Range.getBegin());
  SourceLocation SpellingEnd = Sources.getSpellingLoc(Range.getEnd());
  std::pair<FileID, unsigned> Start = Sources.getDecomposedLoc(SpellingBegin);
  std::pair<FileID, unsigned> End = Sources.getDecomposedLoc(SpellingEnd);
  if (Start.first != End.first) return -1;
  if (Range.isTokenRange())
    End.second += Lexer::MeasureTokenLength(SpellingEnd, Sources,
                                            LangOptions());
  return End.second - Start.second;
}

/// \brief Retrieve the name of the immediate macro expansion.
///
/// This routine starts from a source location, and finds the name of the macro
/// responsible for its immediate expansion. It looks through any intervening
/// macro argument expansions to compute this. It returns a StringRef which
/// refers to the SourceManager-owned buffer of the source where that macro
/// name is spelled. Thus, the result shouldn't out-live that SourceManager.
///
/// This differs from Lexer::getImmediateMacroName in that any macro argument
/// location will result in the topmost function macro that accepted it.
/// e.g.
/// \code
///   MAC1( MAC2(foo) )
/// \endcode
/// for location of 'foo' token, this function will return "MAC1" while
/// Lexer::getImmediateMacroName will return "MAC2".
static StringRef getImmediateMacroName(SourceLocation Loc,
                                       const SourceManager &SM,
                                       const LangOptions &LangOpts) {
   assert(Loc.isMacroID() && "Only reasonble to call this on macros");
   // Walk past macro argument expanions.
   while (SM.isMacroArgExpansion(Loc))
     Loc = SM.getImmediateExpansionRange(Loc).first;

   // If the macro's spelling has no FileID, then it's actually a token paste
   // or stringization (or similar) and not a macro at all.
   if (!SM.getFileEntryForID(SM.getFileID(SM.getSpellingLoc(Loc))))
     return StringRef();

   // Find the spelling location of the start of the non-argument expansion
   // range. This is where the macro name was spelled in order to begin
   // expanding this macro.
   Loc = SM.getSpellingLoc(SM.getImmediateExpansionRange(Loc).first);

   // Dig out the buffer where the macro name was spelled and the extents of the
   // name so that we can render it into the expansion note.
   std::pair<FileID, unsigned> ExpansionInfo = SM.getDecomposedLoc(Loc);
   unsigned MacroTokenLength = Lexer::MeasureTokenLength(Loc, SM, LangOpts);
   StringRef ExpansionBuffer = SM.getBufferData(ExpansionInfo.first);
   return ExpansionBuffer.substr(ExpansionInfo.second, MacroTokenLength);
}

unsigned RawComment::getBeginLine(const SourceManager &SM) const {
  if (BeginLineValid)
    return BeginLine;

  std::pair<FileID, unsigned> LocInfo = SM.getDecomposedLoc(Range.getBegin());
  BeginLine = SM.getLineNumber(LocInfo.first, LocInfo.second);
  BeginLineValid = true;
  return BeginLine;
}

void Replacement::setFromSourceLocation(const SourceManager &Sources,
                                        SourceLocation Start, unsigned Length,
                                        StringRef ReplacementText) {
  const std::pair<FileID, unsigned> DecomposedLocation =
      Sources.getDecomposedLoc(Start);
  const FileEntry *Entry = Sources.getFileEntryForID(DecomposedLocation.first);
  this->FilePath = Entry ? Entry->getName() : InvalidLocation;
  this->ReplacementRange = Range(DecomposedLocation.second, Length);
  this->ReplacementText = ReplacementText;
}

// This function is an adaptation from StringLiteral::getLocationOfByte in llvm-3.7.1\src\tools\clang\lib\AST\Expr.cpp
std::vector<std::string>
splitStringLiteral(StringLiteral *S, const SourceManager &SM, const LangOptions &Features, const TargetInfo &Target)
{
    // Loop over all of the tokens in this string until we find the one that
    // contains the byte we're looking for.
    unsigned TokNo = 0;

    std::vector<std::string> result;
    for (TokNo = 0; TokNo < S->getNumConcatenated(); ++TokNo)
    {
        SourceLocation StrTokLoc = S->getStrTokenLoc(TokNo);

        // Get the spelling of the string so that we can get the data that makes up
        // the string literal, not the identifier for the macro it is potentially
        // expanded through.
        SourceLocation StrTokSpellingLoc = SM.getSpellingLoc(StrTokLoc);

        // Re-lex the token to get its length and original spelling.
        std::pair<FileID, unsigned> LocInfo = SM.getDecomposedLoc(StrTokSpellingLoc);
        bool Invalid = false;
        StringRef Buffer = SM.getBufferData(LocInfo.first, &Invalid);
        if (Invalid)
            continue; // We ignore this part

        const char *StrData = Buffer.data() + LocInfo.second;

        // Create a lexer starting at the beginning of this token.
        Lexer TheLexer(SM.getLocForStartOfFile(LocInfo.first), Features,
            Buffer.begin(), StrData, Buffer.end());
        Token TheTok;
        TheLexer.LexFromRawLexer(TheTok);
        if (TheTok.isAnyIdentifier())
        {
            // It should not be, since we are parsing inside a string literal, but it can happen with special macros such as __func__
            // of __PRETTY_FUNCTION__ that are not resolved at this time. In that case, we just ignore them...
            continue;
        }
        // Get the spelling of the token.
        SmallString<32> SpellingBuffer;
        SpellingBuffer.resize(TheTok.getLength());

        bool StringInvalid = false;
        const char *SpellingPtr = &SpellingBuffer[0];
        unsigned TokLen = Lexer::getSpelling(TheTok, SpellingPtr, SM, Features, &StringInvalid);
        if (StringInvalid)
            continue;

        const char *SpellingStart = SpellingPtr;
        const char *SpellingEnd = SpellingPtr + TokLen;
        result.push_back(std::string(SpellingStart, SpellingEnd));

    }
    return result;
}

void Replacement::setFromSourceLocation(SourceManager &Sources,
                                        SourceLocation Start, unsigned Length,
                                        llvm::StringRef ReplacementText) {
  const std::pair<FileID, unsigned> DecomposedLocation =
      Sources.getDecomposedLoc(Start);
  const FileEntry *Entry = Sources.getFileEntryForID(DecomposedLocation.first);
  this->FilePath = Entry != NULL ? Entry->getName() : InvalidLocation;
  this->Offset = DecomposedLocation.second;
  this->Length = Length;
  this->ReplacementText = ReplacementText;
}

/// getLocationOfByte - Return a source location that points to the specified
/// byte of this string literal.
///
/// Strings are amazingly complex.  They can be formed from multiple tokens and
/// can have escape sequences in them in addition to the usual trigraph and
/// escaped newline business.  This routine handles this complexity.
///
SourceLocation
StringLiteral::getLocationOfByte(unsigned ByteNo, const SourceManager &SM,
                  const LangOptions &Features, const TargetInfo &Target) const {

  // Loop over all of the tokens in this string until we find the one that
  // contains the byte we're looking for.
  unsigned TokNo = 0;
  while (1) {
    assert(TokNo < getNumConcatenated() && "Invalid byte number!");
    SourceLocation StrTokLoc = getStrTokenLoc(TokNo);
    
    // Get the spelling of the string so that we can get the data that makes up
    // the string literal, not the identifier for the macro it is potentially
    // expanded through.
    SourceLocation StrTokSpellingLoc = SM.getSpellingLoc(StrTokLoc);
    
    // Re-lex the token to get its length and original spelling.
    std::pair<FileID, unsigned> LocInfo =SM.getDecomposedLoc(StrTokSpellingLoc);
    bool Invalid = false;
    llvm::StringRef Buffer = SM.getBufferData(LocInfo.first, &Invalid);
    if (Invalid)
      return StrTokSpellingLoc;
    
    const char *StrData = Buffer.data()+LocInfo.second;
    
    // Create a langops struct and enable trigraphs.  This is sufficient for
    // relexing tokens.
    LangOptions LangOpts;
    
    // Create a lexer starting at the beginning of this token.
    Lexer TheLexer(StrTokSpellingLoc, Features, Buffer.begin(), StrData,
                   Buffer.end());
    Token TheTok;
    TheLexer.LexFromRawLexer(TheTok);
    
    // Use the StringLiteralParser to compute the length of the string in bytes.
    StringLiteralParser SLP(&TheTok, 1, SM, Features, Target);
    unsigned TokNumBytes = SLP.GetStringLength();
    
    // If the byte is in this token, return the location of the byte.
    if (ByteNo < TokNumBytes ||
        (ByteNo == TokNumBytes && TokNo == getNumConcatenated())) {
      unsigned Offset = SLP.getOffsetOfStringByte(TheTok, ByteNo); 
      
      // Now that we know the offset of the token in the spelling, use the
      // preprocessor to get the offset in the original source.
      return Lexer::AdvanceToTokenCharacter(StrTokLoc, Offset, SM, Features);
    }
    
    // Move to the next string token.
    ++TokNo;
    ByteNo -= TokNumBytes;
  }
}

static bool onlyWhitespaceBetween(SourceManager &SM,
                                  SourceLocation Loc1, SourceLocation Loc2,
                                  unsigned MaxNewlinesAllowed) {
  std::pair<FileID, unsigned> Loc1Info = SM.getDecomposedLoc(Loc1);
  std::pair<FileID, unsigned> Loc2Info = SM.getDecomposedLoc(Loc2);

  // Question does not make sense if locations are in different files.
  if (Loc1Info.first != Loc2Info.first)
    return false;

  bool Invalid = false;
  const char *Buffer = SM.getBufferData(Loc1Info.first, &Invalid).data();
  if (Invalid)
    return false;

  unsigned NumNewlines = 0;
  assert(Loc1Info.second <= Loc2Info.second && "Loc1 after Loc2!");
  // Look for non-whitespace characters and remember any newlines seen.
  for (unsigned I = Loc1Info.second; I != Loc2Info.second; ++I) {
    switch (Buffer[I]) {
    default:
      return false;
    case ' ':
    case '\t':
    case '\f':
    case '\v':
      break;
    case '\r':
    case '\n':
      ++NumNewlines;

      // Check if we have found more than the maximum allowed number of
      // newlines.
      if (NumNewlines > MaxNewlinesAllowed)
        return false;

      // Collapse \r\n and \n\r into a single newline.
      if (I + 1 != Loc2Info.second &&
          (Buffer[I + 1] == '\n' || Buffer[I + 1] == '\r') &&
          Buffer[I] != Buffer[I + 1])
        ++I;
      break;
    }
  }

  return true;
}

// Checks if 'typedef' keyword can be removed - we do it only if
// it is the only declaration in a declaration chain.
static bool CheckRemoval(SourceManager &SM, SourceLocation StartLoc,
                         ASTContext &Context) {
  assert(StartLoc.isFileID() && "StartLoc must not be in a macro");
  std::pair<FileID, unsigned> LocInfo = SM.getDecomposedLoc(StartLoc);
  StringRef File = SM.getBufferData(LocInfo.first);
  const char *TokenBegin = File.data() + LocInfo.second;
  Lexer DeclLexer(SM.getLocForStartOfFile(LocInfo.first), Context.getLangOpts(),
                  File.begin(), TokenBegin, File.end());

  Token Tok;
  int ParenLevel = 0;
  bool FoundTypedef = false;

  while (!DeclLexer.LexFromRawLexer(Tok) && !Tok.is(tok::semi)) {
    switch (Tok.getKind()) {
    case tok::l_brace:
    case tok::r_brace:
      // This might be the `typedef struct {...} T;` case.
      return false;
    case tok::l_paren:
      ParenLevel++;
      break;
    case tok::r_paren:
      ParenLevel--;
      break;
    case tok::comma:
      if (ParenLevel == 0) {
        // If there is comma and we are not between open parenthesis then it is
        // two or more declarations in this chain.
        return false;
      }
      break;
    case tok::raw_identifier:
      if (Tok.getRawIdentifier() == "typedef") {
        FoundTypedef = true;
      }
      break;
    default:
      break;
    }
  }

  // Sanity check against weird macro cases.
  return FoundTypedef;
}

void Replacement::setFromSourceLocation(const SourceManager &Sources,
                                        SourceLocation Start, unsigned Length,
                                        StringRef ReplacementText) {
    const std::pair<FileID, unsigned> DecomposedLocation =
        Sources.getDecomposedLoc(Start);
    const FileEntry *Entry = Sources.getFileEntryForID(DecomposedLocation.first);
    if (Entry) {
        // Make FilePath absolute so replacements can be applied correctly when
        // relative paths for files are used.
        llvm::SmallString<256> FilePath(Entry->getName());
        std::error_code EC = llvm::sys::fs::make_absolute(FilePath);
        this->FilePath = EC ? FilePath.c_str() : Entry->getName();
    } else {
        this->FilePath = InvalidLocation;
    }
    this->ReplacementRange = Range(DecomposedLocation.second, Length);
    this->ReplacementText = ReplacementText;
}

void TextDiagnostic::emitParseableFixits(ArrayRef<FixItHint> Hints,
                                         const SourceManager &SM) {
  if (!DiagOpts->ShowParseableFixits)
    return;

  // We follow FixItRewriter's example in not (yet) handling
  // fix-its in macros.
  for (ArrayRef<FixItHint>::iterator I = Hints.begin(), E = Hints.end();
       I != E; ++I) {
    if (I->RemoveRange.isInvalid() ||
        I->RemoveRange.getBegin().isMacroID() ||
        I->RemoveRange.getEnd().isMacroID())
      return;
  }

  for (ArrayRef<FixItHint>::iterator I = Hints.begin(), E = Hints.end();
       I != E; ++I) {
    SourceLocation BLoc = I->RemoveRange.getBegin();
    SourceLocation ELoc = I->RemoveRange.getEnd();

    std::pair<FileID, unsigned> BInfo = SM.getDecomposedLoc(BLoc);
    std::pair<FileID, unsigned> EInfo = SM.getDecomposedLoc(ELoc);

    // Adjust for token ranges.
    if (I->RemoveRange.isTokenRange())
      EInfo.second += Lexer::MeasureTokenLength(ELoc, SM, LangOpts);

    // We specifically do not do word-wrapping or tab-expansion here,
    // because this is supposed to be easy to parse.
    PresumedLoc PLoc = SM.getPresumedLoc(BLoc);
    if (PLoc.isInvalid())
      break;

    OS << "fix-it:\"";
    OS.write_escaped(PLoc.getFilename());
    OS << "\":{" << SM.getLineNumber(BInfo.first, BInfo.second)
      << ':' << SM.getColumnNumber(BInfo.first, BInfo.second)
      << '-' << SM.getLineNumber(EInfo.first, EInfo.second)
      << ':' << SM.getColumnNumber(EInfo.first, EInfo.second)
      << "}:\"";
    OS.write_escaped(I->CodeToInsert);
    OS << "\"\n";
  }
}

// Re-lex the tokens to get precise locations to insert 'override' and remove
// 'virtual'.
static SmallVector<Token, 16> ParseTokens(CharSourceRange Range,
                                          const SourceManager &Sources,
                                          LangOptions LangOpts) {
  std::pair<FileID, unsigned> LocInfo =
      Sources.getDecomposedLoc(Range.getBegin());
  StringRef File = Sources.getBufferData(LocInfo.first);
  const char *TokenBegin = File.data() + LocInfo.second;
  Lexer RawLexer(Sources.getLocForStartOfFile(LocInfo.first), LangOpts,
                 File.begin(), TokenBegin, File.end());
  SmallVector<Token, 16> Tokens;
  Token Tok;
  while (!RawLexer.LexFromRawLexer(Tok)) {
    if (Tok.is(tok::semi) || Tok.is(tok::l_brace))
      break;
    if (Sources.isBeforeInTranslationUnit(Range.getEnd(), Tok.getLocation()))
      break;
    Tokens.push_back(Tok);
  }
  return Tokens;
}

RawComment::RawComment(const SourceManager &SourceMgr, SourceRange SR,
                       bool Merged, bool ParseAllComments) :
    Range(SR), RawTextValid(false), BriefTextValid(false),
    IsAttached(false), IsTrailingComment(false), IsAlmostTrailingComment(false),
    ParseAllComments(ParseAllComments) {
  // Extract raw comment text, if possible.
  if (SR.getBegin() == SR.getEnd() || getRawText(SourceMgr).empty()) {
    Kind = RCK_Invalid;
    return;
  }

  // Guess comment kind.
  std::pair<CommentKind, bool> K = getCommentKind(RawText, ParseAllComments);

  // Guess whether an ordinary comment is trailing.
  if (ParseAllComments && isOrdinaryKind(K.first)) {
    FileID BeginFileID;
    unsigned BeginOffset;
    std::tie(BeginFileID, BeginOffset) =
        SourceMgr.getDecomposedLoc(Range.getBegin());
    if (BeginOffset != 0) {
      bool Invalid = false;
      const char *Buffer =
          SourceMgr.getBufferData(BeginFileID, &Invalid).data();
      IsTrailingComment |=
          (!Invalid && !onlyWhitespaceOnLineBefore(Buffer, BeginOffset));
    }
  }

  if (!Merged) {
    Kind = K.first;
    IsTrailingComment |= K.second;

    IsAlmostTrailingComment = RawText.startswith("//<") ||
                                 RawText.startswith("/*<");
  } else {
    Kind = RCK_Merged;
    IsTrailingComment =
        IsTrailingComment || mergedCommentIsTrailingComment(RawText);
  }
}