masqmail

view src/parse.c @ 426:d48d3622b64f

Updated AUTHORS and THANKS files.
author markus schnalke <meillo@marmaro.de>
date Wed, 30 May 2012 11:49:40 +0200
parents cff967b2f51e
children
line source
1 /*
2 ** MasqMail
3 ** Copyright (C) 1999-2001 Oliver Kurth
4 ** Copyright (C) 2010 markus schnalke <meillo@marmaro.de>
5 **
6 ** This program is free software; you can redistribute it and/or modify
7 ** it under the terms of the GNU General Public License as published by
8 ** the Free Software Foundation; either version 2 of the License, or
9 ** (at your option) any later version.
10 **
11 ** This program is distributed in the hope that it will be useful,
12 ** but WITHOUT ANY WARRANTY; without even the implied warranty of
13 ** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 ** GNU General Public License for more details.
15 **
16 ** You should have received a copy of the GNU General Public License
17 ** along with this program; if not, write to the Free Software
18 ** Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
19 */
21 #ifndef PARSE_TEST
22 #include "masqmail.h"
23 #endif
25 /*
26 ** This is really dangerous. I hope that I was careful enough,
27 ** but maybe there is some malformed address possible that causes
28 ** this to segfault or be caught in endless loops.
30 ** If you find something like that, PLEASE mail the string to me
31 ** (no matter how idiotic it is), so that I can debug that.
32 ** Those things really should not happen.
33 */
35 static gchar *specials = "()<>@,;:\\\".[]`";
37 char *parse_error = NULL;
39 static gchar*
40 skip_comment(gchar *p)
41 {
43 #ifdef PARSE_TEST
44 g_print("skip_comment: %s\n", p);
45 #endif
47 p++;
48 while (*p && *p != ')') {
49 p++;
50 if (*p == '(') {
51 p = skip_comment(p);
52 }
53 }
54 p++;
56 return p;
57 }
59 static gboolean
60 read_word(gchar *p, gchar **b, gchar **e)
61 {
62 #ifdef PARSE_TEST
63 g_print("read_word: %s\n", p);
64 #endif
65 /* eat leading spaces */
66 while (*p && isspace(*p)) {
67 p++;
68 }
70 *b = p;
71 /* b = &p; */
72 if (*p == '\"') {
73 /* quoted-string */
74 p++;
75 while (*p && (*p != '\"')) {
76 p++;
77 }
78 p++;
79 } else {
80 /* atom */
81 while (*p && !strchr(specials, *p) && !iscntrl(*p) && !isspace(*p)) {
82 p++;
83 }
84 }
85 *e = p;
86 return TRUE;
87 }
89 static gboolean
90 read_word_with_dots(gchar *p, gchar **b, gchar **e)
91 {
92 gchar *b0 = p;
94 #ifdef PARSE_TEST
95 g_print("read_word_with_dots: %s\n", p);
96 #endif
97 while (TRUE) {
98 if (!read_word(p, b, e)) {
99 return FALSE;
100 }
101 p = *e;
102 if (*p != '.') {
103 break;
104 }
105 p++;
106 }
107 *b = b0;
108 *e = p;
109 return TRUE;
110 }
112 static gboolean
113 read_domain(gchar *p, gchar **b, gchar **e)
114 {
115 #ifdef PARSE_TEST
116 g_print("read_domain: %s\n", p);
117 #endif
118 *b = p;
119 if (*p != '[') {
120 while (isalnum(*p) || (*p == '-') || (*p == '.')) {
121 p++;
122 }
123 } else {
124 p++;
125 while (isalpha(*p) || (*p == '.')) {
126 p++;
127 }
128 if (*p != ']') {
129 parse_error = g_strdup_printf("']' expected at end of literal address %s", *b);
130 return FALSE;
131 }
132 p++;
133 }
134 *e = p;
135 return TRUE;
136 }
138 gboolean
139 parse_address_rfc822(gchar *string, gchar **local_begin, gchar **local_end,
140 gchar **domain_begin, gchar **domain_end, gchar **address_end)
141 {
142 gint angle_brackets = 0;
144 gchar *p = string;
145 gchar *b, *e;
147 *local_begin = *local_end = NULL;
148 *domain_begin = *domain_end = NULL;
150 /* might be some memory left from previous call: */
151 if (parse_error) {
152 g_free(parse_error);
153 parse_error = NULL;
154 }
156 /* leading spaces and angle brackets */
157 while (*p && (isspace(*p) || (*p == '<'))) {
158 if (*p == '<') {
159 angle_brackets++;
160 }
161 p++;
162 }
164 if (!*p) {
165 return FALSE;
166 }
168 while (TRUE) {
169 if (!read_word_with_dots(p, &b, &e)) {
170 return FALSE;
171 }
173 p = e;
174 #ifdef PARSE_TEST
175 g_print("after read_word_with_dots: %s\n", p);
176 #endif
177 /* eat white spaces and comments */
178 while ((*p && (isspace(*p))) || (*p == '(')) {
179 if (*p == '(') {
180 if (!(p = skip_comment(p))) {
181 parse_error = g_strdup("missing right bracket ')'");
182 return FALSE;
183 }
184 } else {
185 p++;
186 }
187 }
188 /*
189 ** we now have a non-space char that is not
190 ** the beginning of a comment
191 */
193 if (*p == '@' || *p == ',') {
194 /* the last word was the local_part of an addr-spec */
195 *local_begin = b;
196 *local_end = e;
197 #ifdef PARSE_TEST
198 g_print("found local part: %s\n", *local_begin);
199 #endif
200 if (*p == '@') {
201 p++; /* skip @ */
202 /* now the domain */
203 if (!read_domain(p, &b, &e)) {
204 return FALSE;
205 }
206 p = e;
207 *domain_begin = b;
208 *domain_end = e;
209 } else {
210 /* unqualified? */
211 /* something like `To: alice, bob' with -t */
212 *domain_begin = *domain_end = NULL;
213 }
214 break;
216 } else if (*p == '<') {
217 /* addr-spec follows */
218 while (isspace(*p) || (*p == '<')) {
219 if (*p == '<') {
220 angle_brackets++;
221 }
222 p++;
223 }
224 if (!read_word_with_dots(p, &b, &e)) {
225 return FALSE;
226 }
227 p = e;
228 *local_begin = b;
229 *local_end = e;
230 #ifdef PARSE_TEST
231 g_print("found local part: %s\n", *local_begin);
232 #endif
233 if (*p == '@') {
234 p++;
235 if (!read_domain(p, &b, &e)) {
236 return FALSE;
237 }
238 p = e;
239 *domain_begin = b;
240 *domain_end = e;
241 } else {
242 /* may be unqualified address */
243 *domain_begin = *domain_end = NULL;
244 }
245 break;
247 } else if (!*p || *p == '>') {
248 *local_begin = b;
249 *local_end = e;
250 #ifdef PARSE_TEST
251 g_print("found local part: %s\n", *local_begin);
252 #endif
253 *domain_begin = *domain_end = NULL;
254 break;
256 } else if (strchr(specials, *p) || iscntrl(*p) || isspace(*p)) {
257 parse_error = g_strdup_printf("unexpected character: %c", *p);
258 #ifdef PARSE_TEST
259 g_print("unexpected character: %c", *p);
260 #endif
261 return FALSE;
262 }
263 }
265 /* trailing spaces and angle brackets */
266 #ifdef PARSE_TEST
267 g_print("down counting trailing '>'\n");
268 #endif
269 while (*p && (isspace(*p) || (*p == '>'))) {
270 if (*p == '>') {
271 angle_brackets--;
272 }
273 p++;
274 }
276 *address_end = p;
278 if (angle_brackets > 0) {
279 parse_error = g_strdup("missing '>' at end of string");
280 return FALSE;
281 } else if (angle_brackets < 0) {
282 parse_error = g_strdup("superfluous '>' at end of string");
283 return FALSE;
284 }
286 /* we successfully parsed the address */
287 return TRUE;
288 }
290 gboolean
291 parse_address_rfc821(gchar *string, gchar **local_begin, gchar **local_end,
292 gchar **domain_begin, gchar **domain_end, gchar **address_end)
293 {
294 gint angle_brackets = 0;
296 gchar *p = string;
297 gchar *b, *e;
299 *local_begin = *local_end = NULL;
300 *domain_begin = *domain_end = NULL;
302 /* might be some memory left from previous call: */
303 if (parse_error != NULL) {
304 g_free(parse_error);
305 parse_error = NULL;
306 }
308 /* leading spaces and angle brackets */
309 while (*p && (isspace(*p) || (*p == '<'))) {
310 if (*p == '<') {
311 angle_brackets++;
312 }
313 p++;
314 }
316 if (!*p) {
317 return FALSE;
318 }
320 while (TRUE) {
321 if (!read_word_with_dots(p, &b, &e)) {
322 return FALSE;
323 }
325 p = e;
326 #ifdef PARSE_TEST
327 g_print("after read_word_with_dots: %s\n", p);
328 #endif
329 *local_begin = b;
330 *local_end = e;
331 #ifdef PARSE_TEST
332 g_print("found local part: %s\n", *local_begin);
333 g_print("local_end = %s\n", *local_end);
334 #endif
335 if (!(*p) || isspace(*p) || (*p == '>')) {
336 /* unqualified ? */
337 domain_begin = domain_end = NULL;
338 break;
339 } else if (*p == '@') {
340 p++;
341 if (read_domain(p, &b, &e)) {
342 p = e;
343 *domain_begin = b;
344 *domain_end = e;
345 }
346 break;
347 } else {
348 parse_error = g_strdup_printf ("unexpected character after local part '%c'", *p);
349 return FALSE;
350 }
351 }
353 /* trailing spaces and angle brackets */
354 #ifdef PARSE_TEST
355 g_print("down counting trailing '>'\n");
356 #endif
357 while (*p && (isspace(*p) || (*p == '>'))) {
358 if (*p == '>') {
359 angle_brackets--;
360 }
361 p++;
362 }
363 *address_end = p;
365 if (angle_brackets > 0) {
366 parse_error = g_strdup("missing '>' at end of string");
367 return FALSE;
368 } else if (angle_brackets < 0) {
369 parse_error = g_strdup("superfluous '>' at end of string");
370 return FALSE;
371 }
373 /* we successfully parsed the address */
374 return TRUE;
375 }
377 /*
378 ** allocate address, reading from string.
379 ** On failure, returns NULL.
380 ** after call, end contains a pointer to the end of the parsed string
381 ** end may be NULL, if we are not interested.
382 **
383 ** parses both rfc 821 and rfc 822 addresses, depending on flag is_rfc821
384 */
385 address*
386 _create_address(gchar *string, gchar **end, gboolean is_rfc821)
387 {
388 gchar *loc_beg, *loc_end;
389 gchar *dom_beg, *dom_end;
390 gchar *addr_end;
391 gboolean ret;
393 if (!string) {
394 return NULL;
395 }
396 while (isspace(*string)) {
397 string++;
398 }
399 /* TODO: what about (string == NULL)? */
400 if (string && (string[0] == '\0')) {
401 address *addr = g_malloc(sizeof(address));
402 addr->address = g_strdup("");
403 addr->local_part = g_strdup("");
404 /*
405 ** 'NULL' address: failure notice
406 ** "": will *not* be qualified with a hostname
407 */
408 addr->domain = g_strdup("");
409 return addr;
410 }
412 if (is_rfc821) {
413 ret = parse_address_rfc821(string, &loc_beg, &loc_end, &dom_beg, &dom_end, &addr_end);
414 } else {
415 ret = parse_address_rfc822(string, &loc_beg, &loc_end, &dom_beg, &dom_end, &addr_end);
416 }
417 if (!ret) {
418 return NULL;
419 }
420 if (*loc_beg == '|') {
421 parse_error = g_strdup("no pipe allowed for RFC 822/821 address");
422 return NULL;
423 }
425 address *addr = g_malloc(sizeof(address));
426 memset(addr, 0, sizeof(address));
428 gchar *p = addr_end;
429 while (*p && (*p != ',')) {
430 /* it seems as if we do this for the code in rewrite.c */
431 p++;
432 }
433 addr->address = g_strstrip(g_strndup(string, p - string));
434 addr->local_part = g_strndup(loc_beg, loc_end - loc_beg);
436 #ifdef PARSE_TEST
437 g_print("addr->local_part = %s\n", addr->local_part);
438 #endif
440 if (dom_beg != NULL) {
441 addr->domain = g_strndup(dom_beg, dom_end - dom_beg);
442 } else if (addr->local_part[0] == '\0') {
443 /*
444 ** 'NULL' address: failure notice
445 ** "": will *not* be qualified with a hostname
446 */
447 addr->domain = g_strdup("");
448 } else {
449 addr->domain = NULL;
450 }
452 if (end) {
453 *end = p;
454 }
456 DEBUG(6) debugf("_create_address(): address: `%s'\n", addr->address);
457 DEBUG(6) debugf("_create_address(): local_part: `%s'\n", addr->local_part);
458 DEBUG(6) debugf("_create_address(): domain: `%s'\n", addr->domain);
460 #ifndef PARSE_TEST
461 addr_unmark_delivered(addr);
462 #endif
464 return addr;
465 }
467 address*
468 create_address_rfc822(gchar *string, gchar **end)
469 {
470 return _create_address(string, end, FALSE);
471 }
473 address*
474 create_address_rfc821(gchar *string, gchar **end)
475 {
476 return _create_address(string, end, TRUE);
477 }
479 GList*
480 addr_list_append_rfc822(GList *addr_list, gchar *string, gchar *domain)
481 {
482 gchar *p = string;
483 gchar *end;
485 while (*p) {
486 #ifdef PARSE_TEST
487 g_print("string: %s\n", p);
488 #endif
490 address *addr = _create_address(p, &end, FALSE);
491 if (!addr) {
492 break;
493 }
495 #ifdef PARSE_TEST
496 g_print("addr: %s (%s<@>%s)", addr->address, addr->local_part, addr->domain);
497 #endif
498 if (domain && !addr->domain) {
499 addr->domain = g_strdup(domain);
500 }
501 #ifdef PARSE_TEST
502 g_print(" (%s<@>%s)\n", addr->local_part, addr->domain);
503 #endif
505 addr_list = g_list_append(addr_list, addr);
506 p = end;
508 while (*p == ',' || isspace(*p)) {
509 p++;
510 }
511 }
512 return addr_list;
513 }