Lexer.h revision 263508
1//===--- Lexer.h - C Language Family Lexer ----------------------*- 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 Lexer interface. 11// 12//===----------------------------------------------------------------------===// 13 14#ifndef LLVM_CLANG_LEXER_H 15#define LLVM_CLANG_LEXER_H 16 17#include "clang/Basic/LangOptions.h" 18#include "clang/Lex/PreprocessorLexer.h" 19#include "llvm/ADT/SmallVector.h" 20#include <cassert> 21#include <string> 22 23namespace clang { 24class DiagnosticsEngine; 25class SourceManager; 26class Preprocessor; 27class DiagnosticBuilder; 28 29/// ConflictMarkerKind - Kinds of conflict marker which the lexer might be 30/// recovering from. 31enum ConflictMarkerKind { 32 /// Not within a conflict marker. 33 CMK_None, 34 /// A normal or diff3 conflict marker, initiated by at least 7 "<"s, 35 /// separated by at least 7 "="s or "|"s, and terminated by at least 7 ">"s. 36 CMK_Normal, 37 /// A Perforce-style conflict marker, initiated by 4 ">"s, 38 /// separated by 4 "="s, and terminated by 4 "<"s. 39 CMK_Perforce 40}; 41 42/// Lexer - This provides a simple interface that turns a text buffer into a 43/// stream of tokens. This provides no support for file reading or buffering, 44/// or buffering/seeking of tokens, only forward lexing is supported. It relies 45/// on the specified Preprocessor object to handle preprocessor directives, etc. 46class Lexer : public PreprocessorLexer { 47 virtual void anchor(); 48 49 //===--------------------------------------------------------------------===// 50 // Constant configuration values for this lexer. 51 const char *BufferStart; // Start of the buffer. 52 const char *BufferEnd; // End of the buffer. 53 SourceLocation FileLoc; // Location for start of file. 54 LangOptions LangOpts; // LangOpts enabled by this language (cache). 55 bool Is_PragmaLexer; // True if lexer for _Pragma handling. 56 57 //===--------------------------------------------------------------------===// 58 // Context-specific lexing flags set by the preprocessor. 59 // 60 61 /// ExtendedTokenMode - The lexer can optionally keep comments and whitespace 62 /// and return them as tokens. This is used for -C and -CC modes, and 63 /// whitespace preservation can be useful for some clients that want to lex 64 /// the file in raw mode and get every character from the file. 65 /// 66 /// When this is set to 2 it returns comments and whitespace. When set to 1 67 /// it returns comments, when it is set to 0 it returns normal tokens only. 68 unsigned char ExtendedTokenMode; 69 70 //===--------------------------------------------------------------------===// 71 // Context that changes as the file is lexed. 72 // NOTE: any state that mutates when in raw mode must have save/restore code 73 // in Lexer::isNextPPTokenLParen. 74 75 // BufferPtr - Current pointer into the buffer. This is the next character 76 // to be lexed. 77 const char *BufferPtr; 78 79 // IsAtStartOfLine - True if the next lexed token should get the "start of 80 // line" flag set on it. 81 bool IsAtStartOfLine; 82 83 bool IsAtPhysicalStartOfLine; 84 85 bool HasLeadingSpace; 86 87 bool HasLeadingEmptyMacro; 88 89 // CurrentConflictMarkerState - The kind of conflict marker we are handling. 90 ConflictMarkerKind CurrentConflictMarkerState; 91 92 Lexer(const Lexer &) LLVM_DELETED_FUNCTION; 93 void operator=(const Lexer &) LLVM_DELETED_FUNCTION; 94 friend class Preprocessor; 95 96 void InitLexer(const char *BufStart, const char *BufPtr, const char *BufEnd); 97public: 98 99 /// Lexer constructor - Create a new lexer object for the specified buffer 100 /// with the specified preprocessor managing the lexing process. This lexer 101 /// assumes that the associated file buffer and Preprocessor objects will 102 /// outlive it, so it doesn't take ownership of either of them. 103 Lexer(FileID FID, const llvm::MemoryBuffer *InputBuffer, Preprocessor &PP); 104 105 /// Lexer constructor - Create a new raw lexer object. This object is only 106 /// suitable for calls to 'LexFromRawLexer'. This lexer assumes that the 107 /// text range will outlive it, so it doesn't take ownership of it. 108 Lexer(SourceLocation FileLoc, const LangOptions &LangOpts, 109 const char *BufStart, const char *BufPtr, const char *BufEnd); 110 111 /// Lexer constructor - Create a new raw lexer object. This object is only 112 /// suitable for calls to 'LexFromRawLexer'. This lexer assumes that the 113 /// text range will outlive it, so it doesn't take ownership of it. 114 Lexer(FileID FID, const llvm::MemoryBuffer *InputBuffer, 115 const SourceManager &SM, const LangOptions &LangOpts); 116 117 /// Create_PragmaLexer: Lexer constructor - Create a new lexer object for 118 /// _Pragma expansion. This has a variety of magic semantics that this method 119 /// sets up. It returns a new'd Lexer that must be delete'd when done. 120 static Lexer *Create_PragmaLexer(SourceLocation SpellingLoc, 121 SourceLocation ExpansionLocStart, 122 SourceLocation ExpansionLocEnd, 123 unsigned TokLen, Preprocessor &PP); 124 125 126 /// getLangOpts - Return the language features currently enabled. 127 /// NOTE: this lexer modifies features as a file is parsed! 128 const LangOptions &getLangOpts() const { return LangOpts; } 129 130 /// getFileLoc - Return the File Location for the file we are lexing out of. 131 /// The physical location encodes the location where the characters come from, 132 /// the virtual location encodes where we should *claim* the characters came 133 /// from. Currently this is only used by _Pragma handling. 134 SourceLocation getFileLoc() const { return FileLoc; } 135 136private: 137 /// Lex - Return the next token in the file. If this is the end of file, it 138 /// return the tok::eof token. This implicitly involves the preprocessor. 139 bool Lex(Token &Result); 140 141public: 142 /// isPragmaLexer - Returns true if this Lexer is being used to lex a pragma. 143 bool isPragmaLexer() const { return Is_PragmaLexer; } 144 145private: 146 /// IndirectLex - An indirect call to 'Lex' that can be invoked via 147 /// the PreprocessorLexer interface. 148 void IndirectLex(Token &Result) { Lex(Result); } 149 150public: 151 /// LexFromRawLexer - Lex a token from a designated raw lexer (one with no 152 /// associated preprocessor object. Return true if the 'next character to 153 /// read' pointer points at the end of the lexer buffer, false otherwise. 154 bool LexFromRawLexer(Token &Result) { 155 assert(LexingRawMode && "Not already in raw mode!"); 156 Lex(Result); 157 // Note that lexing to the end of the buffer doesn't implicitly delete the 158 // lexer when in raw mode. 159 return BufferPtr == BufferEnd; 160 } 161 162 /// isKeepWhitespaceMode - Return true if the lexer should return tokens for 163 /// every character in the file, including whitespace and comments. This 164 /// should only be used in raw mode, as the preprocessor is not prepared to 165 /// deal with the excess tokens. 166 bool isKeepWhitespaceMode() const { 167 return ExtendedTokenMode > 1; 168 } 169 170 /// SetKeepWhitespaceMode - This method lets clients enable or disable 171 /// whitespace retention mode. 172 void SetKeepWhitespaceMode(bool Val) { 173 assert((!Val || LexingRawMode || LangOpts.TraditionalCPP) && 174 "Can only retain whitespace in raw mode or -traditional-cpp"); 175 ExtendedTokenMode = Val ? 2 : 0; 176 } 177 178 /// inKeepCommentMode - Return true if the lexer should return comments as 179 /// tokens. 180 bool inKeepCommentMode() const { 181 return ExtendedTokenMode > 0; 182 } 183 184 /// SetCommentRetentionMode - Change the comment retention mode of the lexer 185 /// to the specified mode. This is really only useful when lexing in raw 186 /// mode, because otherwise the lexer needs to manage this. 187 void SetCommentRetentionState(bool Mode) { 188 assert(!isKeepWhitespaceMode() && 189 "Can't play with comment retention state when retaining whitespace"); 190 ExtendedTokenMode = Mode ? 1 : 0; 191 } 192 193 /// Sets the extended token mode back to its initial value, according to the 194 /// language options and preprocessor. This controls whether the lexer 195 /// produces comment and whitespace tokens. 196 /// 197 /// This requires the lexer to have an associated preprocessor. A standalone 198 /// lexer has nothing to reset to. 199 void resetExtendedTokenMode(); 200 201 /// Gets source code buffer. 202 StringRef getBuffer() const { 203 return StringRef(BufferStart, BufferEnd - BufferStart); 204 } 205 206 /// ReadToEndOfLine - Read the rest of the current preprocessor line as an 207 /// uninterpreted string. This switches the lexer out of directive mode. 208 void ReadToEndOfLine(SmallVectorImpl<char> *Result = 0); 209 210 211 /// Diag - Forwarding function for diagnostics. This translate a source 212 /// position in the current buffer into a SourceLocation object for rendering. 213 DiagnosticBuilder Diag(const char *Loc, unsigned DiagID) const; 214 215 /// getSourceLocation - Return a source location identifier for the specified 216 /// offset in the current file. 217 SourceLocation getSourceLocation(const char *Loc, unsigned TokLen = 1) const; 218 219 /// getSourceLocation - Return a source location for the next character in 220 /// the current file. 221 SourceLocation getSourceLocation() { return getSourceLocation(BufferPtr); } 222 223 /// \brief Return the current location in the buffer. 224 const char *getBufferLocation() const { return BufferPtr; } 225 226 /// Stringify - Convert the specified string into a C string by escaping '\' 227 /// and " characters. This does not add surrounding ""'s to the string. 228 /// If Charify is true, this escapes the ' character instead of ". 229 static std::string Stringify(const std::string &Str, bool Charify = false); 230 231 /// Stringify - Convert the specified string into a C string by escaping '\' 232 /// and " characters. This does not add surrounding ""'s to the string. 233 static void Stringify(SmallVectorImpl<char> &Str); 234 235 236 /// getSpelling - This method is used to get the spelling of a token into a 237 /// preallocated buffer, instead of as an std::string. The caller is required 238 /// to allocate enough space for the token, which is guaranteed to be at least 239 /// Tok.getLength() bytes long. The length of the actual result is returned. 240 /// 241 /// Note that this method may do two possible things: it may either fill in 242 /// the buffer specified with characters, or it may *change the input pointer* 243 /// to point to a constant buffer with the data already in it (avoiding a 244 /// copy). The caller is not allowed to modify the returned buffer pointer 245 /// if an internal buffer is returned. 246 static unsigned getSpelling(const Token &Tok, const char *&Buffer, 247 const SourceManager &SourceMgr, 248 const LangOptions &LangOpts, 249 bool *Invalid = 0); 250 251 /// getSpelling() - Return the 'spelling' of the Tok token. The spelling of a 252 /// token is the characters used to represent the token in the source file 253 /// after trigraph expansion and escaped-newline folding. In particular, this 254 /// wants to get the true, uncanonicalized, spelling of things like digraphs 255 /// UCNs, etc. 256 static std::string getSpelling(const Token &Tok, 257 const SourceManager &SourceMgr, 258 const LangOptions &LangOpts, 259 bool *Invalid = 0); 260 261 /// getSpelling - This method is used to get the spelling of the 262 /// token at the given source location. If, as is usually true, it 263 /// is not necessary to copy any data, then the returned string may 264 /// not point into the provided buffer. 265 /// 266 /// This method lexes at the expansion depth of the given 267 /// location and does not jump to the expansion or spelling 268 /// location. 269 static StringRef getSpelling(SourceLocation loc, 270 SmallVectorImpl<char> &buffer, 271 const SourceManager &SourceMgr, 272 const LangOptions &LangOpts, 273 bool *invalid = 0); 274 275 /// MeasureTokenLength - Relex the token at the specified location and return 276 /// its length in bytes in the input file. If the token needs cleaning (e.g. 277 /// includes a trigraph or an escaped newline) then this count includes bytes 278 /// that are part of that. 279 static unsigned MeasureTokenLength(SourceLocation Loc, 280 const SourceManager &SM, 281 const LangOptions &LangOpts); 282 283 /// \brief Relex the token at the specified location. 284 /// \returns true if there was a failure, false on success. 285 static bool getRawToken(SourceLocation Loc, Token &Result, 286 const SourceManager &SM, 287 const LangOptions &LangOpts, 288 bool IgnoreWhiteSpace = false); 289 290 /// \brief Given a location any where in a source buffer, find the location 291 /// that corresponds to the beginning of the token in which the original 292 /// source location lands. 293 static SourceLocation GetBeginningOfToken(SourceLocation Loc, 294 const SourceManager &SM, 295 const LangOptions &LangOpts); 296 297 /// AdvanceToTokenCharacter - If the current SourceLocation specifies a 298 /// location at the start of a token, return a new location that specifies a 299 /// character within the token. This handles trigraphs and escaped newlines. 300 static SourceLocation AdvanceToTokenCharacter(SourceLocation TokStart, 301 unsigned Character, 302 const SourceManager &SM, 303 const LangOptions &LangOpts); 304 305 /// \brief Computes the source location just past the end of the 306 /// token at this source location. 307 /// 308 /// This routine can be used to produce a source location that 309 /// points just past the end of the token referenced by \p Loc, and 310 /// is generally used when a diagnostic needs to point just after a 311 /// token where it expected something different that it received. If 312 /// the returned source location would not be meaningful (e.g., if 313 /// it points into a macro), this routine returns an invalid 314 /// source location. 315 /// 316 /// \param Offset an offset from the end of the token, where the source 317 /// location should refer to. The default offset (0) produces a source 318 /// location pointing just past the end of the token; an offset of 1 produces 319 /// a source location pointing to the last character in the token, etc. 320 static SourceLocation getLocForEndOfToken(SourceLocation Loc, unsigned Offset, 321 const SourceManager &SM, 322 const LangOptions &LangOpts); 323 324 /// \brief Returns true if the given MacroID location points at the first 325 /// token of the macro expansion. 326 /// 327 /// \param MacroBegin If non-null and function returns true, it is set to 328 /// begin location of the macro. 329 static bool isAtStartOfMacroExpansion(SourceLocation loc, 330 const SourceManager &SM, 331 const LangOptions &LangOpts, 332 SourceLocation *MacroBegin = 0); 333 334 /// \brief Returns true if the given MacroID location points at the last 335 /// token of the macro expansion. 336 /// 337 /// \param MacroEnd If non-null and function returns true, it is set to 338 /// end location of the macro. 339 static bool isAtEndOfMacroExpansion(SourceLocation loc, 340 const SourceManager &SM, 341 const LangOptions &LangOpts, 342 SourceLocation *MacroEnd = 0); 343 344 /// \brief Accepts a range and returns a character range with file locations. 345 /// 346 /// Returns a null range if a part of the range resides inside a macro 347 /// expansion or the range does not reside on the same FileID. 348 /// 349 /// This function is trying to deal with macros and return a range based on 350 /// file locations. The cases where it can successfully handle macros are: 351 /// 352 /// -begin or end range lies at the start or end of a macro expansion, in 353 /// which case the location will be set to the expansion point, e.g: 354 /// \#define M 1 2 355 /// a M 356 /// If you have a range [a, 2] (where 2 came from the macro), the function 357 /// will return a range for "a M" 358 /// if you have range [a, 1], the function will fail because the range 359 /// overlaps with only a part of the macro 360 /// 361 /// -The macro is a function macro and the range can be mapped to the macro 362 /// arguments, e.g: 363 /// \#define M 1 2 364 /// \#define FM(x) x 365 /// FM(a b M) 366 /// if you have range [b, 2], the function will return the file range "b M" 367 /// inside the macro arguments. 368 /// if you have range [a, 2], the function will return the file range 369 /// "FM(a b M)" since the range includes all of the macro expansion. 370 static CharSourceRange makeFileCharRange(CharSourceRange Range, 371 const SourceManager &SM, 372 const LangOptions &LangOpts); 373 374 /// \brief Returns a string for the source that the range encompasses. 375 static StringRef getSourceText(CharSourceRange Range, 376 const SourceManager &SM, 377 const LangOptions &LangOpts, 378 bool *Invalid = 0); 379 380 /// \brief Retrieve the name of the immediate macro expansion. 381 /// 382 /// This routine starts from a source location, and finds the name of the macro 383 /// responsible for its immediate expansion. It looks through any intervening 384 /// macro argument expansions to compute this. It returns a StringRef which 385 /// refers to the SourceManager-owned buffer of the source where that macro 386 /// name is spelled. Thus, the result shouldn't out-live that SourceManager. 387 static StringRef getImmediateMacroName(SourceLocation Loc, 388 const SourceManager &SM, 389 const LangOptions &LangOpts); 390 391 /// \brief Compute the preamble of the given file. 392 /// 393 /// The preamble of a file contains the initial comments, include directives, 394 /// and other preprocessor directives that occur before the code in this 395 /// particular file actually begins. The preamble of the main source file is 396 /// a potential prefix header. 397 /// 398 /// \param Buffer The memory buffer containing the file's contents. 399 /// 400 /// \param MaxLines If non-zero, restrict the length of the preamble 401 /// to fewer than this number of lines. 402 /// 403 /// \returns The offset into the file where the preamble ends and the rest 404 /// of the file begins along with a boolean value indicating whether 405 /// the preamble ends at the beginning of a new line. 406 static std::pair<unsigned, bool> 407 ComputePreamble(const llvm::MemoryBuffer *Buffer, const LangOptions &LangOpts, 408 unsigned MaxLines = 0); 409 410 /// \brief Checks that the given token is the first token that occurs after 411 /// the given location (this excludes comments and whitespace). Returns the 412 /// location immediately after the specified token. If the token is not found 413 /// or the location is inside a macro, the returned source location will be 414 /// invalid. 415 static SourceLocation findLocationAfterToken(SourceLocation loc, 416 tok::TokenKind TKind, 417 const SourceManager &SM, 418 const LangOptions &LangOpts, 419 bool SkipTrailingWhitespaceAndNewLine); 420 421 /// \brief Returns true if the given character could appear in an identifier. 422 static bool isIdentifierBodyChar(char c, const LangOptions &LangOpts); 423 424 /// getCharAndSizeNoWarn - Like the getCharAndSize method, but does not ever 425 /// emit a warning. 426 static inline char getCharAndSizeNoWarn(const char *Ptr, unsigned &Size, 427 const LangOptions &LangOpts) { 428 // If this is not a trigraph and not a UCN or escaped newline, return 429 // quickly. 430 if (isObviouslySimpleCharacter(Ptr[0])) { 431 Size = 1; 432 return *Ptr; 433 } 434 435 Size = 0; 436 return getCharAndSizeSlowNoWarn(Ptr, Size, LangOpts); 437 } 438 439 //===--------------------------------------------------------------------===// 440 // Internal implementation interfaces. 441private: 442 443 /// LexTokenInternal - Internal interface to lex a preprocessing token. Called 444 /// by Lex. 445 /// 446 bool LexTokenInternal(Token &Result, bool TokAtPhysicalStartOfLine); 447 448 bool CheckUnicodeWhitespace(Token &Result, uint32_t C, const char *CurPtr); 449 450 /// Given that a token begins with the Unicode character \p C, figure out 451 /// what kind of token it is and dispatch to the appropriate lexing helper 452 /// function. 453 bool LexUnicode(Token &Result, uint32_t C, const char *CurPtr); 454 455 /// FormTokenWithChars - When we lex a token, we have identified a span 456 /// starting at BufferPtr, going to TokEnd that forms the token. This method 457 /// takes that range and assigns it to the token as its location and size. In 458 /// addition, since tokens cannot overlap, this also updates BufferPtr to be 459 /// TokEnd. 460 void FormTokenWithChars(Token &Result, const char *TokEnd, 461 tok::TokenKind Kind) { 462 unsigned TokLen = TokEnd-BufferPtr; 463 Result.setLength(TokLen); 464 Result.setLocation(getSourceLocation(BufferPtr, TokLen)); 465 Result.setKind(Kind); 466 BufferPtr = TokEnd; 467 } 468 469 /// isNextPPTokenLParen - Return 1 if the next unexpanded token will return a 470 /// tok::l_paren token, 0 if it is something else and 2 if there are no more 471 /// tokens in the buffer controlled by this lexer. 472 unsigned isNextPPTokenLParen(); 473 474 //===--------------------------------------------------------------------===// 475 // Lexer character reading interfaces. 476 477 // This lexer is built on two interfaces for reading characters, both of which 478 // automatically provide phase 1/2 translation. getAndAdvanceChar is used 479 // when we know that we will be reading a character from the input buffer and 480 // that this character will be part of the result token. This occurs in (f.e.) 481 // string processing, because we know we need to read until we find the 482 // closing '"' character. 483 // 484 // The second interface is the combination of getCharAndSize with 485 // ConsumeChar. getCharAndSize reads a phase 1/2 translated character, 486 // returning it and its size. If the lexer decides that this character is 487 // part of the current token, it calls ConsumeChar on it. This two stage 488 // approach allows us to emit diagnostics for characters (e.g. warnings about 489 // trigraphs), knowing that they only are emitted if the character is 490 // consumed. 491 492 /// isObviouslySimpleCharacter - Return true if the specified character is 493 /// obviously the same in translation phase 1 and translation phase 3. This 494 /// can return false for characters that end up being the same, but it will 495 /// never return true for something that needs to be mapped. 496 static bool isObviouslySimpleCharacter(char C) { 497 return C != '?' && C != '\\'; 498 } 499 500 /// getAndAdvanceChar - Read a single 'character' from the specified buffer, 501 /// advance over it, and return it. This is tricky in several cases. Here we 502 /// just handle the trivial case and fall-back to the non-inlined 503 /// getCharAndSizeSlow method to handle the hard case. 504 inline char getAndAdvanceChar(const char *&Ptr, Token &Tok) { 505 // If this is not a trigraph and not a UCN or escaped newline, return 506 // quickly. 507 if (isObviouslySimpleCharacter(Ptr[0])) return *Ptr++; 508 509 unsigned Size = 0; 510 char C = getCharAndSizeSlow(Ptr, Size, &Tok); 511 Ptr += Size; 512 return C; 513 } 514 515 /// ConsumeChar - When a character (identified by getCharAndSize) is consumed 516 /// and added to a given token, check to see if there are diagnostics that 517 /// need to be emitted or flags that need to be set on the token. If so, do 518 /// it. 519 const char *ConsumeChar(const char *Ptr, unsigned Size, Token &Tok) { 520 // Normal case, we consumed exactly one token. Just return it. 521 if (Size == 1) 522 return Ptr+Size; 523 524 // Otherwise, re-lex the character with a current token, allowing 525 // diagnostics to be emitted and flags to be set. 526 Size = 0; 527 getCharAndSizeSlow(Ptr, Size, &Tok); 528 return Ptr+Size; 529 } 530 531 /// getCharAndSize - Peek a single 'character' from the specified buffer, 532 /// get its size, and return it. This is tricky in several cases. Here we 533 /// just handle the trivial case and fall-back to the non-inlined 534 /// getCharAndSizeSlow method to handle the hard case. 535 inline char getCharAndSize(const char *Ptr, unsigned &Size) { 536 // If this is not a trigraph and not a UCN or escaped newline, return 537 // quickly. 538 if (isObviouslySimpleCharacter(Ptr[0])) { 539 Size = 1; 540 return *Ptr; 541 } 542 543 Size = 0; 544 return getCharAndSizeSlow(Ptr, Size); 545 } 546 547 /// getCharAndSizeSlow - Handle the slow/uncommon case of the getCharAndSize 548 /// method. 549 char getCharAndSizeSlow(const char *Ptr, unsigned &Size, Token *Tok = 0); 550 551 /// getEscapedNewLineSize - Return the size of the specified escaped newline, 552 /// or 0 if it is not an escaped newline. P[-1] is known to be a "\" on entry 553 /// to this function. 554 static unsigned getEscapedNewLineSize(const char *P); 555 556 /// SkipEscapedNewLines - If P points to an escaped newline (or a series of 557 /// them), skip over them and return the first non-escaped-newline found, 558 /// otherwise return P. 559 static const char *SkipEscapedNewLines(const char *P); 560 561 /// getCharAndSizeSlowNoWarn - Same as getCharAndSizeSlow, but never emits a 562 /// diagnostic. 563 static char getCharAndSizeSlowNoWarn(const char *Ptr, unsigned &Size, 564 const LangOptions &LangOpts); 565 566 //===--------------------------------------------------------------------===// 567 // Other lexer functions. 568 569 void SkipBytes(unsigned Bytes, bool StartOfLine); 570 571 void PropagateLineStartLeadingSpaceInfo(Token &Result); 572 573 const char *LexUDSuffix(Token &Result, const char *CurPtr, 574 bool IsStringLiteral); 575 576 // Helper functions to lex the remainder of a token of the specific type. 577 bool LexIdentifier (Token &Result, const char *CurPtr); 578 bool LexNumericConstant (Token &Result, const char *CurPtr); 579 bool LexStringLiteral (Token &Result, const char *CurPtr, 580 tok::TokenKind Kind); 581 bool LexRawStringLiteral (Token &Result, const char *CurPtr, 582 tok::TokenKind Kind); 583 bool LexAngledStringLiteral(Token &Result, const char *CurPtr); 584 bool LexCharConstant (Token &Result, const char *CurPtr, 585 tok::TokenKind Kind); 586 bool LexEndOfFile (Token &Result, const char *CurPtr); 587 bool SkipWhitespace (Token &Result, const char *CurPtr, 588 bool &TokAtPhysicalStartOfLine); 589 bool SkipLineComment (Token &Result, const char *CurPtr, 590 bool &TokAtPhysicalStartOfLine); 591 bool SkipBlockComment (Token &Result, const char *CurPtr, 592 bool &TokAtPhysicalStartOfLine); 593 bool SaveLineComment (Token &Result, const char *CurPtr); 594 595 bool IsStartOfConflictMarker(const char *CurPtr); 596 bool HandleEndOfConflictMarker(const char *CurPtr); 597 598 bool isCodeCompletionPoint(const char *CurPtr) const; 599 void cutOffLexing() { BufferPtr = BufferEnd; } 600 601 bool isHexaLiteral(const char *Start, const LangOptions &LangOpts); 602 603 604 /// Read a universal character name. 605 /// 606 /// \param CurPtr The position in the source buffer after the initial '\'. 607 /// If the UCN is syntactically well-formed (but not necessarily 608 /// valid), this parameter will be updated to point to the 609 /// character after the UCN. 610 /// \param SlashLoc The position in the source buffer of the '\'. 611 /// \param Tok The token being formed. Pass \c NULL to suppress diagnostics 612 /// and handle token formation in the caller. 613 /// 614 /// \return The Unicode codepoint specified by the UCN, or 0 if the UCN is 615 /// invalid. 616 uint32_t tryReadUCN(const char *&CurPtr, const char *SlashLoc, Token *Tok); 617}; 618 619 620} // end namespace clang 621 622#endif 623