TokenLexer.h revision d0f5e443ceed37e1a72015bd109db8e5fc6206cf 
1//===--- TokenLexer.h - Lex from a token buffer -----------------*- C++ -*-===//
2//
3//                     The LLVM Compiler Infrastructure
4//
5// This file is distributed under the University of Illinois Open Source
6// License. See LICENSE.TXT for details.
7//
8//===----------------------------------------------------------------------===//
9//
10// This file defines the TokenLexer interface.
11//
12//===----------------------------------------------------------------------===//
13
14#ifndef LLVM_CLANG_TOKENLEXER_H
15#define LLVM_CLANG_TOKENLEXER_H
16
17#include "clang/Basic/SourceLocation.h"
18
19namespace clang {
20  class MacroInfo;
21  class Preprocessor;
22  class Token;
23  class MacroArgs;
24
25/// TokenLexer - This implements a lexer that returns token from a macro body
26/// or token stream instead of lexing from a character buffer.  This is used for
27/// macro expansion and _Pragma handling, for example.
28///
29class TokenLexer {
30  /// Macro - The macro we are expanding from.  This is null if expanding a
31  /// token stream.
32  ///
33  MacroInfo *Macro;
34
35  /// ActualArgs - The actual arguments specified for a function-like macro, or
36  /// null.  The TokenLexer owns the pointed-to object.
37  MacroArgs *ActualArgs;
38
39  /// PP - The current preprocessor object we are expanding for.
40  ///
41  Preprocessor &PP;
42
43  /// Tokens - This is the pointer to an array of tokens that the macro is
44  /// defined to, with arguments expanded for function-like macros.  If this is
45  /// a token stream, these are the tokens we are returning.  This points into
46  /// the macro definition we are lexing from, a cache buffer that is owned by
47  /// the preprocessor, or some other buffer that we may or may not own
48  /// (depending on OwnsTokens).
49  /// Note that if it points into Preprocessor's cache buffer, the Preprocessor
50  /// may update the pointer as needed.
51  const Token *Tokens;
52  friend class Preprocessor;
53
54  /// NumTokens - This is the length of the Tokens array.
55  ///
56  unsigned NumTokens;
57
58  /// CurToken - This is the next token that Lex will return.
59  ///
60  unsigned CurToken;
61
62  /// ExpandLocStart/End - The source location range where this macro was
63  /// expanded.
64  SourceLocation ExpandLocStart, ExpandLocEnd;
65
66  /// \brief Source location pointing at the source location entry chunk that
67  /// was reserved for the current macro expansion.
68  SourceLocation MacroExpansionStart;
69
70  /// \brief The offset of the macro expansion in the
71  /// "source location address space".
72  unsigned MacroStartSLocOffset;
73
74  /// \brief FileID/offset of the start of the macro definition.
75  std::pair<FileID, unsigned> MacroDefStartInfo;
76
77  /// Lexical information about the expansion point of the macro: the identifier
78  /// that the macro expanded from had these properties.
79  bool AtStartOfLine : 1;
80  bool HasLeadingSpace : 1;
81
82  /// OwnsTokens - This is true if this TokenLexer allocated the Tokens
83  /// array, and thus needs to free it when destroyed.  For simple object-like
84  /// macros (for example) we just point into the token buffer of the macro
85  /// definition, we don't make a copy of it.
86  bool OwnsTokens : 1;
87
88  /// DisableMacroExpansion - This is true when tokens lexed from the TokenLexer
89  /// should not be subject to further macro expansion.
90  bool DisableMacroExpansion : 1;
91
92  TokenLexer(const TokenLexer&);  // DO NOT IMPLEMENT
93  void operator=(const TokenLexer&); // DO NOT IMPLEMENT
94public:
95  /// Create a TokenLexer for the specified macro with the specified actual
96  /// arguments.  Note that this ctor takes ownership of the ActualArgs pointer.
97  /// ILEnd specifies the location of the ')' for a function-like macro or the
98  /// identifier for an object-like macro.
99  TokenLexer(Token &Tok, SourceLocation ILEnd, MacroArgs *ActualArgs,
100             Preprocessor &pp)
101    : Macro(0), ActualArgs(0), PP(pp), OwnsTokens(false) {
102    Init(Tok, ILEnd, ActualArgs);
103  }
104
105  /// Init - Initialize this TokenLexer to expand from the specified macro
106  /// with the specified argument information.  Note that this ctor takes
107  /// ownership of the ActualArgs pointer.  ILEnd specifies the location of the
108  /// ')' for a function-like macro or the identifier for an object-like macro.
109  void Init(Token &Tok, SourceLocation ILEnd, MacroArgs *ActualArgs);
110
111  /// Create a TokenLexer for the specified token stream.  If 'OwnsTokens' is
112  /// specified, this takes ownership of the tokens and delete[]'s them when
113  /// the token lexer is empty.
114  TokenLexer(const Token *TokArray, unsigned NumToks, bool DisableExpansion,
115             bool ownsTokens, Preprocessor &pp)
116    : Macro(0), ActualArgs(0), PP(pp), OwnsTokens(false) {
117    Init(TokArray, NumToks, DisableExpansion, ownsTokens);
118  }
119
120  /// Init - Initialize this TokenLexer with the specified token stream.
121  /// This does not take ownership of the specified token vector.
122  ///
123  /// DisableExpansion is true when macro expansion of tokens lexed from this
124  /// stream should be disabled.
125  void Init(const Token *TokArray, unsigned NumToks,
126            bool DisableMacroExpansion, bool OwnsTokens);
127
128  ~TokenLexer() { destroy(); }
129
130  /// isNextTokenLParen - If the next token lexed will pop this macro off the
131  /// expansion stack, return 2.  If the next unexpanded token is a '(', return
132  /// 1, otherwise return 0.
133  unsigned isNextTokenLParen() const;
134
135  /// Lex - Lex and return a token from this macro stream.
136  void Lex(Token &Tok);
137
138  /// isParsingPreprocessorDirective - Return true if we are in the middle of a
139  /// preprocessor directive.
140  bool isParsingPreprocessorDirective() const;
141
142private:
143  void destroy();
144
145  /// isAtEnd - Return true if the next lex call will pop this macro off the
146  /// include stack.
147  bool isAtEnd() const {
148    return CurToken == NumTokens;
149  }
150
151  /// PasteTokens - Tok is the LHS of a ## operator, and CurToken is the ##
152  /// operator.  Read the ## and RHS, and paste the LHS/RHS together.  If there
153  /// are is another ## after it, chomp it iteratively.  Return the result as
154  /// Tok.  If this returns true, the caller should immediately return the
155  /// token.
156  bool PasteTokens(Token &Tok);
157
158  /// Expand the arguments of a function-like macro so that we can quickly
159  /// return preexpanded tokens from Tokens.
160  void ExpandFunctionArguments();
161
162  /// HandleMicrosoftCommentPaste - In microsoft compatibility mode, /##/ pastes
163  /// together to form a comment that comments out everything in the current
164  /// macro, other active macros, and anything left on the current physical
165  /// source line of the expanded buffer.  Handle this by returning the
166  /// first token on the next line.
167  void HandleMicrosoftCommentPaste(Token &Tok);
168
169  /// \brief If \arg loc is a FileID and points inside the current macro
170  /// definition, returns the appropriate source location pointing at the
171  /// macro expansion source location entry.
172  SourceLocation getMacroExpansionLocation(SourceLocation loc) const;
173};
174
175}  // end namespace clang
176
177#endif
178