1//===- TokenLexer.h - Lex from a token buffer -------------------*- C++ -*-===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9// This file defines the TokenLexer interface.
10//
11//===----------------------------------------------------------------------===//
12
13#ifndef LLVM_CLANG_LEX_TOKENLEXER_H
14#define LLVM_CLANG_LEX_TOKENLEXER_H
15
16#include "clang/Basic/SourceLocation.h"
17#include "llvm/ADT/ArrayRef.h"
18
19namespace clang {
20
21class MacroArgs;
22class MacroInfo;
23class Preprocessor;
24class Token;
25class VAOptExpansionContext;
26
27/// TokenLexer - This implements a lexer that returns tokens from a macro body
28/// or token stream instead of lexing from a character buffer.  This is used for
29/// macro expansion and _Pragma handling, for example.
30class TokenLexer {
31  friend class Preprocessor;
32
33  /// The macro we are expanding from. This is null if expanding a token stream.
34  MacroInfo *Macro = nullptr;
35
36  /// The actual arguments specified for a function-like macro, or null. The
37  /// TokenLexer owns the pointed-to object.
38  MacroArgs *ActualArgs = nullptr;
39
40  /// The current preprocessor object we are expanding for.
41  Preprocessor &PP;
42
43  /// 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
53  /// This is the length of the Tokens array.
54  unsigned NumTokens;
55
56  /// This is the index of the next token that Lex will return.
57  unsigned CurTokenIdx;
58
59  /// The source location range where this macro was expanded.
60  SourceLocation ExpandLocStart, ExpandLocEnd;
61
62  /// Source location pointing at the source location entry chunk that
63  /// was reserved for the current macro expansion.
64  SourceLocation MacroExpansionStart;
65
66  /// The offset of the macro expansion in the
67  /// "source location address space".
68  unsigned MacroStartSLocOffset;
69
70  /// Location of the macro definition.
71  SourceLocation MacroDefStart;
72
73  /// Length of the macro definition.
74  unsigned MacroDefLength;
75
76  /// Lexical information about the expansion point of the macro: the identifier
77  /// that the macro expanded from had these properties.
78  bool AtStartOfLine : 1;
79  bool HasLeadingSpace : 1;
80
81  // When this is true, the next token appended to the
82  // output list during function argument expansion will get a leading space,
83  // regardless of whether it had one to begin with or not. This is used for
84  // placemarker support. If still true after function argument expansion, the
85  // leading space will be applied to the first token following the macro
86  // expansion.
87  bool NextTokGetsSpace : 1;
88
89  /// This is true if this TokenLexer allocated the Tokens
90  /// array, and thus needs to free it when destroyed.  For simple object-like
91  /// macros (for example) we just point into the token buffer of the macro
92  /// definition, we don't make a copy of it.
93  bool OwnsTokens : 1;
94
95  /// This is true when tokens lexed from the TokenLexer
96  /// should not be subject to further macro expansion.
97  bool DisableMacroExpansion : 1;
98
99  /// When true, the produced tokens have Token::IsReinjected flag set.
100  /// See the flag documentation for details.
101  bool IsReinject : 1;
102
103public:
104  /// Create a TokenLexer for the specified macro with the specified actual
105  /// arguments.  Note that this ctor takes ownership of the ActualArgs pointer.
106  /// ILEnd specifies the location of the ')' for a function-like macro or the
107  /// identifier for an object-like macro.
108  TokenLexer(Token &Tok, SourceLocation ILEnd, MacroInfo *MI,
109             MacroArgs *ActualArgs, Preprocessor &pp)
110      : PP(pp), OwnsTokens(false) {
111    Init(Tok, ILEnd, MI, ActualArgs);
112  }
113
114  /// Create a TokenLexer for the specified token stream.  If 'OwnsTokens' is
115  /// specified, this takes ownership of the tokens and delete[]'s them when
116  /// the token lexer is empty.
117  TokenLexer(const Token *TokArray, unsigned NumToks, bool DisableExpansion,
118             bool ownsTokens, bool isReinject, Preprocessor &pp)
119      : PP(pp), OwnsTokens(false) {
120    Init(TokArray, NumToks, DisableExpansion, ownsTokens, isReinject);
121  }
122
123  TokenLexer(const TokenLexer &) = delete;
124  TokenLexer &operator=(const TokenLexer &) = delete;
125  ~TokenLexer() { destroy(); }
126
127  /// Initialize this TokenLexer to expand from the specified macro
128  /// with the specified argument information.  Note that this ctor takes
129  /// ownership of the ActualArgs pointer.  ILEnd specifies the location of the
130  /// ')' for a function-like macro or the identifier for an object-like macro.
131  void Init(Token &Tok, SourceLocation ELEnd, MacroInfo *MI,
132            MacroArgs *Actuals);
133
134  /// Initialize this TokenLexer with the specified token stream.
135  /// This does not take ownership of the specified token vector.
136  ///
137  /// DisableExpansion is true when macro expansion of tokens lexed from this
138  /// stream should be disabled.
139  void Init(const Token *TokArray, unsigned NumToks, bool DisableMacroExpansion,
140            bool OwnsTokens, bool IsReinject);
141
142  /// If the next token lexed will pop this macro off the
143  /// expansion stack, return 2.  If the next unexpanded token is a '(', return
144  /// 1, otherwise return 0.
145  unsigned isNextTokenLParen() const;
146
147  /// Lex and return a token from this macro stream.
148  bool Lex(Token &Tok);
149
150  /// isParsingPreprocessorDirective - Return true if we are in the middle of a
151  /// preprocessor directive.
152  bool isParsingPreprocessorDirective() const;
153
154private:
155  void destroy();
156
157  /// Return true if the next lex call will pop this macro off the include
158  /// stack.
159  bool isAtEnd() const {
160    return CurTokenIdx == NumTokens;
161  }
162
163  /// Concatenates the next (sub-)sequence of \p Tokens separated by '##'
164  /// starting with LHSTok - stopping when we encounter a token that is neither
165  /// '##' nor preceded by '##'.  Places the result back into \p LHSTok and sets
166  /// \p CurIdx to point to the token following the last one that was pasted.
167  ///
168  /// Also performs the MSVC extension wide-literal token pasting involved with:
169  ///       \code L #macro-arg. \endcode
170  ///
171  /// \param[in,out] LHSTok - Contains the token to the left of '##' in \p
172  /// Tokens upon entry and will contain the resulting concatenated Token upon
173  /// exit.
174  ///
175  /// \param[in] TokenStream - The stream of Tokens we are lexing from.
176  ///
177  /// \param[in,out] CurIdx - Upon entry, \pTokens[\pCurIdx] must equal '##'
178  /// (with the exception of the MSVC extension mentioned above).  Upon exit, it
179  /// is set to the index of the token following the last token that was
180  /// concatenated together.
181  ///
182  /// \returns If this returns true, the caller should immediately return the
183  /// token.
184  bool pasteTokens(Token &LHSTok, ArrayRef<Token> TokenStream,
185                   unsigned int &CurIdx);
186
187  /// Calls pasteTokens above, passing in the '*this' object's Tokens and
188  /// CurTokenIdx data members.
189  bool pasteTokens(Token &Tok);
190
191
192  /// Takes the tail sequence of tokens within ReplacementToks that represent
193  /// the just expanded __VA_OPT__ tokens (possibly zero tokens) and transforms
194  /// them into a string.  \p VCtx is used to determine which token represents
195  /// the first __VA_OPT__ replacement token.
196  ///
197  /// \param[in,out] ResultToks - Contains the current Replacement Tokens
198  /// (prior to rescanning and token pasting), the tail end of which represents
199  /// the tokens just expanded through __VA_OPT__ processing.  These (sub)
200  /// sequence of tokens are folded into one stringified token.
201  ///
202  /// \param[in] VCtx - contains relevant contextual information about the
203  /// state of the tokens around and including the __VA_OPT__ token, necessary
204  /// for stringification.
205  void stringifyVAOPTContents(SmallVectorImpl<Token> &ResultToks,
206                              const VAOptExpansionContext &VCtx,
207                              SourceLocation VAOPTClosingParenLoc);
208
209  /// Expand the arguments of a function-like macro so that we can quickly
210  /// return preexpanded tokens from Tokens.
211  void ExpandFunctionArguments();
212
213  /// In microsoft compatibility mode, /##/ pastes
214  /// together to form a comment that comments out everything in the current
215  /// macro, other active macros, and anything left on the current physical
216  /// source line of the expanded buffer.  Handle this by returning the
217  /// first token on the next line.
218  void HandleMicrosoftCommentPaste(Token &Tok, SourceLocation OpLoc);
219
220  /// If \p loc is a FileID and points inside the current macro
221  /// definition, returns the appropriate source location pointing at the
222  /// macro expansion source location entry.
223  SourceLocation getExpansionLocForMacroDefLoc(SourceLocation loc) const;
224
225  /// Creates SLocEntries and updates the locations of macro argument
226  /// tokens to their new expanded locations.
227  ///
228  /// \param ArgIdSpellLoc the location of the macro argument id inside the
229  /// macro definition.
230  void updateLocForMacroArgTokens(SourceLocation ArgIdSpellLoc,
231                                  Token *begin_tokens, Token *end_tokens);
232
233  /// Remove comma ahead of __VA_ARGS__, if present, according to compiler
234  /// dialect settings.  Returns true if the comma is removed.
235  bool MaybeRemoveCommaBeforeVaArgs(SmallVectorImpl<Token> &ResultToks,
236                                    bool HasPasteOperator,
237                                    MacroInfo *Macro, unsigned MacroArgNo,
238                                    Preprocessor &PP);
239
240  void PropagateLineStartLeadingSpaceInfo(Token &Result);
241};
242
243} // namespace clang
244
245#endif // LLVM_CLANG_LEX_TOKENLEXER_H
246