1 /* GLIB - Library of useful routines for C programming
2 * Copyright (C) 1995-1997 Peter Mattis, Spencer Kimball and Josh MacDonald
4 * GScanner: Flexible lexical scanner for general purpose.
5 * Copyright (C) 1997, 1998 Tim Janik
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 of the License, or (at your option) any later version.
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.
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with this library; if not, write to the
19 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
20 * Boston, MA 02111-1307, USA.
24 * Modified by the GLib Team and others 1997-2000. See the AUTHORS
25 * file for a list of people on the GLib Team. See the ChangeLog
26 * files for a list of changes. These files are distributed with
27 * GLib at ftp://ftp.gtk.org/pub/gtk/.
47 #include "gprintfint.h"
48 #include "gstrfuncs.h"
50 #include "gtestutils.h"
53 #include <io.h> /* For _read() */
59 * @title: Lexical Scanner
60 * @short_description: a general purpose lexical scanner
62 * The #GScanner and its associated functions provide a
63 * general purpose lexical scanner.
68 * @scanner: a #GScanner
69 * @message: the message
70 * @error: %TRUE if the message signals an error,
71 * %FALSE if it signals a warning.
73 * Specifies the type of the message handler function.
79 * The set of lowercase ASCII alphabet characters.
80 * Used for specifying valid identifier characters
87 * The set of uppercase ASCII alphabet characters.
88 * Used for specifying valid identifier characters
95 * The set of uppercase ISO 8859-1 alphabet characters
96 * which are not ASCII characters.
97 * Used for specifying valid identifier characters
104 * The set of lowercase ISO 8859-1 alphabet characters
105 * which are not ASCII characters.
106 * Used for specifying valid identifier characters
107 * in #GScannerConfig.
112 * @G_TOKEN_EOF: the end of the file
113 * @G_TOKEN_LEFT_PAREN: a '(' character
114 * @G_TOKEN_LEFT_CURLY: a '{' character
115 * @G_TOKEN_LEFT_BRACE: a '[' character
116 * @G_TOKEN_RIGHT_CURLY: a '}' character
117 * @G_TOKEN_RIGHT_PAREN: a ')' character
118 * @G_TOKEN_RIGHT_BRACE: a ']' character
119 * @G_TOKEN_EQUAL_SIGN: a '=' character
120 * @G_TOKEN_COMMA: a ',' character
121 * @G_TOKEN_NONE: not a token
122 * @G_TOKEN_ERROR: an error occurred
123 * @G_TOKEN_CHAR: a character
124 * @G_TOKEN_BINARY: a binary integer
125 * @G_TOKEN_OCTAL: an octal integer
126 * @G_TOKEN_INT: an integer
127 * @G_TOKEN_HEX: a hex integer
128 * @G_TOKEN_FLOAT: a floating point number
129 * @G_TOKEN_STRING: a string
130 * @G_TOKEN_SYMBOL: a symbol
131 * @G_TOKEN_IDENTIFIER: an identifier
132 * @G_TOKEN_IDENTIFIER_NULL: a null identifier
133 * @G_TOKEN_COMMENT_SINGLE: one line comment
134 * @G_TOKEN_COMMENT_MULTI: multi line comment
136 * The possible types of token returned from each
137 * g_scanner_get_next_token() call.
142 * @v_symbol: token symbol value
143 * @v_identifier: token identifier value
144 * @v_binary: token binary integer value
145 * @v_octal: octal integer value
146 * @v_int: integer value
147 * @v_int64: 64-bit integer value
148 * @v_float: floating point value
149 * @v_hex: hex integer value
150 * @v_string: string value
151 * @v_comment: comment value
152 * @v_char: character value
153 * @v_error: error value
155 * A union holding the value of the token.
160 * @G_ERR_UNKNOWN: unknown error
161 * @G_ERR_UNEXP_EOF: unexpected end of file
162 * @G_ERR_UNEXP_EOF_IN_STRING: unterminated string constant
163 * @G_ERR_UNEXP_EOF_IN_COMMENT: unterminated comment
164 * @G_ERR_NON_DIGIT_IN_CONST: non-digit character in a number
165 * @G_ERR_DIGIT_RADIX: digit beyond radix in a number
166 * @G_ERR_FLOAT_RADIX: non-decimal floating point number
167 * @G_ERR_FLOAT_MALFORMED: malformed floating point number
169 * The possible errors, used in the @v_error field
170 * of #GTokenValue, when the token is a %G_TOKEN_ERROR.
176 * @max_parse_errors: unused
177 * @parse_errors: g_scanner_error() increments this field
178 * @input_name: name of input stream, featured by the default message handler
179 * @qdata: quarked data
180 * @config: link into the scanner configuration
181 * @token: token parsed by the last g_scanner_get_next_token()
182 * @value: value of the last token from g_scanner_get_next_token()
183 * @line: line number of the last token from g_scanner_get_next_token()
184 * @position: char number of the last token from g_scanner_get_next_token()
185 * @next_token: token parsed by the last g_scanner_peek_next_token()
186 * @next_value: value of the last token from g_scanner_peek_next_token()
187 * @next_line: line number of the last token from g_scanner_peek_next_token()
188 * @next_position: char number of the last token from g_scanner_peek_next_token()
189 * @msg_handler: handler function for _warn and _error
191 * The data structure representing a lexical scanner.
193 * You should set @input_name after creating the scanner, since
194 * it is used by the default message handler when displaying
195 * warnings and errors. If you are scanning a file, the filename
196 * would be a good choice.
198 * The @user_data and @max_parse_errors fields are not used.
199 * If you need to associate extra data with the scanner you
200 * can place them here.
202 * If you want to use your own message handler you can set the
203 * @msg_handler field. The type of the message handler function
204 * is declared by #GScannerMsgFunc.
209 * @cset_skip_characters: specifies which characters should be skipped
210 * by the scanner (the default is the whitespace characters: space,
211 * tab, carriage-return and line-feed).
212 * @cset_identifier_first: specifies the characters which can start
213 * identifiers (the default is #G_CSET_a_2_z, "_", and #G_CSET_A_2_Z).
214 * @cset_identifier_nth: specifies the characters which can be used
215 * in identifiers, after the first character (the default is
216 * #G_CSET_a_2_z, "_0123456789", #G_CSET_A_2_Z, #G_CSET_LATINS,
218 * @cpair_comment_single: specifies the characters at the start and
219 * end of single-line comments. The default is "#\n" which means
220 * that single-line comments start with a '#' and continue until
221 * a '\n' (end of line).
222 * @case_sensitive: specifies if symbols are case sensitive (the
223 * default is %FALSE).
224 * @skip_comment_multi: specifies if multi-line comments are skipped
225 * and not returned as tokens (the default is %TRUE).
226 * @skip_comment_single: specifies if single-line comments are skipped
227 * and not returned as tokens (the default is %TRUE).
228 * @scan_comment_multi: specifies if multi-line comments are recognized
229 * (the default is %TRUE).
230 * @scan_identifier: specifies if identifiers are recognized (the
232 * @scan_identifier_1char: specifies if single-character
233 * identifiers are recognized (the default is %FALSE).
234 * @scan_identifier_NULL: specifies if %NULL is reported as
235 * %G_TOKEN_IDENTIFIER_NULL (the default is %FALSE).
236 * @scan_symbols: specifies if symbols are recognized (the default
238 * @scan_binary: specifies if binary numbers are recognized (the
239 * default is %FALSE).
240 * @scan_octal: specifies if octal numbers are recognized (the
242 * @scan_float: specifies if floating point numbers are recognized
243 * (the default is %TRUE).
244 * @scan_hex: specifies if hexadecimal numbers are recognized (the
246 * @scan_hex_dollar: specifies if '$' is recognized as a prefix for
247 * hexadecimal numbers (the default is %FALSE).
248 * @scan_string_sq: specifies if strings can be enclosed in single
249 * quotes (the default is %TRUE).
250 * @scan_string_dq: specifies if strings can be enclosed in double
251 * quotes (the default is %TRUE).
252 * @numbers_2_int: specifies if binary, octal and hexadecimal numbers
253 * are reported as #G_TOKEN_INT (the default is %TRUE).
254 * @int_2_float: specifies if all numbers are reported as %G_TOKEN_FLOAT
255 * (the default is %FALSE).
256 * @identifier_2_string: specifies if identifiers are reported as strings
257 * (the default is %FALSE).
258 * @char_2_token: specifies if characters are reported by setting
259 * <literal>token = ch</literal> or as %G_TOKEN_CHAR (the default
261 * @symbol_2_token: specifies if symbols are reported by setting
262 * <literal>token = v_symbol</literal> or as %G_TOKEN_SYMBOL (the
263 * default is %FALSE).
264 * @scope_0_fallback: specifies if a symbol is searched for in the
265 * default scope in addition to the current scope (the default is %FALSE).
266 * @store_int64: use value.v_int64 rather than v_int
268 * Specifies the #GScanner parser configuration. Most settings can
269 * be changed during the parsing phase and will affect the lexical
270 * parsing of the next unpeeked token.
273 /* --- defines --- */
274 #define to_lower(c) ( \
276 ( (((guchar)(c))>='A' && ((guchar)(c))<='Z') * ('a'-'A') ) | \
277 ( (((guchar)(c))>=192 && ((guchar)(c))<=214) * (224-192) ) | \
278 ( (((guchar)(c))>=216 && ((guchar)(c))<=222) * (248-216) ) | \
282 #define READ_BUFFER_SIZE (4000)
285 /* --- typedefs --- */
286 typedef struct _GScannerKey GScannerKey
;
296 /* --- variables --- */
297 static const GScannerConfig g_scanner_config_template
=
301 ) /* cset_skip_characters */,
306 ) /* cset_identifier_first */,
314 ) /* cset_identifier_nth */,
315 ( "#\n" ) /* cpair_comment_single */,
317 FALSE
/* case_sensitive */,
319 TRUE
/* skip_comment_multi */,
320 TRUE
/* skip_comment_single */,
321 TRUE
/* scan_comment_multi */,
322 TRUE
/* scan_identifier */,
323 FALSE
/* scan_identifier_1char */,
324 FALSE
/* scan_identifier_NULL */,
325 TRUE
/* scan_symbols */,
326 FALSE
/* scan_binary */,
327 TRUE
/* scan_octal */,
328 TRUE
/* scan_float */,
330 FALSE
/* scan_hex_dollar */,
331 TRUE
/* scan_string_sq */,
332 TRUE
/* scan_string_dq */,
333 TRUE
/* numbers_2_int */,
334 FALSE
/* int_2_float */,
335 FALSE
/* identifier_2_string */,
336 TRUE
/* char_2_token */,
337 FALSE
/* symbol_2_token */,
338 FALSE
/* scope_0_fallback */,
339 FALSE
/* store_int64 */,
343 /* --- prototypes --- */
345 GScannerKey
* g_scanner_lookup_internal (GScanner
*scanner
,
347 const gchar
*symbol
);
348 static gboolean
g_scanner_key_equal (gconstpointer v1
,
350 static guint
g_scanner_key_hash (gconstpointer v
);
351 static void g_scanner_get_token_ll (GScanner
*scanner
,
353 GTokenValue
*value_p
,
356 static void g_scanner_get_token_i (GScanner
*scanner
,
358 GTokenValue
*value_p
,
362 static guchar
g_scanner_peek_next_char (GScanner
*scanner
);
363 static guchar
g_scanner_get_char (GScanner
*scanner
,
366 static void g_scanner_msg_handler (GScanner
*scanner
,
371 /* --- functions --- */
373 g_scanner_char_2_num (guchar c
,
376 if (c
>= '0' && c
<= '9')
378 else if (c
>= 'A' && c
<= 'Z')
380 else if (c
>= 'a' && c
<= 'z')
393 * @config_templ: the initial scanner settings
395 * Creates a new #GScanner.
397 * The @config_templ structure specifies the initial settings
398 * of the scanner, which are copied into the #GScanner
399 * @config field. If you pass %NULL then the default settings
402 * Returns: the new #GScanner
405 g_scanner_new (const GScannerConfig
*config_templ
)
410 config_templ
= &g_scanner_config_template
;
412 scanner
= g_new0 (GScanner
, 1);
414 scanner
->user_data
= NULL
;
415 scanner
->max_parse_errors
= 1;
416 scanner
->parse_errors
= 0;
417 scanner
->input_name
= NULL
;
418 g_datalist_init (&scanner
->qdata
);
420 scanner
->config
= g_new0 (GScannerConfig
, 1);
422 scanner
->config
->case_sensitive
= config_templ
->case_sensitive
;
423 scanner
->config
->cset_skip_characters
= config_templ
->cset_skip_characters
;
424 if (!scanner
->config
->cset_skip_characters
)
425 scanner
->config
->cset_skip_characters
= "";
426 scanner
->config
->cset_identifier_first
= config_templ
->cset_identifier_first
;
427 scanner
->config
->cset_identifier_nth
= config_templ
->cset_identifier_nth
;
428 scanner
->config
->cpair_comment_single
= config_templ
->cpair_comment_single
;
429 scanner
->config
->skip_comment_multi
= config_templ
->skip_comment_multi
;
430 scanner
->config
->skip_comment_single
= config_templ
->skip_comment_single
;
431 scanner
->config
->scan_comment_multi
= config_templ
->scan_comment_multi
;
432 scanner
->config
->scan_identifier
= config_templ
->scan_identifier
;
433 scanner
->config
->scan_identifier_1char
= config_templ
->scan_identifier_1char
;
434 scanner
->config
->scan_identifier_NULL
= config_templ
->scan_identifier_NULL
;
435 scanner
->config
->scan_symbols
= config_templ
->scan_symbols
;
436 scanner
->config
->scan_binary
= config_templ
->scan_binary
;
437 scanner
->config
->scan_octal
= config_templ
->scan_octal
;
438 scanner
->config
->scan_float
= config_templ
->scan_float
;
439 scanner
->config
->scan_hex
= config_templ
->scan_hex
;
440 scanner
->config
->scan_hex_dollar
= config_templ
->scan_hex_dollar
;
441 scanner
->config
->scan_string_sq
= config_templ
->scan_string_sq
;
442 scanner
->config
->scan_string_dq
= config_templ
->scan_string_dq
;
443 scanner
->config
->numbers_2_int
= config_templ
->numbers_2_int
;
444 scanner
->config
->int_2_float
= config_templ
->int_2_float
;
445 scanner
->config
->identifier_2_string
= config_templ
->identifier_2_string
;
446 scanner
->config
->char_2_token
= config_templ
->char_2_token
;
447 scanner
->config
->symbol_2_token
= config_templ
->symbol_2_token
;
448 scanner
->config
->scope_0_fallback
= config_templ
->scope_0_fallback
;
449 scanner
->config
->store_int64
= config_templ
->store_int64
;
451 scanner
->token
= G_TOKEN_NONE
;
452 scanner
->value
.v_int64
= 0;
454 scanner
->position
= 0;
456 scanner
->next_token
= G_TOKEN_NONE
;
457 scanner
->next_value
.v_int64
= 0;
458 scanner
->next_line
= 1;
459 scanner
->next_position
= 0;
461 scanner
->symbol_table
= g_hash_table_new (g_scanner_key_hash
, g_scanner_key_equal
);
462 scanner
->input_fd
= -1;
463 scanner
->text
= NULL
;
464 scanner
->text_end
= NULL
;
465 scanner
->buffer
= NULL
;
466 scanner
->scope_id
= 0;
468 scanner
->msg_handler
= g_scanner_msg_handler
;
474 g_scanner_free_value (GTokenType
*token_p
,
475 GTokenValue
*value_p
)
480 case G_TOKEN_IDENTIFIER
:
481 case G_TOKEN_IDENTIFIER_NULL
:
482 case G_TOKEN_COMMENT_SINGLE
:
483 case G_TOKEN_COMMENT_MULTI
:
484 g_free (value_p
->v_string
);
491 *token_p
= G_TOKEN_NONE
;
495 g_scanner_destroy_symbol_table_entry (gpointer _key
,
499 GScannerKey
*key
= _key
;
501 g_free (key
->symbol
);
507 * @scanner: a #GScanner
509 * Frees all memory used by the #GScanner.
512 g_scanner_destroy (GScanner
*scanner
)
514 g_return_if_fail (scanner
!= NULL
);
516 g_datalist_clear (&scanner
->qdata
);
517 g_hash_table_foreach (scanner
->symbol_table
,
518 g_scanner_destroy_symbol_table_entry
, NULL
);
519 g_hash_table_destroy (scanner
->symbol_table
);
520 g_scanner_free_value (&scanner
->token
, &scanner
->value
);
521 g_scanner_free_value (&scanner
->next_token
, &scanner
->next_value
);
522 g_free (scanner
->config
);
523 g_free (scanner
->buffer
);
528 g_scanner_msg_handler (GScanner
*scanner
,
532 g_return_if_fail (scanner
!= NULL
);
534 _g_fprintf (stderr
, "%s:%d: ",
535 scanner
->input_name
? scanner
->input_name
: "<memory>",
538 _g_fprintf (stderr
, "error: ");
539 _g_fprintf (stderr
, "%s\n", message
);
544 * @scanner: a #GScanner
545 * @format: the message format. See the printf() documentation
546 * @...: the parameters to insert into the format string
548 * Outputs an error message, via the #GScanner message handler.
551 g_scanner_error (GScanner
*scanner
,
555 g_return_if_fail (scanner
!= NULL
);
556 g_return_if_fail (format
!= NULL
);
558 scanner
->parse_errors
++;
560 if (scanner
->msg_handler
)
565 va_start (args
, format
);
566 string
= g_strdup_vprintf (format
, args
);
569 scanner
->msg_handler (scanner
, string
, TRUE
);
577 * @scanner: a #GScanner
578 * @format: the message format. See the printf() documentation
579 * @...: the parameters to insert into the format string
581 * Outputs a warning message, via the #GScanner message handler.
584 g_scanner_warn (GScanner
*scanner
,
588 g_return_if_fail (scanner
!= NULL
);
589 g_return_if_fail (format
!= NULL
);
591 if (scanner
->msg_handler
)
596 va_start (args
, format
);
597 string
= g_strdup_vprintf (format
, args
);
600 scanner
->msg_handler (scanner
, string
, FALSE
);
607 g_scanner_key_equal (gconstpointer v1
,
610 const GScannerKey
*key1
= v1
;
611 const GScannerKey
*key2
= v2
;
613 return (key1
->scope_id
== key2
->scope_id
) && (strcmp (key1
->symbol
, key2
->symbol
) == 0);
617 g_scanner_key_hash (gconstpointer v
)
619 const GScannerKey
*key
= v
;
624 for (c
= key
->symbol
; *c
; c
++)
625 h
= (h
<< 5) - h
+ *c
;
630 static inline GScannerKey
*
631 g_scanner_lookup_internal (GScanner
*scanner
,
638 key
.scope_id
= scope_id
;
640 if (!scanner
->config
->case_sensitive
)
645 key
.symbol
= g_new (gchar
, strlen (symbol
) + 1);
646 for (d
= key
.symbol
, c
= symbol
; *c
; c
++, d
++)
649 key_p
= g_hash_table_lookup (scanner
->symbol_table
, &key
);
654 key
.symbol
= (gchar
*) symbol
;
655 key_p
= g_hash_table_lookup (scanner
->symbol_table
, &key
);
662 * g_scanner_add_symbol:
663 * @scanner: a #GScanner
664 * @symbol: the symbol to add
665 * @value: the value of the symbol
667 * Adds a symbol to the default scope.
669 * Deprecated: 2.2: Use g_scanner_scope_add_symbol() instead.
673 * g_scanner_scope_add_symbol:
674 * @scanner: a #GScanner
675 * @scope_id: the scope id
676 * @symbol: the symbol to add
677 * @value: the value of the symbol
679 * Adds a symbol to the given scope.
682 g_scanner_scope_add_symbol (GScanner
*scanner
,
689 g_return_if_fail (scanner
!= NULL
);
690 g_return_if_fail (symbol
!= NULL
);
692 key
= g_scanner_lookup_internal (scanner
, scope_id
, symbol
);
696 key
= g_new (GScannerKey
, 1);
697 key
->scope_id
= scope_id
;
698 key
->symbol
= g_strdup (symbol
);
700 if (!scanner
->config
->case_sensitive
)
711 g_hash_table_insert (scanner
->symbol_table
, key
, key
);
718 * g_scanner_remove_symbol:
719 * @scanner: a #GScanner
720 * @symbol: the symbol to remove
722 * Removes a symbol from the default scope.
724 * Deprecated: 2.2: Use g_scanner_scope_remove_symbol() instead.
728 * g_scanner_scope_remove_symbol:
729 * @scanner: a #GScanner
730 * @scope_id: the scope id
731 * @symbol: the symbol to remove
733 * Removes a symbol from a scope.
736 g_scanner_scope_remove_symbol (GScanner
*scanner
,
742 g_return_if_fail (scanner
!= NULL
);
743 g_return_if_fail (symbol
!= NULL
);
745 key
= g_scanner_lookup_internal (scanner
, scope_id
, symbol
);
749 g_hash_table_remove (scanner
->symbol_table
, key
);
750 g_free (key
->symbol
);
756 * g_scanner_freeze_symbol_table:
757 * @scanner: a #GScanner
759 * There is no reason to use this macro, since it does nothing.
761 * Deprecated: 2.2: This macro does nothing.
765 * g_scanner_thaw_symbol_table:
766 * @scanner: a #GScanner
768 * There is no reason to use this macro, since it does nothing.
770 * Deprecated: 2.2: This macro does nothing.
774 * g_scanner_lookup_symbol:
775 * @scanner: a #GScanner
776 * @symbol: the symbol to look up
778 * Looks up a symbol in the current scope and return its value.
779 * If the symbol is not bound in the current scope, %NULL is
782 * Returns: the value of @symbol in the current scope, or %NULL
783 * if @symbol is not bound in the current scope
786 g_scanner_lookup_symbol (GScanner
*scanner
,
792 g_return_val_if_fail (scanner
!= NULL
, NULL
);
797 scope_id
= scanner
->scope_id
;
798 key
= g_scanner_lookup_internal (scanner
, scope_id
, symbol
);
799 if (!key
&& scope_id
&& scanner
->config
->scope_0_fallback
)
800 key
= g_scanner_lookup_internal (scanner
, 0, symbol
);
809 * g_scanner_scope_lookup_symbol:
810 * @scanner: a #GScanner
811 * @scope_id: the scope id
812 * @symbol: the symbol to look up
814 * Looks up a symbol in a scope and return its value. If the
815 * symbol is not bound in the scope, %NULL is returned.
817 * Returns: the value of @symbol in the given scope, or %NULL
818 * if @symbol is not bound in the given scope.
822 g_scanner_scope_lookup_symbol (GScanner
*scanner
,
828 g_return_val_if_fail (scanner
!= NULL
, NULL
);
833 key
= g_scanner_lookup_internal (scanner
, scope_id
, symbol
);
842 * g_scanner_set_scope:
843 * @scanner: a #GScanner
844 * @scope_id: the new scope id
846 * Sets the current scope.
848 * Returns: the old scope id
851 g_scanner_set_scope (GScanner
*scanner
,
856 g_return_val_if_fail (scanner
!= NULL
, 0);
858 old_scope_id
= scanner
->scope_id
;
859 scanner
->scope_id
= scope_id
;
865 g_scanner_foreach_internal (gpointer _key
,
876 func
= (GHFunc
) d
[0];
881 if (key
->scope_id
== *scope_id
)
882 func (key
->symbol
, key
->value
, user_data
);
886 * g_scanner_foreach_symbol:
887 * @scanner: a #GScanner
888 * @func: the function to call with each symbol
889 * @data: data to pass to the function
891 * Calls a function for each symbol in the default scope.
893 * Deprecated: 2.2: Use g_scanner_scope_foreach_symbol() instead.
897 * g_scanner_scope_foreach_symbol:
898 * @scanner: a #GScanner
899 * @scope_id: the scope id
900 * @func: the function to call for each symbol/value pair
901 * @user_data: user data to pass to the function
903 * Calls the given function for each of the symbol/value pairs
904 * in the given scope of the #GScanner. The function is passed
905 * the symbol and value of each pair, and the given @user_data
909 g_scanner_scope_foreach_symbol (GScanner
*scanner
,
916 g_return_if_fail (scanner
!= NULL
);
918 d
[0] = (gpointer
) func
;
922 g_hash_table_foreach (scanner
->symbol_table
, g_scanner_foreach_internal
, d
);
926 * g_scanner_peek_next_token:
927 * @scanner: a #GScanner
929 * Parses the next token, without removing it from the input stream.
930 * The token data is placed in the @next_token, @next_value, @next_line,
931 * and @next_position fields of the #GScanner structure.
933 * Note that, while the token is not removed from the input stream
934 * (i.e. the next call to g_scanner_get_next_token() will return the
935 * same token), it will not be reevaluated. This can lead to surprising
936 * results when changing scope or the scanner configuration after peeking
937 * the next token. Getting the next token after switching the scope or
938 * configuration will return whatever was peeked before, regardless of
939 * any symbols that may have been added or removed in the new scope.
941 * Returns: the type of the token
944 g_scanner_peek_next_token (GScanner
*scanner
)
946 g_return_val_if_fail (scanner
!= NULL
, G_TOKEN_EOF
);
948 if (scanner
->next_token
== G_TOKEN_NONE
)
950 scanner
->next_line
= scanner
->line
;
951 scanner
->next_position
= scanner
->position
;
952 g_scanner_get_token_i (scanner
,
953 &scanner
->next_token
,
954 &scanner
->next_value
,
956 &scanner
->next_position
);
959 return scanner
->next_token
;
963 * g_scanner_get_next_token:
964 * @scanner: a #GScanner
966 * Parses the next token just like g_scanner_peek_next_token()
967 * and also removes it from the input stream. The token data is
968 * placed in the @token, @value, @line, and @position fields of
969 * the #GScanner structure.
971 * Returns: the type of the token
974 g_scanner_get_next_token (GScanner
*scanner
)
976 g_return_val_if_fail (scanner
!= NULL
, G_TOKEN_EOF
);
978 if (scanner
->next_token
!= G_TOKEN_NONE
)
980 g_scanner_free_value (&scanner
->token
, &scanner
->value
);
982 scanner
->token
= scanner
->next_token
;
983 scanner
->value
= scanner
->next_value
;
984 scanner
->line
= scanner
->next_line
;
985 scanner
->position
= scanner
->next_position
;
986 scanner
->next_token
= G_TOKEN_NONE
;
989 g_scanner_get_token_i (scanner
,
995 return scanner
->token
;
999 * g_scanner_cur_token:
1000 * @scanner: a #GScanner
1002 * Gets the current token type. This is simply the @token
1003 * field in the #GScanner structure.
1005 * Returns: the current token type
1008 g_scanner_cur_token (GScanner
*scanner
)
1010 g_return_val_if_fail (scanner
!= NULL
, G_TOKEN_EOF
);
1012 return scanner
->token
;
1016 * g_scanner_cur_value:
1017 * @scanner: a #GScanner
1019 * Gets the current token value. This is simply the @value
1020 * field in the #GScanner structure.
1022 * Returns: the current token value
1025 g_scanner_cur_value (GScanner
*scanner
)
1031 g_return_val_if_fail (scanner
!= NULL
, v
);
1033 /* MSC isn't capable of handling return scanner->value; ? */
1041 * g_scanner_cur_line:
1042 * @scanner: a #GScanner
1044 * Returns the current line in the input stream (counting
1045 * from 1). This is the line of the last token parsed via
1046 * g_scanner_get_next_token().
1048 * Returns: the current line
1051 g_scanner_cur_line (GScanner
*scanner
)
1053 g_return_val_if_fail (scanner
!= NULL
, 0);
1055 return scanner
->line
;
1059 * g_scanner_cur_position:
1060 * @scanner: a #GScanner
1062 * Returns the current position in the current line (counting
1063 * from 0). This is the position of the last token parsed via
1064 * g_scanner_get_next_token().
1066 * Returns: the current position on the line
1069 g_scanner_cur_position (GScanner
*scanner
)
1071 g_return_val_if_fail (scanner
!= NULL
, 0);
1073 return scanner
->position
;
1078 * @scanner: a #GScanner
1080 * Returns %TRUE if the scanner has reached the end of
1081 * the file or text buffer.
1083 * Returns: %TRUE if the scanner has reached the end of
1084 * the file or text buffer
1087 g_scanner_eof (GScanner
*scanner
)
1089 g_return_val_if_fail (scanner
!= NULL
, TRUE
);
1091 return scanner
->token
== G_TOKEN_EOF
|| scanner
->token
== G_TOKEN_ERROR
;
1095 * g_scanner_input_file:
1096 * @scanner: a #GScanner
1097 * @input_fd: a file descriptor
1099 * Prepares to scan a file.
1102 g_scanner_input_file (GScanner
*scanner
,
1105 g_return_if_fail (scanner
!= NULL
);
1106 g_return_if_fail (input_fd
>= 0);
1108 if (scanner
->input_fd
>= 0)
1109 g_scanner_sync_file_offset (scanner
);
1111 scanner
->token
= G_TOKEN_NONE
;
1112 scanner
->value
.v_int64
= 0;
1114 scanner
->position
= 0;
1115 scanner
->next_token
= G_TOKEN_NONE
;
1117 scanner
->input_fd
= input_fd
;
1118 scanner
->text
= NULL
;
1119 scanner
->text_end
= NULL
;
1121 if (!scanner
->buffer
)
1122 scanner
->buffer
= g_new (gchar
, READ_BUFFER_SIZE
+ 1);
1126 * g_scanner_input_text:
1127 * @scanner: a #GScanner
1128 * @text: the text buffer to scan
1129 * @text_len: the length of the text buffer
1131 * Prepares to scan a text buffer.
1134 g_scanner_input_text (GScanner
*scanner
,
1138 g_return_if_fail (scanner
!= NULL
);
1140 g_return_if_fail (text
!= NULL
);
1144 if (scanner
->input_fd
>= 0)
1145 g_scanner_sync_file_offset (scanner
);
1147 scanner
->token
= G_TOKEN_NONE
;
1148 scanner
->value
.v_int64
= 0;
1150 scanner
->position
= 0;
1151 scanner
->next_token
= G_TOKEN_NONE
;
1153 scanner
->input_fd
= -1;
1154 scanner
->text
= text
;
1155 scanner
->text_end
= text
+ text_len
;
1157 if (scanner
->buffer
)
1159 g_free (scanner
->buffer
);
1160 scanner
->buffer
= NULL
;
1165 g_scanner_peek_next_char (GScanner
*scanner
)
1167 if (scanner
->text
< scanner
->text_end
)
1169 return *scanner
->text
;
1171 else if (scanner
->input_fd
>= 0)
1176 buffer
= scanner
->buffer
;
1179 count
= read (scanner
->input_fd
, buffer
, READ_BUFFER_SIZE
);
1181 while (count
== -1 && (errno
== EINTR
|| errno
== EAGAIN
));
1185 scanner
->input_fd
= -1;
1191 scanner
->text
= buffer
;
1192 scanner
->text_end
= buffer
+ count
;
1202 * g_scanner_sync_file_offset:
1203 * @scanner: a #GScanner
1205 * Rewinds the filedescriptor to the current buffer position
1206 * and blows the file read ahead buffer. This is useful for
1207 * third party uses of the scanners filedescriptor, which hooks
1208 * onto the current scanning position.
1211 g_scanner_sync_file_offset (GScanner
*scanner
)
1213 g_return_if_fail (scanner
!= NULL
);
1215 /* for file input, rewind the filedescriptor to the current
1216 * buffer position and blow the file read ahead buffer. useful
1217 * for third party uses of our file descriptor, which hooks
1218 * onto the current scanning position.
1221 if (scanner
->input_fd
>= 0 && scanner
->text_end
> scanner
->text
)
1225 buffered
= scanner
->text_end
- scanner
->text
;
1226 if (lseek (scanner
->input_fd
, - buffered
, SEEK_CUR
) >= 0)
1228 /* we succeeded, blow our buffer's contents now */
1229 scanner
->text
= NULL
;
1230 scanner
->text_end
= NULL
;
1238 g_scanner_get_char (GScanner
*scanner
,
1244 if (scanner
->text
< scanner
->text_end
)
1245 fchar
= *(scanner
->text
++);
1246 else if (scanner
->input_fd
>= 0)
1251 buffer
= scanner
->buffer
;
1254 count
= read (scanner
->input_fd
, buffer
, READ_BUFFER_SIZE
);
1256 while (count
== -1 && (errno
== EINTR
|| errno
== EAGAIN
));
1260 scanner
->input_fd
= -1;
1265 scanner
->text
= buffer
+ 1;
1266 scanner
->text_end
= buffer
+ count
;
1270 g_scanner_sync_file_offset (scanner
);
1271 scanner
->text_end
= scanner
->text
;
1272 scanner
->input_fd
= -1;
1293 * g_scanner_unexp_token:
1294 * @scanner: a #GScanner
1295 * @expected_token: the expected token
1296 * @identifier_spec: a string describing how the scanner's user
1297 * refers to identifiers (%NULL defaults to "identifier").
1298 * This is used if @expected_token is %G_TOKEN_IDENTIFIER or
1299 * %G_TOKEN_IDENTIFIER_NULL.
1300 * @symbol_spec: a string describing how the scanner's user refers
1301 * to symbols (%NULL defaults to "symbol"). This is used if
1302 * @expected_token is %G_TOKEN_SYMBOL or any token value greater
1303 * than %G_TOKEN_LAST.
1304 * @symbol_name: the name of the symbol, if the scanner's current
1305 * token is a symbol.
1306 * @message: a message string to output at the end of the
1307 * warning/error, or %NULL.
1308 * @is_error: if %TRUE it is output as an error. If %FALSE it is
1309 * output as a warning.
1311 * Outputs a message through the scanner's msg_handler,
1312 * resulting from an unexpected token in the input stream.
1313 * Note that you should not call g_scanner_peek_next_token()
1314 * followed by g_scanner_unexp_token() without an intermediate
1315 * call to g_scanner_get_next_token(), as g_scanner_unexp_token()
1316 * evaluates the scanner's current token (not the peeked token)
1317 * to construct part of the message.
1320 g_scanner_unexp_token (GScanner
*scanner
,
1321 GTokenType expected_token
,
1322 const gchar
*identifier_spec
,
1323 const gchar
*symbol_spec
,
1324 const gchar
*symbol_name
,
1325 const gchar
*message
,
1328 gchar
*token_string
;
1329 guint token_string_len
;
1330 gchar
*expected_string
;
1331 guint expected_string_len
;
1332 gchar
*message_prefix
;
1333 gboolean print_unexp
;
1334 void (*msg_handler
) (GScanner
*, const gchar
*, ...);
1336 g_return_if_fail (scanner
!= NULL
);
1339 msg_handler
= g_scanner_error
;
1341 msg_handler
= g_scanner_warn
;
1343 if (!identifier_spec
)
1344 identifier_spec
= "identifier";
1346 symbol_spec
= "symbol";
1348 token_string_len
= 56;
1349 token_string
= g_new (gchar
, token_string_len
+ 1);
1350 expected_string_len
= 64;
1351 expected_string
= g_new (gchar
, expected_string_len
+ 1);
1354 switch (scanner
->token
)
1357 _g_snprintf (token_string
, token_string_len
, "end of file");
1361 if (scanner
->token
>= 1 && scanner
->token
<= 255)
1363 if ((scanner
->token
>= ' ' && scanner
->token
<= '~') ||
1364 strchr (scanner
->config
->cset_identifier_first
, scanner
->token
) ||
1365 strchr (scanner
->config
->cset_identifier_nth
, scanner
->token
))
1366 _g_snprintf (token_string
, token_string_len
, "character '%c'", scanner
->token
);
1368 _g_snprintf (token_string
, token_string_len
, "character '\\%o'", scanner
->token
);
1371 else if (!scanner
->config
->symbol_2_token
)
1373 _g_snprintf (token_string
, token_string_len
, "(unknown) token <%d>", scanner
->token
);
1377 case G_TOKEN_SYMBOL
:
1378 if (expected_token
== G_TOKEN_SYMBOL
||
1379 (scanner
->config
->symbol_2_token
&&
1380 expected_token
> G_TOKEN_LAST
))
1381 print_unexp
= FALSE
;
1383 _g_snprintf (token_string
,
1386 print_unexp
? "" : "invalid ",
1390 _g_snprintf (token_string
,
1393 print_unexp
? "" : "invalid ",
1398 print_unexp
= FALSE
;
1399 expected_token
= G_TOKEN_NONE
;
1400 switch (scanner
->value
.v_error
)
1402 case G_ERR_UNEXP_EOF
:
1403 _g_snprintf (token_string
, token_string_len
, "scanner: unexpected end of file");
1406 case G_ERR_UNEXP_EOF_IN_STRING
:
1407 _g_snprintf (token_string
, token_string_len
, "scanner: unterminated string constant");
1410 case G_ERR_UNEXP_EOF_IN_COMMENT
:
1411 _g_snprintf (token_string
, token_string_len
, "scanner: unterminated comment");
1414 case G_ERR_NON_DIGIT_IN_CONST
:
1415 _g_snprintf (token_string
, token_string_len
, "scanner: non digit in constant");
1418 case G_ERR_FLOAT_RADIX
:
1419 _g_snprintf (token_string
, token_string_len
, "scanner: invalid radix for floating constant");
1422 case G_ERR_FLOAT_MALFORMED
:
1423 _g_snprintf (token_string
, token_string_len
, "scanner: malformed floating constant");
1426 case G_ERR_DIGIT_RADIX
:
1427 _g_snprintf (token_string
, token_string_len
, "scanner: digit is beyond radix");
1432 _g_snprintf (token_string
, token_string_len
, "scanner: unknown error");
1438 _g_snprintf (token_string
, token_string_len
, "character '%c'", scanner
->value
.v_char
);
1441 case G_TOKEN_IDENTIFIER
:
1442 case G_TOKEN_IDENTIFIER_NULL
:
1443 if (expected_token
== G_TOKEN_IDENTIFIER
||
1444 expected_token
== G_TOKEN_IDENTIFIER_NULL
)
1445 print_unexp
= FALSE
;
1446 _g_snprintf (token_string
,
1449 print_unexp
? "" : "invalid ",
1451 scanner
->token
== G_TOKEN_IDENTIFIER
? scanner
->value
.v_string
: "null");
1454 case G_TOKEN_BINARY
:
1458 if (scanner
->config
->store_int64
)
1459 _g_snprintf (token_string
, token_string_len
, "number '%" G_GUINT64_FORMAT
"'", scanner
->value
.v_int64
);
1461 _g_snprintf (token_string
, token_string_len
, "number '%lu'", scanner
->value
.v_int
);
1465 _g_snprintf (token_string
, token_string_len
, "number '%.3f'", scanner
->value
.v_float
);
1468 case G_TOKEN_STRING
:
1469 if (expected_token
== G_TOKEN_STRING
)
1470 print_unexp
= FALSE
;
1471 _g_snprintf (token_string
,
1473 "%s%sstring constant \"%s\"",
1474 print_unexp
? "" : "invalid ",
1475 scanner
->value
.v_string
[0] == 0 ? "empty " : "",
1476 scanner
->value
.v_string
);
1477 token_string
[token_string_len
- 2] = '"';
1478 token_string
[token_string_len
- 1] = 0;
1481 case G_TOKEN_COMMENT_SINGLE
:
1482 case G_TOKEN_COMMENT_MULTI
:
1483 _g_snprintf (token_string
, token_string_len
, "comment");
1487 /* somehow the user's parsing code is screwed, there isn't much
1488 * we can do about it.
1489 * Note, a common case to trigger this is
1490 * g_scanner_peek_next_token(); g_scanner_unexp_token();
1491 * without an intermediate g_scanner_get_next_token().
1493 g_assert_not_reached ();
1498 switch (expected_token
)
1500 gboolean need_valid
;
1503 _g_snprintf (expected_string
, expected_string_len
, "end of file");
1506 if (expected_token
>= 1 && expected_token
<= 255)
1508 if ((expected_token
>= ' ' && expected_token
<= '~') ||
1509 strchr (scanner
->config
->cset_identifier_first
, expected_token
) ||
1510 strchr (scanner
->config
->cset_identifier_nth
, expected_token
))
1511 _g_snprintf (expected_string
, expected_string_len
, "character '%c'", expected_token
);
1513 _g_snprintf (expected_string
, expected_string_len
, "character '\\%o'", expected_token
);
1516 else if (!scanner
->config
->symbol_2_token
)
1518 _g_snprintf (expected_string
, expected_string_len
, "(unknown) token <%d>", expected_token
);
1522 case G_TOKEN_SYMBOL
:
1523 need_valid
= (scanner
->token
== G_TOKEN_SYMBOL
||
1524 (scanner
->config
->symbol_2_token
&&
1525 scanner
->token
> G_TOKEN_LAST
));
1526 _g_snprintf (expected_string
,
1527 expected_string_len
,
1529 need_valid
? "valid " : "",
1531 /* FIXME: should we attempt to lookup the symbol_name for symbol_2_token? */
1534 _g_snprintf (expected_string
, expected_string_len
, "%scharacter",
1535 scanner
->token
== G_TOKEN_CHAR
? "valid " : "");
1537 case G_TOKEN_BINARY
:
1539 _g_snprintf (expected_string
, expected_string_len
, "%snumber (%s)",
1540 scanner
->token
== expected_token
? "valid " : "", tstring
);
1544 _g_snprintf (expected_string
, expected_string_len
, "%snumber (%s)",
1545 scanner
->token
== expected_token
? "valid " : "", tstring
);
1548 tstring
= "integer";
1549 _g_snprintf (expected_string
, expected_string_len
, "%snumber (%s)",
1550 scanner
->token
== expected_token
? "valid " : "", tstring
);
1553 tstring
= "hexadecimal";
1554 _g_snprintf (expected_string
, expected_string_len
, "%snumber (%s)",
1555 scanner
->token
== expected_token
? "valid " : "", tstring
);
1559 _g_snprintf (expected_string
, expected_string_len
, "%snumber (%s)",
1560 scanner
->token
== expected_token
? "valid " : "", tstring
);
1562 case G_TOKEN_STRING
:
1563 _g_snprintf (expected_string
,
1564 expected_string_len
,
1565 "%sstring constant",
1566 scanner
->token
== G_TOKEN_STRING
? "valid " : "");
1568 case G_TOKEN_IDENTIFIER
:
1569 case G_TOKEN_IDENTIFIER_NULL
:
1570 need_valid
= (scanner
->token
== G_TOKEN_IDENTIFIER_NULL
||
1571 scanner
->token
== G_TOKEN_IDENTIFIER
);
1572 _g_snprintf (expected_string
,
1573 expected_string_len
,
1575 need_valid
? "valid " : "",
1578 case G_TOKEN_COMMENT_SINGLE
:
1579 tstring
= "single-line";
1580 _g_snprintf (expected_string
, expected_string_len
, "%scomment (%s)",
1581 scanner
->token
== expected_token
? "valid " : "", tstring
);
1583 case G_TOKEN_COMMENT_MULTI
:
1584 tstring
= "multi-line";
1585 _g_snprintf (expected_string
, expected_string_len
, "%scomment (%s)",
1586 scanner
->token
== expected_token
? "valid " : "", tstring
);
1590 /* this is handled upon printout */
1594 if (message
&& message
[0] != 0)
1595 message_prefix
= " - ";
1598 message_prefix
= "";
1601 if (expected_token
== G_TOKEN_ERROR
)
1603 msg_handler (scanner
,
1604 "failure around %s%s%s",
1609 else if (expected_token
== G_TOKEN_NONE
)
1612 msg_handler (scanner
,
1613 "unexpected %s%s%s",
1618 msg_handler (scanner
,
1627 msg_handler (scanner
,
1628 "unexpected %s, expected %s%s%s",
1634 msg_handler (scanner
,
1635 "%s, expected %s%s%s",
1642 g_free (token_string
);
1643 g_free (expected_string
);
1647 g_scanner_get_token_i (GScanner
*scanner
,
1648 GTokenType
*token_p
,
1649 GTokenValue
*value_p
,
1655 g_scanner_free_value (token_p
, value_p
);
1656 g_scanner_get_token_ll (scanner
, token_p
, value_p
, line_p
, position_p
);
1658 while (((*token_p
> 0 && *token_p
< 256) &&
1659 strchr (scanner
->config
->cset_skip_characters
, *token_p
)) ||
1660 (*token_p
== G_TOKEN_CHAR
&&
1661 strchr (scanner
->config
->cset_skip_characters
, value_p
->v_char
)) ||
1662 (*token_p
== G_TOKEN_COMMENT_MULTI
&&
1663 scanner
->config
->skip_comment_multi
) ||
1664 (*token_p
== G_TOKEN_COMMENT_SINGLE
&&
1665 scanner
->config
->skip_comment_single
));
1669 case G_TOKEN_IDENTIFIER
:
1670 if (scanner
->config
->identifier_2_string
)
1671 *token_p
= G_TOKEN_STRING
;
1674 case G_TOKEN_SYMBOL
:
1675 if (scanner
->config
->symbol_2_token
)
1676 *token_p
= (GTokenType
) value_p
->v_symbol
;
1679 case G_TOKEN_BINARY
:
1682 if (scanner
->config
->numbers_2_int
)
1683 *token_p
= G_TOKEN_INT
;
1690 if (*token_p
== G_TOKEN_INT
&&
1691 scanner
->config
->int_2_float
)
1693 *token_p
= G_TOKEN_FLOAT
;
1694 if (scanner
->config
->store_int64
)
1697 /* work around error C2520, see gvaluetransform.c */
1698 value_p
->v_float
= (__int64
)value_p
->v_int64
;
1700 value_p
->v_float
= value_p
->v_int64
;
1704 value_p
->v_float
= value_p
->v_int
;
1711 g_scanner_get_token_ll (GScanner
*scanner
,
1712 GTokenType
*token_p
,
1713 GTokenValue
*value_p
,
1717 GScannerConfig
*config
;
1719 gboolean in_comment_multi
;
1720 gboolean in_comment_single
;
1721 gboolean in_string_sq
;
1722 gboolean in_string_dq
;
1727 config
= scanner
->config
;
1728 (*value_p
).v_int64
= 0;
1730 if ((scanner
->text
>= scanner
->text_end
&& scanner
->input_fd
< 0) ||
1731 scanner
->token
== G_TOKEN_EOF
)
1733 *token_p
= G_TOKEN_EOF
;
1737 in_comment_multi
= FALSE
;
1738 in_comment_single
= FALSE
;
1739 in_string_sq
= FALSE
;
1740 in_string_dq
= FALSE
;
1743 do /* while (ch != 0) */
1745 gboolean dotted_float
= FALSE
;
1747 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1750 token
= G_TOKEN_NONE
;
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
1756 if (config
->scan_identifier
&&
1757 ch
&& strchr (config
->cset_identifier_first
, ch
))
1758 goto identifier_precedence
;
1763 token
= G_TOKEN_EOF
;
1769 if (!config
->scan_comment_multi
||
1770 g_scanner_peek_next_char (scanner
) != '*')
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)
1778 if (ch
== '*' && g_scanner_peek_next_char (scanner
) == '/')
1780 g_scanner_get_char (scanner
, line_p
, position_p
);
1781 in_comment_multi
= FALSE
;
1785 gstring
= g_string_append_c (gstring
, ch
);
1791 if (!config
->scan_string_sq
)
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)
1800 in_string_sq
= FALSE
;
1804 gstring
= g_string_append_c (gstring
, ch
);
1810 if (!config
->scan_string_dq
)
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)
1819 in_string_dq
= FALSE
;
1826 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1836 gstring
= g_string_append_c (gstring
, '\\');
1840 gstring
= g_string_append_c (gstring
, '\n');
1844 gstring
= g_string_append_c (gstring
, '\t');
1848 gstring
= g_string_append_c (gstring
, '\r');
1852 gstring
= g_string_append_c (gstring
, '\b');
1856 gstring
= g_string_append_c (gstring
, '\f');
1868 fchar
= g_scanner_peek_next_char (scanner
);
1869 if (fchar
>= '0' && fchar
<= '7')
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')
1876 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1877 i
= i
* 8 + ch
- '0';
1880 gstring
= g_string_append_c (gstring
, i
);
1884 gstring
= g_string_append_c (gstring
, ch
);
1889 gstring
= g_string_append_c (gstring
, ch
);
1896 if (!config
->scan_float
)
1898 token
= G_TOKEN_FLOAT
;
1899 dotted_float
= TRUE
;
1900 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1901 goto number_parsing
;
1904 if (!config
->scan_hex_dollar
)
1906 token
= G_TOKEN_HEX
;
1907 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1908 goto number_parsing
;
1911 if (config
->scan_octal
)
1912 token
= G_TOKEN_OCTAL
;
1914 token
= G_TOKEN_INT
;
1915 ch
= g_scanner_peek_next_char (scanner
);
1916 if (config
->scan_hex
&& (ch
== 'x' || ch
== 'X'))
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
);
1923 token
= G_TOKEN_ERROR
;
1924 value
.v_error
= G_ERR_UNEXP_EOF
;
1928 if (g_scanner_char_2_num (ch
, 16) < 0)
1930 token
= G_TOKEN_ERROR
;
1931 value
.v_error
= G_ERR_DIGIT_RADIX
;
1936 else if (config
->scan_binary
&& (ch
== 'b' || ch
== 'B'))
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
);
1943 token
= G_TOKEN_ERROR
;
1944 value
.v_error
= G_ERR_UNEXP_EOF
;
1948 if (g_scanner_char_2_num (ch
, 10) < 0)
1950 token
= G_TOKEN_ERROR
;
1951 value
.v_error
= G_ERR_NON_DIGIT_IN_CONST
;
1970 gboolean in_number
= TRUE
;
1973 if (token
== G_TOKEN_NONE
)
1974 token
= G_TOKEN_INT
;
1976 gstring
= g_string_new (dotted_float
? "0." : "");
1977 gstring
= g_string_append_c (gstring
, ch
);
1979 do /* while (in_number) */
1983 is_E
= token
== G_TOKEN_FLOAT
&& (ch
== 'e' || ch
== 'E');
1985 ch
= g_scanner_peek_next_char (scanner
);
1987 if (g_scanner_char_2_num (ch
, 36) >= 0 ||
1988 (config
->scan_float
&& ch
== '.') ||
1989 (is_E
&& (ch
== '+' || ch
== '-')))
1991 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
1996 if (token
!= G_TOKEN_INT
&& token
!= G_TOKEN_OCTAL
)
1998 value
.v_error
= token
== G_TOKEN_FLOAT
? G_ERR_FLOAT_MALFORMED
: G_ERR_FLOAT_RADIX
;
1999 token
= G_TOKEN_ERROR
;
2004 token
= G_TOKEN_FLOAT
;
2005 gstring
= g_string_append_c (gstring
, ch
);
2019 gstring
= g_string_append_c (gstring
, ch
);
2024 if (token
!= G_TOKEN_FLOAT
)
2026 token
= G_TOKEN_ERROR
;
2027 value
.v_error
= G_ERR_NON_DIGIT_IN_CONST
;
2031 gstring
= g_string_append_c (gstring
, ch
);
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
))
2042 token
= G_TOKEN_ERROR
;
2043 value
.v_error
= G_ERR_NON_DIGIT_IN_CONST
;
2048 if (token
!= G_TOKEN_HEX
)
2049 token
= G_TOKEN_FLOAT
;
2050 gstring
= g_string_append_c (gstring
, ch
);
2055 if (token
!= G_TOKEN_HEX
)
2057 token
= G_TOKEN_ERROR
;
2058 value
.v_error
= G_ERR_NON_DIGIT_IN_CONST
;
2062 gstring
= g_string_append_c (gstring
, ch
);
2072 if (token
== G_TOKEN_FLOAT
)
2073 value
.v_float
= g_strtod (gstring
->str
, &endptr
);
2079 case G_TOKEN_BINARY
:
2080 ui64
= g_ascii_strtoull (gstring
->str
, &endptr
, 2);
2083 ui64
= g_ascii_strtoull (gstring
->str
, &endptr
, 8);
2086 ui64
= g_ascii_strtoull (gstring
->str
, &endptr
, 10);
2089 ui64
= g_ascii_strtoull (gstring
->str
, &endptr
, 16);
2093 if (scanner
->config
->store_int64
)
2094 value
.v_int64
= ui64
;
2098 if (endptr
&& *endptr
)
2100 token
= G_TOKEN_ERROR
;
2101 if (*endptr
== 'e' || *endptr
== 'E')
2102 value
.v_error
= G_ERR_NON_DIGIT_IN_CONST
;
2104 value
.v_error
= G_ERR_DIGIT_RADIX
;
2106 g_string_free (gstring
, TRUE
);
2109 } /* number_parsing:... */
2115 if (config
->cpair_comment_single
&&
2116 ch
== config
->cpair_comment_single
[0])
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
);
2124 if (ch
== config
->cpair_comment_single
[1])
2126 in_comment_single
= FALSE
;
2131 gstring
= g_string_append_c (gstring
, ch
);
2132 ch
= g_scanner_get_char (scanner
, line_p
, position_p
);
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
;
2139 else if (config
->scan_identifier
&& ch
&&
2140 strchr (config
->cset_identifier_first
, ch
))
2142 identifier_precedence
:
2144 if (config
->cset_identifier_nth
&& ch
&&
2145 strchr (config
->cset_identifier_nth
,
2146 g_scanner_peek_next_char (scanner
)))
2148 token
= G_TOKEN_IDENTIFIER
;
2149 gstring
= g_string_new (NULL
);
2150 gstring
= g_string_append_c (gstring
, ch
);
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
);
2157 while (ch
&& strchr (config
->cset_identifier_nth
, ch
));
2160 else if (config
->scan_identifier_1char
)
2162 token
= G_TOKEN_IDENTIFIER
;
2163 value
.v_identifier
= g_new0 (gchar
, 2);
2164 value
.v_identifier
[0] = ch
;
2170 if (config
->char_2_token
)
2174 token
= G_TOKEN_CHAR
;
2179 } /* default_case:... */
2182 g_assert (ch
== 0 && token
!= G_TOKEN_NONE
); /* paranoid */
2186 if (in_comment_multi
|| in_comment_single
||
2187 in_string_sq
|| in_string_dq
)
2189 token
= G_TOKEN_ERROR
;
2192 g_string_free (gstring
, TRUE
);
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
;
2204 value
.v_string
= g_string_free (gstring
, FALSE
);
2208 if (token
== G_TOKEN_IDENTIFIER
)
2210 if (config
->scan_symbols
)
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
);
2222 g_free (value
.v_identifier
);
2223 token
= G_TOKEN_SYMBOL
;
2224 value
.v_symbol
= key
->value
;
2228 if (token
== G_TOKEN_IDENTIFIER
&&
2229 config
->scan_identifier_NULL
&&
2230 strlen (value
.v_identifier
) == 4)
2232 gchar
*null_upper
= "NULL";
2233 gchar
*null_lower
= "null";
2235 if (scanner
->config
->case_sensitive
)
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
;
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
;