1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
|
// Copyright (c) 2013 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "tools/gn/tokenizer.h"
#include "base/logging.h"
#include "tools/gn/input_file.h"
namespace {
bool IsNumberChar(char c) {
return c >= '0' && c <= '9';
}
bool CouldBeTwoCharOperatorBegin(char c) {
return c == '<' || c == '>' || c == '!' || c == '=' || c == '-' ||
c == '+' || c == '|' || c == '&';
}
bool CouldBeTwoCharOperatorEnd(char c) {
return c == '=' || c == '|' || c == '&';
}
bool CouldBeOneCharOperator(char c) {
return c == '=' || c == '<' || c == '>' || c == '+' || c == '!' ||
c == ':' || c == '|' || c == '&' || c == '-';
}
bool CouldBeOperator(char c) {
return CouldBeOneCharOperator(c) || CouldBeTwoCharOperatorBegin(c);
}
bool IsScoperChar(char c) {
return c == '(' || c == ')' || c == '[' || c == ']' || c == '{' || c == '}';
}
Token::Type GetSpecificOperatorType(base::StringPiece value) {
if (value == "=")
return Token::EQUAL;
if (value == "+")
return Token::PLUS;
if (value == "-")
return Token::MINUS;
if (value == "+=")
return Token::PLUS_EQUALS;
if (value == "-=")
return Token::MINUS_EQUALS;
if (value == "==")
return Token::EQUAL_EQUAL;
if (value == "!=")
return Token::NOT_EQUAL;
if (value == "<=")
return Token::LESS_EQUAL;
if (value == ">=")
return Token::GREATER_EQUAL;
if (value == "<")
return Token::LESS_THAN;
if (value == ">")
return Token::GREATER_THAN;
if (value == "&&")
return Token::BOOLEAN_AND;
if (value == "||")
return Token::BOOLEAN_OR;
if (value == "!")
return Token::BANG;
return Token::INVALID;
}
} // namespace
Tokenizer::Tokenizer(const InputFile* input_file, Err* err)
: input_file_(input_file),
input_(input_file->contents()),
err_(err),
cur_(0),
line_number_(1),
char_in_line_(1) {
}
Tokenizer::~Tokenizer() {
}
// static
std::vector<Token> Tokenizer::Tokenize(const InputFile* input_file, Err* err) {
Tokenizer t(input_file, err);
return t.Run();
}
std::vector<Token> Tokenizer::Run() {
DCHECK(tokens_.empty());
while (!done()) {
AdvanceToNextToken();
if (done())
break;
Location location = GetCurrentLocation();
Token::Type type = ClassifyCurrent();
if (type == Token::INVALID) {
*err_ = GetErrorForInvalidToken(location);
break;
}
size_t token_begin = cur_;
AdvanceToEndOfToken(location, type);
if (has_error())
break;
size_t token_end = cur_;
base::StringPiece token_value(&input_.data()[token_begin],
token_end - token_begin);
if (type == Token::UNCLASSIFIED_OPERATOR)
type = GetSpecificOperatorType(token_value);
if (type == Token::IDENTIFIER) {
if (token_value == "if")
type = Token::IF;
else if (token_value == "else")
type = Token::ELSE;
else if (token_value == "true")
type = Token::TRUE_TOKEN;
else if (token_value == "false")
type = Token::FALSE_TOKEN;
}
// TODO(brettw) This just strips comments from the token stream. This
// is probably wrong, they should be removed at a later stage so we can
// do things like rewrite the file. But this makes the parser simpler and
// is OK for now.
if (type != Token::COMMENT)
tokens_.push_back(Token(location, type, token_value));
}
if (err_->has_error())
tokens_.clear();
return tokens_;
}
// static
size_t Tokenizer::ByteOffsetOfNthLine(const base::StringPiece& buf, int n) {
int cur_line = 1;
size_t cur_byte = 0;
DCHECK(n > 0);
if (n == 1)
return 0;
while (cur_byte < buf.size()) {
if (IsNewline(buf, cur_byte)) {
cur_line++;
if (cur_line == n)
return cur_byte + 1;
}
cur_byte++;
}
return -1;
}
// static
bool Tokenizer::IsNewline(const base::StringPiece& buffer, size_t offset) {
DCHECK(offset < buffer.size());
// We may need more logic here to handle different line ending styles.
return buffer[offset] == '\n';
}
void Tokenizer::AdvanceToNextToken() {
while (!at_end() && IsCurrentWhitespace())
Advance();
}
Token::Type Tokenizer::ClassifyCurrent() const {
DCHECK(!at_end());
char next_char = cur_char();
if (next_char >= '0' && next_char <= '9')
return Token::INTEGER;
if (next_char == '"')
return Token::STRING;
// Note: '-' handled specially below.
if (next_char != '-' && CouldBeOperator(next_char))
return Token::UNCLASSIFIED_OPERATOR;
if (IsIdentifierFirstChar(next_char))
return Token::IDENTIFIER;
if (next_char == '[')
return Token::LEFT_BRACKET;
if (next_char == ']')
return Token::RIGHT_BRACKET;
if (next_char == '(')
return Token::LEFT_PAREN;
if (next_char == ')')
return Token::RIGHT_PAREN;
if (next_char == '{')
return Token::LEFT_BRACE;
if (next_char == '}')
return Token::RIGHT_BRACE;
if (next_char == ',')
return Token::COMMA;
if (next_char == '#')
return Token::COMMENT;
// For the case of '-' differentiate between a negative number and anything
// else.
if (next_char == '-') {
if (!CanIncrement())
return Token::UNCLASSIFIED_OPERATOR; // Just the minus before end of
// file.
char following_char = input_[cur_ + 1];
if (following_char >= '0' && following_char <= '9')
return Token::INTEGER;
return Token::UNCLASSIFIED_OPERATOR;
}
return Token::INVALID;
}
void Tokenizer::AdvanceToEndOfToken(const Location& location,
Token::Type type) {
switch (type) {
case Token::INTEGER:
do {
Advance();
} while (!at_end() && IsNumberChar(cur_char()));
if (!at_end()) {
// Require the char after a number to be some kind of space, scope,
// or operator.
char c = cur_char();
if (!IsCurrentWhitespace() && !CouldBeOperator(c) &&
!IsScoperChar(c) && c != ',') {
*err_ = Err(GetCurrentLocation(),
"This is not a valid number.",
"Learn to count.");
// Highlight the number.
err_->AppendRange(LocationRange(location, GetCurrentLocation()));
}
}
break;
case Token::STRING: {
char initial = cur_char();
Advance(); // Advance past initial "
for (;;) {
if (at_end()) {
*err_ = Err(LocationRange(location,
Location(input_file_, line_number_, char_in_line_)),
"Unterminated string literal.",
"Don't leave me hanging like this!");
break;
}
if (IsCurrentStringTerminator(initial)) {
Advance(); // Skip past last "
break;
} else if (cur_char() == '\n') {
*err_ = Err(LocationRange(location,
GetCurrentLocation()),
"Newline in string constant.");
}
Advance();
}
break;
}
case Token::UNCLASSIFIED_OPERATOR:
// Some operators are two characters, some are one.
if (CouldBeTwoCharOperatorBegin(cur_char())) {
if (CanIncrement() && CouldBeTwoCharOperatorEnd(input_[cur_ + 1]))
Advance();
}
Advance();
break;
case Token::IDENTIFIER:
while (!at_end() && IsIdentifierContinuingChar(cur_char()))
Advance();
break;
case Token::LEFT_BRACKET:
case Token::RIGHT_BRACKET:
case Token::LEFT_BRACE:
case Token::RIGHT_BRACE:
case Token::LEFT_PAREN:
case Token::RIGHT_PAREN:
case Token::COMMA:
Advance(); // All are one char.
break;
case Token::COMMENT:
// Eat to EOL.
while (!at_end() && !IsCurrentNewline())
Advance();
break;
case Token::INVALID:
default:
*err_ = Err(location, "Everything is all messed up",
"Please insert system disk in drive A: and press any key.");
NOTREACHED();
return;
}
}
bool Tokenizer::IsCurrentWhitespace() const {
DCHECK(!at_end());
char c = input_[cur_];
// Note that tab (0x09) is illegal.
return c == 0x0A || c == 0x0B || c == 0x0C || c == 0x0D || c == 0x20;
}
bool Tokenizer::IsCurrentStringTerminator(char quote_char) const {
DCHECK(!at_end());
if (cur_char() != quote_char)
return false;
// Check for escaping. \" is not a string terminator, but \\" is. Count
// the number of preceeding backslashes.
int num_backslashes = 0;
for (int i = static_cast<int>(cur_) - 1; i >= 0 && input_[i] == '\\'; i--)
num_backslashes++;
// Even backslashes mean that they were escaping each other and don't count
// as escaping this quote.
return (num_backslashes % 2) == 0;
}
bool Tokenizer::IsCurrentNewline() const {
return IsNewline(input_, cur_);
}
void Tokenizer::Advance() {
DCHECK(cur_ < input_.size());
if (IsCurrentNewline()) {
line_number_++;
char_in_line_ = 1;
} else {
char_in_line_++;
}
cur_++;
}
Location Tokenizer::GetCurrentLocation() const {
return Location(input_file_, line_number_, char_in_line_);
}
Err Tokenizer::GetErrorForInvalidToken(const Location& location) const {
std::string help;
if (cur_char() == ';') {
// Semicolon.
help = "Semicolons are not needed, delete this one.";
} else if (cur_char() == '\t') {
// Tab.
help = "You got a tab character in here. Tabs are evil. "
"Convert to spaces.";
} else if (cur_char() == '/' && cur_ + 1 < input_.size() &&
(input_[cur_ + 1] == '/' || input_[cur_ + 1] == '*')) {
// Different types of comments.
help = "Comments should start with # instead";
} else {
help = "I have no idea what this is.";
}
return Err(location, "Invalid token.", help);
}
|