Token.h revision b28d6de75e0cb27ce3106ce6052f87ad0ab276d7
1//===--- Token.h - Token interface ------------------------------*- 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 Token interface. 11// 12//===----------------------------------------------------------------------===// 13 14#ifndef LLVM_CLANG_TOKEN_H 15#define LLVM_CLANG_TOKEN_H 16 17#include "clang/Basic/TokenKinds.h" 18#include "clang/Basic/SourceLocation.h" 19 20namespace clang { 21 22class IdentifierInfo; 23 24/// Token - This structure provides full information about a lexed token. 25/// It is not intended to be space efficient, it is intended to return as much 26/// information as possible about each returned token. This is expected to be 27/// compressed into a smaller form if memory footprint is important. 28/// 29/// The parser can create a special "annotation token" representing a stream of 30/// tokens that were parsed and semantically resolved, e.g.: "foo::MyClass<int>" 31/// can be represented by a single typename annotation token that carries 32/// information about the SourceRange of the tokens and the type object. 33class Token { 34 /// The location of the token. 35 SourceLocation Loc; 36 37 // Conceptually these next two fields could be in a union with 38 // access depending on isAnnotationToken(). However, this causes gcc 39 // 4.2 to pessimize LexTokenInternal, a very performance critical 40 // routine. Keeping as separate members with casts until a more 41 // beautiful fix presents itself. 42 43 /// UintData - This holds either the length of the token text, when 44 /// a normal token, or the end of the SourceRange when an annotation 45 /// token. 46 unsigned UintData; 47 48 /// PtrData - For normal tokens, this points to the uniqued 49 /// information for the identifier (if an identifier token) or 50 /// null. For annotation tokens, this points to information specific 51 /// to the annotation token. 52 void *PtrData; 53 54 /// Kind - The actual flavor of token this is. 55 /// 56 unsigned Kind : 8; // DON'T make Kind a 'tok::TokenKind'; 57 // MSVC will treat it as a signed char and 58 // TokenKinds > 127 won't be handled correctly. 59 60 /// Flags - Bits we track about this token, members of the TokenFlags enum. 61 unsigned Flags : 8; 62public: 63 64 // Various flags set per token: 65 enum TokenFlags { 66 StartOfLine = 0x01, // At start of line or only after whitespace. 67 LeadingSpace = 0x02, // Whitespace exists before this token. 68 DisableExpand = 0x04, // This identifier may never be macro expanded. 69 NeedsCleaning = 0x08 // Contained an escaped newline or trigraph. 70 }; 71 72 tok::TokenKind getKind() const { return (tok::TokenKind)Kind; } 73 void setKind(tok::TokenKind K) { Kind = K; } 74 75 /// is/isNot - Predicates to check if this token is a specific kind, as in 76 /// "if (Tok.is(tok::l_brace)) {...}". 77 bool is(tok::TokenKind K) const { return Kind == (unsigned) K; } 78 bool isNot(tok::TokenKind K) const { return Kind != (unsigned) K; } 79 80 bool isAnnotationToken() const { 81 return is(tok::annot_qualtypename) || is(tok::annot_cxxscope); 82 } 83 84 /// getLocation - Return a source location identifier for the specified 85 /// offset in the current file. 86 SourceLocation getLocation() const { return Loc; } 87 unsigned getLength() const { 88 assert(!isAnnotationToken() && "Used Length on annotation token"); 89 return UintData; 90 } 91 92 void setLocation(SourceLocation L) { Loc = L; } 93 void setLength(unsigned Len) { UintData = Len; } 94 95 SourceLocation getAnnotationEndLoc() const { 96 assert(isAnnotationToken() && "Used AnnotEndLocID on non-annotation token"); 97 return SourceLocation::getFromRawEncoding(UintData); 98 } 99 void setAnnotationEndLoc(SourceLocation L) { 100 assert(isAnnotationToken() && "Used AnnotEndLocID on non-annotation token"); 101 UintData = L.getRawEncoding(); 102 } 103 104 /// getAnnotationRange - SourceRange of the group of tokens that this 105 /// annotation token represents. 106 SourceRange getAnnotationRange() const { 107 return SourceRange(getLocation(), getAnnotationEndLoc()); 108 } 109 void setAnnotationRange(SourceRange R) { 110 setLocation(R.getBegin()); 111 setAnnotationEndLoc(R.getEnd()); 112 } 113 114 const char *getName() const { 115 return tok::getTokenName( (tok::TokenKind) Kind); 116 } 117 118 /// startToken - Reset all flags to cleared. 119 /// 120 void startToken() { 121 Flags = 0; 122 PtrData = 0; 123 Loc = SourceLocation(); 124 } 125 126 IdentifierInfo *getIdentifierInfo() const { 127 assert(!isAnnotationToken() && "Used IdentInfo on annotation token"); 128 return (IdentifierInfo*) PtrData; 129 } 130 void setIdentifierInfo(IdentifierInfo *II) { 131 PtrData = (void*) II; 132 } 133 134 void *getAnnotationValue() const { 135 assert(isAnnotationToken() && "Used AnnotVal on non-annotation token"); 136 return PtrData; 137 } 138 void setAnnotationValue(void *val) { 139 assert(isAnnotationToken() && "Used AnnotVal on non-annotation token"); 140 PtrData = val; 141 } 142 143 /// setFlag - Set the specified flag. 144 void setFlag(TokenFlags Flag) { 145 Flags |= Flag; 146 } 147 148 /// clearFlag - Unset the specified flag. 149 void clearFlag(TokenFlags Flag) { 150 Flags &= ~Flag; 151 } 152 153 /// getFlags - Return the internal represtation of the flags. 154 /// Only intended for low-level operations such as writing tokens to 155 // disk. 156 unsigned getFlags() const { 157 return Flags; 158 } 159 160 /// setFlagValue - Set a flag to either true or false. 161 void setFlagValue(TokenFlags Flag, bool Val) { 162 if (Val) 163 setFlag(Flag); 164 else 165 clearFlag(Flag); 166 } 167 168 /// isAtStartOfLine - Return true if this token is at the start of a line. 169 /// 170 bool isAtStartOfLine() const { return (Flags & StartOfLine) ? true : false; } 171 172 /// hasLeadingSpace - Return true if this token has whitespace before it. 173 /// 174 bool hasLeadingSpace() const { return (Flags & LeadingSpace) ? true : false; } 175 176 /// isExpandDisabled - Return true if this identifier token should never 177 /// be expanded in the future, due to C99 6.10.3.4p2. 178 bool isExpandDisabled() const { 179 return (Flags & DisableExpand) ? true : false; 180 } 181 182 /// isObjCAtKeyword - Return true if we have an ObjC keyword identifier. 183 bool isObjCAtKeyword(tok::ObjCKeywordKind objcKey) const; 184 185 /// getObjCKeywordID - Return the ObjC keyword kind. 186 tok::ObjCKeywordKind getObjCKeywordID() const; 187 188 /// needsCleaning - Return true if this token has trigraphs or escaped 189 /// newlines in it. 190 /// 191 bool needsCleaning() const { return (Flags & NeedsCleaning) ? true : false; } 192}; 193 194/// PPConditionalInfo - Information about the conditional stack (#if directives) 195/// currently active. 196struct PPConditionalInfo { 197 /// IfLoc - Location where the conditional started. 198 /// 199 SourceLocation IfLoc; 200 201 /// WasSkipping - True if this was contained in a skipping directive, e.g. 202 /// in a "#if 0" block. 203 bool WasSkipping; 204 205 /// FoundNonSkip - True if we have emitted tokens already, and now we're in 206 /// an #else block or something. Only useful in Skipping blocks. 207 bool FoundNonSkip; 208 209 /// FoundElse - True if we've seen a #else in this block. If so, 210 /// #elif/#else directives are not allowed. 211 bool FoundElse; 212}; 213 214} // end namespace clang 215 216#endif 217