1 //===-- llvm/MC/MCAsmLexer.h - Abstract Asm Lexer Interface -----*- C++ -*-===//
3 // The LLVM Compiler Infrastructure
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
10 #ifndef LLVM_MC_MCPARSER_MCASMLEXER_H
11 #define LLVM_MC_MCPARSER_MCASMLEXER_H
13 #include "llvm/ADT/APInt.h"
14 #include "llvm/ADT/StringRef.h"
15 #include "llvm/Support/Compiler.h"
16 #include "llvm/Support/DataTypes.h"
17 #include "llvm/Support/SMLoc.h"
21 /// Target independent representation for an assembler token.
34 BigNum, // larger than 64 bits
46 LParen, RParen, LBrac, RBrac, LCurly, RCurly,
47 Star, Dot, Comma, Dollar, Equal, EqualEqual,
49 Pipe, PipePipe, Caret,
50 Amp, AmpAmp, Exclaim, ExclaimEqual, Percent, Hash,
51 Less, LessEqual, LessLess, LessGreater,
52 Greater, GreaterEqual, GreaterGreater, At
58 /// A reference to the entire token contents; this is always a pointer into
59 /// a memory buffer owned by the source manager.
66 AsmToken(TokenKind Kind, StringRef Str, APInt IntVal)
67 : Kind(Kind), Str(Str), IntVal(IntVal) {}
68 AsmToken(TokenKind Kind, StringRef Str, int64_t IntVal = 0)
69 : Kind(Kind), Str(Str), IntVal(64, IntVal, true) {}
71 TokenKind getKind() const { return Kind; }
72 bool is(TokenKind K) const { return Kind == K; }
73 bool isNot(TokenKind K) const { return Kind != K; }
76 SMLoc getEndLoc() const;
77 SMRange getLocRange() const;
79 /// Get the contents of a string token (without quotes).
80 StringRef getStringContents() const {
81 assert(Kind == String && "This token isn't a string!");
82 return Str.slice(1, Str.size() - 1);
85 /// Get the identifier string for the current token, which should be an
86 /// identifier or a string. This gets the portion of the string which should
87 /// be used as the identifier, e.g., it does not include the quotes on
89 StringRef getIdentifier() const {
90 if (Kind == Identifier)
92 return getStringContents();
95 /// Get the string for the current token, this includes all characters (for
96 /// example, the quotes on strings) in the token.
98 /// The returned StringRef points into the source manager's memory buffer, and
99 /// is safe to store across calls to Lex().
100 StringRef getString() const { return Str; }
102 // FIXME: Don't compute this in advance, it makes every token larger, and is
103 // also not generally what we want (it is nicer for recovery etc. to lex 123br
104 // as a single token, then diagnose as an invalid number).
105 int64_t getIntVal() const {
106 assert(Kind == Integer && "This token isn't an integer!");
107 return IntVal.getZExtValue();
110 APInt getAPIntVal() const {
111 assert((Kind == Integer || Kind == BigNum) &&
112 "This token isn't an integer!");
117 /// Generic assembler lexer interface, for use by target specific assembly
120 /// The current token, stored in the base class for faster access.
123 /// The location and description of the current error
127 MCAsmLexer(const MCAsmLexer &) = delete;
128 void operator=(const MCAsmLexer &) = delete;
129 protected: // Can only create subclasses.
130 const char *TokStart;
132 bool AllowAtInIdentifier;
136 virtual AsmToken LexToken() = 0;
138 void SetError(SMLoc errLoc, const std::string &err) {
144 virtual ~MCAsmLexer();
146 /// Consume the next token from the input stream and return it.
148 /// The lexer will continuosly return the end-of-file token once the end of
149 /// the main input file has been reached.
150 const AsmToken &Lex() {
151 return CurTok = LexToken();
154 virtual StringRef LexUntilEndOfStatement() = 0;
156 /// Get the current source location.
157 SMLoc getLoc() const;
159 /// Get the current (last) lexed token.
160 const AsmToken &getTok() const {
164 /// Look ahead at the next token to be lexed.
165 const AsmToken peekTok(bool ShouldSkipSpace = true) {
168 MutableArrayRef<AsmToken> Buf(Tok);
169 size_t ReadCount = peekTokens(Buf, ShouldSkipSpace);
171 assert(ReadCount == 1);
177 /// Look ahead an arbitrary number of tokens.
178 virtual size_t peekTokens(MutableArrayRef<AsmToken> Buf,
179 bool ShouldSkipSpace = true) = 0;
181 /// Get the current error location
186 /// Get the current error string
187 const std::string &getErr() {
191 /// Get the kind of current token.
192 AsmToken::TokenKind getKind() const { return CurTok.getKind(); }
194 /// Check if the current token has kind \p K.
195 bool is(AsmToken::TokenKind K) const { return CurTok.is(K); }
197 /// Check if the current token has kind \p K.
198 bool isNot(AsmToken::TokenKind K) const { return CurTok.isNot(K); }
200 /// Set whether spaces should be ignored by the lexer
201 void setSkipSpace(bool val) { SkipSpace = val; }
203 bool getAllowAtInIdentifier() { return AllowAtInIdentifier; }
204 void setAllowAtInIdentifier(bool v) { AllowAtInIdentifier = v; }
207 } // End llvm namespace