1 //===--- Token.h - Token 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 // This file defines the Token interface.
12 //===----------------------------------------------------------------------===//
14 #ifndef LLVM_CLANG_TOKEN_H
15 #define LLVM_CLANG_TOKEN_H
17 #include "clang/Basic/TemplateKinds.h"
18 #include "clang/Basic/TokenKinds.h"
19 #include "clang/Basic/SourceLocation.h"
20 #include "clang/Basic/OperatorKinds.h"
27 /// Token - This structure provides full information about a lexed token.
28 /// It is not intended to be space efficient, it is intended to return as much
29 /// information as possible about each returned token. This is expected to be
30 /// compressed into a smaller form if memory footprint is important.
32 /// The parser can create a special "annotation token" representing a stream of
33 /// tokens that were parsed and semantically resolved, e.g.: "foo::MyClass<int>"
34 /// can be represented by a single typename annotation token that carries
35 /// information about the SourceRange of the tokens and the type object.
37 /// The location of the token.
40 // Conceptually these next two fields could be in a union. However, this
41 // causes gcc 4.2 to pessimize LexTokenInternal, a very performance critical
42 // routine. Keeping as separate members with casts until a more beautiful fix
45 /// UintData - This holds either the length of the token text, when
46 /// a normal token, or the end of the SourceRange when an annotation
50 /// PtrData - This is a union of four different pointer types, which depends
51 /// on what type of token this is:
52 /// Identifiers, keywords, etc:
53 /// This is an IdentifierInfo*, which contains the uniqued identifier
55 /// Literals: isLiteral() returns true.
56 /// This is a pointer to the start of the token in a text buffer, which
57 /// may be dirty (have trigraphs / escaped newlines).
58 /// Annotations (resolved type names, C++ scopes, etc): isAnnotation().
59 /// This is a pointer to sema-specific data for the annotation token.
64 /// Kind - The actual flavor of token this is.
68 /// Flags - Bits we track about this token, members of the TokenFlags enum.
72 // Various flags set per token:
74 StartOfLine = 0x01, // At start of line or only after whitespace.
75 LeadingSpace = 0x02, // Whitespace exists before this token.
76 DisableExpand = 0x04, // This identifier may never be macro expanded.
77 NeedsCleaning = 0x08, // Contained an escaped newline or trigraph.
78 LeadingEmptyMacro = 0x10, // Empty macro exists before this token.
79 HasUDSuffix = 0x20 // This string or character literal has a ud-suffix.
82 tok::TokenKind getKind() const { return (tok::TokenKind)Kind; }
83 void setKind(tok::TokenKind K) { Kind = K; }
85 /// is/isNot - Predicates to check if this token is a specific kind, as in
86 /// "if (Tok.is(tok::l_brace)) {...}".
87 bool is(tok::TokenKind K) const { return Kind == (unsigned) K; }
88 bool isNot(tok::TokenKind K) const { return Kind != (unsigned) K; }
90 /// \brief Return true if this is a raw identifier (when lexing
91 /// in raw mode) or a non-keyword identifier (when lexing in non-raw mode).
92 bool isAnyIdentifier() const {
93 return tok::isAnyIdentifier(getKind());
96 /// \brief Return true if this is a "literal", like a numeric
97 /// constant, string, etc.
98 bool isLiteral() const {
99 return tok::isLiteral(getKind());
102 /// \brief Return true if this is any of tok::annot_* kind tokens.
103 bool isAnnotation() const {
104 return tok::isAnnotation(getKind());
107 /// \brief Return a source location identifier for the specified
108 /// offset in the current file.
109 SourceLocation getLocation() const { return Loc; }
110 unsigned getLength() const {
111 assert(!isAnnotation() && "Annotation tokens have no length field");
115 void setLocation(SourceLocation L) { Loc = L; }
116 void setLength(unsigned Len) {
117 assert(!isAnnotation() && "Annotation tokens have no length field");
121 SourceLocation getAnnotationEndLoc() const {
122 assert(isAnnotation() && "Used AnnotEndLocID on non-annotation token");
123 return SourceLocation::getFromRawEncoding(UintData);
125 void setAnnotationEndLoc(SourceLocation L) {
126 assert(isAnnotation() && "Used AnnotEndLocID on non-annotation token");
127 UintData = L.getRawEncoding();
130 SourceLocation getLastLoc() const {
131 return isAnnotation() ? getAnnotationEndLoc() : getLocation();
134 /// \brief SourceRange of the group of tokens that this annotation token
136 SourceRange getAnnotationRange() const {
137 return SourceRange(getLocation(), getAnnotationEndLoc());
139 void setAnnotationRange(SourceRange R) {
140 setLocation(R.getBegin());
141 setAnnotationEndLoc(R.getEnd());
144 const char *getName() const {
145 return tok::getTokenName( (tok::TokenKind) Kind);
148 /// \brief Reset all flags to cleared.
154 Loc = SourceLocation();
157 IdentifierInfo *getIdentifierInfo() const {
158 assert(isNot(tok::raw_identifier) &&
159 "getIdentifierInfo() on a tok::raw_identifier token!");
160 assert(!isAnnotation() &&
161 "getIdentifierInfo() on an annotation token!");
162 if (isLiteral()) return 0;
163 return (IdentifierInfo*) PtrData;
165 void setIdentifierInfo(IdentifierInfo *II) {
166 PtrData = (void*) II;
169 /// getRawIdentifierData - For a raw identifier token (i.e., an identifier
170 /// lexed in raw mode), returns a pointer to the start of it in the text
171 /// buffer if known, null otherwise.
172 const char *getRawIdentifierData() const {
173 assert(is(tok::raw_identifier));
174 return reinterpret_cast<const char*>(PtrData);
176 void setRawIdentifierData(const char *Ptr) {
177 assert(is(tok::raw_identifier));
178 PtrData = const_cast<char*>(Ptr);
181 /// getLiteralData - For a literal token (numeric constant, string, etc), this
182 /// returns a pointer to the start of it in the text buffer if known, null
184 const char *getLiteralData() const {
185 assert(isLiteral() && "Cannot get literal data of non-literal");
186 return reinterpret_cast<const char*>(PtrData);
188 void setLiteralData(const char *Ptr) {
189 assert(isLiteral() && "Cannot set literal data of non-literal");
190 PtrData = const_cast<char*>(Ptr);
193 void *getAnnotationValue() const {
194 assert(isAnnotation() && "Used AnnotVal on non-annotation token");
197 void setAnnotationValue(void *val) {
198 assert(isAnnotation() && "Used AnnotVal on non-annotation token");
202 /// \brief Set the specified flag.
203 void setFlag(TokenFlags Flag) {
207 /// \brief Unset the specified flag.
208 void clearFlag(TokenFlags Flag) {
212 /// \brief Return the internal represtation of the flags.
214 /// This is only intended for low-level operations such as writing tokens to
216 unsigned getFlags() const {
220 /// \brief Set a flag to either true or false.
221 void setFlagValue(TokenFlags Flag, bool Val) {
228 /// isAtStartOfLine - Return true if this token is at the start of a line.
230 bool isAtStartOfLine() const { return (Flags & StartOfLine) ? true : false; }
232 /// \brief Return true if this token has whitespace before it.
234 bool hasLeadingSpace() const { return (Flags & LeadingSpace) ? true : false; }
236 /// \brief Return true if this identifier token should never
237 /// be expanded in the future, due to C99 6.10.3.4p2.
238 bool isExpandDisabled() const {
239 return (Flags & DisableExpand) ? true : false;
242 /// \brief Return true if we have an ObjC keyword identifier.
243 bool isObjCAtKeyword(tok::ObjCKeywordKind objcKey) const;
245 /// \brief Return the ObjC keyword kind.
246 tok::ObjCKeywordKind getObjCKeywordID() const;
248 /// \brief Return true if this token has trigraphs or escaped newlines in it.
249 bool needsCleaning() const { return (Flags & NeedsCleaning) ? true : false; }
251 /// \brief Return true if this token has an empty macro before it.
253 bool hasLeadingEmptyMacro() const {
254 return (Flags & LeadingEmptyMacro) ? true : false;
257 /// \brief Return true if this token is a string or character literal which
259 bool hasUDSuffix() const { return (Flags & HasUDSuffix) ? true : false; }
262 /// \brief Information about the conditional stack (\#if directives)
263 /// currently active.
264 struct PPConditionalInfo {
265 /// \brief Location where the conditional started.
266 SourceLocation IfLoc;
268 /// \brief True if this was contained in a skipping directive, e.g.,
269 /// in a "\#if 0" block.
272 /// \brief True if we have emitted tokens already, and now we're in
273 /// an \#else block or something. Only useful in Skipping blocks.
276 /// \brief True if we've seen a \#else in this block. If so,
277 /// \#elif/\#else directives are not allowed.
281 } // end namespace clang
285 struct isPodLike<clang::Token> { static const bool value = true; };
286 } // end namespace llvm