1 /* GLIB - Library of useful routines for C programming
2 * Copyright (C) 1995-1997 Peter Mattis, Spencer Kimball and Josh MacDonald
3 *
4 * GScanner: Flexible lexical scanner for general purpose.
5 * Copyright (C) 1997, 1998 Tim Janik
6 *
7 * This library is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * This library is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with this library; if not, see <http://www.gnu.org/licenses/>.
19 */
20
21 /*
22 * Modified by the GLib Team and others 1997-2000. See the AUTHORS
23 * file for a list of people on the GLib Team. See the ChangeLog
24 * files for a list of changes. These files are distributed with
25 * GLib at ftp://ftp.gtk.org/pub/gtk/.
26 */
27
28 /*
29 * MT safe
30 */
31
32 #include "config.h"
33
34 #include <errno.h>
35 #include <stdlib.h>
36 #include <stdarg.h>
37 #include <string.h>
38 #include <stdio.h>
39
40 #include "gscanner.h"
41
42 #include "gprintfint.h"
43 #include "gstrfuncs.h"
44 #include "gstring.h"
45 #include "gtestutils.h"
46
47 #ifdef G_OS_UNIX
48 #include <unistd.h>
49 #endif
50 #ifdef G_OS_WIN32
51 #include <io.h>
52 #endif
53
54
55 /**
56 * SECTION:scanner
57 * @title: Lexical Scanner
58 * @short_description: a general purpose lexical scanner
59 *
60 * The #GScanner and its associated functions provide a
61 * general purpose lexical scanner.
62 */
63
64 /**
65 * GScannerMsgFunc:
66 * @scanner: a #GScanner
67 * @message: the message
68 * @error: %TRUE if the message signals an error,
69 * %FALSE if it signals a warning.
70 *
71 * Specifies the type of the message handler function.
72 */
73
74 /**
75 * G_CSET_a_2_z:
76 *
77 * The set of lowercase ASCII alphabet characters.
78 * Used for specifying valid identifier characters
79 * in #GScannerConfig.
80 */
81
82 /**
83 * G_CSET_A_2_Z:
84 *
85 * The set of uppercase ASCII alphabet characters.
86 * Used for specifying valid identifier characters
87 * in #GScannerConfig.
88 */
89
90 /**
91 * G_CSET_DIGITS:
92 *
93 * The set of ASCII digits.
94 * Used for specifying valid identifier characters
95 * in #GScannerConfig.
96 */
97
98 /**
99 * G_CSET_LATINC:
100 *
101 * The set of uppercase ISO 8859-1 alphabet characters
102 * which are not ASCII characters.
103 * Used for specifying valid identifier characters
104 * in #GScannerConfig.
105 */
106
107 /**
108 * G_CSET_LATINS:
109 *
110 * The set of lowercase ISO 8859-1 alphabet characters
111 * which are not ASCII characters.
112 * Used for specifying valid identifier characters
113 * in #GScannerConfig.
114 */
115
116 /**
117 * GTokenType:
118 * @G_TOKEN_EOF: the end of the file
119 * @G_TOKEN_LEFT_PAREN: a '(' character
120 * @G_TOKEN_LEFT_CURLY: a '{' character
121 * @G_TOKEN_LEFT_BRACE: a '[' character
122 * @G_TOKEN_RIGHT_CURLY: a '}' character
123 * @G_TOKEN_RIGHT_PAREN: a ')' character
124 * @G_TOKEN_RIGHT_BRACE: a ']' character
125 * @G_TOKEN_EQUAL_SIGN: a '=' character
126 * @G_TOKEN_COMMA: a ',' character
127 * @G_TOKEN_NONE: not a token
128 * @G_TOKEN_ERROR: an error occurred
129 * @G_TOKEN_CHAR: a character
130 * @G_TOKEN_BINARY: a binary integer
131 * @G_TOKEN_OCTAL: an octal integer
132 * @G_TOKEN_INT: an integer
133 * @G_TOKEN_HEX: a hex integer
134 * @G_TOKEN_FLOAT: a floating point number
135 * @G_TOKEN_STRING: a string
136 * @G_TOKEN_SYMBOL: a symbol
137 * @G_TOKEN_IDENTIFIER: an identifier
138 * @G_TOKEN_IDENTIFIER_NULL: a null identifier
139 * @G_TOKEN_COMMENT_SINGLE: one line comment
140 * @G_TOKEN_COMMENT_MULTI: multi line comment
141 *
142 * The possible types of token returned from each
143 * g_scanner_get_next_token() call.
144 */
145
146 /**
147 * GTokenValue:
148 * @v_symbol: token symbol value
149 * @v_identifier: token identifier value
150 * @v_binary: token binary integer value
151 * @v_octal: octal integer value
152 * @v_int: integer value
153 * @v_int64: 64-bit integer value
154 * @v_float: floating point value
155 * @v_hex: hex integer value
156 * @v_string: string value
157 * @v_comment: comment value
158 * @v_char: character value
159 * @v_error: error value
160 *
161 * A union holding the value of the token.
162 */
163
164 /**
165 * GErrorType:
166 * @G_ERR_UNKNOWN: unknown error
167 * @G_ERR_UNEXP_EOF: unexpected end of file
168 * @G_ERR_UNEXP_EOF_IN_STRING: unterminated string constant
169 * @G_ERR_UNEXP_EOF_IN_COMMENT: unterminated comment
170 * @G_ERR_NON_DIGIT_IN_CONST: non-digit character in a number
171 * @G_ERR_DIGIT_RADIX: digit beyond radix in a number
172 * @G_ERR_FLOAT_RADIX: non-decimal floating point number
173 * @G_ERR_FLOAT_MALFORMED: malformed floating point number
174 *
175 * The possible errors, used in the @v_error field
176 * of #GTokenValue, when the token is a %G_TOKEN_ERROR.
177 */
178
179 /**
180 * GScanner:
181 * @user_data: unused
182 * @max_parse_errors: unused
183 * @parse_errors: g_scanner_error() increments this field
184 * @input_name: name of input stream, featured by the default message handler
185 * @qdata: quarked data
186 * @config: link into the scanner configuration
187 * @token: token parsed by the last g_scanner_get_next_token()
188 * @value: value of the last token from g_scanner_get_next_token()
189 * @line: line number of the last token from g_scanner_get_next_token()
190 * @position: char number of the last token from g_scanner_get_next_token()
191 * @next_token: token parsed by the last g_scanner_peek_next_token()
192 * @next_value: value of the last token from g_scanner_peek_next_token()
193 * @next_line: line number of the last token from g_scanner_peek_next_token()
194 * @next_position: char number of the last token from g_scanner_peek_next_token()
195 * @msg_handler: handler function for _warn and _error
196 *
197 * The data structure representing a lexical scanner.
198 *
199 * You should set @input_name after creating the scanner, since
200 * it is used by the default message handler when displaying
201 * warnings and errors. If you are scanning a file, the filename
202 * would be a good choice.
203 *
204 * The @user_data and @max_parse_errors fields are not used.
205 * If you need to associate extra data with the scanner you
206 * can place them here.
207 *
208 * If you want to use your own message handler you can set the
209 * @msg_handler field. The type of the message handler function
210 * is declared by #GScannerMsgFunc.
211 */
212
213 /**
214 * GScannerConfig:
215 * @cset_skip_characters: specifies which characters should be skipped
216 * by the scanner (the default is the whitespace characters: space,
217 * tab, carriage-return and line-feed).
218 * @cset_identifier_first: specifies the characters which can start
219 * identifiers (the default is #G_CSET_a_2_z, "_", and #G_CSET_A_2_Z).
220 * @cset_identifier_nth: specifies the characters which can be used
221 * in identifiers, after the first character (the default is
222 * #G_CSET_a_2_z, "_0123456789", #G_CSET_A_2_Z, #G_CSET_LATINS,
223 * #G_CSET_LATINC).
224 * @cpair_comment_single: specifies the characters at the start and
225 * end of single-line comments. The default is "#\n" which means
226 * that single-line comments start with a '#' and continue until
227 * a '\n' (end of line).
228 * @case_sensitive: specifies if symbols are case sensitive (the
229 * default is %FALSE).
230 * @skip_comment_multi: specifies if multi-line comments are skipped
231 * and not returned as tokens (the default is %TRUE).
232 * @skip_comment_single: specifies if single-line comments are skipped
233 * and not returned as tokens (the default is %TRUE).
234 * @scan_comment_multi: specifies if multi-line comments are recognized
235 * (the default is %TRUE).
236 * @scan_identifier: specifies if identifiers are recognized (the
237 * default is %TRUE).
238 * @scan_identifier_1char: specifies if single-character
239 * identifiers are recognized (the default is %FALSE).
240 * @scan_identifier_NULL: specifies if %NULL is reported as
241 * %G_TOKEN_IDENTIFIER_NULL (the default is %FALSE).
242 * @scan_symbols: specifies if symbols are recognized (the default
243 * is %TRUE).
244 * @scan_binary: specifies if binary numbers are recognized (the
245 * default is %FALSE).
246 * @scan_octal: specifies if octal numbers are recognized (the
247 * default is %TRUE).
248 * @scan_float: specifies if floating point numbers are recognized
249 * (the default is %TRUE).
250 * @scan_hex: specifies if hexadecimal numbers are recognized (the
251 * default is %TRUE).
252 * @scan_hex_dollar: specifies if '$' is recognized as a prefix for
253 * hexadecimal numbers (the default is %FALSE).
254 * @scan_string_sq: specifies if strings can be enclosed in single
255 * quotes (the default is %TRUE).
256 * @scan_string_dq: specifies if strings can be enclosed in double
257 * quotes (the default is %TRUE).
258 * @numbers_2_int: specifies if binary, octal and hexadecimal numbers
259 * are reported as #G_TOKEN_INT (the default is %TRUE).
260 * @int_2_float: specifies if all numbers are reported as %G_TOKEN_FLOAT
261 * (the default is %FALSE).
262 * @identifier_2_string: specifies if identifiers are reported as strings
263 * (the default is %FALSE).
264 * @char_2_token: specifies if characters are reported by setting
265 * `token = ch` or as %G_TOKEN_CHAR (the default is %TRUE).
266 * @symbol_2_token: specifies if symbols are reported by setting
267 * `token = v_symbol` or as %G_TOKEN_SYMBOL (the default is %FALSE).
268 * @scope_0_fallback: specifies if a symbol is searched for in the
269 * default scope in addition to the current scope (the default is %FALSE).
270 * @store_int64: use value.v_int64 rather than v_int
271 *
272 * Specifies the #GScanner parser configuration. Most settings can
273 * be changed during the parsing phase and will affect the lexical
274 * parsing of the next unpeeked token.
275 */
276
277 /* --- defines --- */
278 #define to_lower(c) ( \
279 (guchar) ( \
280 ( (((guchar)(c))>='A' && ((guchar)(c))<='Z') * ('a'-'A') ) | \
281 ( (((guchar)(c))>=192 && ((guchar)(c))<=214) * (224-192) ) | \
282 ( (((guchar)(c))>=216 && ((guchar)(c))<=222) * (248-216) ) | \
283 ((guchar)(c)) \
284 ) \
285 )
286 #define READ_BUFFER_SIZE (4000)
287
288
289 /* --- typedefs --- */
290 typedef struct _GScannerKey GScannerKey;
291
292 struct _GScannerKey
293 {
294 guint scope_id;
295 gchar *symbol;
296 gpointer value;
297 };
298
299
300 /* --- variables --- */
301 static const GScannerConfig g_scanner_config_template =
302 {
303 (
304 " \t\r\n"
305 ) /* cset_skip_characters */,
306 (
307 G_CSET_a_2_z
308 "_"
309 G_CSET_A_2_Z
310 ) /* cset_identifier_first */,
311 (
312 G_CSET_a_2_z
313 "_"
314 G_CSET_A_2_Z
315 G_CSET_DIGITS
316 G_CSET_LATINS
317 G_CSET_LATINC
318 ) /* cset_identifier_nth */,
319 ( "#\n" ) /* cpair_comment_single */,
320
321 FALSE /* case_sensitive */,
322
323 TRUE /* skip_comment_multi */,
324 TRUE /* skip_comment_single */,
325 TRUE /* scan_comment_multi */,
326 TRUE /* scan_identifier */,
327 FALSE /* scan_identifier_1char */,
328 FALSE /* scan_identifier_NULL */,
329 TRUE /* scan_symbols */,
330 FALSE /* scan_binary */,
331 TRUE /* scan_octal */,
332 TRUE /* scan_float */,
333 TRUE /* scan_hex */,
334 FALSE /* scan_hex_dollar */,
335 TRUE /* scan_string_sq */,
336 TRUE /* scan_string_dq */,
337 TRUE /* numbers_2_int */,
338 FALSE /* int_2_float */,
339 FALSE /* identifier_2_string */,
340 TRUE /* char_2_token */,
341 FALSE /* symbol_2_token */,
342 FALSE /* scope_0_fallback */,
343 FALSE /* store_int64 */,
344 0 /* padding_dummy */
345 };
346
347
348 /* --- prototypes --- */
349 static inline
350 GScannerKey* g_scanner_lookup_internal (GScanner *scanner,
351 guint scope_id,
352 const gchar *symbol);
353 static gboolean g_scanner_key_equal (gconstpointer v1,
354 gconstpointer v2);
355 static guint g_scanner_key_hash (gconstpointer v);
356 static void g_scanner_get_token_ll (GScanner *scanner,
357 GTokenType *token_p,
358 GTokenValue *value_p,
359 guint *line_p,
360 guint *position_p);
361 static void g_scanner_get_token_i (GScanner *scanner,
362 GTokenType *token_p,
363 GTokenValue *value_p,
364 guint *line_p,
365 guint *position_p);
366
367 static guchar g_scanner_peek_next_char (GScanner *scanner);
368 static guchar g_scanner_get_char (GScanner *scanner,
369 guint *line_p,
370 guint *position_p);
371 static void g_scanner_msg_handler (GScanner *scanner,
372 gchar *message,
373 gboolean is_error);
374
375
376 /* --- functions --- */
377 static inline gint
g_scanner_char_2_num(guchar c,guchar base)378 g_scanner_char_2_num (guchar c,
379 guchar base)
380 {
381 if (c >= '0' && c <= '9')
382 c -= '0';
383 else if (c >= 'A' && c <= 'Z')
384 c -= 'A' - 10;
385 else if (c >= 'a' && c <= 'z')
386 c -= 'a' - 10;
387 else
388 return -1;
389
390 if (c < base)
391 return c;
392
393 return -1;
394 }
395
396 /**
397 * g_scanner_new:
398 * @config_templ: the initial scanner settings
399 *
400 * Creates a new #GScanner.
401 *
402 * The @config_templ structure specifies the initial settings
403 * of the scanner, which are copied into the #GScanner
404 * @config field. If you pass %NULL then the default settings
405 * are used.
406 *
407 * Returns: the new #GScanner
408 */
409 GScanner *
g_scanner_new(const GScannerConfig * config_templ)410 g_scanner_new (const GScannerConfig *config_templ)
411 {
412 GScanner *scanner;
413
414 if (!config_templ)
415 config_templ = &g_scanner_config_template;
416
417 scanner = g_new0 (GScanner, 1);
418
419 scanner->user_data = NULL;
420 scanner->max_parse_errors = 1;
421 scanner->parse_errors = 0;
422 scanner->input_name = NULL;
423 g_datalist_init (&scanner->qdata);
424
425 scanner->config = g_new0 (GScannerConfig, 1);
426
427 scanner->config->case_sensitive = config_templ->case_sensitive;
428 scanner->config->cset_skip_characters = config_templ->cset_skip_characters;
429 if (!scanner->config->cset_skip_characters)
430 scanner->config->cset_skip_characters = "";
431 scanner->config->cset_identifier_first = config_templ->cset_identifier_first;
432 scanner->config->cset_identifier_nth = config_templ->cset_identifier_nth;
433 scanner->config->cpair_comment_single = config_templ->cpair_comment_single;
434 scanner->config->skip_comment_multi = config_templ->skip_comment_multi;
435 scanner->config->skip_comment_single = config_templ->skip_comment_single;
436 scanner->config->scan_comment_multi = config_templ->scan_comment_multi;
437 scanner->config->scan_identifier = config_templ->scan_identifier;
438 scanner->config->scan_identifier_1char = config_templ->scan_identifier_1char;
439 scanner->config->scan_identifier_NULL = config_templ->scan_identifier_NULL;
440 scanner->config->scan_symbols = config_templ->scan_symbols;
441 scanner->config->scan_binary = config_templ->scan_binary;
442 scanner->config->scan_octal = config_templ->scan_octal;
443 scanner->config->scan_float = config_templ->scan_float;
444 scanner->config->scan_hex = config_templ->scan_hex;
445 scanner->config->scan_hex_dollar = config_templ->scan_hex_dollar;
446 scanner->config->scan_string_sq = config_templ->scan_string_sq;
447 scanner->config->scan_string_dq = config_templ->scan_string_dq;
448 scanner->config->numbers_2_int = config_templ->numbers_2_int;
449 scanner->config->int_2_float = config_templ->int_2_float;
450 scanner->config->identifier_2_string = config_templ->identifier_2_string;
451 scanner->config->char_2_token = config_templ->char_2_token;
452 scanner->config->symbol_2_token = config_templ->symbol_2_token;
453 scanner->config->scope_0_fallback = config_templ->scope_0_fallback;
454 scanner->config->store_int64 = config_templ->store_int64;
455
456 scanner->token = G_TOKEN_NONE;
457 scanner->value.v_int64 = 0;
458 scanner->line = 1;
459 scanner->position = 0;
460
461 scanner->next_token = G_TOKEN_NONE;
462 scanner->next_value.v_int64 = 0;
463 scanner->next_line = 1;
464 scanner->next_position = 0;
465
466 scanner->symbol_table = g_hash_table_new (g_scanner_key_hash, g_scanner_key_equal);
467 scanner->input_fd = -1;
468 scanner->text = NULL;
469 scanner->text_end = NULL;
470 scanner->buffer = NULL;
471 scanner->scope_id = 0;
472
473 scanner->msg_handler = g_scanner_msg_handler;
474
475 return scanner;
476 }
477
478 static inline void
g_scanner_free_value(GTokenType * token_p,GTokenValue * value_p)479 g_scanner_free_value (GTokenType *token_p,
480 GTokenValue *value_p)
481 {
482 switch (*token_p)
483 {
484 case G_TOKEN_STRING:
485 case G_TOKEN_IDENTIFIER:
486 case G_TOKEN_IDENTIFIER_NULL:
487 case G_TOKEN_COMMENT_SINGLE:
488 case G_TOKEN_COMMENT_MULTI:
489 g_free (value_p->v_string);
490 break;
491
492 default:
493 break;
494 }
495
496 *token_p = G_TOKEN_NONE;
497 }
498
499 static void
g_scanner_destroy_symbol_table_entry(gpointer _key,gpointer _value,gpointer _data)500 g_scanner_destroy_symbol_table_entry (gpointer _key,
501 gpointer _value,
502 gpointer _data)
503 {
504 GScannerKey *key = _key;
505
506 g_free (key->symbol);
507 g_free (key);
508 }
509
510 /**
511 * g_scanner_destroy:
512 * @scanner: a #GScanner
513 *
514 * Frees all memory used by the #GScanner.
515 */
516 void
g_scanner_destroy(GScanner * scanner)517 g_scanner_destroy (GScanner *scanner)
518 {
519 g_return_if_fail (scanner != NULL);
520
521 g_datalist_clear (&scanner->qdata);
522 g_hash_table_foreach (scanner->symbol_table,
523 g_scanner_destroy_symbol_table_entry, NULL);
524 g_hash_table_destroy (scanner->symbol_table);
525 g_scanner_free_value (&scanner->token, &scanner->value);
526 g_scanner_free_value (&scanner->next_token, &scanner->next_value);
527 g_free (scanner->config);
528 g_free (scanner->buffer);
529 g_free (scanner);
530 }
531
532 static void
g_scanner_msg_handler(GScanner * scanner,gchar * message,gboolean is_error)533 g_scanner_msg_handler (GScanner *scanner,
534 gchar *message,
535 gboolean is_error)
536 {
537 g_return_if_fail (scanner != NULL);
538
539 _g_fprintf (stderr, "%s:%d: ",
540 scanner->input_name ? scanner->input_name : "<memory>",
541 scanner->line);
542 if (is_error)
543 _g_fprintf (stderr, "error: ");
544 _g_fprintf (stderr, "%s\n", message);
545 }
546
547 /**
548 * g_scanner_error:
549 * @scanner: a #GScanner
550 * @format: the message format. See the printf() documentation
551 * @...: the parameters to insert into the format string
552 *
553 * Outputs an error message, via the #GScanner message handler.
554 */
555 void
g_scanner_error(GScanner * scanner,const gchar * format,...)556 g_scanner_error (GScanner *scanner,
557 const gchar *format,
558 ...)
559 {
560 g_return_if_fail (scanner != NULL);
561 g_return_if_fail (format != NULL);
562
563 scanner->parse_errors++;
564
565 if (scanner->msg_handler)
566 {
567 va_list args;
568 gchar *string;
569
570 va_start (args, format);
571 string = g_strdup_vprintf (format, args);
572 va_end (args);
573
574 scanner->msg_handler (scanner, string, TRUE);
575
576 g_free (string);
577 }
578 }
579
580 /**
581 * g_scanner_warn:
582 * @scanner: a #GScanner
583 * @format: the message format. See the printf() documentation
584 * @...: the parameters to insert into the format string
585 *
586 * Outputs a warning message, via the #GScanner message handler.
587 */
588 void
g_scanner_warn(GScanner * scanner,const gchar * format,...)589 g_scanner_warn (GScanner *scanner,
590 const gchar *format,
591 ...)
592 {
593 g_return_if_fail (scanner != NULL);
594 g_return_if_fail (format != NULL);
595
596 if (scanner->msg_handler)
597 {
598 va_list args;
599 gchar *string;
600
601 va_start (args, format);
602 string = g_strdup_vprintf (format, args);
603 va_end (args);
604
605 scanner->msg_handler (scanner, string, FALSE);
606
607 g_free (string);
608 }
609 }
610
611 static gboolean
g_scanner_key_equal(gconstpointer v1,gconstpointer v2)612 g_scanner_key_equal (gconstpointer v1,
613 gconstpointer v2)
614 {
615 const GScannerKey *key1 = v1;
616 const GScannerKey *key2 = v2;
617
618 return (key1->scope_id == key2->scope_id) && (strcmp (key1->symbol, key2->symbol) == 0);
619 }
620
621 static guint
g_scanner_key_hash(gconstpointer v)622 g_scanner_key_hash (gconstpointer v)
623 {
624 const GScannerKey *key = v;
625 gchar *c;
626 guint h;
627
628 h = key->scope_id;
629 for (c = key->symbol; *c; c++)
630 h = (h << 5) - h + *c;
631
632 return h;
633 }
634
635 static inline GScannerKey*
g_scanner_lookup_internal(GScanner * scanner,guint scope_id,const gchar * symbol)636 g_scanner_lookup_internal (GScanner *scanner,
637 guint scope_id,
638 const gchar *symbol)
639 {
640 GScannerKey *key_p;
641 GScannerKey key;
642
643 key.scope_id = scope_id;
644
645 if (!scanner->config->case_sensitive)
646 {
647 gchar *d;
648 const gchar *c;
649
650 key.symbol = g_new (gchar, strlen (symbol) + 1);
651 for (d = key.symbol, c = symbol; *c; c++, d++)
652 *d = to_lower (*c);
653 *d = 0;
654 key_p = g_hash_table_lookup (scanner->symbol_table, &key);
655 g_free (key.symbol);
656 }
657 else
658 {
659 key.symbol = (gchar*) symbol;
660 key_p = g_hash_table_lookup (scanner->symbol_table, &key);
661 }
662
663 return key_p;
664 }
665
666 /**
667 * g_scanner_add_symbol:
668 * @scanner: a #GScanner
669 * @symbol: the symbol to add
670 * @value: the value of the symbol
671 *
672 * Adds a symbol to the default scope.
673 *
674 * Deprecated: 2.2: Use g_scanner_scope_add_symbol() instead.
675 */
676
677 /**
678 * g_scanner_scope_add_symbol:
679 * @scanner: a #GScanner
680 * @scope_id: the scope id
681 * @symbol: the symbol to add
682 * @value: the value of the symbol
683 *
684 * Adds a symbol to the given scope.
685 */
686 void
g_scanner_scope_add_symbol(GScanner * scanner,guint scope_id,const gchar * symbol,gpointer value)687 g_scanner_scope_add_symbol (GScanner *scanner,
688 guint scope_id,
689 const gchar *symbol,
690 gpointer value)
691 {
692 GScannerKey *key;
693
694 g_return_if_fail (scanner != NULL);
695 g_return_if_fail (symbol != NULL);
696
697 key = g_scanner_lookup_internal (scanner, scope_id, symbol);
698
699 if (!key)
700 {
701 key = g_new (GScannerKey, 1);
702 key->scope_id = scope_id;
703 key->symbol = g_strdup (symbol);
704 key->value = value;
705 if (!scanner->config->case_sensitive)
706 {
707 gchar *c;
708
709 c = key->symbol;
710 while (*c != 0)
711 {
712 *c = to_lower (*c);
713 c++;
714 }
715 }
716 g_hash_table_add (scanner->symbol_table, key);
717 }
718 else
719 key->value = value;
720 }
721
722 /**
723 * g_scanner_remove_symbol:
724 * @scanner: a #GScanner
725 * @symbol: the symbol to remove
726 *
727 * Removes a symbol from the default scope.
728 *
729 * Deprecated: 2.2: Use g_scanner_scope_remove_symbol() instead.
730 */
731
732 /**
733 * g_scanner_scope_remove_symbol:
734 * @scanner: a #GScanner
735 * @scope_id: the scope id
736 * @symbol: the symbol to remove
737 *
738 * Removes a symbol from a scope.
739 */
740 void
g_scanner_scope_remove_symbol(GScanner * scanner,guint scope_id,const gchar * symbol)741 g_scanner_scope_remove_symbol (GScanner *scanner,
742 guint scope_id,
743 const gchar *symbol)
744 {
745 GScannerKey *key;
746
747 g_return_if_fail (scanner != NULL);
748 g_return_if_fail (symbol != NULL);
749
750 key = g_scanner_lookup_internal (scanner, scope_id, symbol);
751
752 if (key)
753 {
754 g_hash_table_remove (scanner->symbol_table, key);
755 g_free (key->symbol);
756 g_free (key);
757 }
758 }
759
760 /**
761 * g_scanner_freeze_symbol_table:
762 * @scanner: a #GScanner
763 *
764 * There is no reason to use this macro, since it does nothing.
765 *
766 * Deprecated: 2.2: This macro does nothing.
767 */
768
769 /**
770 * g_scanner_thaw_symbol_table:
771 * @scanner: a #GScanner
772 *
773 * There is no reason to use this macro, since it does nothing.
774 *
775 * Deprecated: 2.2: This macro does nothing.
776 */
777
778 /**
779 * g_scanner_lookup_symbol:
780 * @scanner: a #GScanner
781 * @symbol: the symbol to look up
782 *
783 * Looks up a symbol in the current scope and return its value.
784 * If the symbol is not bound in the current scope, %NULL is
785 * returned.
786 *
787 * Returns: the value of @symbol in the current scope, or %NULL
788 * if @symbol is not bound in the current scope
789 */
790 gpointer
g_scanner_lookup_symbol(GScanner * scanner,const gchar * symbol)791 g_scanner_lookup_symbol (GScanner *scanner,
792 const gchar *symbol)
793 {
794 GScannerKey *key;
795 guint scope_id;
796
797 g_return_val_if_fail (scanner != NULL, NULL);
798
799 if (!symbol)
800 return NULL;
801
802 scope_id = scanner->scope_id;
803 key = g_scanner_lookup_internal (scanner, scope_id, symbol);
804 if (!key && scope_id && scanner->config->scope_0_fallback)
805 key = g_scanner_lookup_internal (scanner, 0, symbol);
806
807 if (key)
808 return key->value;
809 else
810 return NULL;
811 }
812
813 /**
814 * g_scanner_scope_lookup_symbol:
815 * @scanner: a #GScanner
816 * @scope_id: the scope id
817 * @symbol: the symbol to look up
818 *
819 * Looks up a symbol in a scope and return its value. If the
820 * symbol is not bound in the scope, %NULL is returned.
821 *
822 * Returns: the value of @symbol in the given scope, or %NULL
823 * if @symbol is not bound in the given scope.
824 *
825 */
826 gpointer
g_scanner_scope_lookup_symbol(GScanner * scanner,guint scope_id,const gchar * symbol)827 g_scanner_scope_lookup_symbol (GScanner *scanner,
828 guint scope_id,
829 const gchar *symbol)
830 {
831 GScannerKey *key;
832
833 g_return_val_if_fail (scanner != NULL, NULL);
834
835 if (!symbol)
836 return NULL;
837
838 key = g_scanner_lookup_internal (scanner, scope_id, symbol);
839
840 if (key)
841 return key->value;
842 else
843 return NULL;
844 }
845
846 /**
847 * g_scanner_set_scope:
848 * @scanner: a #GScanner
849 * @scope_id: the new scope id
850 *
851 * Sets the current scope.
852 *
853 * Returns: the old scope id
854 */
855 guint
g_scanner_set_scope(GScanner * scanner,guint scope_id)856 g_scanner_set_scope (GScanner *scanner,
857 guint scope_id)
858 {
859 guint old_scope_id;
860
861 g_return_val_if_fail (scanner != NULL, 0);
862
863 old_scope_id = scanner->scope_id;
864 scanner->scope_id = scope_id;
865
866 return old_scope_id;
867 }
868
869 static void
g_scanner_foreach_internal(gpointer _key,gpointer _value,gpointer _user_data)870 g_scanner_foreach_internal (gpointer _key,
871 gpointer _value,
872 gpointer _user_data)
873 {
874 GScannerKey *key;
875 gpointer *d;
876 GHFunc func;
877 gpointer user_data;
878 guint *scope_id;
879
880 d = _user_data;
881 func = (GHFunc) d[0];
882 user_data = d[1];
883 scope_id = d[2];
884 key = _value;
885
886 if (key->scope_id == *scope_id)
887 func (key->symbol, key->value, user_data);
888 }
889
890 /**
891 * g_scanner_foreach_symbol:
892 * @scanner: a #GScanner
893 * @func: the function to call with each symbol
894 * @data: data to pass to the function
895 *
896 * Calls a function for each symbol in the default scope.
897 *
898 * Deprecated: 2.2: Use g_scanner_scope_foreach_symbol() instead.
899 */
900
901 /**
902 * g_scanner_scope_foreach_symbol:
903 * @scanner: a #GScanner
904 * @scope_id: the scope id
905 * @func: the function to call for each symbol/value pair
906 * @user_data: user data to pass to the function
907 *
908 * Calls the given function for each of the symbol/value pairs
909 * in the given scope of the #GScanner. The function is passed
910 * the symbol and value of each pair, and the given @user_data
911 * parameter.
912 */
913 void
g_scanner_scope_foreach_symbol(GScanner * scanner,guint scope_id,GHFunc func,gpointer user_data)914 g_scanner_scope_foreach_symbol (GScanner *scanner,
915 guint scope_id,
916 GHFunc func,
917 gpointer user_data)
918 {
919 gpointer d[3];
920
921 g_return_if_fail (scanner != NULL);
922
923 d[0] = (gpointer) func;
924 d[1] = user_data;
925 d[2] = &scope_id;
926
927 g_hash_table_foreach (scanner->symbol_table, g_scanner_foreach_internal, d);
928 }
929
930 /**
931 * g_scanner_peek_next_token:
932 * @scanner: a #GScanner
933 *
934 * Parses the next token, without removing it from the input stream.
935 * The token data is placed in the @next_token, @next_value, @next_line,
936 * and @next_position fields of the #GScanner structure.
937 *
938 * Note that, while the token is not removed from the input stream
939 * (i.e. the next call to g_scanner_get_next_token() will return the
940 * same token), it will not be reevaluated. This can lead to surprising
941 * results when changing scope or the scanner configuration after peeking
942 * the next token. Getting the next token after switching the scope or
943 * configuration will return whatever was peeked before, regardless of
944 * any symbols that may have been added or removed in the new scope.
945 *
946 * Returns: the type of the token
947 */
948 GTokenType
g_scanner_peek_next_token(GScanner * scanner)949 g_scanner_peek_next_token (GScanner *scanner)
950 {
951 g_return_val_if_fail (scanner != NULL, G_TOKEN_EOF);
952
953 if (scanner->next_token == G_TOKEN_NONE)
954 {
955 scanner->next_line = scanner->line;
956 scanner->next_position = scanner->position;
957 g_scanner_get_token_i (scanner,
958 &scanner->next_token,
959 &scanner->next_value,
960 &scanner->next_line,
961 &scanner->next_position);
962 }
963
964 return scanner->next_token;
965 }
966
967 /**
968 * g_scanner_get_next_token:
969 * @scanner: a #GScanner
970 *
971 * Parses the next token just like g_scanner_peek_next_token()
972 * and also removes it from the input stream. The token data is
973 * placed in the @token, @value, @line, and @position fields of
974 * the #GScanner structure.
975 *
976 * Returns: the type of the token
977 */
978 GTokenType
g_scanner_get_next_token(GScanner * scanner)979 g_scanner_get_next_token (GScanner *scanner)
980 {
981 g_return_val_if_fail (scanner != NULL, G_TOKEN_EOF);
982
983 if (scanner->next_token != G_TOKEN_NONE)
984 {
985 g_scanner_free_value (&scanner->token, &scanner->value);
986
987 scanner->token = scanner->next_token;
988 scanner->value = scanner->next_value;
989 scanner->line = scanner->next_line;
990 scanner->position = scanner->next_position;
991 scanner->next_token = G_TOKEN_NONE;
992 }
993 else
994 g_scanner_get_token_i (scanner,
995 &scanner->token,
996 &scanner->value,
997 &scanner->line,
998 &scanner->position);
999
1000 return scanner->token;
1001 }
1002
1003 /**
1004 * g_scanner_cur_token:
1005 * @scanner: a #GScanner
1006 *
1007 * Gets the current token type. This is simply the @token
1008 * field in the #GScanner structure.
1009 *
1010 * Returns: the current token type
1011 */
1012 GTokenType
g_scanner_cur_token(GScanner * scanner)1013 g_scanner_cur_token (GScanner *scanner)
1014 {
1015 g_return_val_if_fail (scanner != NULL, G_TOKEN_EOF);
1016
1017 return scanner->token;
1018 }
1019
1020 /**
1021 * g_scanner_cur_value:
1022 * @scanner: a #GScanner
1023 *
1024 * Gets the current token value. This is simply the @value
1025 * field in the #GScanner structure.
1026 *
1027 * Returns: the current token value
1028 */
1029 GTokenValue
g_scanner_cur_value(GScanner * scanner)1030 g_scanner_cur_value (GScanner *scanner)
1031 {
1032 GTokenValue v;
1033
1034 v.v_int64 = 0;
1035
1036 g_return_val_if_fail (scanner != NULL, v);
1037
1038 /* MSC isn't capable of handling return scanner->value; ? */
1039
1040 v = scanner->value;
1041
1042 return v;
1043 }
1044
1045 /**
1046 * g_scanner_cur_line:
1047 * @scanner: a #GScanner
1048 *
1049 * Returns the current line in the input stream (counting
1050 * from 1). This is the line of the last token parsed via
1051 * g_scanner_get_next_token().
1052 *
1053 * Returns: the current line
1054 */
1055 guint
g_scanner_cur_line(GScanner * scanner)1056 g_scanner_cur_line (GScanner *scanner)
1057 {
1058 g_return_val_if_fail (scanner != NULL, 0);
1059
1060 return scanner->line;
1061 }
1062
1063 /**
1064 * g_scanner_cur_position:
1065 * @scanner: a #GScanner
1066 *
1067 * Returns the current position in the current line (counting
1068 * from 0). This is the position of the last token parsed via
1069 * g_scanner_get_next_token().
1070 *
1071 * Returns: the current position on the line
1072 */
1073 guint
g_scanner_cur_position(GScanner * scanner)1074 g_scanner_cur_position (GScanner *scanner)
1075 {
1076 g_return_val_if_fail (scanner != NULL, 0);
1077
1078 return scanner->position;
1079 }
1080
1081 /**
1082 * g_scanner_eof:
1083 * @scanner: a #GScanner
1084 *
1085 * Returns %TRUE if the scanner has reached the end of
1086 * the file or text buffer.
1087 *
1088 * Returns: %TRUE if the scanner has reached the end of
1089 * the file or text buffer
1090 */
1091 gboolean
g_scanner_eof(GScanner * scanner)1092 g_scanner_eof (GScanner *scanner)
1093 {
1094 g_return_val_if_fail (scanner != NULL, TRUE);
1095
1096 return scanner->token == G_TOKEN_EOF || scanner->token == G_TOKEN_ERROR;
1097 }
1098
1099 /**
1100 * g_scanner_input_file:
1101 * @scanner: a #GScanner
1102 * @input_fd: a file descriptor
1103 *
1104 * Prepares to scan a file.
1105 */
1106 void
g_scanner_input_file(GScanner * scanner,gint input_fd)1107 g_scanner_input_file (GScanner *scanner,
1108 gint input_fd)
1109 {
1110 g_return_if_fail (scanner != NULL);
1111 g_return_if_fail (input_fd >= 0);
1112
1113 if (scanner->input_fd >= 0)
1114 g_scanner_sync_file_offset (scanner);
1115
1116 scanner->token = G_TOKEN_NONE;
1117 scanner->value.v_int64 = 0;
1118 scanner->line = 1;
1119 scanner->position = 0;
1120 scanner->next_token = G_TOKEN_NONE;
1121
1122 scanner->input_fd = input_fd;
1123 scanner->text = NULL;
1124 scanner->text_end = NULL;
1125
1126 if (!scanner->buffer)
1127 scanner->buffer = g_new (gchar, READ_BUFFER_SIZE + 1);
1128 }
1129
1130 /**
1131 * g_scanner_input_text:
1132 * @scanner: a #GScanner
1133 * @text: the text buffer to scan
1134 * @text_len: the length of the text buffer
1135 *
1136 * Prepares to scan a text buffer.
1137 */
1138 void
g_scanner_input_text(GScanner * scanner,const gchar * text,guint text_len)1139 g_scanner_input_text (GScanner *scanner,
1140 const gchar *text,
1141 guint text_len)
1142 {
1143 g_return_if_fail (scanner != NULL);
1144 if (text_len)
1145 g_return_if_fail (text != NULL);
1146 else
1147 text = NULL;
1148
1149 if (scanner->input_fd >= 0)
1150 g_scanner_sync_file_offset (scanner);
1151
1152 scanner->token = G_TOKEN_NONE;
1153 scanner->value.v_int64 = 0;
1154 scanner->line = 1;
1155 scanner->position = 0;
1156 scanner->next_token = G_TOKEN_NONE;
1157
1158 scanner->input_fd = -1;
1159 scanner->text = text;
1160 scanner->text_end = text + text_len;
1161
1162 if (scanner->buffer)
1163 {
1164 g_free (scanner->buffer);
1165 scanner->buffer = NULL;
1166 }
1167 }
1168
1169 static guchar
g_scanner_peek_next_char(GScanner * scanner)1170 g_scanner_peek_next_char (GScanner *scanner)
1171 {
1172 if (scanner->text < scanner->text_end)
1173 {
1174 return *scanner->text;
1175 }
1176 else if (scanner->input_fd >= 0)
1177 {
1178 gint count;
1179 gchar *buffer;
1180
1181 buffer = scanner->buffer;
1182 do
1183 {
1184 count = read (scanner->input_fd, buffer, READ_BUFFER_SIZE);
1185 }
1186 while (count == -1 && (errno == EINTR || errno == EAGAIN));
1187
1188 if (count < 1)
1189 {
1190 scanner->input_fd = -1;
1191
1192 return 0;
1193 }
1194 else
1195 {
1196 scanner->text = buffer;
1197 scanner->text_end = buffer + count;
1198
1199 return *buffer;
1200 }
1201 }
1202 else
1203 return 0;
1204 }
1205
1206 /**
1207 * g_scanner_sync_file_offset:
1208 * @scanner: a #GScanner
1209 *
1210 * Rewinds the filedescriptor to the current buffer position
1211 * and blows the file read ahead buffer. This is useful for
1212 * third party uses of the scanners filedescriptor, which hooks
1213 * onto the current scanning position.
1214 */
1215 void
g_scanner_sync_file_offset(GScanner * scanner)1216 g_scanner_sync_file_offset (GScanner *scanner)
1217 {
1218 g_return_if_fail (scanner != NULL);
1219
1220 /* for file input, rewind the filedescriptor to the current
1221 * buffer position and blow the file read ahead buffer. useful
1222 * for third party uses of our file descriptor, which hooks
1223 * onto the current scanning position.
1224 */
1225
1226 if (scanner->input_fd >= 0 && scanner->text_end > scanner->text)
1227 {
1228 gint buffered;
1229
1230 buffered = scanner->text_end - scanner->text;
1231 if (lseek (scanner->input_fd, - buffered, SEEK_CUR) >= 0)
1232 {
1233 /* we succeeded, blow our buffer's contents now */
1234 scanner->text = NULL;
1235 scanner->text_end = NULL;
1236 }
1237 else
1238 errno = 0;
1239 }
1240 }
1241
1242 static guchar
g_scanner_get_char(GScanner * scanner,guint * line_p,guint * position_p)1243 g_scanner_get_char (GScanner *scanner,
1244 guint *line_p,
1245 guint *position_p)
1246 {
1247 guchar fchar;
1248
1249 if (scanner->text < scanner->text_end)
1250 fchar = *(scanner->text++);
1251 else if (scanner->input_fd >= 0)
1252 {
1253 gint count;
1254 gchar *buffer;
1255
1256 buffer = scanner->buffer;
1257 do
1258 {
1259 count = read (scanner->input_fd, buffer, READ_BUFFER_SIZE);
1260 }
1261 while (count == -1 && (errno == EINTR || errno == EAGAIN));
1262
1263 if (count < 1)
1264 {
1265 scanner->input_fd = -1;
1266 fchar = 0;
1267 }
1268 else
1269 {
1270 scanner->text = buffer + 1;
1271 scanner->text_end = buffer + count;
1272 fchar = *buffer;
1273 if (!fchar)
1274 {
1275 g_scanner_sync_file_offset (scanner);
1276 scanner->text_end = scanner->text;
1277 scanner->input_fd = -1;
1278 }
1279 }
1280 }
1281 else
1282 fchar = 0;
1283
1284 if (fchar == '\n')
1285 {
1286 (*position_p) = 0;
1287 (*line_p)++;
1288 }
1289 else if (fchar)
1290 {
1291 (*position_p)++;
1292 }
1293
1294 return fchar;
1295 }
1296
1297 /**
1298 * g_scanner_unexp_token:
1299 * @scanner: a #GScanner
1300 * @expected_token: the expected token
1301 * @identifier_spec: a string describing how the scanner's user
1302 * refers to identifiers (%NULL defaults to "identifier").
1303 * This is used if @expected_token is %G_TOKEN_IDENTIFIER or
1304 * %G_TOKEN_IDENTIFIER_NULL.
1305 * @symbol_spec: a string describing how the scanner's user refers
1306 * to symbols (%NULL defaults to "symbol"). This is used if
1307 * @expected_token is %G_TOKEN_SYMBOL or any token value greater
1308 * than %G_TOKEN_LAST.
1309 * @symbol_name: the name of the symbol, if the scanner's current
1310 * token is a symbol.
1311 * @message: a message string to output at the end of the
1312 * warning/error, or %NULL.
1313 * @is_error: if %TRUE it is output as an error. If %FALSE it is
1314 * output as a warning.
1315 *
1316 * Outputs a message through the scanner's msg_handler,
1317 * resulting from an unexpected token in the input stream.
1318 * Note that you should not call g_scanner_peek_next_token()
1319 * followed by g_scanner_unexp_token() without an intermediate
1320 * call to g_scanner_get_next_token(), as g_scanner_unexp_token()
1321 * evaluates the scanner's current token (not the peeked token)
1322 * to construct part of the message.
1323 */
1324 void
g_scanner_unexp_token(GScanner * scanner,GTokenType expected_token,const gchar * identifier_spec,const gchar * symbol_spec,const gchar * symbol_name,const gchar * message,gint is_error)1325 g_scanner_unexp_token (GScanner *scanner,
1326 GTokenType expected_token,
1327 const gchar *identifier_spec,
1328 const gchar *symbol_spec,
1329 const gchar *symbol_name,
1330 const gchar *message,
1331 gint is_error)
1332 {
1333 gchar *token_string;
1334 guint token_string_len;
1335 gchar *expected_string;
1336 guint expected_string_len;
1337 gchar *message_prefix;
1338 gboolean print_unexp;
1339 void (*msg_handler) (GScanner*, const gchar*, ...);
1340
1341 g_return_if_fail (scanner != NULL);
1342
1343 if (is_error)
1344 msg_handler = g_scanner_error;
1345 else
1346 msg_handler = g_scanner_warn;
1347
1348 if (!identifier_spec)
1349 identifier_spec = "identifier";
1350 if (!symbol_spec)
1351 symbol_spec = "symbol";
1352
1353 token_string_len = 56;
1354 token_string = g_new (gchar, token_string_len + 1);
1355 expected_string_len = 64;
1356 expected_string = g_new (gchar, expected_string_len + 1);
1357 print_unexp = TRUE;
1358
1359 switch (scanner->token)
1360 {
1361 case G_TOKEN_EOF:
1362 _g_snprintf (token_string, token_string_len, "end of file");
1363 break;
1364
1365 default:
1366 if (scanner->token >= 1 && scanner->token <= 255)
1367 {
1368 if ((scanner->token >= ' ' && scanner->token <= '~') ||
1369 strchr (scanner->config->cset_identifier_first, scanner->token) ||
1370 strchr (scanner->config->cset_identifier_nth, scanner->token))
1371 _g_snprintf (token_string, token_string_len, "character '%c'", scanner->token);
1372 else
1373 _g_snprintf (token_string, token_string_len, "character '\\%o'", scanner->token);
1374 break;
1375 }
1376 else if (!scanner->config->symbol_2_token)
1377 {
1378 _g_snprintf (token_string, token_string_len, "(unknown) token <%d>", scanner->token);
1379 break;
1380 }
1381 /* fall through */
1382 case G_TOKEN_SYMBOL:
1383 if (expected_token == G_TOKEN_SYMBOL ||
1384 (scanner->config->symbol_2_token &&
1385 expected_token > G_TOKEN_LAST))
1386 print_unexp = FALSE;
1387 if (symbol_name)
1388 _g_snprintf (token_string,
1389 token_string_len,
1390 "%s%s '%s'",
1391 print_unexp ? "" : "invalid ",
1392 symbol_spec,
1393 symbol_name);
1394 else
1395 _g_snprintf (token_string,
1396 token_string_len,
1397 "%s%s",
1398 print_unexp ? "" : "invalid ",
1399 symbol_spec);
1400 break;
1401
1402 case G_TOKEN_ERROR:
1403 print_unexp = FALSE;
1404 expected_token = G_TOKEN_NONE;
1405 switch (scanner->value.v_error)
1406 {
1407 case G_ERR_UNEXP_EOF:
1408 _g_snprintf (token_string, token_string_len, "scanner: unexpected end of file");
1409 break;
1410
1411 case G_ERR_UNEXP_EOF_IN_STRING:
1412 _g_snprintf (token_string, token_string_len, "scanner: unterminated string constant");
1413 break;
1414
1415 case G_ERR_UNEXP_EOF_IN_COMMENT:
1416 _g_snprintf (token_string, token_string_len, "scanner: unterminated comment");
1417 break;
1418
1419 case G_ERR_NON_DIGIT_IN_CONST:
1420 _g_snprintf (token_string, token_string_len, "scanner: non digit in constant");
1421 break;
1422
1423 case G_ERR_FLOAT_RADIX:
1424 _g_snprintf (token_string, token_string_len, "scanner: invalid radix for floating constant");
1425 break;
1426
1427 case G_ERR_FLOAT_MALFORMED:
1428 _g_snprintf (token_string, token_string_len, "scanner: malformed floating constant");
1429 break;
1430
1431 case G_ERR_DIGIT_RADIX:
1432 _g_snprintf (token_string, token_string_len, "scanner: digit is beyond radix");
1433 break;
1434
1435 case G_ERR_UNKNOWN:
1436 default:
1437 _g_snprintf (token_string, token_string_len, "scanner: unknown error");
1438 break;
1439 }
1440 break;
1441
1442 case G_TOKEN_CHAR:
1443 _g_snprintf (token_string, token_string_len, "character '%c'", scanner->value.v_char);
1444 break;
1445
1446 case G_TOKEN_IDENTIFIER:
1447 case G_TOKEN_IDENTIFIER_NULL:
1448 if (expected_token == G_TOKEN_IDENTIFIER ||
1449 expected_token == G_TOKEN_IDENTIFIER_NULL)
1450 print_unexp = FALSE;
1451 _g_snprintf (token_string,
1452 token_string_len,
1453 "%s%s '%s'",
1454 print_unexp ? "" : "invalid ",
1455 identifier_spec,
1456 scanner->token == G_TOKEN_IDENTIFIER ? scanner->value.v_string : "null");
1457 break;
1458
1459 case G_TOKEN_BINARY:
1460 case G_TOKEN_OCTAL:
1461 case G_TOKEN_INT:
1462 case G_TOKEN_HEX:
1463 if (scanner->config->store_int64)
1464 _g_snprintf (token_string, token_string_len, "number '%" G_GUINT64_FORMAT "'", scanner->value.v_int64);
1465 else
1466 _g_snprintf (token_string, token_string_len, "number '%lu'", scanner->value.v_int);
1467 break;
1468
1469 case G_TOKEN_FLOAT:
1470 _g_snprintf (token_string, token_string_len, "number '%.3f'", scanner->value.v_float);
1471 break;
1472
1473 case G_TOKEN_STRING:
1474 if (expected_token == G_TOKEN_STRING)
1475 print_unexp = FALSE;
1476 _g_snprintf (token_string,
1477 token_string_len,
1478 "%s%sstring constant \"%s\"",
1479 print_unexp ? "" : "invalid ",
1480 scanner->value.v_string[0] == 0 ? "empty " : "",
1481 scanner->value.v_string);
1482 token_string[token_string_len - 2] = '"';
1483 token_string[token_string_len - 1] = 0;
1484 break;
1485
1486 case G_TOKEN_COMMENT_SINGLE:
1487 case G_TOKEN_COMMENT_MULTI:
1488 _g_snprintf (token_string, token_string_len, "comment");
1489 break;
1490
1491 case G_TOKEN_NONE:
1492 /* somehow the user's parsing code is screwed, there isn't much
1493 * we can do about it.
1494 * Note, a common case to trigger this is
1495 * g_scanner_peek_next_token(); g_scanner_unexp_token();
1496 * without an intermediate g_scanner_get_next_token().
1497 */
1498 g_assert_not_reached ();
1499 break;
1500 }
1501
1502
1503 switch (expected_token)
1504 {
1505 gboolean need_valid;
1506 gchar *tstring;
1507 case G_TOKEN_EOF:
1508 _g_snprintf (expected_string, expected_string_len, "end of file");
1509 break;
1510 default:
1511 if (expected_token >= 1 && expected_token <= 255)
1512 {
1513 if ((expected_token >= ' ' && expected_token <= '~') ||
1514 strchr (scanner->config->cset_identifier_first, expected_token) ||
1515 strchr (scanner->config->cset_identifier_nth, expected_token))
1516 _g_snprintf (expected_string, expected_string_len, "character '%c'", expected_token);
1517 else
1518 _g_snprintf (expected_string, expected_string_len, "character '\\%o'", expected_token);
1519 break;
1520 }
1521 else if (!scanner->config->symbol_2_token)
1522 {
1523 _g_snprintf (expected_string, expected_string_len, "(unknown) token <%d>", expected_token);
1524 break;
1525 }
1526 /* fall through */
1527 case G_TOKEN_SYMBOL:
1528 need_valid = (scanner->token == G_TOKEN_SYMBOL ||
1529 (scanner->config->symbol_2_token &&
1530 scanner->token > G_TOKEN_LAST));
1531 _g_snprintf (expected_string,
1532 expected_string_len,
1533 "%s%s",
1534 need_valid ? "valid " : "",
1535 symbol_spec);
1536 /* FIXME: should we attempt to look up the symbol_name for symbol_2_token? */
1537 break;
1538 case G_TOKEN_CHAR:
1539 _g_snprintf (expected_string, expected_string_len, "%scharacter",
1540 scanner->token == G_TOKEN_CHAR ? "valid " : "");
1541 break;
1542 case G_TOKEN_BINARY:
1543 tstring = "binary";
1544 _g_snprintf (expected_string, expected_string_len, "%snumber (%s)",
1545 scanner->token == expected_token ? "valid " : "", tstring);
1546 break;
1547 case G_TOKEN_OCTAL:
1548 tstring = "octal";
1549 _g_snprintf (expected_string, expected_string_len, "%snumber (%s)",
1550 scanner->token == expected_token ? "valid " : "", tstring);
1551 break;
1552 case G_TOKEN_INT:
1553 tstring = "integer";
1554 _g_snprintf (expected_string, expected_string_len, "%snumber (%s)",
1555 scanner->token == expected_token ? "valid " : "", tstring);
1556 break;
1557 case G_TOKEN_HEX:
1558 tstring = "hexadecimal";
1559 _g_snprintf (expected_string, expected_string_len, "%snumber (%s)",
1560 scanner->token == expected_token ? "valid " : "", tstring);
1561 break;
1562 case G_TOKEN_FLOAT:
1563 tstring = "float";
1564 _g_snprintf (expected_string, expected_string_len, "%snumber (%s)",
1565 scanner->token == expected_token ? "valid " : "", tstring);
1566 break;
1567 case G_TOKEN_STRING:
1568 _g_snprintf (expected_string,
1569 expected_string_len,
1570 "%sstring constant",
1571 scanner->token == G_TOKEN_STRING ? "valid " : "");
1572 break;
1573 case G_TOKEN_IDENTIFIER:
1574 case G_TOKEN_IDENTIFIER_NULL:
1575 need_valid = (scanner->token == G_TOKEN_IDENTIFIER_NULL ||
1576 scanner->token == G_TOKEN_IDENTIFIER);
1577 _g_snprintf (expected_string,
1578 expected_string_len,
1579 "%s%s",
1580 need_valid ? "valid " : "",
1581 identifier_spec);
1582 break;
1583 case G_TOKEN_COMMENT_SINGLE:
1584 tstring = "single-line";
1585 _g_snprintf (expected_string, expected_string_len, "%scomment (%s)",
1586 scanner->token == expected_token ? "valid " : "", tstring);
1587 break;
1588 case G_TOKEN_COMMENT_MULTI:
1589 tstring = "multi-line";
1590 _g_snprintf (expected_string, expected_string_len, "%scomment (%s)",
1591 scanner->token == expected_token ? "valid " : "", tstring);
1592 break;
1593 case G_TOKEN_NONE:
1594 case G_TOKEN_ERROR:
1595 /* this is handled upon printout */
1596 break;
1597 }
1598
1599 if (message && message[0] != 0)
1600 message_prefix = " - ";
1601 else
1602 {
1603 message_prefix = "";
1604 message = "";
1605 }
1606 if (expected_token == G_TOKEN_ERROR)
1607 {
1608 msg_handler (scanner,
1609 "failure around %s%s%s",
1610 token_string,
1611 message_prefix,
1612 message);
1613 }
1614 else if (expected_token == G_TOKEN_NONE)
1615 {
1616 if (print_unexp)
1617 msg_handler (scanner,
1618 "unexpected %s%s%s",
1619 token_string,
1620 message_prefix,
1621 message);
1622 else
1623 msg_handler (scanner,
1624 "%s%s%s",
1625 token_string,
1626 message_prefix,
1627 message);
1628 }
1629 else
1630 {
1631 if (print_unexp)
1632 msg_handler (scanner,
1633 "unexpected %s, expected %s%s%s",
1634 token_string,
1635 expected_string,
1636 message_prefix,
1637 message);
1638 else
1639 msg_handler (scanner,
1640 "%s, expected %s%s%s",
1641 token_string,
1642 expected_string,
1643 message_prefix,
1644 message);
1645 }
1646
1647 g_free (token_string);
1648 g_free (expected_string);
1649 }
1650
1651 static void
g_scanner_get_token_i(GScanner * scanner,GTokenType * token_p,GTokenValue * value_p,guint * line_p,guint * position_p)1652 g_scanner_get_token_i (GScanner *scanner,
1653 GTokenType *token_p,
1654 GTokenValue *value_p,
1655 guint *line_p,
1656 guint *position_p)
1657 {
1658 do
1659 {
1660 g_scanner_free_value (token_p, value_p);
1661 g_scanner_get_token_ll (scanner, token_p, value_p, line_p, position_p);
1662 }
1663 while (((*token_p > 0 && *token_p < 256) &&
1664 strchr (scanner->config->cset_skip_characters, *token_p)) ||
1665 (*token_p == G_TOKEN_CHAR &&
1666 strchr (scanner->config->cset_skip_characters, value_p->v_char)) ||
1667 (*token_p == G_TOKEN_COMMENT_MULTI &&
1668 scanner->config->skip_comment_multi) ||
1669 (*token_p == G_TOKEN_COMMENT_SINGLE &&
1670 scanner->config->skip_comment_single));
1671
1672 switch (*token_p)
1673 {
1674 case G_TOKEN_IDENTIFIER:
1675 if (scanner->config->identifier_2_string)
1676 *token_p = G_TOKEN_STRING;
1677 break;
1678
1679 case G_TOKEN_SYMBOL:
1680 if (scanner->config->symbol_2_token)
1681 *token_p = (GTokenType) value_p->v_symbol;
1682 break;
1683
1684 case G_TOKEN_BINARY:
1685 case G_TOKEN_OCTAL:
1686 case G_TOKEN_HEX:
1687 if (scanner->config->numbers_2_int)
1688 *token_p = G_TOKEN_INT;
1689 break;
1690
1691 default:
1692 break;
1693 }
1694
1695 if (*token_p == G_TOKEN_INT &&
1696 scanner->config->int_2_float)
1697 {
1698 *token_p = G_TOKEN_FLOAT;
1699 if (scanner->config->store_int64)
1700 {
1701 value_p->v_float = value_p->v_int64;
1702 }
1703 else
1704 value_p->v_float = value_p->v_int;
1705 }
1706
1707 errno = 0;
1708 }
1709
1710 static void
g_scanner_get_token_ll(GScanner * scanner,GTokenType * token_p,GTokenValue * value_p,guint * line_p,guint * position_p)1711 g_scanner_get_token_ll (GScanner *scanner,
1712 GTokenType *token_p,
1713 GTokenValue *value_p,
1714 guint *line_p,
1715 guint *position_p)
1716 {
1717 GScannerConfig *config;
1718 GTokenType token;
1719 gboolean in_comment_multi;
1720 gboolean in_comment_single;
1721 gboolean in_string_sq;
1722 gboolean in_string_dq;
1723 GString *gstring;
1724 GTokenValue value;
1725 guchar ch;
1726
1727 config = scanner->config;
1728 (*value_p).v_int64 = 0;
1729
1730 if ((scanner->text >= scanner->text_end && scanner->input_fd < 0) ||
1731 scanner->token == G_TOKEN_EOF)
1732 {
1733 *token_p = G_TOKEN_EOF;
1734 return;
1735 }
1736
1737 in_comment_multi = FALSE;
1738 in_comment_single = FALSE;
1739 in_string_sq = FALSE;
1740 in_string_dq = FALSE;
1741 gstring = NULL;
1742
1743 do /* while (ch != 0) */
1744 {
1745 gboolean dotted_float = FALSE;
1746
1747 ch = g_scanner_get_char (scanner, line_p, position_p);
1748
1749 value.v_int64 = 0;
1750 token = G_TOKEN_NONE;
1751
1752 /* this is *evil*, but needed ;(
1753 * we first check for identifier first character, because it
1754 * might interfere with other key chars like slashes or numbers
1755 */
1756 if (config->scan_identifier &&
1757 ch && strchr (config->cset_identifier_first, ch))
1758 goto identifier_precedence;
1759
1760 switch (ch)
1761 {
1762 case 0:
1763 token = G_TOKEN_EOF;
1764 (*position_p)++;
1765 /* ch = 0; */
1766 break;
1767
1768 case '/':
1769 if (!config->scan_comment_multi ||
1770 g_scanner_peek_next_char (scanner) != '*')
1771 goto default_case;
1772 g_scanner_get_char (scanner, line_p, position_p);
1773 token = G_TOKEN_COMMENT_MULTI;
1774 in_comment_multi = TRUE;
1775 gstring = g_string_new (NULL);
1776 while ((ch = g_scanner_get_char (scanner, line_p, position_p)) != 0)
1777 {
1778 if (ch == '*' && g_scanner_peek_next_char (scanner) == '/')
1779 {
1780 g_scanner_get_char (scanner, line_p, position_p);
1781 in_comment_multi = FALSE;
1782 break;
1783 }
1784 else
1785 gstring = g_string_append_c (gstring, ch);
1786 }
1787 ch = 0;
1788 break;
1789
1790 case '\'':
1791 if (!config->scan_string_sq)
1792 goto default_case;
1793 token = G_TOKEN_STRING;
1794 in_string_sq = TRUE;
1795 gstring = g_string_new (NULL);
1796 while ((ch = g_scanner_get_char (scanner, line_p, position_p)) != 0)
1797 {
1798 if (ch == '\'')
1799 {
1800 in_string_sq = FALSE;
1801 break;
1802 }
1803 else
1804 gstring = g_string_append_c (gstring, ch);
1805 }
1806 ch = 0;
1807 break;
1808
1809 case '"':
1810 if (!config->scan_string_dq)
1811 goto default_case;
1812 token = G_TOKEN_STRING;
1813 in_string_dq = TRUE;
1814 gstring = g_string_new (NULL);
1815 while ((ch = g_scanner_get_char (scanner, line_p, position_p)) != 0)
1816 {
1817 if (ch == '"')
1818 {
1819 in_string_dq = FALSE;
1820 break;
1821 }
1822 else
1823 {
1824 if (ch == '\\')
1825 {
1826 ch = g_scanner_get_char (scanner, line_p, position_p);
1827 switch (ch)
1828 {
1829 guint i;
1830 guint fchar;
1831
1832 case 0:
1833 break;
1834
1835 case '\\':
1836 gstring = g_string_append_c (gstring, '\\');
1837 break;
1838
1839 case 'n':
1840 gstring = g_string_append_c (gstring, '\n');
1841 break;
1842
1843 case 't':
1844 gstring = g_string_append_c (gstring, '\t');
1845 break;
1846
1847 case 'r':
1848 gstring = g_string_append_c (gstring, '\r');
1849 break;
1850
1851 case 'b':
1852 gstring = g_string_append_c (gstring, '\b');
1853 break;
1854
1855 case 'f':
1856 gstring = g_string_append_c (gstring, '\f');
1857 break;
1858
1859 case '0':
1860 case '1':
1861 case '2':
1862 case '3':
1863 case '4':
1864 case '5':
1865 case '6':
1866 case '7':
1867 i = ch - '0';
1868 fchar = g_scanner_peek_next_char (scanner);
1869 if (fchar >= '0' && fchar <= '7')
1870 {
1871 ch = g_scanner_get_char (scanner, line_p, position_p);
1872 i = i * 8 + ch - '0';
1873 fchar = g_scanner_peek_next_char (scanner);
1874 if (fchar >= '0' && fchar <= '7')
1875 {
1876 ch = g_scanner_get_char (scanner, line_p, position_p);
1877 i = i * 8 + ch - '0';
1878 }
1879 }
1880 gstring = g_string_append_c (gstring, i);
1881 break;
1882
1883 default:
1884 gstring = g_string_append_c (gstring, ch);
1885 break;
1886 }
1887 }
1888 else
1889 gstring = g_string_append_c (gstring, ch);
1890 }
1891 }
1892 ch = 0;
1893 break;
1894
1895 case '.':
1896 if (!config->scan_float)
1897 goto default_case;
1898 token = G_TOKEN_FLOAT;
1899 dotted_float = TRUE;
1900 ch = g_scanner_get_char (scanner, line_p, position_p);
1901 goto number_parsing;
1902
1903 case '$':
1904 if (!config->scan_hex_dollar)
1905 goto default_case;
1906 token = G_TOKEN_HEX;
1907 ch = g_scanner_get_char (scanner, line_p, position_p);
1908 goto number_parsing;
1909
1910 case '0':
1911 if (config->scan_octal)
1912 token = G_TOKEN_OCTAL;
1913 else
1914 token = G_TOKEN_INT;
1915 ch = g_scanner_peek_next_char (scanner);
1916 if (config->scan_hex && (ch == 'x' || ch == 'X'))
1917 {
1918 token = G_TOKEN_HEX;
1919 g_scanner_get_char (scanner, line_p, position_p);
1920 ch = g_scanner_get_char (scanner, line_p, position_p);
1921 if (ch == 0)
1922 {
1923 token = G_TOKEN_ERROR;
1924 value.v_error = G_ERR_UNEXP_EOF;
1925 (*position_p)++;
1926 break;
1927 }
1928 if (g_scanner_char_2_num (ch, 16) < 0)
1929 {
1930 token = G_TOKEN_ERROR;
1931 value.v_error = G_ERR_DIGIT_RADIX;
1932 ch = 0;
1933 break;
1934 }
1935 }
1936 else if (config->scan_binary && (ch == 'b' || ch == 'B'))
1937 {
1938 token = G_TOKEN_BINARY;
1939 g_scanner_get_char (scanner, line_p, position_p);
1940 ch = g_scanner_get_char (scanner, line_p, position_p);
1941 if (ch == 0)
1942 {
1943 token = G_TOKEN_ERROR;
1944 value.v_error = G_ERR_UNEXP_EOF;
1945 (*position_p)++;
1946 break;
1947 }
1948 if (g_scanner_char_2_num (ch, 10) < 0)
1949 {
1950 token = G_TOKEN_ERROR;
1951 value.v_error = G_ERR_NON_DIGIT_IN_CONST;
1952 ch = 0;
1953 break;
1954 }
1955 }
1956 else
1957 ch = '0';
1958 /* fall through */
1959 case '1':
1960 case '2':
1961 case '3':
1962 case '4':
1963 case '5':
1964 case '6':
1965 case '7':
1966 case '8':
1967 case '9':
1968 number_parsing:
1969 {
1970 gboolean in_number = TRUE;
1971 gchar *endptr;
1972
1973 if (token == G_TOKEN_NONE)
1974 token = G_TOKEN_INT;
1975
1976 gstring = g_string_new (dotted_float ? "0." : "");
1977 gstring = g_string_append_c (gstring, ch);
1978
1979 do /* while (in_number) */
1980 {
1981 gboolean is_E;
1982
1983 is_E = token == G_TOKEN_FLOAT && (ch == 'e' || ch == 'E');
1984
1985 ch = g_scanner_peek_next_char (scanner);
1986
1987 if (g_scanner_char_2_num (ch, 36) >= 0 ||
1988 (config->scan_float && ch == '.') ||
1989 (is_E && (ch == '+' || ch == '-')))
1990 {
1991 ch = g_scanner_get_char (scanner, line_p, position_p);
1992
1993 switch (ch)
1994 {
1995 case '.':
1996 if (token != G_TOKEN_INT && token != G_TOKEN_OCTAL)
1997 {
1998 value.v_error = token == G_TOKEN_FLOAT ? G_ERR_FLOAT_MALFORMED : G_ERR_FLOAT_RADIX;
1999 token = G_TOKEN_ERROR;
2000 in_number = FALSE;
2001 }
2002 else
2003 {
2004 token = G_TOKEN_FLOAT;
2005 gstring = g_string_append_c (gstring, ch);
2006 }
2007 break;
2008
2009 case '0':
2010 case '1':
2011 case '2':
2012 case '3':
2013 case '4':
2014 case '5':
2015 case '6':
2016 case '7':
2017 case '8':
2018 case '9':
2019 gstring = g_string_append_c (gstring, ch);
2020 break;
2021
2022 case '-':
2023 case '+':
2024 if (token != G_TOKEN_FLOAT)
2025 {
2026 token = G_TOKEN_ERROR;
2027 value.v_error = G_ERR_NON_DIGIT_IN_CONST;
2028 in_number = FALSE;
2029 }
2030 else
2031 gstring = g_string_append_c (gstring, ch);
2032 break;
2033
2034 case 'e':
2035 case 'E':
2036 if ((token != G_TOKEN_HEX && !config->scan_float) ||
2037 (token != G_TOKEN_HEX &&
2038 token != G_TOKEN_OCTAL &&
2039 token != G_TOKEN_FLOAT &&
2040 token != G_TOKEN_INT))
2041 {
2042 token = G_TOKEN_ERROR;
2043 value.v_error = G_ERR_NON_DIGIT_IN_CONST;
2044 in_number = FALSE;
2045 }
2046 else
2047 {
2048 if (token != G_TOKEN_HEX)
2049 token = G_TOKEN_FLOAT;
2050 gstring = g_string_append_c (gstring, ch);
2051 }
2052 break;
2053
2054 default:
2055 if (token != G_TOKEN_HEX)
2056 {
2057 token = G_TOKEN_ERROR;
2058 value.v_error = G_ERR_NON_DIGIT_IN_CONST;
2059 in_number = FALSE;
2060 }
2061 else
2062 gstring = g_string_append_c (gstring, ch);
2063 break;
2064 }
2065 }
2066 else
2067 in_number = FALSE;
2068 }
2069 while (in_number);
2070
2071 endptr = NULL;
2072 if (token == G_TOKEN_FLOAT)
2073 value.v_float = g_strtod (gstring->str, &endptr);
2074 else
2075 {
2076 guint64 ui64 = 0;
2077 switch (token)
2078 {
2079 case G_TOKEN_BINARY:
2080 ui64 = g_ascii_strtoull (gstring->str, &endptr, 2);
2081 break;
2082 case G_TOKEN_OCTAL:
2083 ui64 = g_ascii_strtoull (gstring->str, &endptr, 8);
2084 break;
2085 case G_TOKEN_INT:
2086 ui64 = g_ascii_strtoull (gstring->str, &endptr, 10);
2087 break;
2088 case G_TOKEN_HEX:
2089 ui64 = g_ascii_strtoull (gstring->str, &endptr, 16);
2090 break;
2091 default: ;
2092 }
2093 if (scanner->config->store_int64)
2094 value.v_int64 = ui64;
2095 else
2096 value.v_int = ui64;
2097 }
2098 if (endptr && *endptr)
2099 {
2100 token = G_TOKEN_ERROR;
2101 if (*endptr == 'e' || *endptr == 'E')
2102 value.v_error = G_ERR_NON_DIGIT_IN_CONST;
2103 else
2104 value.v_error = G_ERR_DIGIT_RADIX;
2105 }
2106 g_string_free (gstring, TRUE);
2107 gstring = NULL;
2108 ch = 0;
2109 } /* number_parsing:... */
2110 break;
2111
2112 default:
2113 default_case:
2114 {
2115 if (config->cpair_comment_single &&
2116 ch == config->cpair_comment_single[0])
2117 {
2118 token = G_TOKEN_COMMENT_SINGLE;
2119 in_comment_single = TRUE;
2120 gstring = g_string_new (NULL);
2121 ch = g_scanner_get_char (scanner, line_p, position_p);
2122 while (ch != 0)
2123 {
2124 if (ch == config->cpair_comment_single[1])
2125 {
2126 in_comment_single = FALSE;
2127 ch = 0;
2128 break;
2129 }
2130
2131 gstring = g_string_append_c (gstring, ch);
2132 ch = g_scanner_get_char (scanner, line_p, position_p);
2133 }
2134 /* ignore a missing newline at EOF for single line comments */
2135 if (in_comment_single &&
2136 config->cpair_comment_single[1] == '\n')
2137 in_comment_single = FALSE;
2138 }
2139 else if (config->scan_identifier && ch &&
2140 strchr (config->cset_identifier_first, ch))
2141 {
2142 identifier_precedence:
2143
2144 if (config->cset_identifier_nth && ch &&
2145 strchr (config->cset_identifier_nth,
2146 g_scanner_peek_next_char (scanner)))
2147 {
2148 token = G_TOKEN_IDENTIFIER;
2149 gstring = g_string_new (NULL);
2150 gstring = g_string_append_c (gstring, ch);
2151 do
2152 {
2153 ch = g_scanner_get_char (scanner, line_p, position_p);
2154 gstring = g_string_append_c (gstring, ch);
2155 ch = g_scanner_peek_next_char (scanner);
2156 }
2157 while (ch && strchr (config->cset_identifier_nth, ch));
2158 ch = 0;
2159 }
2160 else if (config->scan_identifier_1char)
2161 {
2162 token = G_TOKEN_IDENTIFIER;
2163 value.v_identifier = g_new0 (gchar, 2);
2164 value.v_identifier[0] = ch;
2165 ch = 0;
2166 }
2167 }
2168 if (ch)
2169 {
2170 if (config->char_2_token)
2171 token = ch;
2172 else
2173 {
2174 token = G_TOKEN_CHAR;
2175 value.v_char = ch;
2176 }
2177 ch = 0;
2178 }
2179 } /* default_case:... */
2180 break;
2181 }
2182 g_assert (ch == 0 && token != G_TOKEN_NONE); /* paranoid */
2183 }
2184 while (ch != 0);
2185
2186 if (in_comment_multi || in_comment_single ||
2187 in_string_sq || in_string_dq)
2188 {
2189 token = G_TOKEN_ERROR;
2190 if (gstring)
2191 {
2192 g_string_free (gstring, TRUE);
2193 gstring = NULL;
2194 }
2195 (*position_p)++;
2196 if (in_comment_multi || in_comment_single)
2197 value.v_error = G_ERR_UNEXP_EOF_IN_COMMENT;
2198 else /* (in_string_sq || in_string_dq) */
2199 value.v_error = G_ERR_UNEXP_EOF_IN_STRING;
2200 }
2201
2202 if (gstring)
2203 {
2204 value.v_string = g_string_free (gstring, FALSE);
2205 gstring = NULL;
2206 }
2207
2208 if (token == G_TOKEN_IDENTIFIER)
2209 {
2210 if (config->scan_symbols)
2211 {
2212 GScannerKey *key;
2213 guint scope_id;
2214
2215 scope_id = scanner->scope_id;
2216 key = g_scanner_lookup_internal (scanner, scope_id, value.v_identifier);
2217 if (!key && scope_id && scanner->config->scope_0_fallback)
2218 key = g_scanner_lookup_internal (scanner, 0, value.v_identifier);
2219
2220 if (key)
2221 {
2222 g_free (value.v_identifier);
2223 token = G_TOKEN_SYMBOL;
2224 value.v_symbol = key->value;
2225 }
2226 }
2227
2228 if (token == G_TOKEN_IDENTIFIER &&
2229 config->scan_identifier_NULL &&
2230 strlen (value.v_identifier) == 4)
2231 {
2232 gchar *null_upper = "NULL";
2233 gchar *null_lower = "null";
2234
2235 if (scanner->config->case_sensitive)
2236 {
2237 if (value.v_identifier[0] == null_upper[0] &&
2238 value.v_identifier[1] == null_upper[1] &&
2239 value.v_identifier[2] == null_upper[2] &&
2240 value.v_identifier[3] == null_upper[3])
2241 token = G_TOKEN_IDENTIFIER_NULL;
2242 }
2243 else
2244 {
2245 if ((value.v_identifier[0] == null_upper[0] ||
2246 value.v_identifier[0] == null_lower[0]) &&
2247 (value.v_identifier[1] == null_upper[1] ||
2248 value.v_identifier[1] == null_lower[1]) &&
2249 (value.v_identifier[2] == null_upper[2] ||
2250 value.v_identifier[2] == null_lower[2]) &&
2251 (value.v_identifier[3] == null_upper[3] ||
2252 value.v_identifier[3] == null_lower[3]))
2253 token = G_TOKEN_IDENTIFIER_NULL;
2254 }
2255 }
2256 }
2257
2258 *token_p = token;
2259 *value_p = value;
2260 }
2261