summaryrefslogtreecommitdiff
path: root/clang/lib/Lex/MacroArgs.cpp
blob: e2b251a4499da82f721049d451a1200bf9f31d98 (about) (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
//===--- TokenLexer.cpp - Lex from a token stream -------------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements the TokenLexer interface.
//
//===----------------------------------------------------------------------===//

#include "MacroArgs.h"
#include "clang/Lex/MacroInfo.h"
#include "clang/Lex/Preprocessor.h"
#include "clang/Lex/LexDiagnostic.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/Support/SaveAndRestore.h"
#include <algorithm>

using namespace clang;

/// MacroArgs ctor function - This destroys the vector passed in.
MacroArgs *MacroArgs::create(const MacroInfo *MI,
                             llvm::ArrayRef<Token> UnexpArgTokens,
                             bool VarargsElided, Preprocessor &PP) {
  assert(MI->isFunctionLike() &&
         "Can't have args for an object-like macro!");
  MacroArgs **ResultEnt = 0;
  unsigned ClosestMatch = ~0U;
  
  // See if we have an entry with a big enough argument list to reuse on the
  // free list.  If so, reuse it.
  for (MacroArgs **Entry = &PP.MacroArgCache; *Entry;
       Entry = &(*Entry)->ArgCache)
    if ((*Entry)->NumUnexpArgTokens >= UnexpArgTokens.size() &&
        (*Entry)->NumUnexpArgTokens < ClosestMatch) {
      ResultEnt = Entry;
      
      // If we have an exact match, use it.
      if ((*Entry)->NumUnexpArgTokens == UnexpArgTokens.size())
        break;
      // Otherwise, use the best fit.
      ClosestMatch = (*Entry)->NumUnexpArgTokens;
    }
  
  MacroArgs *Result;
  if (ResultEnt == 0) {
    // Allocate memory for a MacroArgs object with the lexer tokens at the end.
    Result = (MacroArgs*)malloc(sizeof(MacroArgs) + 
                                UnexpArgTokens.size() * sizeof(Token));
    // Construct the MacroArgs object.
    new (Result) MacroArgs(UnexpArgTokens.size(), VarargsElided);
  } else {
    Result = *ResultEnt;
    // Unlink this node from the preprocessors singly linked list.
    *ResultEnt = Result->ArgCache;
    Result->NumUnexpArgTokens = UnexpArgTokens.size();
    Result->VarargsElided = VarargsElided;
  }

  // Copy the actual unexpanded tokens to immediately after the result ptr.
  if (!UnexpArgTokens.empty())
    std::copy(UnexpArgTokens.begin(), UnexpArgTokens.end(), 
              const_cast<Token*>(Result->getUnexpArgument(0)));

  return Result;
}

/// destroy - Destroy and deallocate the memory for this object.
///
void MacroArgs::destroy(Preprocessor &PP) {
  StringifiedArgs.clear();

  // Don't clear PreExpArgTokens, just clear the entries.  Clearing the entries
  // would deallocate the element vectors.
  for (unsigned i = 0, e = PreExpArgTokens.size(); i != e; ++i)
    PreExpArgTokens[i].clear();
  
  // Add this to the preprocessor's free list.
  ArgCache = PP.MacroArgCache;
  PP.MacroArgCache = this;
}

/// deallocate - This should only be called by the Preprocessor when managing
/// its freelist.
MacroArgs *MacroArgs::deallocate() {
  MacroArgs *Next = ArgCache;
  
  // Run the dtor to deallocate the vectors.
  this->~MacroArgs();
  // Release the memory for the object.
  free(this);
  
  return Next;
}


/// getArgLength - Given a pointer to an expanded or unexpanded argument,
/// return the number of tokens, not counting the EOF, that make up the
/// argument.
unsigned MacroArgs::getArgLength(const Token *ArgPtr) {
  unsigned NumArgTokens = 0;
  for (; ArgPtr->isNot(tok::eof); ++ArgPtr)
    ++NumArgTokens;
  return NumArgTokens;
}


/// getUnexpArgument - Return the unexpanded tokens for the specified formal.
///
const Token *MacroArgs::getUnexpArgument(unsigned Arg) const {
  // The unexpanded argument tokens start immediately after the MacroArgs object
  // in memory.
  const Token *Start = (const Token *)(this+1);
  const Token *Result = Start;
  // Scan to find Arg.
  for (; Arg; ++Result) {
    assert(Result < Start+NumUnexpArgTokens && "Invalid arg #");
    if (Result->is(tok::eof))
      --Arg;
  }
  assert(Result < Start+NumUnexpArgTokens && "Invalid arg #");
  return Result;
}


/// ArgNeedsPreexpansion - If we can prove that the argument won't be affected
/// by pre-expansion, return false.  Otherwise, conservatively return true.
bool MacroArgs::ArgNeedsPreexpansion(const Token *ArgTok,
                                     Preprocessor &PP) const {
  // If there are no identifiers in the argument list, or if the identifiers are
  // known to not be macros, pre-expansion won't modify it.
  for (; ArgTok->isNot(tok::eof); ++ArgTok)
    if (IdentifierInfo *II = ArgTok->getIdentifierInfo()) {
      if (II->hasMacroDefinition() && PP.getMacroInfo(II)->isEnabled())
        // Return true even though the macro could be a function-like macro
        // without a following '(' token.
        return true;
    }
  return false;
}

/// getPreExpArgument - Return the pre-expanded form of the specified
/// argument.
const std::vector<Token> &
MacroArgs::getPreExpArgument(unsigned Arg, const MacroInfo *MI, 
                             Preprocessor &PP) {
  assert(Arg < MI->getNumArgs() && "Invalid argument number!");

  // If we have already computed this, return it.
  if (PreExpArgTokens.size() < MI->getNumArgs())
    PreExpArgTokens.resize(MI->getNumArgs());
  
  std::vector<Token> &Result = PreExpArgTokens[Arg];
  if (!Result.empty()) return Result;

  SaveAndRestore<bool> PreExpandingMacroArgs(PP.InMacroArgPreExpansion, true);

  const Token *AT = getUnexpArgument(Arg);
  unsigned NumToks = getArgLength(AT)+1;  // Include the EOF.

  // Otherwise, we have to pre-expand this argument, populating Result.  To do
  // this, we set up a fake TokenLexer to lex from the unexpanded argument
  // list.  With this installed, we lex expanded tokens until we hit the EOF
  // token at the end of the unexp list.
  PP.EnterTokenStream(AT, NumToks, false /*disable expand*/,
                      false /*owns tokens*/);

  // Lex all of the macro-expanded tokens into Result.
  do {
    Result.push_back(Token());
    Token &Tok = Result.back();
    PP.Lex(Tok);
  } while (Result.back().isNot(tok::eof));

  // Pop the token stream off the top of the stack.  We know that the internal
  // pointer inside of it is to the "end" of the token stream, but the stack
  // will not otherwise be popped until the next token is lexed.  The problem is
  // that the token may be lexed sometime after the vector of tokens itself is
  // destroyed, which would be badness.
  if (PP.InCachingLexMode())
    PP.ExitCachingLexMode();
  PP.RemoveTopOfLexerStack();
  return Result;
}


/// StringifyArgument - Implement C99 6.10.3.2p2, converting a sequence of
/// tokens into the literal string token that should be produced by the C #
/// preprocessor operator.  If Charify is true, then it should be turned into
/// a character literal for the Microsoft charize (#@) extension.
///
Token MacroArgs::StringifyArgument(const Token *ArgToks,
                                   Preprocessor &PP, bool Charify,
                                   SourceLocation ExpansionLocStart,
                                   SourceLocation ExpansionLocEnd) {
  Token Tok;
  Tok.startToken();
  Tok.setKind(Charify ? tok::char_constant : tok::string_literal);

  const Token *ArgTokStart = ArgToks;

  // Stringify all the tokens.
  SmallString<128> Result;
  Result += "\"";

  bool isFirst = true;
  for (; ArgToks->isNot(tok::eof); ++ArgToks) {
    const Token &Tok = *ArgToks;
    if (!isFirst && (Tok.hasLeadingSpace() || Tok.isAtStartOfLine()))
      Result += ' ';
    isFirst = false;

    // If this is a string or character constant, escape the token as specified
    // by 6.10.3.2p2.
    if (Tok.is(tok::string_literal) ||       // "foo"
        Tok.is(tok::wide_string_literal) ||  // L"foo"
        Tok.is(tok::utf8_string_literal) ||  // u8"foo"
        Tok.is(tok::utf16_string_literal) || // u"foo"
        Tok.is(tok::utf32_string_literal) || // U"foo"
        Tok.is(tok::char_constant) ||        // 'x'
        Tok.is(tok::wide_char_constant) ||   // L'x'.
        Tok.is(tok::utf16_char_constant) ||  // u'x'.
        Tok.is(tok::utf32_char_constant)) {  // U'x'.
      bool Invalid = false;
      std::string TokStr = PP.getSpelling(Tok, &Invalid);
      if (!Invalid) {
        std::string Str = Lexer::Stringify(TokStr);
        Result.append(Str.begin(), Str.end());
      }
    } else if (Tok.is(tok::code_completion)) {
      PP.CodeCompleteNaturalLanguage();
    } else {
      // Otherwise, just append the token.  Do some gymnastics to get the token
      // in place and avoid copies where possible.
      unsigned CurStrLen = Result.size();
      Result.resize(CurStrLen+Tok.getLength());
      const char *BufPtr = &Result[CurStrLen];
      bool Invalid = false;
      unsigned ActualTokLen = PP.getSpelling(Tok, BufPtr, &Invalid);

      if (!Invalid) {
        // If getSpelling returned a pointer to an already uniqued version of
        // the string instead of filling in BufPtr, memcpy it onto our string.
        if (BufPtr != &Result[CurStrLen])
          memcpy(&Result[CurStrLen], BufPtr, ActualTokLen);

        // If the token was dirty, the spelling may be shorter than the token.
        if (ActualTokLen != Tok.getLength())
          Result.resize(CurStrLen+ActualTokLen);
      }
    }
  }

  // If the last character of the string is a \, and if it isn't escaped, this
  // is an invalid string literal, diagnose it as specified in C99.
  if (Result.back() == '\\') {
    // Count the number of consequtive \ characters.  If even, then they are
    // just escaped backslashes, otherwise it's an error.
    unsigned FirstNonSlash = Result.size()-2;
    // Guaranteed to find the starting " if nothing else.
    while (Result[FirstNonSlash] == '\\')
      --FirstNonSlash;
    if ((Result.size()-1-FirstNonSlash) & 1) {
      // Diagnose errors for things like: #define F(X) #X   /   F(\)
      PP.Diag(ArgToks[-1], diag::pp_invalid_string_literal);
      Result.pop_back();  // remove one of the \'s.
    }
  }
  Result += '"';

  // If this is the charify operation and the result is not a legal character
  // constant, diagnose it.
  if (Charify) {
    // First step, turn double quotes into single quotes:
    Result[0] = '\'';
    Result[Result.size()-1] = '\'';

    // Check for bogus character.
    bool isBad = false;
    if (Result.size() == 3)
      isBad = Result[1] == '\'';   // ''' is not legal. '\' already fixed above.
    else
      isBad = (Result.size() != 4 || Result[1] != '\\');  // Not '\x'

    if (isBad) {
      PP.Diag(ArgTokStart[0], diag::err_invalid_character_to_charify);
      Result = "' '";  // Use something arbitrary, but legal.
    }
  }

  PP.CreateString(&Result[0], Result.size(), Tok,
                  ExpansionLocStart, ExpansionLocEnd);
  return Tok;
}

/// getStringifiedArgument - Compute, cache, and return the specified argument
/// that has been 'stringified' as required by the # operator.
const Token &MacroArgs::getStringifiedArgument(unsigned ArgNo,
                                               Preprocessor &PP,
                                               SourceLocation ExpansionLocStart,
                                               SourceLocation ExpansionLocEnd) {
  assert(ArgNo < NumUnexpArgTokens && "Invalid argument number!");
  if (StringifiedArgs.empty()) {
    StringifiedArgs.resize(getNumArguments());
    memset((void*)&StringifiedArgs[0], 0,
           sizeof(StringifiedArgs[0])*getNumArguments());
  }
  if (StringifiedArgs[ArgNo].isNot(tok::string_literal))
    StringifiedArgs[ArgNo] = StringifyArgument(getUnexpArgument(ArgNo), PP,
                                               /*Charify=*/false,
                                               ExpansionLocStart,
                                               ExpansionLocEnd);
  return StringifiedArgs[ArgNo];
}