natalie_parser 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/CHANGELOG.md +22 -0
- data/Dockerfile +26 -0
- data/Gemfile +10 -0
- data/LICENSE +21 -0
- data/README.md +55 -0
- data/Rakefile +242 -0
- data/ext/natalie_parser/extconf.rb +9 -0
- data/ext/natalie_parser/mri_creator.hpp +139 -0
- data/ext/natalie_parser/natalie_parser.cpp +144 -0
- data/include/natalie_parser/creator/debug_creator.hpp +113 -0
- data/include/natalie_parser/creator.hpp +108 -0
- data/include/natalie_parser/lexer/interpolated_string_lexer.hpp +64 -0
- data/include/natalie_parser/lexer/regexp_lexer.hpp +37 -0
- data/include/natalie_parser/lexer/word_array_lexer.hpp +57 -0
- data/include/natalie_parser/lexer.hpp +135 -0
- data/include/natalie_parser/node/alias_node.hpp +35 -0
- data/include/natalie_parser/node/arg_node.hpp +74 -0
- data/include/natalie_parser/node/array_node.hpp +34 -0
- data/include/natalie_parser/node/array_pattern_node.hpp +28 -0
- data/include/natalie_parser/node/assignment_node.hpp +34 -0
- data/include/natalie_parser/node/back_ref_node.hpp +28 -0
- data/include/natalie_parser/node/begin_block_node.hpp +25 -0
- data/include/natalie_parser/node/begin_node.hpp +52 -0
- data/include/natalie_parser/node/begin_rescue_node.hpp +47 -0
- data/include/natalie_parser/node/bignum_node.hpp +37 -0
- data/include/natalie_parser/node/block_node.hpp +55 -0
- data/include/natalie_parser/node/block_pass_node.hpp +33 -0
- data/include/natalie_parser/node/break_node.hpp +32 -0
- data/include/natalie_parser/node/call_node.hpp +85 -0
- data/include/natalie_parser/node/case_in_node.hpp +40 -0
- data/include/natalie_parser/node/case_node.hpp +52 -0
- data/include/natalie_parser/node/case_when_node.hpp +43 -0
- data/include/natalie_parser/node/class_node.hpp +39 -0
- data/include/natalie_parser/node/colon2_node.hpp +44 -0
- data/include/natalie_parser/node/colon3_node.hpp +34 -0
- data/include/natalie_parser/node/constant_node.hpp +26 -0
- data/include/natalie_parser/node/def_node.hpp +55 -0
- data/include/natalie_parser/node/defined_node.hpp +33 -0
- data/include/natalie_parser/node/encoding_node.hpp +26 -0
- data/include/natalie_parser/node/end_block_node.hpp +25 -0
- data/include/natalie_parser/node/evaluate_to_string_node.hpp +37 -0
- data/include/natalie_parser/node/false_node.hpp +23 -0
- data/include/natalie_parser/node/fixnum_node.hpp +36 -0
- data/include/natalie_parser/node/float_node.hpp +36 -0
- data/include/natalie_parser/node/hash_node.hpp +34 -0
- data/include/natalie_parser/node/hash_pattern_node.hpp +27 -0
- data/include/natalie_parser/node/identifier_node.hpp +123 -0
- data/include/natalie_parser/node/if_node.hpp +43 -0
- data/include/natalie_parser/node/infix_op_node.hpp +46 -0
- data/include/natalie_parser/node/interpolated_node.hpp +33 -0
- data/include/natalie_parser/node/interpolated_regexp_node.hpp +28 -0
- data/include/natalie_parser/node/interpolated_shell_node.hpp +22 -0
- data/include/natalie_parser/node/interpolated_string_node.hpp +31 -0
- data/include/natalie_parser/node/interpolated_symbol_key_node.hpp +18 -0
- data/include/natalie_parser/node/interpolated_symbol_node.hpp +28 -0
- data/include/natalie_parser/node/iter_node.hpp +45 -0
- data/include/natalie_parser/node/keyword_arg_node.hpp +25 -0
- data/include/natalie_parser/node/keyword_splat_node.hpp +38 -0
- data/include/natalie_parser/node/logical_and_node.hpp +40 -0
- data/include/natalie_parser/node/logical_or_node.hpp +40 -0
- data/include/natalie_parser/node/match_node.hpp +38 -0
- data/include/natalie_parser/node/module_node.hpp +32 -0
- data/include/natalie_parser/node/multiple_assignment_arg_node.hpp +32 -0
- data/include/natalie_parser/node/multiple_assignment_node.hpp +37 -0
- data/include/natalie_parser/node/next_node.hpp +37 -0
- data/include/natalie_parser/node/nil_node.hpp +23 -0
- data/include/natalie_parser/node/nil_sexp_node.hpp +23 -0
- data/include/natalie_parser/node/node.hpp +155 -0
- data/include/natalie_parser/node/node_with_args.hpp +47 -0
- data/include/natalie_parser/node/not_match_node.hpp +35 -0
- data/include/natalie_parser/node/not_node.hpp +37 -0
- data/include/natalie_parser/node/nth_ref_node.hpp +27 -0
- data/include/natalie_parser/node/op_assign_accessor_node.hpp +74 -0
- data/include/natalie_parser/node/op_assign_and_node.hpp +34 -0
- data/include/natalie_parser/node/op_assign_node.hpp +47 -0
- data/include/natalie_parser/node/op_assign_or_node.hpp +34 -0
- data/include/natalie_parser/node/pin_node.hpp +33 -0
- data/include/natalie_parser/node/range_node.hpp +52 -0
- data/include/natalie_parser/node/redo_node.hpp +20 -0
- data/include/natalie_parser/node/regexp_node.hpp +36 -0
- data/include/natalie_parser/node/retry_node.hpp +20 -0
- data/include/natalie_parser/node/return_node.hpp +34 -0
- data/include/natalie_parser/node/safe_call_node.hpp +31 -0
- data/include/natalie_parser/node/sclass_node.hpp +37 -0
- data/include/natalie_parser/node/self_node.hpp +23 -0
- data/include/natalie_parser/node/shadow_arg_node.hpp +40 -0
- data/include/natalie_parser/node/shell_node.hpp +32 -0
- data/include/natalie_parser/node/splat_node.hpp +39 -0
- data/include/natalie_parser/node/splat_value_node.hpp +32 -0
- data/include/natalie_parser/node/stabby_proc_node.hpp +29 -0
- data/include/natalie_parser/node/string_node.hpp +42 -0
- data/include/natalie_parser/node/super_node.hpp +44 -0
- data/include/natalie_parser/node/symbol_key_node.hpp +19 -0
- data/include/natalie_parser/node/symbol_node.hpp +30 -0
- data/include/natalie_parser/node/to_array_node.hpp +33 -0
- data/include/natalie_parser/node/true_node.hpp +23 -0
- data/include/natalie_parser/node/unary_op_node.hpp +41 -0
- data/include/natalie_parser/node/undef_node.hpp +31 -0
- data/include/natalie_parser/node/until_node.hpp +21 -0
- data/include/natalie_parser/node/while_node.hpp +52 -0
- data/include/natalie_parser/node/yield_node.hpp +29 -0
- data/include/natalie_parser/node.hpp +89 -0
- data/include/natalie_parser/parser.hpp +218 -0
- data/include/natalie_parser/token.hpp +842 -0
- data/include/tm/defer.hpp +34 -0
- data/include/tm/hashmap.hpp +826 -0
- data/include/tm/macros.hpp +16 -0
- data/include/tm/optional.hpp +223 -0
- data/include/tm/owned_ptr.hpp +186 -0
- data/include/tm/recursion_guard.hpp +156 -0
- data/include/tm/shared_ptr.hpp +259 -0
- data/include/tm/string.hpp +1447 -0
- data/include/tm/tests.hpp +78 -0
- data/include/tm/vector.hpp +796 -0
- data/lib/natalie_parser/sexp.rb +36 -0
- data/lib/natalie_parser/version.rb +5 -0
- data/lib/natalie_parser.rb +3 -0
- data/natalie_parser.gemspec +23 -0
- data/src/lexer/interpolated_string_lexer.cpp +88 -0
- data/src/lexer/regexp_lexer.cpp +95 -0
- data/src/lexer/word_array_lexer.cpp +134 -0
- data/src/lexer.cpp +1703 -0
- data/src/node/alias_node.cpp +11 -0
- data/src/node/assignment_node.cpp +33 -0
- data/src/node/begin_node.cpp +29 -0
- data/src/node/begin_rescue_node.cpp +33 -0
- data/src/node/class_node.cpp +22 -0
- data/src/node/interpolated_regexp_node.cpp +19 -0
- data/src/node/interpolated_shell_node.cpp +25 -0
- data/src/node/interpolated_string_node.cpp +111 -0
- data/src/node/interpolated_symbol_node.cpp +25 -0
- data/src/node/match_node.cpp +14 -0
- data/src/node/module_node.cpp +21 -0
- data/src/node/multiple_assignment_node.cpp +37 -0
- data/src/node/node.cpp +10 -0
- data/src/node/node_with_args.cpp +35 -0
- data/src/node/op_assign_node.cpp +36 -0
- data/src/node/string_node.cpp +33 -0
- data/src/parser.cpp +2972 -0
- data/src/token.cpp +27 -0
- metadata +186 -0
data/src/lexer.cpp
ADDED
@@ -0,0 +1,1703 @@
|
|
1
|
+
#include <errno.h>
|
2
|
+
#include <limits>
|
3
|
+
#include <stdlib.h>
|
4
|
+
|
5
|
+
#include "natalie_parser/lexer.hpp"
|
6
|
+
#include "natalie_parser/lexer/interpolated_string_lexer.hpp"
|
7
|
+
#include "natalie_parser/lexer/regexp_lexer.hpp"
|
8
|
+
#include "natalie_parser/lexer/word_array_lexer.hpp"
|
9
|
+
#include "natalie_parser/token.hpp"
|
10
|
+
|
11
|
+
namespace NatalieParser {
|
12
|
+
|
13
|
+
SharedPtr<Vector<Token>> Lexer::tokens() {
|
14
|
+
SharedPtr<Vector<Token>> tokens = new Vector<Token> {};
|
15
|
+
bool skip_next_newline = false;
|
16
|
+
Token last_doc_token;
|
17
|
+
for (;;) {
|
18
|
+
auto token = next_token();
|
19
|
+
if (token.is_comment())
|
20
|
+
continue;
|
21
|
+
|
22
|
+
if (token.is_doc()) {
|
23
|
+
if (last_doc_token)
|
24
|
+
last_doc_token.literal_string()->append(*token.literal_string());
|
25
|
+
else
|
26
|
+
last_doc_token = token;
|
27
|
+
continue;
|
28
|
+
}
|
29
|
+
|
30
|
+
// get rid of newlines after certain tokens
|
31
|
+
if (skip_next_newline) {
|
32
|
+
if (token.is_newline())
|
33
|
+
continue;
|
34
|
+
else
|
35
|
+
skip_next_newline = false;
|
36
|
+
}
|
37
|
+
|
38
|
+
// get rid of newlines before certain tokens
|
39
|
+
while (token.can_follow_collapsible_newline() && !tokens->is_empty() && tokens->last().is_newline())
|
40
|
+
tokens->pop();
|
41
|
+
|
42
|
+
if (last_doc_token) {
|
43
|
+
if (token.can_have_doc()) {
|
44
|
+
token.set_doc(last_doc_token.literal_string());
|
45
|
+
last_doc_token = {};
|
46
|
+
} else if (!token.is_end_of_line()) {
|
47
|
+
last_doc_token = {};
|
48
|
+
}
|
49
|
+
}
|
50
|
+
|
51
|
+
tokens->push(token);
|
52
|
+
|
53
|
+
m_last_token = token;
|
54
|
+
|
55
|
+
if (token.is_eof())
|
56
|
+
return tokens;
|
57
|
+
if (!token.is_valid())
|
58
|
+
return tokens;
|
59
|
+
if (token.can_precede_collapsible_newline())
|
60
|
+
skip_next_newline = true;
|
61
|
+
};
|
62
|
+
TM_UNREACHABLE();
|
63
|
+
}
|
64
|
+
|
65
|
+
Token Lexer::next_token() {
|
66
|
+
if (m_nested_lexer) {
|
67
|
+
auto token = m_nested_lexer->next_token();
|
68
|
+
if (token.is_eof()) {
|
69
|
+
if (m_nested_lexer->alters_parent_cursor_position()) {
|
70
|
+
m_index = m_nested_lexer->m_index;
|
71
|
+
m_cursor_line = m_nested_lexer->m_cursor_line;
|
72
|
+
m_cursor_column = m_nested_lexer->m_cursor_column;
|
73
|
+
}
|
74
|
+
delete m_nested_lexer;
|
75
|
+
m_nested_lexer = nullptr;
|
76
|
+
} else {
|
77
|
+
return token;
|
78
|
+
}
|
79
|
+
}
|
80
|
+
m_whitespace_precedes = skip_whitespace();
|
81
|
+
m_token_line = m_cursor_line;
|
82
|
+
m_token_column = m_cursor_column;
|
83
|
+
return build_next_token();
|
84
|
+
}
|
85
|
+
|
86
|
+
bool is_identifier_char(char c) {
|
87
|
+
if (!c) return false;
|
88
|
+
return isalnum(c) || c == '_' || (unsigned int)c >= 128;
|
89
|
+
}
|
90
|
+
|
91
|
+
bool is_message_suffix(char c) {
|
92
|
+
if (!c) return false;
|
93
|
+
return c == '?' || c == '!';
|
94
|
+
}
|
95
|
+
|
96
|
+
bool is_identifier_char_or_message_suffix(char c) {
|
97
|
+
return is_identifier_char(c) || is_message_suffix(c);
|
98
|
+
}
|
99
|
+
|
100
|
+
bool Lexer::match(size_t bytes, const char *compare) {
|
101
|
+
if (m_index + bytes > m_size)
|
102
|
+
return false;
|
103
|
+
if (strncmp(compare, m_input->c_str() + m_index, bytes) == 0) {
|
104
|
+
if (m_index + bytes < m_size && is_identifier_char_or_message_suffix(m_input->at(m_index + bytes)))
|
105
|
+
return false;
|
106
|
+
advance(bytes);
|
107
|
+
return true;
|
108
|
+
}
|
109
|
+
return false;
|
110
|
+
}
|
111
|
+
|
112
|
+
void Lexer::advance() {
|
113
|
+
auto c = current_char();
|
114
|
+
m_index++;
|
115
|
+
if (c == '\n') {
|
116
|
+
m_cursor_line++;
|
117
|
+
m_cursor_column = 0;
|
118
|
+
} else {
|
119
|
+
m_cursor_column++;
|
120
|
+
}
|
121
|
+
}
|
122
|
+
|
123
|
+
void Lexer::advance(size_t bytes) {
|
124
|
+
for (size_t i = 0; i < bytes; i++) {
|
125
|
+
advance();
|
126
|
+
}
|
127
|
+
}
|
128
|
+
|
129
|
+
// NOTE: this does not work across lines
|
130
|
+
void Lexer::rewind(size_t bytes) {
|
131
|
+
current_char();
|
132
|
+
m_cursor_column -= bytes;
|
133
|
+
m_index -= bytes;
|
134
|
+
}
|
135
|
+
|
136
|
+
bool Lexer::skip_whitespace() {
|
137
|
+
bool whitespace_found = false;
|
138
|
+
char c = current_char();
|
139
|
+
while (c == ' ' || c == '\t' || (c == '\\' && peek() == '\n')) {
|
140
|
+
whitespace_found = true;
|
141
|
+
advance();
|
142
|
+
if (c == '\\') advance();
|
143
|
+
c = current_char();
|
144
|
+
}
|
145
|
+
return whitespace_found;
|
146
|
+
}
|
147
|
+
|
148
|
+
Token Lexer::build_next_token() {
|
149
|
+
if (m_index >= m_size)
|
150
|
+
return Token { Token::Type::Eof, m_file, m_cursor_line, m_cursor_column };
|
151
|
+
if (m_start_char && current_char() == m_start_char) {
|
152
|
+
m_pair_depth++;
|
153
|
+
} else if (m_stop_char && current_char() == m_stop_char) {
|
154
|
+
if (m_pair_depth == 0)
|
155
|
+
return Token { Token::Type::Eof, m_file, m_cursor_line, m_cursor_column };
|
156
|
+
m_pair_depth--;
|
157
|
+
} else if (m_index == 0 && current_char() == '\xEF') {
|
158
|
+
// UTF-8 BOM
|
159
|
+
advance(); // \xEF
|
160
|
+
if (current_char() == '\xBB') advance();
|
161
|
+
if (current_char() == '\xBF') advance();
|
162
|
+
}
|
163
|
+
Token token;
|
164
|
+
switch (current_char()) {
|
165
|
+
case '=': {
|
166
|
+
advance();
|
167
|
+
switch (current_char()) {
|
168
|
+
case '=': {
|
169
|
+
advance();
|
170
|
+
switch (current_char()) {
|
171
|
+
case '=': {
|
172
|
+
advance();
|
173
|
+
return Token { Token::Type::EqualEqualEqual, m_file, m_token_line, m_token_column };
|
174
|
+
}
|
175
|
+
default:
|
176
|
+
return Token { Token::Type::EqualEqual, m_file, m_token_line, m_token_column };
|
177
|
+
}
|
178
|
+
}
|
179
|
+
case '>':
|
180
|
+
advance();
|
181
|
+
return Token { Token::Type::HashRocket, m_file, m_token_line, m_token_column };
|
182
|
+
case '~':
|
183
|
+
advance();
|
184
|
+
return Token { Token::Type::Match, m_file, m_token_line, m_token_column };
|
185
|
+
default:
|
186
|
+
if (m_cursor_column == 1 && match(5, "begin")) {
|
187
|
+
SharedPtr<String> doc = new String("=begin");
|
188
|
+
char c = current_char();
|
189
|
+
do {
|
190
|
+
doc->append_char(c);
|
191
|
+
c = next();
|
192
|
+
} while (c && !(m_cursor_column == 0 && match(4, "=end")));
|
193
|
+
doc->append("=end\n");
|
194
|
+
return Token { Token::Type::Doc, doc, m_file, m_token_line, m_token_column };
|
195
|
+
}
|
196
|
+
auto token = Token { Token::Type::Equal, m_file, m_token_line, m_token_column };
|
197
|
+
token.set_whitespace_precedes(m_whitespace_precedes);
|
198
|
+
return token;
|
199
|
+
}
|
200
|
+
}
|
201
|
+
case '+':
|
202
|
+
advance();
|
203
|
+
switch (current_char()) {
|
204
|
+
case '=':
|
205
|
+
advance();
|
206
|
+
return Token { Token::Type::PlusEqual, m_file, m_token_line, m_token_column };
|
207
|
+
case '@':
|
208
|
+
if (m_last_token.is_def_keyword() || m_last_token.is_dot()) {
|
209
|
+
advance();
|
210
|
+
SharedPtr<String> lit = new String("+@");
|
211
|
+
return Token { Token::Type::BareName, lit, m_file, m_token_line, m_token_column };
|
212
|
+
} else {
|
213
|
+
return Token { Token::Type::Plus, m_file, m_token_line, m_token_column };
|
214
|
+
}
|
215
|
+
default:
|
216
|
+
return Token { Token::Type::Plus, m_file, m_token_line, m_token_column };
|
217
|
+
}
|
218
|
+
case '-':
|
219
|
+
advance();
|
220
|
+
switch (current_char()) {
|
221
|
+
case '>':
|
222
|
+
advance();
|
223
|
+
return Token { Token::Type::Arrow, m_file, m_token_line, m_token_column };
|
224
|
+
case '=':
|
225
|
+
advance();
|
226
|
+
return Token { Token::Type::MinusEqual, m_file, m_token_line, m_token_column };
|
227
|
+
case '@':
|
228
|
+
if (m_last_token.is_def_keyword() || m_last_token.is_dot()) {
|
229
|
+
advance();
|
230
|
+
SharedPtr<String> lit = new String("-@");
|
231
|
+
return Token { Token::Type::BareName, lit, m_file, m_token_line, m_token_column };
|
232
|
+
} else {
|
233
|
+
return Token { Token::Type::Minus, m_file, m_token_line, m_token_column };
|
234
|
+
}
|
235
|
+
default:
|
236
|
+
return Token { Token::Type::Minus, m_file, m_token_line, m_token_column };
|
237
|
+
}
|
238
|
+
case '*':
|
239
|
+
advance();
|
240
|
+
switch (current_char()) {
|
241
|
+
case '*':
|
242
|
+
advance();
|
243
|
+
switch (current_char()) {
|
244
|
+
case '=':
|
245
|
+
advance();
|
246
|
+
return Token { Token::Type::StarStarEqual, m_file, m_token_line, m_token_column };
|
247
|
+
default:
|
248
|
+
return Token { Token::Type::StarStar, m_file, m_token_line, m_token_column };
|
249
|
+
}
|
250
|
+
case '=':
|
251
|
+
advance();
|
252
|
+
return Token { Token::Type::StarEqual, m_file, m_token_line, m_token_column };
|
253
|
+
default:
|
254
|
+
return Token { Token::Type::Star, m_file, m_token_line, m_token_column };
|
255
|
+
}
|
256
|
+
case '/': {
|
257
|
+
advance();
|
258
|
+
if (!m_last_token)
|
259
|
+
return consume_regexp('/', '/');
|
260
|
+
switch (m_last_token.type()) {
|
261
|
+
case Token::Type::Comma:
|
262
|
+
case Token::Type::Doc:
|
263
|
+
case Token::Type::LBracket:
|
264
|
+
case Token::Type::LCurlyBrace:
|
265
|
+
case Token::Type::LParen:
|
266
|
+
case Token::Type::Match:
|
267
|
+
case Token::Type::Newline:
|
268
|
+
return consume_regexp('/', '/');
|
269
|
+
case Token::Type::DefKeyword:
|
270
|
+
return Token { Token::Type::Slash, m_file, m_token_line, m_token_column };
|
271
|
+
default: {
|
272
|
+
switch (current_char()) {
|
273
|
+
case ' ':
|
274
|
+
return Token { Token::Type::Slash, m_file, m_token_line, m_token_column };
|
275
|
+
case '=':
|
276
|
+
advance();
|
277
|
+
return Token { Token::Type::SlashEqual, m_file, m_token_line, m_token_column };
|
278
|
+
default:
|
279
|
+
if (m_whitespace_precedes) {
|
280
|
+
return consume_regexp('/', '/');
|
281
|
+
} else {
|
282
|
+
return Token { Token::Type::Slash, m_file, m_token_line, m_token_column };
|
283
|
+
}
|
284
|
+
}
|
285
|
+
}
|
286
|
+
}
|
287
|
+
}
|
288
|
+
case '%':
|
289
|
+
advance();
|
290
|
+
switch (current_char()) {
|
291
|
+
case '=':
|
292
|
+
advance();
|
293
|
+
return Token { Token::Type::PercentEqual, m_file, m_token_line, m_token_column };
|
294
|
+
case 'q':
|
295
|
+
switch (peek()) {
|
296
|
+
case '[':
|
297
|
+
advance(2);
|
298
|
+
return consume_single_quoted_string('[', ']');
|
299
|
+
case '{':
|
300
|
+
advance(2);
|
301
|
+
return consume_single_quoted_string('{', '}');
|
302
|
+
case '<':
|
303
|
+
advance(2);
|
304
|
+
return consume_single_quoted_string('<', '>');
|
305
|
+
case '(':
|
306
|
+
advance(2);
|
307
|
+
return consume_single_quoted_string('(', ')');
|
308
|
+
default: {
|
309
|
+
char c = peek();
|
310
|
+
if (char_can_be_string_or_regexp_delimiter(c)) {
|
311
|
+
advance(2);
|
312
|
+
return consume_single_quoted_string(c, c);
|
313
|
+
} else {
|
314
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
315
|
+
}
|
316
|
+
}
|
317
|
+
}
|
318
|
+
case 'Q':
|
319
|
+
switch (peek()) {
|
320
|
+
case '[':
|
321
|
+
advance(2);
|
322
|
+
return consume_double_quoted_string('[', ']');
|
323
|
+
case '{':
|
324
|
+
advance(2);
|
325
|
+
return consume_double_quoted_string('{', '}');
|
326
|
+
case '<':
|
327
|
+
advance(2);
|
328
|
+
return consume_double_quoted_string('<', '>');
|
329
|
+
case '(':
|
330
|
+
advance(2);
|
331
|
+
return consume_double_quoted_string('(', ')');
|
332
|
+
default: {
|
333
|
+
char c = peek();
|
334
|
+
if (char_can_be_string_or_regexp_delimiter(c)) {
|
335
|
+
advance(2);
|
336
|
+
return consume_double_quoted_string(c, c);
|
337
|
+
} else {
|
338
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
339
|
+
}
|
340
|
+
}
|
341
|
+
}
|
342
|
+
case 'r':
|
343
|
+
switch (peek()) {
|
344
|
+
case '[':
|
345
|
+
advance(2);
|
346
|
+
return consume_regexp('[', ']');
|
347
|
+
case '{':
|
348
|
+
advance(2);
|
349
|
+
return consume_regexp('{', '}');
|
350
|
+
case '(':
|
351
|
+
advance(2);
|
352
|
+
return consume_regexp('(', ')');
|
353
|
+
case '<':
|
354
|
+
advance(2);
|
355
|
+
return consume_regexp('<', '>');
|
356
|
+
default: {
|
357
|
+
char c = peek();
|
358
|
+
if (char_can_be_string_or_regexp_delimiter(c)) {
|
359
|
+
advance(2);
|
360
|
+
return consume_regexp(c, c);
|
361
|
+
} else {
|
362
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
363
|
+
}
|
364
|
+
}
|
365
|
+
}
|
366
|
+
case 'x':
|
367
|
+
switch (peek()) {
|
368
|
+
case '/': {
|
369
|
+
advance(2);
|
370
|
+
return consume_double_quoted_string('/', '/', Token::Type::InterpolatedShellBegin, Token::Type::InterpolatedShellEnd);
|
371
|
+
}
|
372
|
+
case '[': {
|
373
|
+
advance(2);
|
374
|
+
return consume_double_quoted_string('[', ']', Token::Type::InterpolatedShellBegin, Token::Type::InterpolatedShellEnd);
|
375
|
+
}
|
376
|
+
case '{': {
|
377
|
+
advance(2);
|
378
|
+
return consume_double_quoted_string('{', '}', Token::Type::InterpolatedShellBegin, Token::Type::InterpolatedShellEnd);
|
379
|
+
}
|
380
|
+
case '(': {
|
381
|
+
advance(2);
|
382
|
+
return consume_double_quoted_string('(', ')', Token::Type::InterpolatedShellBegin, Token::Type::InterpolatedShellEnd);
|
383
|
+
}
|
384
|
+
default:
|
385
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
386
|
+
}
|
387
|
+
case 'w':
|
388
|
+
switch (peek()) {
|
389
|
+
case '/':
|
390
|
+
case '|': {
|
391
|
+
char c = next();
|
392
|
+
advance();
|
393
|
+
return consume_quoted_array_without_interpolation(c, c, Token::Type::PercentLowerW);
|
394
|
+
}
|
395
|
+
case '[':
|
396
|
+
advance(2);
|
397
|
+
return consume_quoted_array_without_interpolation('[', ']', Token::Type::PercentLowerW);
|
398
|
+
case '{':
|
399
|
+
advance(2);
|
400
|
+
return consume_quoted_array_without_interpolation('{', '}', Token::Type::PercentLowerW);
|
401
|
+
case '<':
|
402
|
+
advance(2);
|
403
|
+
return consume_quoted_array_without_interpolation('<', '>', Token::Type::PercentLowerW);
|
404
|
+
case '(':
|
405
|
+
advance(2);
|
406
|
+
return consume_quoted_array_without_interpolation('(', ')', Token::Type::PercentLowerW);
|
407
|
+
default:
|
408
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
409
|
+
}
|
410
|
+
case 'W':
|
411
|
+
switch (peek()) {
|
412
|
+
case '/':
|
413
|
+
case '|': {
|
414
|
+
char c = next();
|
415
|
+
advance();
|
416
|
+
return consume_quoted_array_with_interpolation(0, c, Token::Type::PercentUpperW);
|
417
|
+
}
|
418
|
+
case '[':
|
419
|
+
advance(2);
|
420
|
+
return consume_quoted_array_with_interpolation('[', ']', Token::Type::PercentUpperW);
|
421
|
+
case '{':
|
422
|
+
advance(2);
|
423
|
+
return consume_quoted_array_with_interpolation('{', '}', Token::Type::PercentUpperW);
|
424
|
+
case '<':
|
425
|
+
advance(2);
|
426
|
+
return consume_quoted_array_with_interpolation('<', '>', Token::Type::PercentUpperW);
|
427
|
+
case '(':
|
428
|
+
advance(2);
|
429
|
+
return consume_quoted_array_with_interpolation('(', ')', Token::Type::PercentUpperW);
|
430
|
+
default:
|
431
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
432
|
+
}
|
433
|
+
case 'i':
|
434
|
+
switch (peek()) {
|
435
|
+
case '|':
|
436
|
+
case '/': {
|
437
|
+
char c = next();
|
438
|
+
advance();
|
439
|
+
return consume_quoted_array_without_interpolation(c, c, Token::Type::PercentLowerI);
|
440
|
+
}
|
441
|
+
case '[':
|
442
|
+
advance(2);
|
443
|
+
return consume_quoted_array_without_interpolation('[', ']', Token::Type::PercentLowerI);
|
444
|
+
case '{':
|
445
|
+
advance(2);
|
446
|
+
return consume_quoted_array_without_interpolation('{', '}', Token::Type::PercentLowerI);
|
447
|
+
case '<':
|
448
|
+
advance(2);
|
449
|
+
return consume_quoted_array_without_interpolation('<', '>', Token::Type::PercentLowerI);
|
450
|
+
case '(':
|
451
|
+
advance(2);
|
452
|
+
return consume_quoted_array_without_interpolation('(', ')', Token::Type::PercentLowerI);
|
453
|
+
default:
|
454
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
455
|
+
}
|
456
|
+
case 'I':
|
457
|
+
switch (peek()) {
|
458
|
+
case '|':
|
459
|
+
case '/': {
|
460
|
+
char c = next();
|
461
|
+
advance();
|
462
|
+
return consume_quoted_array_with_interpolation(0, c, Token::Type::PercentUpperI);
|
463
|
+
}
|
464
|
+
case '[':
|
465
|
+
advance(2);
|
466
|
+
return consume_quoted_array_with_interpolation('[', ']', Token::Type::PercentUpperI);
|
467
|
+
case '{':
|
468
|
+
advance(2);
|
469
|
+
return consume_quoted_array_with_interpolation('{', '}', Token::Type::PercentUpperI);
|
470
|
+
case '<':
|
471
|
+
advance(2);
|
472
|
+
return consume_quoted_array_with_interpolation('<', '>', Token::Type::PercentUpperI);
|
473
|
+
case '(':
|
474
|
+
advance(2);
|
475
|
+
return consume_quoted_array_with_interpolation('(', ')', Token::Type::PercentUpperI);
|
476
|
+
default:
|
477
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
478
|
+
}
|
479
|
+
case '[':
|
480
|
+
advance();
|
481
|
+
return consume_double_quoted_string('[', ']');
|
482
|
+
case '{':
|
483
|
+
advance();
|
484
|
+
return consume_double_quoted_string('{', '}');
|
485
|
+
case '<':
|
486
|
+
advance();
|
487
|
+
return consume_double_quoted_string('<', '>');
|
488
|
+
case '(':
|
489
|
+
if (m_last_token.type() == Token::Type::DefKeyword || m_last_token.type() == Token::Type::Dot) {
|
490
|
+
// It's a trap! This looks like a %(string) but it's a method def/call!
|
491
|
+
break;
|
492
|
+
}
|
493
|
+
advance();
|
494
|
+
return consume_double_quoted_string('(', ')');
|
495
|
+
default: {
|
496
|
+
auto c = current_char();
|
497
|
+
if (char_can_be_string_or_regexp_delimiter(c)) {
|
498
|
+
advance();
|
499
|
+
return consume_double_quoted_string(c, c);
|
500
|
+
}
|
501
|
+
break;
|
502
|
+
}
|
503
|
+
}
|
504
|
+
return Token { Token::Type::Percent, m_file, m_token_line, m_token_column };
|
505
|
+
case '!':
|
506
|
+
advance();
|
507
|
+
switch (current_char()) {
|
508
|
+
case '=':
|
509
|
+
advance();
|
510
|
+
return Token { Token::Type::NotEqual, m_file, m_token_line, m_token_column };
|
511
|
+
case '~':
|
512
|
+
advance();
|
513
|
+
return Token { Token::Type::NotMatch, m_file, m_token_line, m_token_column };
|
514
|
+
case '@':
|
515
|
+
if (m_last_token.is_def_keyword() || m_last_token.is_dot()) {
|
516
|
+
advance();
|
517
|
+
SharedPtr<String> lit = new String("!@");
|
518
|
+
return Token { Token::Type::BareName, lit, m_file, m_token_line, m_token_column };
|
519
|
+
} else {
|
520
|
+
return Token { Token::Type::Not, m_file, m_token_line, m_token_column };
|
521
|
+
}
|
522
|
+
default:
|
523
|
+
return Token { Token::Type::Not, m_file, m_token_line, m_token_column };
|
524
|
+
}
|
525
|
+
case '<':
|
526
|
+
advance();
|
527
|
+
switch (current_char()) {
|
528
|
+
case '<': {
|
529
|
+
advance();
|
530
|
+
switch (current_char()) {
|
531
|
+
case '~':
|
532
|
+
case '-': {
|
533
|
+
auto next = peek();
|
534
|
+
if (isalpha(next))
|
535
|
+
return consume_heredoc();
|
536
|
+
switch (next) {
|
537
|
+
case '_':
|
538
|
+
case '"':
|
539
|
+
case '`':
|
540
|
+
case '\'':
|
541
|
+
return consume_heredoc();
|
542
|
+
default:
|
543
|
+
return Token { Token::Type::LeftShift, m_file, m_token_line, m_token_column };
|
544
|
+
}
|
545
|
+
}
|
546
|
+
case '=':
|
547
|
+
advance();
|
548
|
+
return Token { Token::Type::LeftShiftEqual, m_file, m_token_line, m_token_column };
|
549
|
+
default:
|
550
|
+
if (!m_whitespace_precedes) {
|
551
|
+
if (token_is_first_on_line())
|
552
|
+
return consume_heredoc();
|
553
|
+
else if (m_last_token.can_precede_heredoc_that_looks_like_left_shift_operator())
|
554
|
+
return consume_heredoc();
|
555
|
+
else
|
556
|
+
return Token { Token::Type::LeftShift, m_file, m_token_line, m_token_column };
|
557
|
+
}
|
558
|
+
if (isalpha(current_char()))
|
559
|
+
return consume_heredoc();
|
560
|
+
switch (current_char()) {
|
561
|
+
case '_':
|
562
|
+
case '"':
|
563
|
+
case '`':
|
564
|
+
case '\'':
|
565
|
+
return consume_heredoc();
|
566
|
+
default:
|
567
|
+
return Token { Token::Type::LeftShift, m_file, m_token_line, m_token_column };
|
568
|
+
}
|
569
|
+
}
|
570
|
+
}
|
571
|
+
case '=':
|
572
|
+
advance();
|
573
|
+
switch (current_char()) {
|
574
|
+
case '>':
|
575
|
+
advance();
|
576
|
+
return Token { Token::Type::Comparison, m_file, m_token_line, m_token_column };
|
577
|
+
default:
|
578
|
+
return Token { Token::Type::LessThanOrEqual, m_file, m_token_line, m_token_column };
|
579
|
+
}
|
580
|
+
default:
|
581
|
+
return Token { Token::Type::LessThan, m_file, m_token_line, m_token_column };
|
582
|
+
}
|
583
|
+
case '>':
|
584
|
+
advance();
|
585
|
+
switch (current_char()) {
|
586
|
+
case '>':
|
587
|
+
advance();
|
588
|
+
switch (current_char()) {
|
589
|
+
case '=':
|
590
|
+
advance();
|
591
|
+
return Token { Token::Type::RightShiftEqual, m_file, m_token_line, m_token_column };
|
592
|
+
default:
|
593
|
+
return Token { Token::Type::RightShift, m_file, m_token_line, m_token_column };
|
594
|
+
}
|
595
|
+
case '=':
|
596
|
+
advance();
|
597
|
+
return Token { Token::Type::GreaterThanOrEqual, m_file, m_token_line, m_token_column };
|
598
|
+
default:
|
599
|
+
return Token { Token::Type::GreaterThan, m_file, m_token_line, m_token_column };
|
600
|
+
}
|
601
|
+
case '&':
|
602
|
+
advance();
|
603
|
+
switch (current_char()) {
|
604
|
+
case '&':
|
605
|
+
advance();
|
606
|
+
switch (current_char()) {
|
607
|
+
case '=':
|
608
|
+
advance();
|
609
|
+
return Token { Token::Type::AmpersandAmpersandEqual, m_file, m_token_line, m_token_column };
|
610
|
+
default:
|
611
|
+
return Token { Token::Type::AmpersandAmpersand, m_file, m_token_line, m_token_column };
|
612
|
+
}
|
613
|
+
case '=':
|
614
|
+
advance();
|
615
|
+
return Token { Token::Type::AmpersandEqual, m_file, m_token_line, m_token_column };
|
616
|
+
case '.':
|
617
|
+
advance();
|
618
|
+
return Token { Token::Type::SafeNavigation, m_file, m_token_line, m_token_column };
|
619
|
+
default:
|
620
|
+
return Token { Token::Type::Ampersand, m_file, m_token_line, m_token_column };
|
621
|
+
}
|
622
|
+
case '|':
|
623
|
+
advance();
|
624
|
+
switch (current_char()) {
|
625
|
+
case '|':
|
626
|
+
advance();
|
627
|
+
switch (current_char()) {
|
628
|
+
case '=':
|
629
|
+
advance();
|
630
|
+
return Token { Token::Type::PipePipeEqual, m_file, m_token_line, m_token_column };
|
631
|
+
default:
|
632
|
+
return Token { Token::Type::PipePipe, m_file, m_token_line, m_token_column };
|
633
|
+
}
|
634
|
+
case '=':
|
635
|
+
advance();
|
636
|
+
return Token { Token::Type::PipeEqual, m_file, m_token_line, m_token_column };
|
637
|
+
default:
|
638
|
+
return Token { Token::Type::Pipe, m_file, m_token_line, m_token_column };
|
639
|
+
}
|
640
|
+
case '^':
|
641
|
+
advance();
|
642
|
+
switch (current_char()) {
|
643
|
+
case '=':
|
644
|
+
advance();
|
645
|
+
return Token { Token::Type::CaretEqual, m_file, m_token_line, m_token_column };
|
646
|
+
default:
|
647
|
+
return Token { Token::Type::Caret, m_file, m_token_line, m_token_column };
|
648
|
+
}
|
649
|
+
case '~':
|
650
|
+
advance();
|
651
|
+
switch (current_char()) {
|
652
|
+
case '@':
|
653
|
+
if (m_last_token.is_def_keyword() || m_last_token.is_dot()) {
|
654
|
+
advance();
|
655
|
+
SharedPtr<String> lit = new String("~@");
|
656
|
+
return Token { Token::Type::BareName, lit, m_file, m_token_line, m_token_column };
|
657
|
+
} else {
|
658
|
+
return Token { Token::Type::Tilde, m_file, m_token_line, m_token_column };
|
659
|
+
}
|
660
|
+
default:
|
661
|
+
return Token { Token::Type::Tilde, m_file, m_token_line, m_token_column };
|
662
|
+
}
|
663
|
+
case '?': {
|
664
|
+
auto c = next();
|
665
|
+
if (isspace(c)) {
|
666
|
+
m_open_ternary = true;
|
667
|
+
return Token { Token::Type::TernaryQuestion, m_file, m_token_line, m_token_column };
|
668
|
+
} else {
|
669
|
+
advance();
|
670
|
+
if (c == '\\') {
|
671
|
+
auto buf = new String();
|
672
|
+
auto result = consume_escaped_byte(*buf);
|
673
|
+
if (!result.first)
|
674
|
+
return Token { result.second, current_char(), m_file, m_token_line, m_token_column };
|
675
|
+
return Token { Token::Type::String, buf, m_file, m_token_line, m_token_column };
|
676
|
+
} else {
|
677
|
+
return Token { Token::Type::String, c, m_file, m_token_line, m_token_column };
|
678
|
+
}
|
679
|
+
}
|
680
|
+
}
|
681
|
+
case ':': {
|
682
|
+
auto c = next();
|
683
|
+
if (c == ':') {
|
684
|
+
advance();
|
685
|
+
return Token { Token::Type::ConstantResolution, m_file, m_token_line, m_token_column };
|
686
|
+
} else if (m_last_token.type() == Token::Type::InterpolatedStringEnd && !m_whitespace_precedes && !m_open_ternary) {
|
687
|
+
return Token { Token::Type::InterpolatedStringSymbolKey, m_file, m_token_line, m_token_column };
|
688
|
+
} else if (c == '"') {
|
689
|
+
advance();
|
690
|
+
return consume_double_quoted_string('"', '"', Token::Type::InterpolatedSymbolBegin, Token::Type::InterpolatedSymbolEnd);
|
691
|
+
} else if (c == '\'') {
|
692
|
+
advance();
|
693
|
+
auto string = consume_single_quoted_string('\'', '\'');
|
694
|
+
return Token { Token::Type::Symbol, string.literal(), m_file, m_token_line, m_token_column };
|
695
|
+
} else if (isspace(c)) {
|
696
|
+
m_open_ternary = false;
|
697
|
+
auto token = Token { Token::Type::TernaryColon, m_file, m_token_line, m_token_column };
|
698
|
+
token.set_whitespace_precedes(m_whitespace_precedes);
|
699
|
+
return token;
|
700
|
+
} else {
|
701
|
+
return consume_symbol();
|
702
|
+
}
|
703
|
+
}
|
704
|
+
case '@':
|
705
|
+
switch (peek()) {
|
706
|
+
case '@': {
|
707
|
+
// kinda janky, but we gotta trick consume_word and then prepend the '@' back on the front
|
708
|
+
advance();
|
709
|
+
auto token = consume_word(Token::Type::ClassVariable);
|
710
|
+
token.set_literal(String::format("@{}", token.literal()));
|
711
|
+
return token;
|
712
|
+
}
|
713
|
+
default:
|
714
|
+
return consume_word(Token::Type::InstanceVariable);
|
715
|
+
}
|
716
|
+
case '$':
|
717
|
+
if (peek() == '&') {
|
718
|
+
advance(2);
|
719
|
+
return Token { Token::Type::BackRef, '&', m_file, m_token_line, m_token_column };
|
720
|
+
} else if (peek() >= '1' && peek() <= '9') {
|
721
|
+
return consume_nth_ref();
|
722
|
+
} else {
|
723
|
+
return consume_global_variable();
|
724
|
+
}
|
725
|
+
case '.':
|
726
|
+
advance();
|
727
|
+
switch (current_char()) {
|
728
|
+
case '.':
|
729
|
+
advance();
|
730
|
+
switch (current_char()) {
|
731
|
+
case '.':
|
732
|
+
advance();
|
733
|
+
return Token { Token::Type::DotDotDot, m_file, m_token_line, m_token_column };
|
734
|
+
default:
|
735
|
+
return Token { Token::Type::DotDot, m_file, m_token_line, m_token_column };
|
736
|
+
}
|
737
|
+
default:
|
738
|
+
return Token { Token::Type::Dot, m_file, m_token_line, m_token_column };
|
739
|
+
}
|
740
|
+
case '{':
|
741
|
+
advance();
|
742
|
+
return Token { Token::Type::LCurlyBrace, m_file, m_token_line, m_token_column };
|
743
|
+
case '[': {
|
744
|
+
advance();
|
745
|
+
switch (current_char()) {
|
746
|
+
case ']':
|
747
|
+
advance();
|
748
|
+
switch (current_char()) {
|
749
|
+
case '=':
|
750
|
+
advance();
|
751
|
+
return Token { Token::Type::LBracketRBracketEqual, m_file, m_token_line, m_token_column };
|
752
|
+
default:
|
753
|
+
auto token = Token { Token::Type::LBracketRBracket, m_file, m_token_line, m_token_column };
|
754
|
+
token.set_whitespace_precedes(m_whitespace_precedes);
|
755
|
+
return token;
|
756
|
+
}
|
757
|
+
default:
|
758
|
+
auto token = Token { Token::Type::LBracket, m_file, m_token_line, m_token_column };
|
759
|
+
token.set_whitespace_precedes(m_whitespace_precedes);
|
760
|
+
return token;
|
761
|
+
}
|
762
|
+
}
|
763
|
+
case '(': {
|
764
|
+
advance();
|
765
|
+
auto token = Token { Token::Type::LParen, m_file, m_token_line, m_token_column };
|
766
|
+
token.set_whitespace_precedes(m_whitespace_precedes);
|
767
|
+
return token;
|
768
|
+
}
|
769
|
+
case '}':
|
770
|
+
advance();
|
771
|
+
return Token { Token::Type::RCurlyBrace, m_file, m_token_line, m_token_column };
|
772
|
+
case ']':
|
773
|
+
advance();
|
774
|
+
return Token { Token::Type::RBracket, m_file, m_token_line, m_token_column };
|
775
|
+
case ')':
|
776
|
+
advance();
|
777
|
+
return Token { Token::Type::RParen, m_file, m_token_line, m_token_column };
|
778
|
+
case '\n': {
|
779
|
+
advance();
|
780
|
+
auto token = Token { Token::Type::Newline, m_file, m_token_line, m_token_column };
|
781
|
+
if (!m_heredoc_stack.is_empty()) {
|
782
|
+
auto new_index = m_heredoc_stack.last();
|
783
|
+
while (m_index < new_index)
|
784
|
+
advance();
|
785
|
+
m_heredoc_stack.clear();
|
786
|
+
}
|
787
|
+
return token;
|
788
|
+
}
|
789
|
+
case ';':
|
790
|
+
advance();
|
791
|
+
return Token { Token::Type::Semicolon, m_file, m_token_line, m_token_column };
|
792
|
+
case ',':
|
793
|
+
advance();
|
794
|
+
return Token { Token::Type::Comma, m_file, m_token_line, m_token_column };
|
795
|
+
case '"':
|
796
|
+
advance();
|
797
|
+
return consume_double_quoted_string('"', '"');
|
798
|
+
case '\'':
|
799
|
+
advance();
|
800
|
+
return consume_single_quoted_string('\'', '\'');
|
801
|
+
case '`': {
|
802
|
+
advance();
|
803
|
+
return consume_double_quoted_string('`', '`', Token::Type::InterpolatedShellBegin, Token::Type::InterpolatedShellEnd);
|
804
|
+
}
|
805
|
+
case '#':
|
806
|
+
if (token_is_first_on_line()) {
|
807
|
+
SharedPtr<String> doc = new String();
|
808
|
+
bool found_comment_marker = true;
|
809
|
+
char c = current_char();
|
810
|
+
while (c) {
|
811
|
+
if (!found_comment_marker) {
|
812
|
+
if (c == '#')
|
813
|
+
found_comment_marker = true;
|
814
|
+
else if (!isspace(c))
|
815
|
+
break;
|
816
|
+
}
|
817
|
+
if (c == '\n' || c == '\r') {
|
818
|
+
doc->append_char(c);
|
819
|
+
found_comment_marker = false;
|
820
|
+
} else if (found_comment_marker)
|
821
|
+
doc->append_char(c);
|
822
|
+
c = next();
|
823
|
+
}
|
824
|
+
return Token { Token::Type::Doc, doc, m_file, m_token_line, m_token_column };
|
825
|
+
} else {
|
826
|
+
char c;
|
827
|
+
do {
|
828
|
+
c = next();
|
829
|
+
} while (c && c != '\n' && c != '\r');
|
830
|
+
return Token { Token::Type::Comment, m_file, m_token_line, m_token_column };
|
831
|
+
}
|
832
|
+
case '0':
|
833
|
+
case '1':
|
834
|
+
case '2':
|
835
|
+
case '3':
|
836
|
+
case '4':
|
837
|
+
case '5':
|
838
|
+
case '6':
|
839
|
+
case '7':
|
840
|
+
case '8':
|
841
|
+
case '9': {
|
842
|
+
auto token = consume_numeric();
|
843
|
+
return token;
|
844
|
+
}
|
845
|
+
};
|
846
|
+
|
847
|
+
Token keyword_token;
|
848
|
+
|
849
|
+
if (!m_last_token.is_dot() && match(4, "self")) {
|
850
|
+
if (current_char() == '.')
|
851
|
+
keyword_token = { Token::Type::SelfKeyword, m_file, m_token_line, m_token_column };
|
852
|
+
else
|
853
|
+
rewind(4);
|
854
|
+
}
|
855
|
+
|
856
|
+
if (!m_last_token.is_dot() && !m_last_token.is_def_keyword()) {
|
857
|
+
if (match(12, "__ENCODING__"))
|
858
|
+
keyword_token = { Token::Type::ENCODINGKeyword, m_file, m_token_line, m_token_column };
|
859
|
+
else if (match(8, "__LINE__"))
|
860
|
+
keyword_token = { Token::Type::LINEKeyword, m_file, m_token_line, m_token_column };
|
861
|
+
else if (match(8, "__FILE__"))
|
862
|
+
keyword_token = { Token::Type::FILEKeyword, m_file, m_token_line, m_token_column };
|
863
|
+
else if (match(5, "BEGIN"))
|
864
|
+
keyword_token = { Token::Type::BEGINKeyword, m_file, m_token_line, m_token_column };
|
865
|
+
else if (match(3, "END"))
|
866
|
+
keyword_token = { Token::Type::ENDKeyword, m_file, m_token_line, m_token_column };
|
867
|
+
else if (match(5, "alias"))
|
868
|
+
keyword_token = { Token::Type::AliasKeyword, m_file, m_token_line, m_token_column };
|
869
|
+
else if (match(3, "and"))
|
870
|
+
keyword_token = { Token::Type::AndKeyword, m_file, m_token_line, m_token_column };
|
871
|
+
else if (match(5, "begin"))
|
872
|
+
keyword_token = { Token::Type::BeginKeyword, m_file, m_token_line, m_token_column };
|
873
|
+
else if (match(5, "break"))
|
874
|
+
keyword_token = { Token::Type::BreakKeyword, m_file, m_token_line, m_token_column };
|
875
|
+
else if (match(4, "case"))
|
876
|
+
keyword_token = { Token::Type::CaseKeyword, m_file, m_token_line, m_token_column };
|
877
|
+
else if (match(5, "class"))
|
878
|
+
keyword_token = { Token::Type::ClassKeyword, m_file, m_token_line, m_token_column };
|
879
|
+
else if (match(8, "defined?"))
|
880
|
+
keyword_token = { Token::Type::DefinedKeyword, m_file, m_token_line, m_token_column };
|
881
|
+
else if (match(3, "def"))
|
882
|
+
keyword_token = { Token::Type::DefKeyword, m_file, m_token_line, m_token_column };
|
883
|
+
else if (match(2, "do"))
|
884
|
+
keyword_token = { Token::Type::DoKeyword, m_file, m_token_line, m_token_column };
|
885
|
+
else if (match(4, "else"))
|
886
|
+
keyword_token = { Token::Type::ElseKeyword, m_file, m_token_line, m_token_column };
|
887
|
+
else if (match(5, "elsif"))
|
888
|
+
keyword_token = { Token::Type::ElsifKeyword, m_file, m_token_line, m_token_column };
|
889
|
+
else if (match(3, "end"))
|
890
|
+
keyword_token = { Token::Type::EndKeyword, m_file, m_token_line, m_token_column };
|
891
|
+
else if (match(6, "ensure"))
|
892
|
+
keyword_token = { Token::Type::EnsureKeyword, m_file, m_token_line, m_token_column };
|
893
|
+
else if (match(5, "false"))
|
894
|
+
keyword_token = { Token::Type::FalseKeyword, m_file, m_token_line, m_token_column };
|
895
|
+
else if (match(3, "for"))
|
896
|
+
keyword_token = { Token::Type::ForKeyword, m_file, m_token_line, m_token_column };
|
897
|
+
else if (match(2, "if"))
|
898
|
+
keyword_token = { Token::Type::IfKeyword, m_file, m_token_line, m_token_column };
|
899
|
+
else if (match(2, "in"))
|
900
|
+
keyword_token = { Token::Type::InKeyword, m_file, m_token_line, m_token_column };
|
901
|
+
else if (match(6, "module"))
|
902
|
+
keyword_token = { Token::Type::ModuleKeyword, m_file, m_token_line, m_token_column };
|
903
|
+
else if (match(4, "next"))
|
904
|
+
keyword_token = { Token::Type::NextKeyword, m_file, m_token_line, m_token_column };
|
905
|
+
else if (match(3, "nil"))
|
906
|
+
keyword_token = { Token::Type::NilKeyword, m_file, m_token_line, m_token_column };
|
907
|
+
else if (match(3, "not"))
|
908
|
+
keyword_token = { Token::Type::NotKeyword, m_file, m_token_line, m_token_column };
|
909
|
+
else if (match(2, "or"))
|
910
|
+
keyword_token = { Token::Type::OrKeyword, m_file, m_token_line, m_token_column };
|
911
|
+
else if (match(4, "redo"))
|
912
|
+
keyword_token = { Token::Type::RedoKeyword, m_file, m_token_line, m_token_column };
|
913
|
+
else if (match(6, "rescue"))
|
914
|
+
keyword_token = { Token::Type::RescueKeyword, m_file, m_token_line, m_token_column };
|
915
|
+
else if (match(5, "retry"))
|
916
|
+
keyword_token = { Token::Type::RetryKeyword, m_file, m_token_line, m_token_column };
|
917
|
+
else if (match(6, "return"))
|
918
|
+
keyword_token = { Token::Type::ReturnKeyword, m_file, m_token_line, m_token_column };
|
919
|
+
else if (match(4, "self"))
|
920
|
+
keyword_token = { Token::Type::SelfKeyword, m_file, m_token_line, m_token_column };
|
921
|
+
else if (match(5, "super"))
|
922
|
+
keyword_token = { Token::Type::SuperKeyword, m_file, m_token_line, m_token_column };
|
923
|
+
else if (match(4, "then"))
|
924
|
+
keyword_token = { Token::Type::ThenKeyword, m_file, m_token_line, m_token_column };
|
925
|
+
else if (match(4, "true"))
|
926
|
+
keyword_token = { Token::Type::TrueKeyword, m_file, m_token_line, m_token_column };
|
927
|
+
else if (match(5, "undef"))
|
928
|
+
keyword_token = { Token::Type::UndefKeyword, m_file, m_token_line, m_token_column };
|
929
|
+
else if (match(6, "unless"))
|
930
|
+
keyword_token = { Token::Type::UnlessKeyword, m_file, m_token_line, m_token_column };
|
931
|
+
else if (match(5, "until"))
|
932
|
+
keyword_token = { Token::Type::UntilKeyword, m_file, m_token_line, m_token_column };
|
933
|
+
else if (match(4, "when"))
|
934
|
+
keyword_token = { Token::Type::WhenKeyword, m_file, m_token_line, m_token_column };
|
935
|
+
else if (match(5, "while"))
|
936
|
+
keyword_token = { Token::Type::WhileKeyword, m_file, m_token_line, m_token_column };
|
937
|
+
else if (match(5, "yield"))
|
938
|
+
keyword_token = { Token::Type::YieldKeyword, m_file, m_token_line, m_token_column };
|
939
|
+
}
|
940
|
+
|
941
|
+
// if a colon comes next, it's not a keyword -- it's a symbol!
|
942
|
+
if (keyword_token && current_char() == ':' && peek() != ':' && !m_open_ternary) {
|
943
|
+
advance(); // :
|
944
|
+
auto name = keyword_token.type_value();
|
945
|
+
return Token { Token::Type::SymbolKey, name, m_file, m_token_line, m_token_column };
|
946
|
+
} else if (keyword_token) {
|
947
|
+
return keyword_token;
|
948
|
+
}
|
949
|
+
|
950
|
+
auto c = current_char();
|
951
|
+
if ((c >= 'a' && c <= 'z') || c == '_') {
|
952
|
+
return consume_bare_name();
|
953
|
+
} else if (c >= 'A' && c <= 'Z') {
|
954
|
+
return consume_constant();
|
955
|
+
} else {
|
956
|
+
auto buf = consume_non_whitespace();
|
957
|
+
auto token = Token { Token::Type::Invalid, buf, m_file, m_token_line, m_token_column };
|
958
|
+
return token;
|
959
|
+
}
|
960
|
+
|
961
|
+
TM_UNREACHABLE();
|
962
|
+
}
|
963
|
+
|
964
|
+
Token Lexer::consume_symbol() {
|
965
|
+
char c = current_char();
|
966
|
+
SharedPtr<String> buf = new String("");
|
967
|
+
auto gobble = [&buf, this](char c) -> char { buf->append_char(c); return next(); };
|
968
|
+
switch (c) {
|
969
|
+
case '@':
|
970
|
+
c = gobble(c);
|
971
|
+
if (c == '@') c = gobble(c);
|
972
|
+
do {
|
973
|
+
c = gobble(c);
|
974
|
+
} while (is_identifier_char(c));
|
975
|
+
break;
|
976
|
+
case '$':
|
977
|
+
c = gobble(c);
|
978
|
+
do {
|
979
|
+
c = gobble(c);
|
980
|
+
} while (is_identifier_char(c));
|
981
|
+
break;
|
982
|
+
case '~':
|
983
|
+
c = gobble(c);
|
984
|
+
if (c == '@') advance();
|
985
|
+
break;
|
986
|
+
case '+':
|
987
|
+
case '-': {
|
988
|
+
c = gobble(c);
|
989
|
+
if (c == '@') gobble(c);
|
990
|
+
break;
|
991
|
+
}
|
992
|
+
case '&':
|
993
|
+
case '|':
|
994
|
+
case '^':
|
995
|
+
case '%':
|
996
|
+
case '/': {
|
997
|
+
gobble(c);
|
998
|
+
break;
|
999
|
+
}
|
1000
|
+
case '*':
|
1001
|
+
c = gobble(c);
|
1002
|
+
if (c == '*')
|
1003
|
+
gobble(c);
|
1004
|
+
break;
|
1005
|
+
case '=':
|
1006
|
+
switch (peek()) {
|
1007
|
+
case '=':
|
1008
|
+
c = gobble(c);
|
1009
|
+
c = gobble(c);
|
1010
|
+
if (c == '=') gobble(c);
|
1011
|
+
break;
|
1012
|
+
case '~':
|
1013
|
+
c = gobble(c);
|
1014
|
+
gobble(c);
|
1015
|
+
break;
|
1016
|
+
default:
|
1017
|
+
return Token { Token::Type::Invalid, c, m_file, m_token_line, m_token_column };
|
1018
|
+
}
|
1019
|
+
break;
|
1020
|
+
case '!':
|
1021
|
+
c = gobble(c);
|
1022
|
+
switch (c) {
|
1023
|
+
case '=':
|
1024
|
+
case '~':
|
1025
|
+
case '@':
|
1026
|
+
gobble(c);
|
1027
|
+
default:
|
1028
|
+
break;
|
1029
|
+
}
|
1030
|
+
break;
|
1031
|
+
case '>':
|
1032
|
+
c = gobble(c);
|
1033
|
+
switch (c) {
|
1034
|
+
case '=':
|
1035
|
+
case '>':
|
1036
|
+
gobble(c);
|
1037
|
+
default:
|
1038
|
+
break;
|
1039
|
+
}
|
1040
|
+
break;
|
1041
|
+
case '<':
|
1042
|
+
c = gobble(c);
|
1043
|
+
switch (c) {
|
1044
|
+
case '=':
|
1045
|
+
c = gobble(c);
|
1046
|
+
if (c == '>') gobble(c);
|
1047
|
+
break;
|
1048
|
+
case '<':
|
1049
|
+
gobble(c);
|
1050
|
+
default:
|
1051
|
+
break;
|
1052
|
+
}
|
1053
|
+
break;
|
1054
|
+
case '[':
|
1055
|
+
if (peek() == ']') {
|
1056
|
+
c = gobble(c);
|
1057
|
+
c = gobble(c);
|
1058
|
+
if (c == '=') gobble(c);
|
1059
|
+
} else {
|
1060
|
+
return Token { Token::Type::Invalid, c, m_file, m_token_line, m_token_column };
|
1061
|
+
}
|
1062
|
+
break;
|
1063
|
+
default:
|
1064
|
+
do {
|
1065
|
+
c = gobble(c);
|
1066
|
+
} while (is_identifier_char(c));
|
1067
|
+
switch (c) {
|
1068
|
+
case '?':
|
1069
|
+
case '!':
|
1070
|
+
case '=':
|
1071
|
+
switch (peek()) {
|
1072
|
+
case '>':
|
1073
|
+
break;
|
1074
|
+
default:
|
1075
|
+
gobble(c);
|
1076
|
+
}
|
1077
|
+
default:
|
1078
|
+
break;
|
1079
|
+
}
|
1080
|
+
}
|
1081
|
+
return Token { Token::Type::Symbol, buf, m_file, m_token_line, m_token_column };
|
1082
|
+
}
|
1083
|
+
|
1084
|
+
Token Lexer::consume_word(Token::Type type) {
|
1085
|
+
char c = current_char();
|
1086
|
+
SharedPtr<String> buf = new String("");
|
1087
|
+
do {
|
1088
|
+
buf->append_char(c);
|
1089
|
+
c = next();
|
1090
|
+
} while (is_identifier_char(c));
|
1091
|
+
switch (c) {
|
1092
|
+
case '?':
|
1093
|
+
case '!':
|
1094
|
+
advance();
|
1095
|
+
buf->append_char(c);
|
1096
|
+
break;
|
1097
|
+
default:
|
1098
|
+
break;
|
1099
|
+
}
|
1100
|
+
return Token { type, buf, m_file, m_token_line, m_token_column };
|
1101
|
+
}
|
1102
|
+
|
1103
|
+
Token Lexer::consume_bare_name() {
|
1104
|
+
auto token = consume_word(Token::Type::BareName);
|
1105
|
+
auto c = current_char();
|
1106
|
+
if (c == ':' && peek() != ':' && m_last_token.can_precede_symbol_key()) {
|
1107
|
+
advance();
|
1108
|
+
token.set_type(Token::Type::SymbolKey);
|
1109
|
+
}
|
1110
|
+
return token;
|
1111
|
+
}
|
1112
|
+
|
1113
|
+
Token Lexer::consume_constant() {
|
1114
|
+
auto token = consume_word(Token::Type::Constant);
|
1115
|
+
auto c = current_char();
|
1116
|
+
if (c == ':' && peek() != ':' && m_last_token.can_precede_symbol_key()) {
|
1117
|
+
advance();
|
1118
|
+
token.set_type(Token::Type::SymbolKey);
|
1119
|
+
}
|
1120
|
+
return token;
|
1121
|
+
}
|
1122
|
+
|
1123
|
+
Token Lexer::consume_global_variable() {
|
1124
|
+
switch (peek()) {
|
1125
|
+
case '?':
|
1126
|
+
case '!':
|
1127
|
+
case '=':
|
1128
|
+
case '@':
|
1129
|
+
case '&':
|
1130
|
+
case '`':
|
1131
|
+
case '\'':
|
1132
|
+
case '"':
|
1133
|
+
case '+':
|
1134
|
+
case '/':
|
1135
|
+
case '\\':
|
1136
|
+
case ';':
|
1137
|
+
case '<':
|
1138
|
+
case '>':
|
1139
|
+
case '$':
|
1140
|
+
case '*':
|
1141
|
+
case '.':
|
1142
|
+
case ',':
|
1143
|
+
case ':':
|
1144
|
+
case '_':
|
1145
|
+
case '~': {
|
1146
|
+
advance();
|
1147
|
+
SharedPtr<String> buf = new String("$");
|
1148
|
+
buf->append_char(current_char());
|
1149
|
+
advance();
|
1150
|
+
return Token { Token::Type::GlobalVariable, buf, m_file, m_token_line, m_token_column };
|
1151
|
+
}
|
1152
|
+
case '-': {
|
1153
|
+
SharedPtr<String> buf = new String("$-");
|
1154
|
+
advance(2);
|
1155
|
+
buf->append_char(current_char());
|
1156
|
+
advance();
|
1157
|
+
return Token { Token::Type::GlobalVariable, buf, m_file, m_token_line, m_token_column };
|
1158
|
+
}
|
1159
|
+
default: {
|
1160
|
+
return consume_word(Token::Type::GlobalVariable);
|
1161
|
+
}
|
1162
|
+
}
|
1163
|
+
}
|
1164
|
+
|
1165
|
+
bool is_valid_heredoc(bool with_dash, SharedPtr<String> doc, String heredoc_name) {
|
1166
|
+
if (!doc->ends_with(heredoc_name))
|
1167
|
+
return false;
|
1168
|
+
if (doc->length() - heredoc_name.length() == 0)
|
1169
|
+
return true;
|
1170
|
+
auto prefix = (*doc)[doc->length() - heredoc_name.length() - 1];
|
1171
|
+
return with_dash ? isspace(prefix) : prefix == '\n';
|
1172
|
+
}
|
1173
|
+
|
1174
|
+
size_t get_heredoc_indent(SharedPtr<String> doc) {
|
1175
|
+
if (doc->is_empty())
|
1176
|
+
return 0;
|
1177
|
+
size_t heredoc_indent = std::numeric_limits<size_t>::max();
|
1178
|
+
size_t line_indent = 0;
|
1179
|
+
bool maybe_blank_line = true;
|
1180
|
+
for (size_t i = 0; i < doc->length(); i++) {
|
1181
|
+
char c = (*doc)[i];
|
1182
|
+
if (c == '\n') {
|
1183
|
+
if (!maybe_blank_line && line_indent < heredoc_indent)
|
1184
|
+
heredoc_indent = line_indent;
|
1185
|
+
line_indent = 0;
|
1186
|
+
maybe_blank_line = true;
|
1187
|
+
} else if (isspace(c)) {
|
1188
|
+
if (maybe_blank_line)
|
1189
|
+
line_indent++;
|
1190
|
+
} else {
|
1191
|
+
maybe_blank_line = false;
|
1192
|
+
}
|
1193
|
+
}
|
1194
|
+
return heredoc_indent;
|
1195
|
+
}
|
1196
|
+
|
1197
|
+
void dedent_heredoc(SharedPtr<String> &doc) {
|
1198
|
+
size_t heredoc_indent = get_heredoc_indent(doc);
|
1199
|
+
if (heredoc_indent == 0)
|
1200
|
+
return;
|
1201
|
+
SharedPtr<String> new_doc = new String("");
|
1202
|
+
size_t line_begin = 0;
|
1203
|
+
for (size_t i = 0; i < doc->length(); i++) {
|
1204
|
+
char c = (*doc)[i];
|
1205
|
+
if (c == '\n') {
|
1206
|
+
line_begin += heredoc_indent;
|
1207
|
+
if (line_begin < i)
|
1208
|
+
new_doc->append(doc->substring(line_begin, i - line_begin));
|
1209
|
+
new_doc->append_char('\n');
|
1210
|
+
line_begin = i + 1;
|
1211
|
+
}
|
1212
|
+
}
|
1213
|
+
doc = new_doc;
|
1214
|
+
}
|
1215
|
+
|
1216
|
+
Token Lexer::consume_heredoc() {
|
1217
|
+
bool with_dash = false;
|
1218
|
+
bool should_dedent = false;
|
1219
|
+
switch (current_char()) {
|
1220
|
+
case '-':
|
1221
|
+
advance();
|
1222
|
+
with_dash = true;
|
1223
|
+
break;
|
1224
|
+
case '~':
|
1225
|
+
advance();
|
1226
|
+
with_dash = true;
|
1227
|
+
should_dedent = true;
|
1228
|
+
break;
|
1229
|
+
}
|
1230
|
+
|
1231
|
+
auto begin_type = Token::Type::InterpolatedStringBegin;
|
1232
|
+
auto end_type = Token::Type::InterpolatedStringEnd;
|
1233
|
+
bool should_interpolate = true;
|
1234
|
+
char delimiter = 0;
|
1235
|
+
String heredoc_name = "";
|
1236
|
+
switch (current_char()) {
|
1237
|
+
case '"':
|
1238
|
+
delimiter = '"';
|
1239
|
+
break;
|
1240
|
+
case '`':
|
1241
|
+
begin_type = Token::Type::InterpolatedShellBegin;
|
1242
|
+
end_type = Token::Type::InterpolatedShellEnd;
|
1243
|
+
delimiter = '`';
|
1244
|
+
break;
|
1245
|
+
case '\'':
|
1246
|
+
should_interpolate = false;
|
1247
|
+
delimiter = '\'';
|
1248
|
+
break;
|
1249
|
+
default:
|
1250
|
+
delimiter = 0;
|
1251
|
+
}
|
1252
|
+
|
1253
|
+
if (delimiter) {
|
1254
|
+
char c = next();
|
1255
|
+
while (c != delimiter) {
|
1256
|
+
switch (c) {
|
1257
|
+
case '\n':
|
1258
|
+
case '\r':
|
1259
|
+
case 0:
|
1260
|
+
return Token { Token::Type::UnterminatedString, "heredoc identifier", m_file, m_token_line, m_token_column };
|
1261
|
+
default:
|
1262
|
+
heredoc_name.append_char(c);
|
1263
|
+
c = next();
|
1264
|
+
}
|
1265
|
+
}
|
1266
|
+
advance();
|
1267
|
+
} else {
|
1268
|
+
heredoc_name = String(consume_word(Token::Type::BareName).literal());
|
1269
|
+
}
|
1270
|
+
|
1271
|
+
SharedPtr<String> doc = new String("");
|
1272
|
+
size_t heredoc_index = m_index;
|
1273
|
+
auto get_char = [&heredoc_index, this]() { return (heredoc_index >= m_size) ? 0 : m_input->at(heredoc_index); };
|
1274
|
+
|
1275
|
+
if (m_heredoc_stack.is_empty()) {
|
1276
|
+
// start consuming the heredoc on the next line
|
1277
|
+
while (get_char() != '\n') {
|
1278
|
+
if (heredoc_index >= m_size)
|
1279
|
+
return Token { Token::Type::UnterminatedString, "heredoc", m_file, m_token_line, m_token_column };
|
1280
|
+
heredoc_index++;
|
1281
|
+
}
|
1282
|
+
heredoc_index++;
|
1283
|
+
} else {
|
1284
|
+
// start consuming the heredoc right after the last one
|
1285
|
+
heredoc_index = m_heredoc_stack.last();
|
1286
|
+
}
|
1287
|
+
|
1288
|
+
// consume the heredoc until we find the delimiter, either '\n' (if << was used) or any whitespace (if <<- was used) followed by "DELIM\n"
|
1289
|
+
for (;;) {
|
1290
|
+
if (heredoc_index >= m_size) {
|
1291
|
+
if (is_valid_heredoc(with_dash, doc, heredoc_name))
|
1292
|
+
break;
|
1293
|
+
return Token { Token::Type::UnterminatedString, doc, m_file, m_token_line, m_token_column };
|
1294
|
+
}
|
1295
|
+
char c = get_char();
|
1296
|
+
heredoc_index++;
|
1297
|
+
if (c == '\n' && is_valid_heredoc(with_dash, doc, heredoc_name))
|
1298
|
+
break;
|
1299
|
+
doc->append_char(c);
|
1300
|
+
}
|
1301
|
+
|
1302
|
+
// chop the delimiter and any trailing space off the string
|
1303
|
+
doc->truncate(doc->length() - heredoc_name.length());
|
1304
|
+
doc->strip_trailing_spaces();
|
1305
|
+
|
1306
|
+
if (should_dedent)
|
1307
|
+
dedent_heredoc(doc);
|
1308
|
+
|
1309
|
+
// We have to keep tokenizing on the line where the heredoc was started, and then jump to the line after the heredoc.
|
1310
|
+
// This index is used to jump to the end of the heredoc later.
|
1311
|
+
m_heredoc_stack.push(heredoc_index);
|
1312
|
+
|
1313
|
+
auto token = Token { Token::Type::String, doc, m_file, m_token_line, m_token_column };
|
1314
|
+
|
1315
|
+
if (should_interpolate) {
|
1316
|
+
m_nested_lexer = new InterpolatedStringLexer { *this, token, end_type };
|
1317
|
+
return Token { begin_type, m_file, m_token_line, m_token_column };
|
1318
|
+
}
|
1319
|
+
|
1320
|
+
return token;
|
1321
|
+
}
|
1322
|
+
|
1323
|
+
Token Lexer::consume_numeric() {
|
1324
|
+
SharedPtr<String> chars = new String;
|
1325
|
+
if (current_char() == '0') {
|
1326
|
+
switch (peek()) {
|
1327
|
+
case 'd':
|
1328
|
+
case 'D': {
|
1329
|
+
advance();
|
1330
|
+
char c = next();
|
1331
|
+
if (!isdigit(c))
|
1332
|
+
return Token { Token::Type::Invalid, c, m_file, m_cursor_line, m_cursor_column };
|
1333
|
+
do {
|
1334
|
+
chars->append_char(c);
|
1335
|
+
c = next();
|
1336
|
+
if (c == '_')
|
1337
|
+
c = next();
|
1338
|
+
} while (isdigit(c));
|
1339
|
+
return chars_to_fixnum_or_bignum_token(chars, 10, 0);
|
1340
|
+
}
|
1341
|
+
case 'o':
|
1342
|
+
case 'O': {
|
1343
|
+
chars->append_char('0');
|
1344
|
+
chars->append_char('o');
|
1345
|
+
advance();
|
1346
|
+
char c = next();
|
1347
|
+
if (!(c >= '0' && c <= '7'))
|
1348
|
+
return Token { Token::Type::Invalid, c, m_file, m_cursor_line, m_cursor_column };
|
1349
|
+
do {
|
1350
|
+
chars->append_char(c);
|
1351
|
+
c = next();
|
1352
|
+
if (c == '_')
|
1353
|
+
c = next();
|
1354
|
+
} while (c >= '0' && c <= '7');
|
1355
|
+
return chars_to_fixnum_or_bignum_token(chars, 8, 2);
|
1356
|
+
}
|
1357
|
+
case 'x':
|
1358
|
+
case 'X': {
|
1359
|
+
chars->append_char('0');
|
1360
|
+
chars->append_char('x');
|
1361
|
+
advance();
|
1362
|
+
char c = next();
|
1363
|
+
if (!isxdigit(c))
|
1364
|
+
return Token { Token::Type::Invalid, c, m_file, m_cursor_line, m_cursor_column };
|
1365
|
+
do {
|
1366
|
+
chars->append_char(c);
|
1367
|
+
c = next();
|
1368
|
+
if (c == '_')
|
1369
|
+
c = next();
|
1370
|
+
} while (isxdigit(c));
|
1371
|
+
return chars_to_fixnum_or_bignum_token(chars, 16, 2);
|
1372
|
+
}
|
1373
|
+
case 'b':
|
1374
|
+
case 'B': {
|
1375
|
+
chars->append_char('0');
|
1376
|
+
chars->append_char('b');
|
1377
|
+
advance();
|
1378
|
+
char c = next();
|
1379
|
+
if (c != '0' && c != '1')
|
1380
|
+
return Token { Token::Type::Invalid, c, m_file, m_cursor_line, m_cursor_column };
|
1381
|
+
do {
|
1382
|
+
chars->append_char(c);
|
1383
|
+
c = next();
|
1384
|
+
if (c == '_')
|
1385
|
+
c = next();
|
1386
|
+
} while (c == '0' || c == '1');
|
1387
|
+
return chars_to_fixnum_or_bignum_token(chars, 2, 2);
|
1388
|
+
}
|
1389
|
+
}
|
1390
|
+
}
|
1391
|
+
char c = current_char();
|
1392
|
+
do {
|
1393
|
+
chars->append_char(c);
|
1394
|
+
c = next();
|
1395
|
+
if (c == '_')
|
1396
|
+
c = next();
|
1397
|
+
} while (isdigit(c));
|
1398
|
+
if ((c == '.' && isdigit(peek())) || (c == 'e' || c == 'E'))
|
1399
|
+
return consume_numeric_as_float(chars);
|
1400
|
+
else
|
1401
|
+
return chars_to_fixnum_or_bignum_token(chars, 10, 0);
|
1402
|
+
}
|
1403
|
+
|
1404
|
+
const long long max_fixnum = std::numeric_limits<long long>::max() / 2; // 63 bits for MRI
|
1405
|
+
|
1406
|
+
Token Lexer::chars_to_fixnum_or_bignum_token(SharedPtr<String> chars, int base, int offset) {
|
1407
|
+
errno = 0;
|
1408
|
+
auto fixnum = strtoll(chars->c_str() + offset, nullptr, base);
|
1409
|
+
if (errno != 0 || fixnum > max_fixnum)
|
1410
|
+
return Token { Token::Type::Bignum, chars, m_file, m_token_line, m_token_column };
|
1411
|
+
else
|
1412
|
+
return Token { Token::Type::Fixnum, fixnum, m_file, m_token_line, m_token_column };
|
1413
|
+
}
|
1414
|
+
|
1415
|
+
Token Lexer::consume_numeric_as_float(SharedPtr<String> chars) {
|
1416
|
+
char c = current_char();
|
1417
|
+
if (c == '.') {
|
1418
|
+
chars->append_char('.');
|
1419
|
+
c = next();
|
1420
|
+
do {
|
1421
|
+
chars->append_char(c);
|
1422
|
+
c = next();
|
1423
|
+
if (c == '_')
|
1424
|
+
c = next();
|
1425
|
+
} while (isdigit(c));
|
1426
|
+
}
|
1427
|
+
if (c == 'e' || c == 'E') {
|
1428
|
+
chars->append_char('e');
|
1429
|
+
c = next();
|
1430
|
+
if (c == '-' || c == '+') {
|
1431
|
+
chars->append_char(c);
|
1432
|
+
c = next();
|
1433
|
+
}
|
1434
|
+
if (!isdigit(c))
|
1435
|
+
return Token { Token::Type::Invalid, c, m_file, m_cursor_line, m_cursor_column };
|
1436
|
+
do {
|
1437
|
+
chars->append_char(c);
|
1438
|
+
c = next();
|
1439
|
+
if (c == '_')
|
1440
|
+
c = next();
|
1441
|
+
} while (isdigit(c));
|
1442
|
+
}
|
1443
|
+
double dbl = atof(chars->c_str());
|
1444
|
+
return Token { Token::Type::Float, dbl, m_file, m_token_line, m_token_column };
|
1445
|
+
}
|
1446
|
+
|
1447
|
+
Token Lexer::consume_nth_ref() {
|
1448
|
+
char c = next();
|
1449
|
+
long long num = 0;
|
1450
|
+
do {
|
1451
|
+
num *= 10;
|
1452
|
+
num += c - '0';
|
1453
|
+
c = next();
|
1454
|
+
} while (isdigit(c));
|
1455
|
+
return Token { Token::Type::NthRef, num, m_file, m_token_line, m_token_column };
|
1456
|
+
}
|
1457
|
+
|
1458
|
+
long long Lexer::consume_hex_number(int max_length, bool allow_underscore) {
|
1459
|
+
char c = current_char();
|
1460
|
+
int length = 0;
|
1461
|
+
long long number = 0;
|
1462
|
+
do {
|
1463
|
+
number *= 16;
|
1464
|
+
if (c >= 'a' && c <= 'f')
|
1465
|
+
number += c - 'a' + 10;
|
1466
|
+
else if (c >= 'A' && c <= 'F')
|
1467
|
+
number += c - 'A' + 10;
|
1468
|
+
else
|
1469
|
+
number += c - '0';
|
1470
|
+
c = next();
|
1471
|
+
if (allow_underscore && c == '_')
|
1472
|
+
c = next();
|
1473
|
+
} while (isxdigit(c) && (max_length == 0 || ++length < max_length));
|
1474
|
+
return number;
|
1475
|
+
}
|
1476
|
+
|
1477
|
+
long long Lexer::consume_octal_number(int max_length, bool allow_underscore) {
|
1478
|
+
char c = current_char();
|
1479
|
+
int length = 0;
|
1480
|
+
long long number = 0;
|
1481
|
+
do {
|
1482
|
+
number *= 8;
|
1483
|
+
number += c - '0';
|
1484
|
+
c = next();
|
1485
|
+
if (allow_underscore && c == '_')
|
1486
|
+
c = next();
|
1487
|
+
} while (c >= '0' && c <= '7' && (max_length == 0 || ++length < max_length));
|
1488
|
+
return number;
|
1489
|
+
}
|
1490
|
+
|
1491
|
+
// public domain
|
1492
|
+
// https://gist.github.com/Miouyouyou/864130e8734afe3f806512b14022226f
|
1493
|
+
void Lexer::utf32_codepoint_to_utf8(String &buf, long long codepoint) {
|
1494
|
+
if (codepoint < 0x80) {
|
1495
|
+
buf.append_char(codepoint);
|
1496
|
+
} else if (codepoint < 0x800) { // 00000yyy yyxxxxxx
|
1497
|
+
buf.append_char(0b11000000 | (codepoint >> 6));
|
1498
|
+
buf.append_char(0b10000000 | (codepoint & 0x3f));
|
1499
|
+
} else if (codepoint < 0x10000) { // zzzzyyyy yyxxxxxx
|
1500
|
+
buf.append_char(0b11100000 | (codepoint >> 12));
|
1501
|
+
buf.append_char(0b10000000 | ((codepoint >> 6) & 0x3f));
|
1502
|
+
buf.append_char(0b10000000 | (codepoint & 0x3f));
|
1503
|
+
} else if (codepoint < 0x200000) { // 000uuuuu zzzzyyyy yyxxxxxx
|
1504
|
+
buf.append_char(0b11110000 | (codepoint >> 18));
|
1505
|
+
buf.append_char(0b10000000 | ((codepoint >> 12) & 0x3f));
|
1506
|
+
buf.append_char(0b10000000 | ((codepoint >> 6) & 0x3f));
|
1507
|
+
buf.append_char(0b10000000 | (codepoint & 0x3f));
|
1508
|
+
} else {
|
1509
|
+
TM_UNREACHABLE();
|
1510
|
+
}
|
1511
|
+
}
|
1512
|
+
|
1513
|
+
std::pair<bool, Token::Type> Lexer::consume_escaped_byte(String &buf) {
|
1514
|
+
auto control_character = [&](bool meta) {
|
1515
|
+
char c = next();
|
1516
|
+
if (c == '-')
|
1517
|
+
c = next();
|
1518
|
+
int num = 0;
|
1519
|
+
if (!meta && c == '\\' && peek() == 'M') {
|
1520
|
+
advance(); // M
|
1521
|
+
c = next();
|
1522
|
+
if (c != '-')
|
1523
|
+
return -1;
|
1524
|
+
meta = true;
|
1525
|
+
c = next();
|
1526
|
+
}
|
1527
|
+
if (c == '?')
|
1528
|
+
num = 127;
|
1529
|
+
else if (c >= ' ' && c <= '>')
|
1530
|
+
num = c - ' ';
|
1531
|
+
else if (c >= '@' && c <= '_')
|
1532
|
+
num = c - '@';
|
1533
|
+
else if (c >= '`' && c <= '~')
|
1534
|
+
num = c - '`';
|
1535
|
+
if (meta)
|
1536
|
+
return num + 128;
|
1537
|
+
else
|
1538
|
+
return num;
|
1539
|
+
};
|
1540
|
+
auto c = current_char();
|
1541
|
+
if (c >= '0' && c <= '7') {
|
1542
|
+
auto number = consume_octal_number(3);
|
1543
|
+
buf.append_char(number);
|
1544
|
+
} else if (c == 'x') {
|
1545
|
+
// hex: 1-2 digits
|
1546
|
+
advance();
|
1547
|
+
auto number = consume_hex_number(2);
|
1548
|
+
buf.append_char(number);
|
1549
|
+
} else if (c == 'u') {
|
1550
|
+
c = next();
|
1551
|
+
if (c == '{') {
|
1552
|
+
c = next();
|
1553
|
+
// unicode characters, space separated, 1-6 hex digits
|
1554
|
+
while (c != '}') {
|
1555
|
+
if (!isxdigit(c))
|
1556
|
+
return { false, Token::Type::InvalidUnicodeEscape };
|
1557
|
+
auto codepoint = consume_hex_number(6);
|
1558
|
+
utf32_codepoint_to_utf8(buf, codepoint);
|
1559
|
+
c = current_char();
|
1560
|
+
while (c == ' ')
|
1561
|
+
c = next();
|
1562
|
+
}
|
1563
|
+
if (c == '}')
|
1564
|
+
advance();
|
1565
|
+
} else {
|
1566
|
+
// unicode: 4 hex digits
|
1567
|
+
auto codepoint = consume_hex_number(4);
|
1568
|
+
utf32_codepoint_to_utf8(buf, codepoint);
|
1569
|
+
}
|
1570
|
+
} else {
|
1571
|
+
switch (c) {
|
1572
|
+
case 'a':
|
1573
|
+
buf.append_char('\a');
|
1574
|
+
break;
|
1575
|
+
case 'b':
|
1576
|
+
buf.append_char('\b');
|
1577
|
+
break;
|
1578
|
+
case 'c':
|
1579
|
+
case 'C': {
|
1580
|
+
int num = control_character(false);
|
1581
|
+
if (num == -1)
|
1582
|
+
return { false, Token::Type::InvalidCharacterEscape };
|
1583
|
+
buf.append_char((unsigned char)num);
|
1584
|
+
break;
|
1585
|
+
}
|
1586
|
+
case 'e':
|
1587
|
+
buf.append_char('\e');
|
1588
|
+
break;
|
1589
|
+
case 'f':
|
1590
|
+
buf.append_char('\f');
|
1591
|
+
break;
|
1592
|
+
case 'M': {
|
1593
|
+
c = next();
|
1594
|
+
if (c != '-')
|
1595
|
+
return { false, Token::Type::InvalidCharacterEscape };
|
1596
|
+
c = next();
|
1597
|
+
int num = 0;
|
1598
|
+
if (c == '\\' && (peek() == 'c' || peek() == 'C')) {
|
1599
|
+
advance();
|
1600
|
+
num = control_character(true);
|
1601
|
+
} else {
|
1602
|
+
num = (int)c + 128;
|
1603
|
+
}
|
1604
|
+
buf.append_char((unsigned char)num);
|
1605
|
+
break;
|
1606
|
+
}
|
1607
|
+
case 'n':
|
1608
|
+
buf.append_char('\n');
|
1609
|
+
break;
|
1610
|
+
case 'r':
|
1611
|
+
buf.append_char('\r');
|
1612
|
+
break;
|
1613
|
+
case 's':
|
1614
|
+
buf.append_char((unsigned char)32);
|
1615
|
+
break;
|
1616
|
+
case 't':
|
1617
|
+
buf.append_char('\t');
|
1618
|
+
break;
|
1619
|
+
case 'v':
|
1620
|
+
buf.append_char('\v');
|
1621
|
+
break;
|
1622
|
+
case '\n':
|
1623
|
+
break;
|
1624
|
+
default:
|
1625
|
+
buf.append_char(c);
|
1626
|
+
break;
|
1627
|
+
}
|
1628
|
+
advance();
|
1629
|
+
}
|
1630
|
+
return { true, Token::Type::String };
|
1631
|
+
}
|
1632
|
+
|
1633
|
+
bool Lexer::token_is_first_on_line() const {
|
1634
|
+
return !m_last_token || m_last_token.is_newline();
|
1635
|
+
}
|
1636
|
+
|
1637
|
+
Token Lexer::consume_double_quoted_string(char start_char, char stop_char, Token::Type begin_type, Token::Type end_type) {
|
1638
|
+
m_nested_lexer = new InterpolatedStringLexer { *this, start_char, stop_char, end_type };
|
1639
|
+
return Token { begin_type, start_char, m_file, m_token_line, m_token_column };
|
1640
|
+
}
|
1641
|
+
|
1642
|
+
Token Lexer::consume_single_quoted_string(char start_char, char stop_char) {
|
1643
|
+
int pair_depth = 0;
|
1644
|
+
SharedPtr<String> buf = new String("");
|
1645
|
+
char c = current_char();
|
1646
|
+
while (c) {
|
1647
|
+
if (c == '\\') {
|
1648
|
+
c = next();
|
1649
|
+
if (c == stop_char || c == '\\') {
|
1650
|
+
buf->append_char(c);
|
1651
|
+
} else {
|
1652
|
+
buf->append_char('\\');
|
1653
|
+
buf->append_char(c);
|
1654
|
+
}
|
1655
|
+
} else if (c == start_char && start_char != stop_char) {
|
1656
|
+
pair_depth++;
|
1657
|
+
buf->append_char(c);
|
1658
|
+
} else if (c == stop_char) {
|
1659
|
+
if (pair_depth > 0) {
|
1660
|
+
pair_depth--;
|
1661
|
+
buf->append_char(c);
|
1662
|
+
} else {
|
1663
|
+
advance(); // '
|
1664
|
+
if (current_char() == ':' && !m_open_ternary) {
|
1665
|
+
advance(); // :
|
1666
|
+
return Token { Token::Type::SymbolKey, buf, m_file, m_token_line, m_token_column };
|
1667
|
+
} else {
|
1668
|
+
return Token { Token::Type::String, buf, m_file, m_token_line, m_token_column };
|
1669
|
+
}
|
1670
|
+
}
|
1671
|
+
} else {
|
1672
|
+
buf->append_char(c);
|
1673
|
+
}
|
1674
|
+
c = next();
|
1675
|
+
}
|
1676
|
+
return Token { Token::Type::UnterminatedString, start_char, m_file, m_token_line, m_token_column };
|
1677
|
+
}
|
1678
|
+
|
1679
|
+
Token Lexer::consume_quoted_array_without_interpolation(char start_char, char stop_char, Token::Type type) {
|
1680
|
+
m_nested_lexer = new WordArrayLexer { *this, start_char, stop_char, false };
|
1681
|
+
return Token { type, start_char, m_file, m_token_line, m_token_column };
|
1682
|
+
}
|
1683
|
+
|
1684
|
+
Token Lexer::consume_quoted_array_with_interpolation(char start_char, char stop_char, Token::Type type) {
|
1685
|
+
m_nested_lexer = new WordArrayLexer { *this, start_char, stop_char, true };
|
1686
|
+
return Token { type, start_char, m_file, m_token_line, m_token_column };
|
1687
|
+
}
|
1688
|
+
|
1689
|
+
Token Lexer::consume_regexp(char start_char, char stop_char) {
|
1690
|
+
m_nested_lexer = new RegexpLexer { *this, start_char, stop_char };
|
1691
|
+
return Token { Token::Type::InterpolatedRegexpBegin, start_char, m_file, m_token_line, m_token_column };
|
1692
|
+
}
|
1693
|
+
|
1694
|
+
SharedPtr<String> Lexer::consume_non_whitespace() {
|
1695
|
+
char c = current_char();
|
1696
|
+
SharedPtr<String> buf = new String("");
|
1697
|
+
do {
|
1698
|
+
buf->append_char(c);
|
1699
|
+
c = next();
|
1700
|
+
} while (c && c != ' ' && c != '\t' && c != '\n' && c != '\r');
|
1701
|
+
return buf;
|
1702
|
+
}
|
1703
|
+
};
|