path: root/include/clang/Lex/Preprocessor.h

//===--- Preprocessor.h - C Language Family Preprocessor --------*- 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 Preprocessor interface.
//
//===----------------------------------------------------------------------===//

#ifndef LLVM_CLANG_LEX_PREPROCESSOR_H
#define LLVM_CLANG_LEX_PREPROCESSOR_H

#include "clang/Lex/Lexer.h"
#include "clang/Lex/PTHLexer.h"
#include "clang/Lex/PPCallbacks.h"
#include "clang/Lex/TokenLexer.h"
#include "clang/Lex/PTHManager.h"
#include "clang/Basic/Diagnostic.h"
#include "clang/Basic/IdentifierTable.h"
#include "clang/Basic/SourceLocation.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/OwningPtr.h"

namespace clang {
  
class SourceManager;
class FileManager;
class FileEntry;
class HeaderSearch;
class PragmaNamespace;
class PragmaHandler;
class ScratchBuffer;
class TargetInfo;
class PPCallbacks;
class DirectoryLookup;
  
/// Preprocessor - This object engages in a tight little dance with the lexer to
/// efficiently preprocess tokens.  Lexers know only about tokens within a
/// single source file, and don't know anything about preprocessor-level issues
/// like the #include stack, token expansion, etc.
///
class Preprocessor {
  Diagnostic        &Diags;
  const LangOptions &Features;
  TargetInfo        &Target;
  FileManager       &FileMgr;
  SourceManager     &SourceMgr;
  ScratchBuffer     *ScratchBuf;
  HeaderSearch      &HeaderInfo;
  
  /// PTH - An optional PTHManager object used for getting tokens from
  ///  a token cache rather than lexing the original source file.
  llvm::OwningPtr<PTHManager> PTH;
    
  /// Identifiers for builtin macros and other builtins.
  IdentifierInfo *Ident__LINE__, *Ident__FILE__;   // __LINE__, __FILE__
  IdentifierInfo *Ident__DATE__, *Ident__TIME__;   // __DATE__, __TIME__
  IdentifierInfo *Ident__INCLUDE_LEVEL__;          // __INCLUDE_LEVEL__
  IdentifierInfo *Ident__BASE_FILE__;              // __BASE_FILE__
  IdentifierInfo *Ident__TIMESTAMP__;              // __TIMESTAMP__
  IdentifierInfo *Ident_Pragma, *Ident__VA_ARGS__; // _Pragma, __VA_ARGS__
  
  SourceLocation DATELoc, TIMELoc;

  enum {
    /// MaxIncludeStackDepth - Maximum depth of #includes.
    MaxAllowedIncludeStackDepth = 200
  };

  // State that is set before the preprocessor begins.
  bool KeepComments : 1;
  bool KeepMacroComments : 1;
  
  // State that changes while the preprocessor runs:
  bool DisableMacroExpansion : 1;  // True if macro expansion is disabled.
  bool InMacroArgs : 1;            // True if parsing fn macro invocation args.

  /// Identifiers - This is mapping/lookup information for all identifiers in
  /// the program, including program keywords.
  IdentifierTable Identifiers;
  
  /// Selectors - This table contains all the selectors in the program. Unlike
  /// IdentifierTable above, this table *isn't* populated by the preprocessor.
  /// It is declared/instantiated here because it's role/lifetime is 
  /// conceptually similar the IdentifierTable. In addition, the current control
  /// flow (in clang::ParseAST()), make it convenient to put here. 
  /// FIXME: Make sure the lifetime of Identifiers/Selectors *isn't* tied to
  /// the lifetime fo the preprocessor.
  SelectorTable Selectors;

  /// PragmaHandlers - This tracks all of the pragmas that the client registered
  /// with this preprocessor.
  PragmaNamespace *PragmaHandlers;
  
  /// CurLexer - This is the current top of the stack that we're lexing from if
  /// not expanding a macro and we are lexing directly from source code.
  ///  Only one of CurLexer, CurPTHLexer, or CurTokenLexer will be non-null.
  llvm::OwningPtr<Lexer> CurLexer;
  
  /// CurPTHLexer - This is the current top of stack that we're lexing from if
  ///  not expanding from a macro and we are lexing from a PTH cache.
  ///  Only one of CurLexer, CurPTHLexer, or CurTokenLexer will be non-null.
  llvm::OwningPtr<PTHLexer> CurPTHLexer;
  
  /// CurPPLexer - This is the current top of the stack what we're lexing from
  ///  if not expanding a macro.  This is an alias for either CurLexer or
  ///  CurPTHLexer.
  PreprocessorLexer* CurPPLexer;
  
  /// CurLookup - The DirectoryLookup structure used to find the current
  /// FileEntry, if CurLexer is non-null and if applicable.  This allows us to
  /// implement #include_next and find directory-specific properties.
  const DirectoryLookup *CurDirLookup;

  /// CurTokenLexer - This is the current macro we are expanding, if we are
  /// expanding a macro.  One of CurLexer and CurTokenLexer must be null.
  llvm::OwningPtr<TokenLexer> CurTokenLexer;
  
  /// IncludeMacroStack - This keeps track of the stack of files currently
  /// #included, and macros currently being expanded from, not counting
  /// CurLexer/CurTokenLexer.
  struct IncludeStackInfo {
    Lexer                 *TheLexer;
    PTHLexer              *ThePTHLexer;
    PreprocessorLexer     *ThePPLexer;
    TokenLexer            *TheTokenLexer;    
    const DirectoryLookup *TheDirLookup;

    IncludeStackInfo(Lexer *L, PTHLexer* P, PreprocessorLexer* PPL,
                     TokenLexer* TL, const DirectoryLookup *D)
      : TheLexer(L), ThePTHLexer(P), ThePPLexer(PPL), TheTokenLexer(TL),
        TheDirLookup(D) {}
  };
  std::vector<IncludeStackInfo> IncludeMacroStack;
  
  /// Callbacks - These are actions invoked when some preprocessor activity is
  /// encountered (e.g. a file is #included, etc).
  PPCallbacks *Callbacks;
  
  /// Macros - For each IdentifierInfo with 'HasMacro' set, we keep a mapping
  /// to the actual definition of the macro.
  llvm::DenseMap<IdentifierInfo*, MacroInfo*> Macros;
  
  // Various statistics we track for performance analysis.
  unsigned NumDirectives, NumIncluded, NumDefined, NumUndefined, NumPragma;
  unsigned NumIf, NumElse, NumEndif;
  unsigned NumEnteredSourceFiles, MaxIncludeStackDepth;
  unsigned NumMacroExpanded, NumFnMacroExpanded, NumBuiltinMacroExpanded;
  unsigned NumFastMacroExpanded, NumTokenPaste, NumFastTokenPaste;
  unsigned NumSkipped;
  
  /// Predefines - This string is the predefined macros that preprocessor
  /// should use from the command line etc.
  std::string Predefines;
  
  /// TokenLexerCache - Cache macro expanders to reduce malloc traffic.
  enum { TokenLexerCacheSize = 8 };
  unsigned NumCachedTokenLexers;
  TokenLexer *TokenLexerCache[TokenLexerCacheSize];

private:  // Cached tokens state.
  typedef std::vector<Token> CachedTokensTy;

  /// CachedTokens - Cached tokens are stored here when we do backtracking or
  /// lookahead. They are "lexed" by the CachingLex() method.
  CachedTokensTy CachedTokens;

  /// CachedLexPos - The position of the cached token that CachingLex() should
  /// "lex" next. If it points beyond the CachedTokens vector, it means that
  /// a normal Lex() should be invoked.
  CachedTokensTy::size_type CachedLexPos;

  /// BacktrackPositions - Stack of backtrack positions, allowing nested
  /// backtracks. The EnableBacktrackAtThisPos() method pushes a position to
  /// indicate where CachedLexPos should be set when the BackTrack() method is
  /// invoked (at which point the last position is popped).
  std::vector<CachedTokensTy::size_type> BacktrackPositions;

public:
  Preprocessor(Diagnostic &diags, const LangOptions &opts, TargetInfo &target,
               SourceManager &SM, HeaderSearch &Headers);
  ~Preprocessor();

  Diagnostic &getDiagnostics() const { return Diags; }
  const LangOptions &getLangOptions() const { return Features; }
  TargetInfo &getTargetInfo() const { return Target; }
  FileManager &getFileManager() const { return FileMgr; }
  SourceManager &getSourceManager() const { return SourceMgr; }
  HeaderSearch &getHeaderSearchInfo() const { return HeaderInfo; }

  IdentifierTable &getIdentifierTable() { return Identifiers; }
  SelectorTable &getSelectorTable() { return Selectors; }
  
  void setPTHManager(PTHManager* pm) { PTH.reset(pm); }

  /// SetCommentRetentionState - Control whether or not the preprocessor retains
  /// comments in output.
  void SetCommentRetentionState(bool KeepComments, bool KeepMacroComments) {
    this->KeepComments = KeepComments | KeepMacroComments;
    this->KeepMacroComments = KeepMacroComments;
  }
  
  bool getCommentRetentionState() const { return KeepComments; }
  
  /// isCurrentLexer - Return true if we are lexing directly from the specified
  /// lexer.
  bool isCurrentLexer(const PreprocessorLexer *L) const {
    return CurPPLexer == L;
  }
  
  /// getCurrentLexer - Return the current file lexer being lexed from.  Note
  /// that this ignores any potentially active macro expansions and _Pragma
  /// expansions going on at the time.
  PreprocessorLexer *getCurrentFileLexer() const;
  
  /// getPPCallbacks/setPPCallbacks - Accessors for preprocessor