Line data Source code
1 : // Copyright 2012 the V8 project authors. All rights reserved.
2 : // Use of this source code is governed by a BSD-style license that can be
3 : // found in the LICENSE file.
4 :
5 : #ifndef V8_PARSING_TOKEN_H_
6 : #define V8_PARSING_TOKEN_H_
7 :
8 : #include "src/base/logging.h"
9 : #include "src/globals.h"
10 : #include "src/utils.h"
11 :
12 : namespace v8 {
13 : namespace internal {
14 :
15 : // TOKEN_LIST takes a list of 3 macros M, all of which satisfy the
16 : // same signature M(name, string, precedence), where name is the
17 : // symbolic token name, string is the corresponding syntactic symbol
18 : // (or nullptr, for literals), and precedence is the precedence (or 0).
19 : // The parameters are invoked for token categories as follows:
20 : //
21 : // T: Non-keyword tokens
22 : // K: Keyword tokens
23 :
24 : // IGNORE_TOKEN is a convenience macro that can be supplied as
25 : // an argument (at any position) for a TOKEN_LIST call. It does
26 : // nothing with tokens belonging to the respective category.
27 :
28 : #define IGNORE_TOKEN(name, string, precedence)
29 :
30 : /* Binary operators */
31 : /* ADD and SUB are at the end since they are UnaryOp */
32 : #define BINARY_OP_TOKEN_LIST(T, E) \
33 : E(T, BIT_OR, "|", 6) \
34 : E(T, BIT_XOR, "^", 7) \
35 : E(T, BIT_AND, "&", 8) \
36 : E(T, SHL, "<<", 11) \
37 : E(T, SAR, ">>", 11) \
38 : E(T, SHR, ">>>", 11) \
39 : E(T, MUL, "*", 13) \
40 : E(T, DIV, "/", 13) \
41 : E(T, MOD, "%", 13) \
42 : E(T, EXP, "**", 14) \
43 : E(T, ADD, "+", 12) \
44 : E(T, SUB, "-", 12)
45 :
46 : #define EXPAND_BINOP_ASSIGN_TOKEN(T, name, string, precedence) \
47 : T(ASSIGN_##name, string "=", 2)
48 :
49 : #define EXPAND_BINOP_TOKEN(T, name, string, precedence) \
50 : T(name, string, precedence)
51 :
52 : #define TOKEN_LIST(T, K) \
53 : \
54 : /* BEGIN PropertyOrCall */ \
55 : /* BEGIN Member */ \
56 : /* BEGIN Template */ \
57 : /* ES6 Template Literals */ \
58 : T(TEMPLATE_SPAN, nullptr, 0) \
59 : T(TEMPLATE_TAIL, nullptr, 0) \
60 : /* END Template */ \
61 : \
62 : /* Punctuators (ECMA-262, section 7.7, page 15). */ \
63 : /* BEGIN Property */ \
64 : T(PERIOD, ".", 0) \
65 : T(LBRACK, "[", 0) \
66 : /* END Property */ \
67 : /* END Member */ \
68 : T(LPAREN, "(", 0) \
69 : /* END PropertyOrCall */ \
70 : T(RPAREN, ")", 0) \
71 : T(RBRACK, "]", 0) \
72 : T(LBRACE, "{", 0) \
73 : T(COLON, ":", 0) \
74 : T(ELLIPSIS, "...", 0) \
75 : T(CONDITIONAL, "?", 3) \
76 : /* BEGIN AutoSemicolon */ \
77 : T(SEMICOLON, ";", 0) \
78 : T(RBRACE, "}", 0) \
79 : /* End of source indicator. */ \
80 : T(EOS, "EOS", 0) \
81 : /* END AutoSemicolon */ \
82 : \
83 : /* BEGIN ArrowOrAssignmentOp */ \
84 : T(ARROW, "=>", 0) \
85 : /* BEGIN AssignmentOp */ \
86 : /* IsAssignmentOp() relies on this block of enum values being */ \
87 : /* contiguous and sorted in the same order! */ \
88 : T(INIT, "=init", 2) /* AST-use only. */ \
89 : T(ASSIGN, "=", 2) \
90 : BINARY_OP_TOKEN_LIST(T, EXPAND_BINOP_ASSIGN_TOKEN) \
91 : /* END AssignmentOp */ \
92 : /* END ArrowOrAssignmentOp */ \
93 : \
94 : /* Binary operators sorted by precedence. */ \
95 : /* IsBinaryOp() relies on this block of enum values */ \
96 : /* being contiguous and sorted in the same order! */ \
97 : T(COMMA, ",", 1) \
98 : T(OR, "||", 4) \
99 : T(AND, "&&", 5) \
100 : \
101 : /* Unary operators, starting at ADD in BINARY_OP_TOKEN_LIST */ \
102 : /* IsUnaryOp() relies on this block of enum values */ \
103 : /* being contiguous and sorted in the same order! */ \
104 : BINARY_OP_TOKEN_LIST(T, EXPAND_BINOP_TOKEN) \
105 : \
106 : T(NOT, "!", 0) \
107 : T(BIT_NOT, "~", 0) \
108 : K(DELETE, "delete", 0) \
109 : K(TYPEOF, "typeof", 0) \
110 : K(VOID, "void", 0) \
111 : \
112 : /* BEGIN IsCountOp */ \
113 : T(INC, "++", 0) \
114 : T(DEC, "--", 0) \
115 : /* END IsCountOp */ \
116 : /* END IsUnaryOrCountOp */ \
117 : \
118 : /* Compare operators sorted by precedence. */ \
119 : /* IsCompareOp() relies on this block of enum values */ \
120 : /* being contiguous and sorted in the same order! */ \
121 : T(EQ, "==", 9) \
122 : T(EQ_STRICT, "===", 9) \
123 : T(NE, "!=", 9) \
124 : T(NE_STRICT, "!==", 9) \
125 : T(LT, "<", 10) \
126 : T(GT, ">", 10) \
127 : T(LTE, "<=", 10) \
128 : T(GTE, ">=", 10) \
129 : K(INSTANCEOF, "instanceof", 10) \
130 : K(IN, "in", 10) \
131 : \
132 : /* Keywords (ECMA-262, section 7.5.2, page 13). */ \
133 : K(BREAK, "break", 0) \
134 : K(CASE, "case", 0) \
135 : K(CATCH, "catch", 0) \
136 : K(CONTINUE, "continue", 0) \
137 : K(DEBUGGER, "debugger", 0) \
138 : K(DEFAULT, "default", 0) \
139 : /* DELETE */ \
140 : K(DO, "do", 0) \
141 : K(ELSE, "else", 0) \
142 : K(FINALLY, "finally", 0) \
143 : K(FOR, "for", 0) \
144 : K(FUNCTION, "function", 0) \
145 : K(IF, "if", 0) \
146 : /* IN */ \
147 : /* INSTANCEOF */ \
148 : K(NEW, "new", 0) \
149 : K(RETURN, "return", 0) \
150 : K(SWITCH, "switch", 0) \
151 : K(THROW, "throw", 0) \
152 : K(TRY, "try", 0) \
153 : /* TYPEOF */ \
154 : K(VAR, "var", 0) \
155 : /* VOID */ \
156 : K(WHILE, "while", 0) \
157 : K(WITH, "with", 0) \
158 : K(THIS, "this", 0) \
159 : \
160 : /* Literals (ECMA-262, section 7.8, page 16). */ \
161 : K(NULL_LITERAL, "null", 0) \
162 : K(TRUE_LITERAL, "true", 0) \
163 : K(FALSE_LITERAL, "false", 0) \
164 : T(NUMBER, nullptr, 0) \
165 : T(SMI, nullptr, 0) \
166 : T(BIGINT, nullptr, 0) \
167 : T(STRING, nullptr, 0) \
168 : \
169 : /* BEGIN Callable */ \
170 : K(SUPER, "super", 0) \
171 : /* BEGIN AnyIdentifier */ \
172 : /* Identifiers (not keywords or future reserved words). */ \
173 : T(IDENTIFIER, nullptr, 0) \
174 : K(ASYNC, "async", 0) \
175 : /* `await` is a reserved word in module code only */ \
176 : K(AWAIT, "await", 0) \
177 : K(YIELD, "yield", 0) \
178 : K(LET, "let", 0) \
179 : K(STATIC, "static", 0) \
180 : /* Future reserved words (ECMA-262, section 7.6.1.2). */ \
181 : T(FUTURE_STRICT_RESERVED_WORD, nullptr, 0) \
182 : T(ESCAPED_STRICT_RESERVED_WORD, nullptr, 0) \
183 : /* END AnyIdentifier */ \
184 : /* END Callable */ \
185 : K(ENUM, "enum", 0) \
186 : K(CLASS, "class", 0) \
187 : K(CONST, "const", 0) \
188 : K(EXPORT, "export", 0) \
189 : K(EXTENDS, "extends", 0) \
190 : K(IMPORT, "import", 0) \
191 : T(PRIVATE_NAME, nullptr, 0) \
192 : \
193 : /* Illegal token - not able to scan. */ \
194 : T(ILLEGAL, "ILLEGAL", 0) \
195 : T(ESCAPED_KEYWORD, nullptr, 0) \
196 : \
197 : /* Scanner-internal use only. */ \
198 : T(WHITESPACE, nullptr, 0) \
199 : T(UNINITIALIZED, nullptr, 0) \
200 : T(REGEXP_LITERAL, nullptr, 0)
201 :
202 : class Token {
203 : public:
204 : // All token values.
205 : #define T(name, string, precedence) name,
206 : enum Value : uint8_t { TOKEN_LIST(T, T) NUM_TOKENS };
207 : #undef T
208 :
209 : // Returns a string corresponding to the C++ token name
210 : // (e.g. "LT" for the token LT).
211 : static const char* Name(Value token) {
212 : DCHECK_GT(NUM_TOKENS, token); // token is unsigned
213 4665 : return name_[token];
214 : }
215 :
216 : class IsKeywordBits : public BitField8<bool, 0, 1> {};
217 : class IsPropertyNameBits : public BitField8<bool, IsKeywordBits::kNext, 1> {};
218 :
219 : // Predicates
220 : static bool IsKeyword(Value token) {
221 : return IsKeywordBits::decode(token_flags[token]);
222 : }
223 :
224 279431 : static bool IsPropertyName(Value token) {
225 15050760 : return IsPropertyNameBits::decode(token_flags[token]);
226 : }
227 :
228 : V8_INLINE static bool IsValidIdentifier(Value token,
229 : LanguageMode language_mode,
230 : bool is_generator,
231 : bool disallow_await) {
232 3067111 : if (V8_LIKELY(IsInRange(token, IDENTIFIER, ASYNC))) return true;
233 133935 : if (token == AWAIT) return !disallow_await;
234 102444 : if (token == YIELD) return !is_generator && is_sloppy(language_mode);
235 74703 : return IsStrictReservedWord(token) && is_sloppy(language_mode);
236 : }
237 :
238 : static bool IsCallable(Value token) {
239 : return IsInRange(token, SUPER, ESCAPED_STRICT_RESERVED_WORD);
240 : }
241 :
242 : static bool IsAutoSemicolon(Value token) {
243 : return IsInRange(token, SEMICOLON, EOS);
244 : }
245 :
246 : static bool IsAnyIdentifier(Value token) {
247 : return IsInRange(token, IDENTIFIER, ESCAPED_STRICT_RESERVED_WORD);
248 : }
249 :
250 74704 : static bool IsStrictReservedWord(Value token) {
251 74704 : return IsInRange(token, YIELD, ESCAPED_STRICT_RESERVED_WORD);
252 : }
253 :
254 : static bool IsLiteral(Value token) {
255 : return IsInRange(token, NULL_LITERAL, STRING);
256 : }
257 :
258 : static bool IsTemplate(Value token) {
259 : return IsInRange(token, TEMPLATE_SPAN, TEMPLATE_TAIL);
260 : }
261 :
262 107473292 : static bool IsMember(Value token) {
263 107473292 : return IsInRange(token, TEMPLATE_SPAN, LBRACK);
264 : }
265 :
266 : static bool IsProperty(Value token) {
267 : return IsInRange(token, PERIOD, LBRACK);
268 : }
269 :
270 106971320 : static bool IsPropertyOrCall(Value token) {
271 106971320 : return IsInRange(token, TEMPLATE_SPAN, LPAREN);
272 : }
273 :
274 : static bool IsArrowOrAssignmentOp(Value token) {
275 : return IsInRange(token, ARROW, ASSIGN_SUB);
276 : }
277 :
278 : static bool IsAssignmentOp(Value token) {
279 : return IsInRange(token, INIT, ASSIGN_SUB);
280 : }
281 :
282 : static bool IsBinaryOp(Value op) { return IsInRange(op, COMMA, SUB); }
283 :
284 : static bool IsCompareOp(Value op) { return IsInRange(op, EQ, IN); }
285 :
286 : static bool IsOrderedRelationalCompareOp(Value op) {
287 : return IsInRange(op, LT, GTE);
288 : }
289 :
290 : static bool IsEqualityOp(Value op) { return IsInRange(op, EQ, EQ_STRICT); }
291 :
292 : static Value BinaryOpForAssignment(Value op) {
293 : DCHECK(IsInRange(op, ASSIGN_BIT_OR, ASSIGN_SUB));
294 98005 : Value result = static_cast<Value>(op - ASSIGN_BIT_OR + BIT_OR);
295 : DCHECK(IsBinaryOp(result));
296 : return result;
297 : }
298 :
299 : static bool IsBitOp(Value op) {
300 550 : return IsInRange(op, BIT_OR, SHR) || op == BIT_NOT;
301 : }
302 :
303 : static bool IsUnaryOp(Value op) { return IsInRange(op, ADD, VOID); }
304 211537494 : static bool IsCountOp(Value op) { return IsInRange(op, INC, DEC); }
305 218005666 : static bool IsUnaryOrCountOp(Value op) { return IsInRange(op, ADD, DEC); }
306 : static bool IsShiftOp(Value op) { return IsInRange(op, SHL, SHR); }
307 :
308 : // Returns a string corresponding to the JS token string
309 : // (.e., "<" for the token LT) or nullptr if the token doesn't
310 : // have a (unique) string (e.g. an IDENTIFIER).
311 : static const char* String(Value token) {
312 : DCHECK_GT(NUM_TOKENS, token); // token is unsigned
313 82991 : return string_[token];
314 : }
315 :
316 : static uint8_t StringLength(Value token) {
317 : DCHECK_GT(NUM_TOKENS, token); // token is unsigned
318 : return string_length_[token];
319 : }
320 :
321 : // Returns the precedence > 0 for binary and compare
322 : // operators; returns 0 otherwise.
323 106804972 : static int Precedence(Value token, bool accept_IN) {
324 : DCHECK_GT(NUM_TOKENS, token); // token is unsigned
325 174259954 : return precedence_[accept_IN][token];
326 : }
327 :
328 : private:
329 : static const char* const name_[NUM_TOKENS];
330 : static const char* const string_[NUM_TOKENS];
331 : static const uint8_t string_length_[NUM_TOKENS];
332 : static const int8_t precedence_[2][NUM_TOKENS];
333 : static const uint8_t token_flags[NUM_TOKENS];
334 : };
335 :
336 : } // namespace internal
337 : } // namespace v8
338 :
339 : #endif // V8_PARSING_TOKEN_H_
|