3 * an API for text tvb parsers
5 * Copyright 2005, Luis E. Garcia Ontanon <luis@ontanon.org>
9 * Wireshark - Network traffic analyzer
10 * By Gerald Combs <gerald@wireshark.org>
11 * Copyright 1998 Gerald Combs
13 * This program is free software; you can redistribute it and/or
14 * modify it under the terms of the GNU General Public License
15 * as published by the Free Software Foundation; either version 2
16 * of the License, or (at your option) any later version.
18 * This program is distributed in the hope that it will be useful,
19 * but WITHOUT ANY WARRANTY; without even the implied warranty of
20 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 * GNU General Public License for more details.
23 * You should have received a copy of the GNU General Public License
24 * along with this program; if not, write to the Free Software
25 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
29 The intention behind this is to ease the writing of dissectors that have to
30 parse text without the need of writing into buffers.
32 It was originally written to avoid using lex and yacc for the xml dissector.
34 the parser is able to look for wanted elements these can be:
37 - a char out of a string of needles
38 - a char not belonging to a string of needles
39 - a sequence of chars that belong to a set of chars
40 - a sequence of chars that do not belong to a set of chars
43 - all the characters up to a certain wanted element (included or excluded)
46 - one of a given group of wanted elements
47 - a sequence of wanted elements
48 - some (at least one) instances of a wanted element
50 Once a wanted element is successfully extracted, by either tvbparse_get or
51 tvbparse_find, the parser will invoke a given callback
52 before and another one after every of its component's subelement's callbacks
55 If tvbparse_get or tvbparse_find fail to extract the wanted element the
56 subelements callbacks are not going to be invoked.
58 The wanted elements are instantiated once by the proto_register_xxx function.
60 The parser is instantiated for every packet and it mantains its state.
62 The element's data is destroyed before the next packet is dissected.
68 #include <epan/tvbuff.h>
71 typedef struct _tvbparse_elem_t tvbparse_elem_t;
72 typedef struct _tvbparse_wanted_t tvbparse_wanted_t;
73 typedef struct _tvbparse_t tvbparse_t;
77 * a callback function to be called before or after an element has been
78 * successfuly extracted.
80 * Note that if the token belongs to a composed token the callbacks of the
81 * components won't be called unless the composed token is successfully
84 * tvbparse_data: the private data of the parser
85 * wanted_data: the private data of the wanted element
86 * elem: the extracted element
88 typedef void (*tvbparse_action_t)(void* tvbparse_data, const void* wanted_data, struct _tvbparse_elem_t* elem);
90 typedef int (*tvbparse_condition_t)
91 (tvbparse_t*, const int,
92 const tvbparse_wanted_t*,
97 TP_UNTIL_INCLUDE, /* last elem is included, its span is spent by the parser */
98 TP_UNTIL_SPEND, /* last elem is not included, but its span is spent by the parser */
99 TP_UNTIL_LEAVE /* last elem is not included, neither its span is spent by the parser */
103 struct _tvbparse_wanted_t {
105 tvbparse_condition_t condition;
109 struct _tvbparse_wanted_t** handle;
116 gboolean (*comp)(void*,const void*);
117 void* (*extract)(tvbuff_t*,guint);
122 const tvbparse_wanted_t* subelem;
126 struct _tvbparse_wanted_t* key;
127 struct _tvbparse_wanted_t* other;
130 const tvbparse_wanted_t* subelem;
141 tvbparse_action_t before;
142 tvbparse_action_t after;
145 /* an instance of a per packet parser */
151 const tvbparse_wanted_t* ignore;
155 /* a matching token returned by either tvbparser_get or tvb_parser_find */
156 struct _tvbparse_elem_t {
165 struct _tvbparse_elem_t* sub;
167 struct _tvbparse_elem_t* next;
168 struct _tvbparse_elem_t* last;
170 const tvbparse_wanted_t* wanted;
175 * definition of wanted token types
177 * the following functions define the tokens we will be able to look for in a tvb
178 * common parameters are:
180 * id: an arbitrary id that will be copied to the eventual token (don't use 0)
181 * private_data: persistent data to be passed to the callback action (wanted_data)
182 * before_cb: an callback function to be called before those of the subelements
183 * after_cb: an callback function to be called after those of the subelements
190 * When looked for it returns a simple element one character long if the char
191 * at the current offset matches one of the the needles.
193 tvbparse_wanted_t* tvbparse_char(const int id,
194 const gchar* needles,
195 const void* private_data,
196 tvbparse_action_t before_cb,
197 tvbparse_action_t after_cb);
200 * a not_char element.
202 * When looked for it returns a simple element one character long if the char
203 * at the current offset does not match one of the the needles.
205 tvbparse_wanted_t* tvbparse_not_char(const int id,
207 const void* private_data,
208 tvbparse_action_t before_cb,
209 tvbparse_action_t after_cb);
214 * When looked for it returns a simple element one or more characters long if
215 * one or more char(s) starting from the current offset match one of the needles.
216 * An element will be returned if at least min_len chars are given (1 if it's 0)
217 * It will get at most max_len chars or as much as it can if max_len is 0.
219 tvbparse_wanted_t* tvbparse_chars(const int id,
222 const gchar* needles,
223 const void* private_data,
224 tvbparse_action_t before_cb,
225 tvbparse_action_t after_cb);
228 * a not_chars element
230 * When looked for it returns a simple element one or more characters long if
231 * one or more char(s) starting from the current offset do not match one of the
233 * An element will be returned if at least min_len chars are given (1 if it's 0)
234 * It will get at most max_len chars or as much as it can if max_len is 0.
236 tvbparse_wanted_t* tvbparse_not_chars(const int id,
239 const gchar* needles,
240 const void* private_data,
241 tvbparse_action_t before_cb,
242 tvbparse_action_t after_cb);
247 * When looked for it returns a simple element if we have the given string at
250 tvbparse_wanted_t* tvbparse_string(const int id,
252 const void* private_data,
253 tvbparse_action_t before_cb,
254 tvbparse_action_t after_cb);
259 * When looked for it returns a simple element if we have a matching string at
262 tvbparse_wanted_t* tvbparse_casestring(const int id,
265 tvbparse_action_t before_cb,
266 tvbparse_action_t after_cb);
271 * When looked for it returns a simple element containing all the characters
272 * found until the first match of the ending element if the ending element is
275 * When looking for until elements it calls tvbparse_find so it can be very slow.
277 * It won't have a subelement, the ending's callbacks won't get called.
281 * op_mode values determine how the terminating element and the current offset
282 * of the parser are handled
284 tvbparse_wanted_t* tvbparse_until(const int id,
285 const void* private_data,
286 tvbparse_action_t before_cb,
287 tvbparse_action_t after_cb,
288 const tvbparse_wanted_t* ending,
289 until_mode_t until_mode);
294 * When looked for it will try to match to the given candidates and return a
295 * composed element whose subelement is the first match.
297 * The list of candidates is terminated with a NULL
300 tvbparse_wanted_t* tvbparse_set_oneof(const int id,
301 const void* private_data,
302 tvbparse_action_t before_cb,
303 tvbparse_action_t after_cb,
309 tvbparse_wanted_t* tvbparse_hashed(const int id,
311 tvbparse_action_t before_cb,
312 tvbparse_action_t after_cb,
313 tvbparse_wanted_t* key,
314 tvbparse_wanted_t* other,
317 void tvbparse_hashed_add(tvbparse_wanted_t* w, ...);
322 * When looked for it will try to match in order all the given candidates. If
323 * every candidate is found in the given order it will return a composed
324 * element whose subelements are the matcheed elemets.
326 * The list of candidates is terminated with a NULL.
329 tvbparse_wanted_t* tvbparse_set_seq(const int id,
330 const void* private_data,
331 tvbparse_action_t before_cb,
332 tvbparse_action_t after_cb,
338 * When looked for it will try to match the given candidate at least min times
339 * and at most max times. If the given candidate is matched at least min times
340 * a composed element is returned.
343 tvbparse_wanted_t* tvbparse_some(const int id,
346 const void* private_data,
347 tvbparse_action_t before_cb,
348 tvbparse_action_t after_cb,
349 const tvbparse_wanted_t* wanted);
351 #define tvbparse_one_or_more(id, private_data, before_cb, after_cb, wanted)\
352 tvbparse_some(id, 1, G_MAXINT, private_data, before_cb, after_cb, wanted)
358 * this is a pointer to a pointer to a wanted element (that might have not
359 * been initialized yet) so that recursive structures
361 tvbparse_wanted_t* tvbparse_handle(tvbparse_wanted_t** handle);
375 tvbparse_wanted_t* tvbparse_ft(int id,
377 tvbparse_action_t before_cb,
378 tvbparse_action_t after_cb,
382 tvbparse_wanted_t* tvbparse_end_of_buffer(int id,
384 tvbparse_action_t before_cb,
385 tvbparse_action_t after_cb);
387 tvbparse_wanted_t* tvbparse_ft_numcmp(int id,
389 tvbparse_action_t before_cb,
390 tvbparse_action_t after_cb,
393 enum ft_cmp_op ft_cmp_op,
399 * this is a composed candidate, that will try to match a quoted string
400 * (included the quotes) including into it every escaped quote.
402 * C strings are matched with tvbparse_quoted(-1,NULL,NULL,NULL,"\"","\\")
404 tvbparse_wanted_t* tvbparse_quoted(const int id,
406 tvbparse_action_t before_cb,
407 tvbparse_action_t after_cb,
412 * a helper callback for quoted strings that will shrink the token to contain
413 * only the string andnot the quotes
415 void tvbparse_shrink_token_cb(void* tvbparse_data,
416 const void* wanted_data,
417 tvbparse_elem_t* tok);
422 /* initialize the parser (at every packet)
423 * tvb: what are we parsing?
425 * len: for how many bytes
426 * private_data: will be passed to the action callbacks
427 * ignore: a wanted token type to be ignored (the associated cb WILL be called when it matches)
429 tvbparse_t* tvbparse_init(tvbuff_t* tvb,
433 const tvbparse_wanted_t* ignore);
435 /* reset the parser */
436 gboolean tvbparse_reset(tvbparse_t* tt, const int offset, int len);
438 guint tvbparse_curr_offset(tvbparse_t* tt);
439 guint tvbparse_len_left(tvbparse_t* tt);
444 * This will look for the wanted token at the current offset or after any given
445 * number of ignored tokens returning FALSE if there's no match or TRUE if there
447 * The parser will be left in its original state and no callbacks will be called.
449 gboolean tvbparse_peek(tvbparse_t* tt,
450 const tvbparse_wanted_t* wanted);
453 * This will look for the wanted token at the current offset or after any given
454 * number of ignored tokens returning NULL if there's no match.
455 * if there is a match it will set the offset of the current parser after
456 * the end of the token
458 tvbparse_elem_t* tvbparse_get(tvbparse_t* tt,
459 const tvbparse_wanted_t* wanted);
462 * Like tvbparse_get but this will look for a wanted token even beyond the
464 * This function is slow.
466 tvbparse_elem_t* tvbparse_find(tvbparse_t* tt,
467 const tvbparse_wanted_t* wanted);
470 void tvbparse_tree_add_elem(proto_tree* tree, tvbparse_elem_t* curr);