yarp 0.6.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/CODE_OF_CONDUCT.md +76 -0
- data/CONTRIBUTING.md +51 -0
- data/LICENSE.md +7 -0
- data/Makefile.in +79 -0
- data/README.md +86 -0
- data/config.h.in +25 -0
- data/config.yml +2147 -0
- data/configure +4487 -0
- data/docs/build_system.md +85 -0
- data/docs/building.md +26 -0
- data/docs/configuration.md +56 -0
- data/docs/design.md +53 -0
- data/docs/encoding.md +116 -0
- data/docs/extension.md +20 -0
- data/docs/fuzzing.md +93 -0
- data/docs/heredocs.md +36 -0
- data/docs/mapping.md +117 -0
- data/docs/ripper.md +36 -0
- data/docs/serialization.md +130 -0
- data/docs/testing.md +55 -0
- data/ext/yarp/api_node.c +3680 -0
- data/ext/yarp/api_pack.c +256 -0
- data/ext/yarp/extconf.rb +131 -0
- data/ext/yarp/extension.c +547 -0
- data/ext/yarp/extension.h +18 -0
- data/include/yarp/ast.h +1412 -0
- data/include/yarp/defines.h +54 -0
- data/include/yarp/diagnostic.h +24 -0
- data/include/yarp/enc/yp_encoding.h +94 -0
- data/include/yarp/node.h +36 -0
- data/include/yarp/pack.h +141 -0
- data/include/yarp/parser.h +389 -0
- data/include/yarp/regexp.h +19 -0
- data/include/yarp/unescape.h +42 -0
- data/include/yarp/util/yp_buffer.h +39 -0
- data/include/yarp/util/yp_char.h +75 -0
- data/include/yarp/util/yp_constant_pool.h +64 -0
- data/include/yarp/util/yp_list.h +67 -0
- data/include/yarp/util/yp_memchr.h +14 -0
- data/include/yarp/util/yp_newline_list.h +54 -0
- data/include/yarp/util/yp_state_stack.h +24 -0
- data/include/yarp/util/yp_string.h +57 -0
- data/include/yarp/util/yp_string_list.h +28 -0
- data/include/yarp/util/yp_strpbrk.h +29 -0
- data/include/yarp/version.h +5 -0
- data/include/yarp.h +69 -0
- data/lib/yarp/lex_compat.rb +759 -0
- data/lib/yarp/node.rb +7428 -0
- data/lib/yarp/pack.rb +185 -0
- data/lib/yarp/ripper_compat.rb +174 -0
- data/lib/yarp/serialize.rb +389 -0
- data/lib/yarp.rb +330 -0
- data/src/diagnostic.c +25 -0
- data/src/enc/yp_big5.c +79 -0
- data/src/enc/yp_euc_jp.c +85 -0
- data/src/enc/yp_gbk.c +88 -0
- data/src/enc/yp_shift_jis.c +83 -0
- data/src/enc/yp_tables.c +509 -0
- data/src/enc/yp_unicode.c +2320 -0
- data/src/enc/yp_windows_31j.c +83 -0
- data/src/node.c +2011 -0
- data/src/pack.c +493 -0
- data/src/prettyprint.c +1782 -0
- data/src/regexp.c +580 -0
- data/src/serialize.c +1576 -0
- data/src/token_type.c +347 -0
- data/src/unescape.c +576 -0
- data/src/util/yp_buffer.c +78 -0
- data/src/util/yp_char.c +229 -0
- data/src/util/yp_constant_pool.c +147 -0
- data/src/util/yp_list.c +50 -0
- data/src/util/yp_memchr.c +31 -0
- data/src/util/yp_newline_list.c +119 -0
- data/src/util/yp_state_stack.c +25 -0
- data/src/util/yp_string.c +207 -0
- data/src/util/yp_string_list.c +32 -0
- data/src/util/yp_strncasecmp.c +20 -0
- data/src/util/yp_strpbrk.c +66 -0
- data/src/yarp.c +13211 -0
- data/yarp.gemspec +100 -0
- metadata +125 -0
@@ -0,0 +1,547 @@
|
|
1
|
+
#include "yarp/extension.h"
|
2
|
+
|
3
|
+
// NOTE: this file should contain only bindings.
|
4
|
+
// All non-trivial logic should be in librubyparser so it can be shared its the various callers.
|
5
|
+
|
6
|
+
VALUE rb_cYARP;
|
7
|
+
VALUE rb_cYARPNode;
|
8
|
+
VALUE rb_cYARPSource;
|
9
|
+
VALUE rb_cYARPToken;
|
10
|
+
VALUE rb_cYARPLocation;
|
11
|
+
|
12
|
+
VALUE rb_cYARPComment;
|
13
|
+
VALUE rb_cYARPParseError;
|
14
|
+
VALUE rb_cYARPParseWarning;
|
15
|
+
VALUE rb_cYARPParseResult;
|
16
|
+
|
17
|
+
/******************************************************************************/
|
18
|
+
/* IO of Ruby code */
|
19
|
+
/******************************************************************************/
|
20
|
+
|
21
|
+
// Check if the given VALUE is a string. If it's nil, then return NULL. If it's
|
22
|
+
// not a string, then raise a type error. Otherwise return the VALUE as a C
|
23
|
+
// string.
|
24
|
+
static const char *
|
25
|
+
check_string(VALUE value) {
|
26
|
+
// If the value is nil, then we don't need to do anything.
|
27
|
+
if (NIL_P(value)) {
|
28
|
+
return NULL;
|
29
|
+
}
|
30
|
+
|
31
|
+
// Check if the value is a string. If it's not, then raise a type error.
|
32
|
+
if (!RB_TYPE_P(value, T_STRING)) {
|
33
|
+
rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (expected String)", rb_obj_class(value));
|
34
|
+
}
|
35
|
+
|
36
|
+
// Otherwise, return the value as a C string.
|
37
|
+
return RSTRING_PTR(value);
|
38
|
+
}
|
39
|
+
|
40
|
+
// Load the contents and size of the given string into the given yp_string_t.
|
41
|
+
static void
|
42
|
+
input_load_string(yp_string_t *input, VALUE string) {
|
43
|
+
// Check if the string is a string. If it's not, then raise a type error.
|
44
|
+
if (!RB_TYPE_P(string, T_STRING)) {
|
45
|
+
rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (expected String)", rb_obj_class(string));
|
46
|
+
}
|
47
|
+
|
48
|
+
yp_string_constant_init(input, RSTRING_PTR(string), RSTRING_LEN(string));
|
49
|
+
}
|
50
|
+
|
51
|
+
/******************************************************************************/
|
52
|
+
/* Serializing the AST */
|
53
|
+
/******************************************************************************/
|
54
|
+
|
55
|
+
// Dump the AST corresponding to the given input to a string.
|
56
|
+
static VALUE
|
57
|
+
dump_input(yp_string_t *input, const char *filepath) {
|
58
|
+
yp_buffer_t buffer;
|
59
|
+
if (!yp_buffer_init(&buffer)) {
|
60
|
+
rb_raise(rb_eNoMemError, "failed to allocate memory");
|
61
|
+
}
|
62
|
+
|
63
|
+
yp_parser_t parser;
|
64
|
+
yp_parser_init(&parser, yp_string_source(input), yp_string_length(input), filepath);
|
65
|
+
|
66
|
+
yp_node_t *node = yp_parse(&parser);
|
67
|
+
yp_serialize(&parser, node, &buffer);
|
68
|
+
|
69
|
+
VALUE result = rb_str_new(buffer.value, buffer.length);
|
70
|
+
yp_node_destroy(&parser, node);
|
71
|
+
yp_buffer_free(&buffer);
|
72
|
+
yp_parser_free(&parser);
|
73
|
+
|
74
|
+
return result;
|
75
|
+
}
|
76
|
+
|
77
|
+
// Dump the AST corresponding to the given string to a string.
|
78
|
+
static VALUE
|
79
|
+
dump(int argc, VALUE *argv, VALUE self) {
|
80
|
+
VALUE string;
|
81
|
+
VALUE filepath;
|
82
|
+
rb_scan_args(argc, argv, "11", &string, &filepath);
|
83
|
+
|
84
|
+
yp_string_t input;
|
85
|
+
input_load_string(&input, string);
|
86
|
+
return dump_input(&input, check_string(filepath));
|
87
|
+
}
|
88
|
+
|
89
|
+
// Dump the AST corresponding to the given file to a string.
|
90
|
+
static VALUE
|
91
|
+
dump_file(VALUE self, VALUE filepath) {
|
92
|
+
yp_string_t input;
|
93
|
+
|
94
|
+
const char *checked = check_string(filepath);
|
95
|
+
if (!yp_string_mapped_init(&input, checked)) return Qnil;
|
96
|
+
|
97
|
+
VALUE value = dump_input(&input, checked);
|
98
|
+
yp_string_free(&input);
|
99
|
+
|
100
|
+
return value;
|
101
|
+
}
|
102
|
+
|
103
|
+
/******************************************************************************/
|
104
|
+
/* Extracting values for the parse result */
|
105
|
+
/******************************************************************************/
|
106
|
+
|
107
|
+
// Extract the comments out of the parser into an array.
|
108
|
+
static VALUE
|
109
|
+
parser_comments(yp_parser_t *parser, VALUE source) {
|
110
|
+
VALUE comments = rb_ary_new();
|
111
|
+
|
112
|
+
for (yp_comment_t *comment = (yp_comment_t *) parser->comment_list.head; comment != NULL; comment = (yp_comment_t *) comment->node.next) {
|
113
|
+
VALUE location_argv[] = {
|
114
|
+
source,
|
115
|
+
LONG2FIX(comment->start - parser->start),
|
116
|
+
LONG2FIX(comment->end - comment->start)
|
117
|
+
};
|
118
|
+
|
119
|
+
VALUE type;
|
120
|
+
switch (comment->type) {
|
121
|
+
case YP_COMMENT_INLINE:
|
122
|
+
type = ID2SYM(rb_intern("inline"));
|
123
|
+
break;
|
124
|
+
case YP_COMMENT_EMBDOC:
|
125
|
+
type = ID2SYM(rb_intern("embdoc"));
|
126
|
+
break;
|
127
|
+
case YP_COMMENT___END__:
|
128
|
+
type = ID2SYM(rb_intern("__END__"));
|
129
|
+
break;
|
130
|
+
default:
|
131
|
+
type = ID2SYM(rb_intern("inline"));
|
132
|
+
break;
|
133
|
+
}
|
134
|
+
|
135
|
+
VALUE comment_argv[] = { type, rb_class_new_instance(3, location_argv, rb_cYARPLocation) };
|
136
|
+
rb_ary_push(comments, rb_class_new_instance(2, comment_argv, rb_cYARPComment));
|
137
|
+
}
|
138
|
+
|
139
|
+
return comments;
|
140
|
+
}
|
141
|
+
|
142
|
+
// Extract the errors out of the parser into an array.
|
143
|
+
static VALUE
|
144
|
+
parser_errors(yp_parser_t *parser, rb_encoding *encoding, VALUE source) {
|
145
|
+
VALUE errors = rb_ary_new();
|
146
|
+
yp_diagnostic_t *error;
|
147
|
+
|
148
|
+
for (error = (yp_diagnostic_t *) parser->error_list.head; error != NULL; error = (yp_diagnostic_t *) error->node.next) {
|
149
|
+
VALUE location_argv[] = {
|
150
|
+
source,
|
151
|
+
LONG2FIX(error->start - parser->start),
|
152
|
+
LONG2FIX(error->end - error->start)
|
153
|
+
};
|
154
|
+
|
155
|
+
VALUE error_argv[] = {
|
156
|
+
rb_enc_str_new_cstr(error->message, encoding),
|
157
|
+
rb_class_new_instance(3, location_argv, rb_cYARPLocation)
|
158
|
+
};
|
159
|
+
|
160
|
+
rb_ary_push(errors, rb_class_new_instance(2, error_argv, rb_cYARPParseError));
|
161
|
+
}
|
162
|
+
|
163
|
+
return errors;
|
164
|
+
}
|
165
|
+
|
166
|
+
// Extract the warnings out of the parser into an array.
|
167
|
+
static VALUE
|
168
|
+
parser_warnings(yp_parser_t *parser, rb_encoding *encoding, VALUE source) {
|
169
|
+
VALUE warnings = rb_ary_new();
|
170
|
+
yp_diagnostic_t *warning;
|
171
|
+
|
172
|
+
for (warning = (yp_diagnostic_t *) parser->warning_list.head; warning != NULL; warning = (yp_diagnostic_t *) warning->node.next) {
|
173
|
+
VALUE location_argv[] = {
|
174
|
+
source,
|
175
|
+
LONG2FIX(warning->start - parser->start),
|
176
|
+
LONG2FIX(warning->end - warning->start)
|
177
|
+
};
|
178
|
+
|
179
|
+
VALUE warning_argv[] = {
|
180
|
+
rb_enc_str_new_cstr(warning->message, encoding),
|
181
|
+
rb_class_new_instance(3, location_argv, rb_cYARPLocation)
|
182
|
+
};
|
183
|
+
|
184
|
+
rb_ary_push(warnings, rb_class_new_instance(2, warning_argv, rb_cYARPParseWarning));
|
185
|
+
}
|
186
|
+
|
187
|
+
return warnings;
|
188
|
+
}
|
189
|
+
|
190
|
+
/******************************************************************************/
|
191
|
+
/* Lexing Ruby code */
|
192
|
+
/******************************************************************************/
|
193
|
+
|
194
|
+
// This struct gets stored in the parser and passed in to the lex callback any
|
195
|
+
// time a new token is found. We use it to store the necessary information to
|
196
|
+
// initialize a Token instance.
|
197
|
+
typedef struct {
|
198
|
+
VALUE source;
|
199
|
+
VALUE tokens;
|
200
|
+
rb_encoding *encoding;
|
201
|
+
} lex_data_t;
|
202
|
+
|
203
|
+
// This is passed as a callback to the parser. It gets called every time a new
|
204
|
+
// token is found. Once found, we initialize a new instance of Token and push it
|
205
|
+
// onto the tokens array.
|
206
|
+
static void
|
207
|
+
lex_token(void *data, yp_parser_t *parser, yp_token_t *token) {
|
208
|
+
lex_data_t *lex_data = (lex_data_t *) parser->lex_callback->data;
|
209
|
+
|
210
|
+
VALUE yields = rb_ary_new_capa(2);
|
211
|
+
rb_ary_push(yields, yp_token_new(parser, token, lex_data->encoding, lex_data->source));
|
212
|
+
rb_ary_push(yields, INT2FIX(parser->lex_state));
|
213
|
+
|
214
|
+
rb_ary_push(lex_data->tokens, yields);
|
215
|
+
}
|
216
|
+
|
217
|
+
// This is called whenever the encoding changes based on the magic comment at
|
218
|
+
// the top of the file. We use it to update the encoding that we are using to
|
219
|
+
// create tokens.
|
220
|
+
static void
|
221
|
+
lex_encoding_changed_callback(yp_parser_t *parser) {
|
222
|
+
lex_data_t *lex_data = (lex_data_t *) parser->lex_callback->data;
|
223
|
+
lex_data->encoding = rb_enc_find(parser->encoding.name);
|
224
|
+
}
|
225
|
+
|
226
|
+
// Return an array of tokens corresponding to the given source.
|
227
|
+
static VALUE
|
228
|
+
lex_input(yp_string_t *input, const char *filepath) {
|
229
|
+
yp_parser_t parser;
|
230
|
+
yp_parser_init(&parser, yp_string_source(input), yp_string_length(input), filepath);
|
231
|
+
yp_parser_register_encoding_changed_callback(&parser, lex_encoding_changed_callback);
|
232
|
+
|
233
|
+
VALUE offsets = rb_ary_new();
|
234
|
+
VALUE source_argv[] = { rb_str_new(yp_string_source(input), yp_string_length(input)), offsets };
|
235
|
+
VALUE source = rb_class_new_instance(2, source_argv, rb_cYARPSource);
|
236
|
+
|
237
|
+
lex_data_t lex_data = {
|
238
|
+
.source = source,
|
239
|
+
.tokens = rb_ary_new(),
|
240
|
+
.encoding = rb_utf8_encoding()
|
241
|
+
};
|
242
|
+
|
243
|
+
lex_data_t *data = &lex_data;
|
244
|
+
yp_lex_callback_t lex_callback = (yp_lex_callback_t) {
|
245
|
+
.data = (void *) data,
|
246
|
+
.callback = lex_token,
|
247
|
+
};
|
248
|
+
|
249
|
+
parser.lex_callback = &lex_callback;
|
250
|
+
yp_node_t *node = yp_parse(&parser);
|
251
|
+
|
252
|
+
// Here we need to update the source range to have the correct newline
|
253
|
+
// offsets. We do it here because we've already created the object and given
|
254
|
+
// it over to all of the tokens.
|
255
|
+
for (size_t index = 0; index < parser.newline_list.size; index++) {
|
256
|
+
rb_ary_push(offsets, INT2FIX(parser.newline_list.offsets[index]));
|
257
|
+
}
|
258
|
+
|
259
|
+
VALUE result_argv[] = {
|
260
|
+
lex_data.tokens,
|
261
|
+
parser_comments(&parser, source),
|
262
|
+
parser_errors(&parser, lex_data.encoding, source),
|
263
|
+
parser_warnings(&parser, lex_data.encoding, source),
|
264
|
+
source
|
265
|
+
};
|
266
|
+
|
267
|
+
VALUE result = rb_class_new_instance(5, result_argv, rb_cYARPParseResult);
|
268
|
+
|
269
|
+
yp_node_destroy(&parser, node);
|
270
|
+
yp_parser_free(&parser);
|
271
|
+
|
272
|
+
return result;
|
273
|
+
}
|
274
|
+
|
275
|
+
// Return an array of tokens corresponding to the given string.
|
276
|
+
static VALUE
|
277
|
+
lex(int argc, VALUE *argv, VALUE self) {
|
278
|
+
VALUE string;
|
279
|
+
VALUE filepath;
|
280
|
+
rb_scan_args(argc, argv, "11", &string, &filepath);
|
281
|
+
|
282
|
+
yp_string_t input;
|
283
|
+
input_load_string(&input, string);
|
284
|
+
return lex_input(&input, check_string(filepath));
|
285
|
+
}
|
286
|
+
|
287
|
+
// Return an array of tokens corresponding to the given file.
|
288
|
+
static VALUE
|
289
|
+
lex_file(VALUE self, VALUE filepath) {
|
290
|
+
yp_string_t input;
|
291
|
+
|
292
|
+
const char *checked = check_string(filepath);
|
293
|
+
if (!yp_string_mapped_init(&input, checked)) return Qnil;
|
294
|
+
|
295
|
+
VALUE value = lex_input(&input, checked);
|
296
|
+
yp_string_free(&input);
|
297
|
+
|
298
|
+
return value;
|
299
|
+
}
|
300
|
+
|
301
|
+
/******************************************************************************/
|
302
|
+
/* Parsing Ruby code */
|
303
|
+
/******************************************************************************/
|
304
|
+
|
305
|
+
// Parse the given input and return a ParseResult instance.
|
306
|
+
static VALUE
|
307
|
+
parse_input(yp_string_t *input, const char *filepath) {
|
308
|
+
yp_parser_t parser;
|
309
|
+
yp_parser_init(&parser, yp_string_source(input), yp_string_length(input), filepath);
|
310
|
+
|
311
|
+
yp_node_t *node = yp_parse(&parser);
|
312
|
+
rb_encoding *encoding = rb_enc_find(parser.encoding.name);
|
313
|
+
|
314
|
+
VALUE source = yp_source_new(&parser);
|
315
|
+
VALUE result_argv[] = {
|
316
|
+
yp_ast_new(&parser, node, encoding),
|
317
|
+
parser_comments(&parser, source),
|
318
|
+
parser_errors(&parser, encoding, source),
|
319
|
+
parser_warnings(&parser, encoding, source),
|
320
|
+
source
|
321
|
+
};
|
322
|
+
|
323
|
+
VALUE result = rb_class_new_instance(5, result_argv, rb_cYARPParseResult);
|
324
|
+
|
325
|
+
yp_node_destroy(&parser, node);
|
326
|
+
yp_parser_free(&parser);
|
327
|
+
|
328
|
+
return result;
|
329
|
+
}
|
330
|
+
|
331
|
+
// Parse the given string and return a ParseResult instance.
|
332
|
+
static VALUE
|
333
|
+
parse(int argc, VALUE *argv, VALUE self) {
|
334
|
+
VALUE string;
|
335
|
+
VALUE filepath;
|
336
|
+
rb_scan_args(argc, argv, "11", &string, &filepath);
|
337
|
+
|
338
|
+
yp_string_t input;
|
339
|
+
input_load_string(&input, string);
|
340
|
+
|
341
|
+
#ifdef YARP_DEBUG_MODE_BUILD
|
342
|
+
size_t length = yp_string_length(&input);
|
343
|
+
char* dup = malloc(length);
|
344
|
+
memcpy(dup, yp_string_source(&input), length);
|
345
|
+
yp_string_constant_init(&input, dup, length);
|
346
|
+
#endif
|
347
|
+
|
348
|
+
VALUE value = parse_input(&input, check_string(filepath));
|
349
|
+
|
350
|
+
#ifdef YARP_DEBUG_MODE_BUILD
|
351
|
+
free(dup);
|
352
|
+
#endif
|
353
|
+
|
354
|
+
return value;
|
355
|
+
}
|
356
|
+
|
357
|
+
// Parse the given file and return a ParseResult instance.
|
358
|
+
static VALUE
|
359
|
+
parse_file(VALUE self, VALUE filepath) {
|
360
|
+
yp_string_t input;
|
361
|
+
|
362
|
+
const char *checked = check_string(filepath);
|
363
|
+
if (!yp_string_mapped_init(&input, checked)) return Qnil;
|
364
|
+
|
365
|
+
VALUE value = parse_input(&input, checked);
|
366
|
+
yp_string_free(&input);
|
367
|
+
|
368
|
+
return value;
|
369
|
+
}
|
370
|
+
|
371
|
+
/******************************************************************************/
|
372
|
+
/* Utility functions exposed to make testing easier */
|
373
|
+
/******************************************************************************/
|
374
|
+
|
375
|
+
// Returns an array of strings corresponding to the named capture groups in the
|
376
|
+
// given source string. If YARP was unable to parse the regular expression, this
|
377
|
+
// function returns nil.
|
378
|
+
static VALUE
|
379
|
+
named_captures(VALUE self, VALUE source) {
|
380
|
+
yp_string_list_t string_list;
|
381
|
+
yp_string_list_init(&string_list);
|
382
|
+
|
383
|
+
if (!yp_regexp_named_capture_group_names(RSTRING_PTR(source), RSTRING_LEN(source), &string_list, false, &yp_encoding_utf_8)) {
|
384
|
+
yp_string_list_free(&string_list);
|
385
|
+
return Qnil;
|
386
|
+
}
|
387
|
+
|
388
|
+
VALUE names = rb_ary_new();
|
389
|
+
for (size_t index = 0; index < string_list.length; index++) {
|
390
|
+
const yp_string_t *string = &string_list.strings[index];
|
391
|
+
rb_ary_push(names, rb_str_new(yp_string_source(string), yp_string_length(string)));
|
392
|
+
}
|
393
|
+
|
394
|
+
yp_string_list_free(&string_list);
|
395
|
+
return names;
|
396
|
+
}
|
397
|
+
|
398
|
+
// Accepts a source string and a type of unescaping and returns the unescaped
|
399
|
+
// version.
|
400
|
+
static VALUE
|
401
|
+
unescape(VALUE source, yp_unescape_type_t unescape_type) {
|
402
|
+
yp_string_t result;
|
403
|
+
|
404
|
+
if (yp_unescape_string(RSTRING_PTR(source), RSTRING_LEN(source), unescape_type, &result)) {
|
405
|
+
VALUE str = rb_str_new(yp_string_source(&result), yp_string_length(&result));
|
406
|
+
yp_string_free(&result);
|
407
|
+
return str;
|
408
|
+
} else {
|
409
|
+
yp_string_free(&result);
|
410
|
+
return Qnil;
|
411
|
+
}
|
412
|
+
}
|
413
|
+
|
414
|
+
// Do not unescape anything in the given string. This is here to provide a
|
415
|
+
// consistent API.
|
416
|
+
static VALUE
|
417
|
+
unescape_none(VALUE self, VALUE source) {
|
418
|
+
return unescape(source, YP_UNESCAPE_NONE);
|
419
|
+
}
|
420
|
+
|
421
|
+
// Minimally unescape the given string. This means effectively unescaping just
|
422
|
+
// the quotes of a string. Returns the unescaped string.
|
423
|
+
static VALUE
|
424
|
+
unescape_minimal(VALUE self, VALUE source) {
|
425
|
+
return unescape(source, YP_UNESCAPE_MINIMAL);
|
426
|
+
}
|
427
|
+
|
428
|
+
// Unescape everything in the given string. Return the unescaped string.
|
429
|
+
static VALUE
|
430
|
+
unescape_all(VALUE self, VALUE source) {
|
431
|
+
return unescape(source, YP_UNESCAPE_ALL);
|
432
|
+
}
|
433
|
+
|
434
|
+
// Return a hash of information about the given source string's memory usage.
|
435
|
+
static VALUE
|
436
|
+
memsize(VALUE self, VALUE string) {
|
437
|
+
yp_parser_t parser;
|
438
|
+
size_t length = RSTRING_LEN(string);
|
439
|
+
yp_parser_init(&parser, RSTRING_PTR(string), length, NULL);
|
440
|
+
|
441
|
+
yp_node_t *node = yp_parse(&parser);
|
442
|
+
yp_memsize_t memsize;
|
443
|
+
yp_node_memsize(node, &memsize);
|
444
|
+
|
445
|
+
yp_node_destroy(&parser, node);
|
446
|
+
yp_parser_free(&parser);
|
447
|
+
|
448
|
+
VALUE result = rb_hash_new();
|
449
|
+
rb_hash_aset(result, ID2SYM(rb_intern("length")), INT2FIX(length));
|
450
|
+
rb_hash_aset(result, ID2SYM(rb_intern("memsize")), INT2FIX(memsize.memsize));
|
451
|
+
rb_hash_aset(result, ID2SYM(rb_intern("node_count")), INT2FIX(memsize.node_count));
|
452
|
+
return result;
|
453
|
+
}
|
454
|
+
|
455
|
+
// Parse the file, but do nothing with the result. This is used to profile the
|
456
|
+
// parser for memory and speed.
|
457
|
+
static VALUE
|
458
|
+
profile_file(VALUE self, VALUE filepath) {
|
459
|
+
yp_string_t input;
|
460
|
+
|
461
|
+
const char *checked = check_string(filepath);
|
462
|
+
if (!yp_string_mapped_init(&input, checked)) return Qnil;
|
463
|
+
|
464
|
+
yp_parser_t parser;
|
465
|
+
yp_parser_init(&parser, yp_string_source(&input), yp_string_length(&input), checked);
|
466
|
+
|
467
|
+
yp_node_t *node = yp_parse(&parser);
|
468
|
+
yp_node_destroy(&parser, node);
|
469
|
+
yp_parser_free(&parser);
|
470
|
+
|
471
|
+
return Qnil;
|
472
|
+
}
|
473
|
+
|
474
|
+
// Parse the file and serialize the result. This is mostly used to test this
|
475
|
+
// path since it is used by client libraries.
|
476
|
+
static VALUE
|
477
|
+
parse_serialize_file_metadata(VALUE self, VALUE filepath, VALUE metadata) {
|
478
|
+
yp_string_t input;
|
479
|
+
yp_buffer_t buffer;
|
480
|
+
yp_buffer_init(&buffer);
|
481
|
+
|
482
|
+
const char *checked = check_string(filepath);
|
483
|
+
if (!yp_string_mapped_init(&input, checked)) return Qnil;
|
484
|
+
|
485
|
+
yp_parse_serialize(yp_string_source(&input), yp_string_length(&input), &buffer, check_string(metadata));
|
486
|
+
VALUE result = rb_str_new(buffer.value, buffer.length);
|
487
|
+
|
488
|
+
yp_buffer_free(&buffer);
|
489
|
+
return result;
|
490
|
+
}
|
491
|
+
|
492
|
+
/******************************************************************************/
|
493
|
+
/* Initialization of the extension */
|
494
|
+
/******************************************************************************/
|
495
|
+
|
496
|
+
RUBY_FUNC_EXPORTED void
|
497
|
+
Init_yarp(void) {
|
498
|
+
// Make sure that the YARP library version matches the expected version.
|
499
|
+
// Otherwise something was compiled incorrectly.
|
500
|
+
if (strcmp(yp_version(), EXPECTED_YARP_VERSION) != 0) {
|
501
|
+
rb_raise(
|
502
|
+
rb_eRuntimeError,
|
503
|
+
"The YARP library version (%s) does not match the expected version (%s)",
|
504
|
+
yp_version(),
|
505
|
+
EXPECTED_YARP_VERSION
|
506
|
+
);
|
507
|
+
}
|
508
|
+
|
509
|
+
// Grab up references to all of the constants that we're going to need to
|
510
|
+
// reference throughout this extension.
|
511
|
+
rb_cYARP = rb_define_module("YARP");
|
512
|
+
rb_cYARPNode = rb_define_class_under(rb_cYARP, "Node", rb_cObject);
|
513
|
+
rb_cYARPSource = rb_define_class_under(rb_cYARP, "Source", rb_cObject);
|
514
|
+
rb_cYARPToken = rb_define_class_under(rb_cYARP, "Token", rb_cObject);
|
515
|
+
rb_cYARPLocation = rb_define_class_under(rb_cYARP, "Location", rb_cObject);
|
516
|
+
rb_cYARPComment = rb_define_class_under(rb_cYARP, "Comment", rb_cObject);
|
517
|
+
rb_cYARPParseError = rb_define_class_under(rb_cYARP, "ParseError", rb_cObject);
|
518
|
+
rb_cYARPParseWarning = rb_define_class_under(rb_cYARP, "ParseWarning", rb_cObject);
|
519
|
+
rb_cYARPParseResult = rb_define_class_under(rb_cYARP, "ParseResult", rb_cObject);
|
520
|
+
|
521
|
+
// Define the version string here so that we can use the constants defined
|
522
|
+
// in yarp.h.
|
523
|
+
rb_define_const(rb_cYARP, "VERSION", rb_str_new2(EXPECTED_YARP_VERSION));
|
524
|
+
|
525
|
+
// First, the functions that have to do with lexing and parsing.
|
526
|
+
rb_define_singleton_method(rb_cYARP, "dump", dump, -1);
|
527
|
+
rb_define_singleton_method(rb_cYARP, "dump_file", dump_file, 1);
|
528
|
+
rb_define_singleton_method(rb_cYARP, "lex", lex, -1);
|
529
|
+
rb_define_singleton_method(rb_cYARP, "lex_file", lex_file, 1);
|
530
|
+
rb_define_singleton_method(rb_cYARP, "parse", parse, -1);
|
531
|
+
rb_define_singleton_method(rb_cYARP, "parse_file", parse_file, 1);
|
532
|
+
|
533
|
+
// Next, the functions that will be called by the parser to perform various
|
534
|
+
// internal tasks. We expose these to make them easier to test.
|
535
|
+
VALUE rb_cYARPDebug = rb_define_module_under(rb_cYARP, "Debug");
|
536
|
+
rb_define_singleton_method(rb_cYARPDebug, "named_captures", named_captures, 1);
|
537
|
+
rb_define_singleton_method(rb_cYARPDebug, "unescape_none", unescape_none, 1);
|
538
|
+
rb_define_singleton_method(rb_cYARPDebug, "unescape_minimal", unescape_minimal, 1);
|
539
|
+
rb_define_singleton_method(rb_cYARPDebug, "unescape_all", unescape_all, 1);
|
540
|
+
rb_define_singleton_method(rb_cYARPDebug, "memsize", memsize, 1);
|
541
|
+
rb_define_singleton_method(rb_cYARPDebug, "profile_file", profile_file, 1);
|
542
|
+
rb_define_singleton_method(rb_cYARPDebug, "parse_serialize_file_metadata", parse_serialize_file_metadata, 2);
|
543
|
+
|
544
|
+
// Next, initialize the other APIs.
|
545
|
+
Init_yarp_api_node();
|
546
|
+
Init_yarp_pack();
|
547
|
+
}
|
@@ -0,0 +1,18 @@
|
|
1
|
+
#ifndef YARP_EXT_NODE_H
|
2
|
+
#define YARP_EXT_NODE_H
|
3
|
+
|
4
|
+
#include <ruby.h>
|
5
|
+
#include <ruby/encoding.h>
|
6
|
+
#include "yarp.h"
|
7
|
+
|
8
|
+
#define EXPECTED_YARP_VERSION "0.6.0"
|
9
|
+
|
10
|
+
VALUE yp_source_new(yp_parser_t *parser);
|
11
|
+
VALUE yp_token_new(yp_parser_t *parser, yp_token_t *token, rb_encoding *encoding, VALUE source);
|
12
|
+
VALUE yp_ast_new(yp_parser_t *parser, yp_node_t *node, rb_encoding *encoding);
|
13
|
+
|
14
|
+
void Init_yarp_api_node(void);
|
15
|
+
void Init_yarp_pack(void);
|
16
|
+
YP_EXPORTED_FUNCTION void Init_yarp(void);
|
17
|
+
|
18
|
+
#endif
|