path: root/lib/Rewrite/Frontend/InclusionRewriter.cpp

//===--- InclusionRewriter.cpp - Rewrite includes into their expansions ---===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This code rewrites include invocations into their expansions.  This gives you
// a file with all included files merged into it.
//
//===----------------------------------------------------------------------===//

#include "clang/Rewrite/Frontend/Rewriters.h"
#include "clang/Basic/SourceManager.h"
#include "clang/Frontend/PreprocessorOutputOptions.h"
#include "clang/Lex/HeaderSearch.h"
#include "clang/Lex/Preprocessor.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/Support/raw_ostream.h"

using namespace clang;
using namespace llvm;

namespace {

class InclusionRewriter : public PPCallbacks {
  /// Information about which #includes were actually performed,
  /// created by preprocessor callbacks.
  struct FileChange {
    const Module *Mod;
    SourceLocation From;
    FileID Id;
    SrcMgr::CharacteristicKind FileType;
    FileChange(SourceLocation From, const Module *Mod) : Mod(Mod), From(From) {
    }
  };
  Preprocessor &PP; ///< Used to find inclusion directives.
  SourceManager &SM; ///< Used to read and manage source files.
  raw_ostream &OS; ///< The destination stream for rewritten contents.
  bool ShowLineMarkers; ///< Show #line markers.
  bool UseLineDirective; ///< Use of line directives or line markers.
  typedef std::map<unsigned, FileChange> FileChangeMap;
  FileChangeMap FileChanges; ///< Tracks which files were included where.
  /// Used transitively for building up the FileChanges mapping over the
  /// various \c PPCallbacks callbacks.
  FileChangeMap::iterator LastInsertedFileChange;
public:
  InclusionRewriter(Preprocessor &PP, raw_ostream &OS, bool ShowLineMarkers);
  bool Process(FileID FileId, SrcMgr::CharacteristicKind FileType);
private:
  virtual void FileChanged(SourceLocation Loc, FileChangeReason Reason,
                           SrcMgr::CharacteristicKind FileType,
                           FileID PrevFID);
  virtual void FileSkipped(const FileEntry &ParentFile,
                           const Token &FilenameTok,
                           SrcMgr::CharacteristicKind FileType);
  virtual void InclusionDirective(SourceLocation HashLoc,
                                  const Token &IncludeTok,
                                  StringRef FileName,
                                  bool IsAngled,
                                  CharSourceRange FilenameRange,
                                  const FileEntry *File,
                                  StringRef SearchPath,
                                  StringRef RelativePath,
                                  const Module *Imported);
  void WriteLineInfo(const char *Filename, int Line,
                     SrcMgr::CharacteristicKind FileType,
                     StringRef EOL, StringRef Extra = StringRef());
  void WriteImplicitModuleImport(const Module *Mod, StringRef EOL);
  void OutputContentUpTo(const MemoryBuffer &FromFile,
                         unsigned &WriteFrom, unsigned WriteTo,
                         StringRef EOL, int &lines,
                         bool EnsureNewline = false);
  void CommentOutDirective(Lexer &DirectivesLex, const Token &StartToken,
                           const MemoryBuffer &FromFile, StringRef EOL,
                           unsigned &NextToWrite, int &Lines);
  bool HandleHasInclude(FileID FileId, Lexer &RawLex,
                        const DirectoryLookup *Lookup, Token &Tok,
                        bool &FileExists);
  const FileChange *FindFileChangeLocation(SourceLocation Loc) const;
  StringRef NextIdentifierName(Lexer &RawLex, Token &RawToken);
};

}  // end anonymous namespace

/// Initializes an InclusionRewriter with a \p PP source and \p OS destination.
InclusionRewriter::InclusionRewriter(Preprocessor &PP, raw_ostream &OS,
                                     bool ShowLineMarkers)
    : PP(PP), SM(PP.getSourceManager()), OS(OS),
    ShowLineMarkers(ShowLineMarkers),
    LastInsertedFileChange(FileChanges.end()) {
  // If we're in microsoft mode, use normal #line instead of line markers.
  UseLineDirective = PP.getLangOpts().MicrosoftExt;
}

/// Write appropriate line information as either #line directives or GNU line
/// markers depending on what mode we're in, including the \p Filename and
/// \p Line we are located at, using the specified \p EOL line separator, and
/// any \p Extra context specifiers in GNU line directives.
void InclusionRewriter::WriteLineInfo(const char *Filename, int Line,
                                      SrcMgr::CharacteristicKind FileType,
                                      StringRef EOL, StringRef Extra) {
  if (!ShowLineMarkers)
    return;
  if (UseLineDirective) {
    OS << "#line" << ' ' << Line << ' ' << '"' << Filename << '"';
  } else {
    // Use GNU linemarkers as described here:
    // http://gcc.gnu.org/onlinedocs/cpp/Preprocessor-Output.html
    OS << '#' << ' ' << Line << ' ' << '"' << Filename << '"';
    if (!Extra.empty())
      OS << Extra;
    if (FileType == SrcMgr::C_System)
      // "`3' This indicates that the following text comes from a system header
      // file, so certain warnings should be suppressed."
      OS << " 3";
    else if (FileType == SrcMgr::C_ExternCSystem)
      // as above for `3', plus "`4' This indicates that the following text
      // should be treated as being wrapped in an implicit extern "C" block."
      OS << " 3 4";
  }
  OS << EOL;
}

void InclusionRewriter::WriteImplicitModuleImport(const Module *Mod,
                                                  StringRef EOL) {
  OS << "@import " << Mod->getFullModuleName() << ";"
     << " /* clang -frewrite-includes: implicit import */" << EOL;
}

/// FileChanged - Whenever the preprocessor enters or exits a #include file
/// it invokes this handler.
void InclusionRewriter::FileChanged(SourceLocation Loc,
                                    FileChangeReason Reason,
                                    SrcMgr::CharacteristicKind NewFileType,
                                    FileID) {
  if (Reason != EnterFile)
    return;
  if (LastInsertedFileChange == FileChanges.end())
    // we didn't reach this file (eg: the main file) via an inclusion directive
    return;
  LastInsertedFileChange->second.Id = FullSourceLoc(Loc, SM).getFileID();
  LastInsertedFileChange->second.FileType = NewFileType;
  LastInsertedFileChange = FileChanges.end();
}

/// Called whenever an inclusion is skipped due to canonical header protection
/// macros.
void InclusionRewriter::FileSkipped(const FileEntry &/*ParentFile*/,
                                    const Token &/*FilenameTok*/,
                                    SrcMgr::CharacteristicKind /*FileType*/) {
  assert(LastInsertedFileChange != FileChanges.end() && "A file, that wasn't "
    "found via an inclusion directive, was skipped");
  FileChanges.erase(LastInsertedFileChange);
  LastInsertedFileChange = FileChanges.end();
}

/// This should be called whenever the preprocessor encounters include
/// directives. It does not say whether the file has been included, but it
/// provides more information about the directive (hash location instead
/// of location inside the included file). It is assumed that the matching
/// FileChanged() or FileSkipped() is called after this.
void InclusionRewriter::InclusionDirective(SourceLocation HashLoc,
                                           const Token &/*IncludeTok*/,
                                           StringRef /*FileName*/,
                                           bool /*IsAngled*/,
                                           CharSourceRange /*FilenameRange*/,
                                           const FileEntry * /*File*/,
                                           StringRef /*SearchPath*/,
                                           StringRef /*RelativePath*/,
                                           const Module *Imported) {
  assert(LastInsertedFileChange == FileChanges.end() && "Another inclusion "
    "directive was found before the previous one was processed");
  std::pair<FileChangeMap::iterator, bool> p = FileChanges.insert(
    std::make_pair(HashLoc.getRawEncoding(), FileChange(HashLoc, Imported)));
  assert(p.second && "Unexpected revisitation of the same include directive");
  if (!Imported)
    LastInsertedFileChange = p.first;
}

/// Simple lookup for a SourceLocation (specifically one denoting the hash in
/// an inclusion directive) in the map of inclusion information, FileChanges.
const InclusionRewriter::FileChange *
InclusionRewriter::FindFileChangeLocation(SourceLocation Loc) const {
  FileChangeMap::const_iterator I = FileChanges.find(Loc.getRawEncoding());
  if (I != FileChanges.end())
    return &I->second;
  return NULL;
}

/// Detect the likely line ending style of \p FromFile by examining the first
/// newline found within it.
static StringRef DetectEOL(const MemoryBuffer &FromFile) {
  // detect what line endings the file uses, so that added content does not mix
  // the style
  const char *Pos = strchr(FromFile.getBufferStar