clang 17.0.0git
Token.h
Go to the documentation of this file.
1//===--- Token.h - Token interface ------------------------------*- C++ -*-===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9// This file defines the Token interface.
10//
11//===----------------------------------------------------------------------===//
12
13#ifndef LLVM_CLANG_LEX_TOKEN_H
14#define LLVM_CLANG_LEX_TOKEN_H
15
18#include "llvm/ADT/ArrayRef.h"
19#include "llvm/ADT/StringRef.h"
20#include <cassert>
21
22namespace clang {
23
24class IdentifierInfo;
25
26/// Token - This structure provides full information about a lexed token.
27/// It is not intended to be space efficient, it is intended to return as much
28/// information as possible about each returned token. This is expected to be
29/// compressed into a smaller form if memory footprint is important.
30///
31/// The parser can create a special "annotation token" representing a stream of
32/// tokens that were parsed and semantically resolved, e.g.: "foo::MyClass<int>"
33/// can be represented by a single typename annotation token that carries
34/// information about the SourceRange of the tokens and the type object.
35class Token {
36 /// The location of the token. This is actually a SourceLocation.
38
39 // Conceptually these next two fields could be in a union. However, this
40 // causes gcc 4.2 to pessimize LexTokenInternal, a very performance critical
41 // routine. Keeping as separate members with casts until a more beautiful fix
42 // presents itself.
43
44 /// UintData - This holds either the length of the token text, when
45 /// a normal token, or the end of the SourceRange when an annotation
46 /// token.
48
49 /// PtrData - This is a union of four different pointer types, which depends
50 /// on what type of token this is:
51 /// Identifiers, keywords, etc:
52 /// This is an IdentifierInfo*, which contains the uniqued identifier
53 /// spelling.
54 /// Literals: isLiteral() returns true.
55 /// This is a pointer to the start of the token in a text buffer, which
56 /// may be dirty (have trigraphs / escaped newlines).
57 /// Annotations (resolved type names, C++ scopes, etc): isAnnotation().
58 /// This is a pointer to sema-specific data for the annotation token.
59 /// Eof:
60 // This is a pointer to a Decl.
61 /// Other:
62 /// This is null.
63 void *PtrData;
64
65 /// Kind - The actual flavor of token this is.
66 tok::TokenKind Kind;
67
68 /// Flags - Bits we track about this token, members of the TokenFlags enum.
69 unsigned short Flags;
70
71public:
72 // Various flags set per token:
74 StartOfLine = 0x01, // At start of line or only after whitespace
75 // (considering the line after macro expansion).
76 LeadingSpace = 0x02, // Whitespace exists before this token (considering
77 // whitespace after macro expansion).
78 DisableExpand = 0x04, // This identifier may never be macro expanded.
79 NeedsCleaning = 0x08, // Contained an escaped newline or trigraph.
80 LeadingEmptyMacro = 0x10, // Empty macro exists before this token.
81 HasUDSuffix = 0x20, // This string or character literal has a ud-suffix.
82 HasUCN = 0x40, // This identifier contains a UCN.
83 IgnoredComma = 0x80, // This comma is not a macro argument separator (MS).
84 StringifiedInMacro = 0x100, // This string or character literal is formed by
85 // macro stringizing or charizing operator.
86 CommaAfterElided = 0x200, // The comma following this token was elided (MS).
87 IsEditorPlaceholder = 0x400, // This identifier is a placeholder.
88 IsReinjected = 0x800, // A phase 4 token that was produced before and
89 // re-added, e.g. via EnterTokenStream. Annotation
90 // tokens are *not* reinjected.
91 };
92
93 tok::TokenKind getKind() const { return Kind; }
94 void setKind(tok::TokenKind K) { Kind = K; }
95
96 /// is/isNot - Predicates to check if this token is a specific kind, as in
97 /// "if (Tok.is(tok::l_brace)) {...}".
98 bool is(tok::TokenKind K) const { return Kind == K; }
99 bool isNot(tok::TokenKind K) const { return Kind != K; }
101 return is(K1) || is(K2);
102 }
103 template <typename... Ts> bool isOneOf(tok::TokenKind K1, Ts... Ks) const {
104 return is(K1) || isOneOf(Ks...);
105 }
106
107 /// Return true if this is a raw identifier (when lexing
108 /// in raw mode) or a non-keyword identifier (when lexing in non-raw mode).
109 bool isAnyIdentifier() const {
111 }
112
113 /// Return true if this is a "literal", like a numeric
114 /// constant, string, etc.
115 bool isLiteral() const {
116 return tok::isLiteral(getKind());
117 }
118
119 /// Return true if this is any of tok::annot_* kind tokens.
120 bool isAnnotation() const {
121 return tok::isAnnotation(getKind());
122 }
123
124 /// Return a source location identifier for the specified
125 /// offset in the current file.
128 }
129 unsigned getLength() const {
130 assert(!isAnnotation() && "Annotation tokens have no length field");
131 return UintData;
132 }
133
135 void setLength(unsigned Len) {
136 assert(!isAnnotation() && "Annotation tokens have no length field");
137 UintData = Len;
138 }
139
141 assert(isAnnotation() && "Used AnnotEndLocID on non-annotation token");
142 return SourceLocation::getFromRawEncoding(UintData ? UintData : Loc);
143 }
145 assert(isAnnotation() && "Used AnnotEndLocID on non-annotation token");
146 UintData = L.getRawEncoding();
147 }
148
151 }
152
156 }
157
158 /// SourceRange of the group of tokens that this annotation token
159 /// represents.
162 }
166 }
167
168 const char *getName() const { return tok::getTokenName(Kind); }
169
170 /// Reset all flags to cleared.
171 void startToken() {
172 Kind = tok::unknown;
173 Flags = 0;
174 PtrData = nullptr;
175 UintData = 0;
177 }
178
179 bool hasPtrData() const { return PtrData != nullptr; }
180
182 assert(isNot(tok::raw_identifier) &&
183 "getIdentifierInfo() on a tok::raw_identifier token!");
184 assert(!isAnnotation() &&
185 "getIdentifierInfo() on an annotation token!");
186 if (isLiteral()) return nullptr;
187 if (is(tok::eof)) return nullptr;
188 return (IdentifierInfo*) PtrData;
189 }
191 PtrData = (void*) II;
192 }
193
194 const void *getEofData() const {
195 assert(is(tok::eof));
196 return reinterpret_cast<const void *>(PtrData);
197 }
198 void setEofData(const void *D) {
199 assert(is(tok::eof));
200 assert(!PtrData);
201 PtrData = const_cast<void *>(D);
202 }
203
204 /// getRawIdentifier - For a raw identifier token (i.e., an identifier
205 /// lexed in raw mode), returns a reference to the text substring in the
206 /// buffer if known.
207 StringRef getRawIdentifier() const {
208 assert(is(tok::raw_identifier));
209 return StringRef(reinterpret_cast<const char *>(PtrData), getLength());
210 }
211 void setRawIdentifierData(const char *Ptr) {
212 assert(is(tok::raw_identifier));
213 PtrData = const_cast<char*>(Ptr);
214 }
215
216 /// getLiteralData - For a literal token (numeric constant, string, etc), this
217 /// returns a pointer to the start of it in the text buffer if known, null
218 /// otherwise.
219 const char *getLiteralData() const {
220 assert(isLiteral() && "Cannot get literal data of non-literal");
221 return reinterpret_cast<const char*>(PtrData);
222 }
223 void setLiteralData(const char *Ptr) {
224 assert(isLiteral() && "Cannot set literal data of non-literal");
225 PtrData = const_cast<char*>(Ptr);
226 }
227
228 void *getAnnotationValue() const {
229 assert(isAnnotation() && "Used AnnotVal on non-annotation token");
230 return PtrData;
231 }
232 void setAnnotationValue(void *val) {
233 assert(isAnnotation() && "Used AnnotVal on non-annotation token");
234 PtrData = val;
235 }
236
237 /// Set the specified flag.
238 void setFlag(TokenFlags Flag) {
239 Flags |= Flag;
240 }
241
242 /// Get the specified flag.
243 bool getFlag(TokenFlags Flag) const {
244 return (Flags & Flag) != 0;
245 }
246
247 /// Unset the specified flag.
249 Flags &= ~Flag;
250 }
251
252 /// Return the internal represtation of the flags.
253 ///
254 /// This is only intended for low-level operations such as writing tokens to
255 /// disk.
256 unsigned getFlags() const {
257 return Flags;
258 }
259
260 /// Set a flag to either true or false.
261 void setFlagValue(TokenFlags Flag, bool Val) {
262 if (Val)
263 setFlag(Flag);
264 else
265 clearFlag(Flag);
266 }
267
268 /// isAtStartOfLine - Return true if this token is at the start of a line.
269 ///
270 bool isAtStartOfLine() const { return getFlag(StartOfLine); }
271
272 /// Return true if this token has whitespace before it.
273 ///
274 bool hasLeadingSpace() const { return getFlag(LeadingSpace); }
275
276 /// Return true if this identifier token should never
277 /// be expanded in the future, due to C99 6.10.3.4p2.
278 bool isExpandDisabled() const { return getFlag(DisableExpand); }
279
280 /// Return true if we have an ObjC keyword identifier.
281 bool isObjCAtKeyword(tok::ObjCKeywordKind objcKey) const;
282
283 /// Return the ObjC keyword kind.
285
286 /// Return true if this token has trigraphs or escaped newlines in it.
287 bool needsCleaning() const { return getFlag(NeedsCleaning); }
288
289 /// Return true if this token has an empty macro before it.
290 ///
292
293 /// Return true if this token is a string or character literal which
294 /// has a ud-suffix.
295 bool hasUDSuffix() const { return getFlag(HasUDSuffix); }
296
297 /// Returns true if this token contains a universal character name.
298 bool hasUCN() const { return getFlag(HasUCN); }
299
300 /// Returns true if this token is formed by macro by stringizing or charizing
301 /// operator.
303
304 /// Returns true if the comma after this token was elided.
305 bool commaAfterElided() const { return getFlag(CommaAfterElided); }
306
307 /// Returns true if this token is an editor placeholder.
308 ///
309 /// Editor placeholders are produced by the code-completion engine and are
310 /// represented as characters between '<#' and '#>' in the source code. The
311 /// lexer uses identifier tokens to represent placeholders.
313};
314
315/// Information about the conditional stack (\#if directives)
316/// currently active.
318 /// Location where the conditional started.
320
321 /// True if this was contained in a skipping directive, e.g.,
322 /// in a "\#if 0" block.
324
325 /// True if we have emitted tokens already, and now we're in
326 /// an \#else block or something. Only useful in Skipping blocks.
328
329 /// True if we've seen a \#else in this block. If so,
330 /// \#elif/\#else directives are not allowed.
332};
333
334// Extra information needed for annonation tokens.
339};
340} // end namespace clang
341
342#endif // LLVM_CLANG_LEX_TOKEN_H
static constexpr bool isOneOf()
Defines the clang::SourceLocation class and associated facilities.
Defines the clang::TokenKind enum and support functions.
One of these records is kept for each identifier that is lexed.
Encodes a location in the source.
static SourceLocation getFromRawEncoding(UIntTy Encoding)
Turn a raw encoding of a SourceLocation object into a real SourceLocation.
SourceLocation getLocWithOffset(IntTy Offset) const
Return a source location with the specified offset from this SourceLocation.
UIntTy getRawEncoding() const
When a SourceLocation itself cannot be used, this returns an (opaque) 32-bit integer encoding for it.
A trivial tuple used to represent a source range.
SourceLocation getEnd() const
SourceLocation getBegin() const
Token - This structure provides full information about a lexed token.
Definition: Token.h:35
IdentifierInfo * getIdentifierInfo() const
Definition: Token.h:181
void setLiteralData(const char *Ptr)
Definition: Token.h:223
bool isAnyIdentifier() const
Return true if this is a raw identifier (when lexing in raw mode) or a non-keyword identifier (when l...
Definition: Token.h:109
SourceLocation getEndLoc() const
Definition: Token.h:153
unsigned getFlags() const
Return the internal represtation of the flags.
Definition: Token.h:256
void setAnnotationEndLoc(SourceLocation L)
Definition: Token.h:144
bool hasUCN() const
Returns true if this token contains a universal character name.
Definition: Token.h:298
void clearFlag(TokenFlags Flag)
Unset the specified flag.
Definition: Token.h:248
bool isLiteral() const
Return true if this is a "literal", like a numeric constant, string, etc.
Definition: Token.h:115
SourceLocation getLocation() const
Return a source location identifier for the specified offset in the current file.
Definition: Token.h:126
const char * getName() const
Definition: Token.h:168
unsigned getLength() const
Definition: Token.h:129
void setLength(unsigned Len)
Definition: Token.h:135
bool isEditorPlaceholder() const
Returns true if this token is an editor placeholder.
Definition: Token.h:312
bool isExpandDisabled() const
Return true if this identifier token should never be expanded in the future, due to C99 6....
Definition: Token.h:278
void setKind(tok::TokenKind K)
Definition: Token.h:94
bool commaAfterElided() const
Returns true if the comma after this token was elided.
Definition: Token.h:305
SourceLocation getAnnotationEndLoc() const
Definition: Token.h:140
tok::ObjCKeywordKind getObjCKeywordID() const
Return the ObjC keyword kind.
Definition: Lexer.cpp:66
bool is(tok::TokenKind K) const
is/isNot - Predicates to check if this token is a specific kind, as in "if (Tok.is(tok::l_brace)) {....
Definition: Token.h:98
void * getAnnotationValue() const
Definition: Token.h:228
bool isOneOf(tok::TokenKind K1, Ts... Ks) const
Definition: Token.h:103
tok::TokenKind getKind() const
Definition: Token.h:93
bool isAtStartOfLine() const
isAtStartOfLine - Return true if this token is at the start of a line.
Definition: Token.h:270
void setEofData(const void *D)
Definition: Token.h:198
bool getFlag(TokenFlags Flag) const
Get the specified flag.
Definition: Token.h:243
@ DisableExpand
Definition: Token.h:78
@ HasUCN
Definition: Token.h:82
@ IsEditorPlaceholder
Definition: Token.h:87
@ IgnoredComma
Definition: Token.h:83
@ IsReinjected
Definition: Token.h:88
@ LeadingEmptyMacro
Definition: Token.h:80
@ LeadingSpace
Definition: Token.h:76
@ StartOfLine
Definition: Token.h:74
@ StringifiedInMacro
Definition: Token.h:84
@ HasUDSuffix
Definition: Token.h:81
@ CommaAfterElided
Definition: Token.h:86
@ NeedsCleaning
Definition: Token.h:79
bool hasLeadingSpace() const
Return true if this token has whitespace before it.
Definition: Token.h:274
SourceRange getAnnotationRange() const
SourceRange of the group of tokens that this annotation token represents.
Definition: Token.h:160
void setLocation(SourceLocation L)
Definition: Token.h:134
bool hasLeadingEmptyMacro() const
Return true if this token has an empty macro before it.
Definition: Token.h:291
void setRawIdentifierData(const char *Ptr)
Definition: Token.h:211
bool isOneOf(tok::TokenKind K1, tok::TokenKind K2) const
Definition: Token.h:100
bool isNot(tok::TokenKind K) const
Definition: Token.h:99
bool hasPtrData() const
Definition: Token.h:179
bool isAnnotation() const
Return true if this is any of tok::annot_* kind tokens.
Definition: Token.h:120
void setAnnotationValue(void *val)
Definition: Token.h:232
const void * getEofData() const
Definition: Token.h:194
bool hasUDSuffix() const
Return true if this token is a string or character literal which has a ud-suffix.
Definition: Token.h:295
bool stringifiedInMacro() const
Returns true if this token is formed by macro by stringizing or charizing operator.
Definition: Token.h:302
bool isObjCAtKeyword(tok::ObjCKeywordKind objcKey) const
Return true if we have an ObjC keyword identifier.
Definition: Lexer.cpp:57
void setAnnotationRange(SourceRange R)
Definition: Token.h:163
void startToken()
Reset all flags to cleared.
Definition: Token.h:171
bool needsCleaning() const
Return true if this token has trigraphs or escaped newlines in it.
Definition: Token.h:287
void setIdentifierInfo(IdentifierInfo *II)
Definition: Token.h:190
SourceLocation getLastLoc() const
Definition: Token.h:149
void setFlagValue(TokenFlags Flag, bool Val)
Set a flag to either true or false.
Definition: Token.h:261
StringRef getRawIdentifier() const
getRawIdentifier - For a raw identifier token (i.e., an identifier lexed in raw mode),...
Definition: Token.h:207
const char * getLiteralData() const
getLiteralData - For a literal token (numeric constant, string, etc), this returns a pointer to the s...
Definition: Token.h:219
void setFlag(TokenFlags Flag)
Set the specified flag.
Definition: Token.h:238
const char * getTokenName(TokenKind Kind) LLVM_READNONE
Determines the name of a token as used within the front end.
Definition: TokenKinds.cpp:24
bool isAnyIdentifier(TokenKind K)
Return true if this is a raw identifier or an identifier kind.
Definition: TokenKinds.h:75
ObjCKeywordKind
Provides a namespace for Objective-C keywords which start with an '@'.
Definition: TokenKinds.h:41
TokenKind
Provides a simple uniform namespace for tokens from all C languages.
Definition: TokenKinds.h:25
bool isLiteral(TokenKind K)
Return true if this is a "literal" kind, like a numeric constant, string, etc.
Definition: TokenKinds.h:89
bool isAnnotation(TokenKind K)
Return true if this is any of tok::annot_* kinds.
Definition: TokenKinds.cpp:58
Information about the conditional stack (#if directives) currently active.
Definition: Token.h:317
bool FoundNonSkip
True if we have emitted tokens already, and now we're in an #else block or something.
Definition: Token.h:327
SourceLocation IfLoc
Location where the conditional started.
Definition: Token.h:319
bool WasSkipping
True if this was contained in a skipping directive, e.g., in a "\#if 0" block.
Definition: Token.h:323
bool FoundElse
True if we've seen a #else in this block.
Definition: Token.h:331
ArrayRef< Token > Toks
Definition: Token.h:338