masqmail-0.2: 8c55886cacd8 src/parse.c

masqmail-0.2

view src/parse.c @ 3:8c55886cacd8

man pages will be maintained in troff now

author	meillo@marmaro.de
date	Fri, 26 Sep 2008 21:40:10 +0200
parents
children	26e34ae9a3e3

line source

1 /* MasqMail

4 This program is free software; you can redistribute it and/or modify

5 it under the terms of the GNU General Public License as published by

6 the Free Software Foundation; either version 2 of the License, or

7 (at your option) any later version.

9 This program is distributed in the hope that it will be useful,

10 but WITHOUT ANY WARRANTY; without even the implied warranty of

11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

12 GNU General Public License for more details.

14 You should have received a copy of the GNU General Public License

15 along with this program; if not, write to the Free Software

16 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.

17 */

19 #ifndef PARSE_TEST

20 #include "masqmail.h"

21 #endif

23 /* This is really dangerous. I hope that I was careful enough,

24 but maybe there is some malformed address possible that causes

25 this to segfault or be caught in endless loops.

27 If you find something like that, PLEASE mail the string to me

28 (no matter how idiotic it is), so that I can debug that.

29 Those things really should not happen.

30 */

32 static gchar *specials = "()<>@,;:\\\".[]`";

34 char *parse_error = NULL;

36 static

37 gchar *skip_comment(gchar *p)

38 {

40 #ifdef PARSE_TEST

41 g_print("skip_comment: %s\n", p);

42 #endif

44 p++;

45 while(*p && *p != ')'){

46 p++;

47 if(*p == '(')

48 p = skip_comment(p);

49 }

50 p++;

52 return p;

53 }

55 static

56 gboolean read_word(gchar *p, gchar **b, gchar **e)

57 {

58 #ifdef PARSE_TEST

59 g_print("read_word: %s\n", p);

60 #endif

61 /* eat leading spaces */

62 while(*p && isspace(*p)) p++;

64 *b = p;

65 /* b = &p;*/

66 if(*p == '\"'){

67 /* quoted-string */

68 p++;

69 while(*p && (*p != '\"')) p++;

70 p++;

71 }else{

72 /* atom */

73 while(*p && !strchr(specials, *p) && !iscntrl(*p) && !isspace(*p))

74 p++;

75 }

76 *e = p;

77 return TRUE;

78 }

80 static

81 gboolean read_word_with_dots(gchar *p, gchar **b, gchar **e)

82 {

83 gchar *b0 = p;

85 #ifdef PARSE_TEST

86 g_print("read_word_with_dots: %s\n", p);

87 #endif

88 while(TRUE){

89 if(!read_word(p, b, e))

90 return FALSE;

91 p = *e;

92 if(*p != '.') break;

93 p++;

94 }

95 *b = b0;

96 *e = p;

97 return TRUE;

98 }

100 static

101 gboolean read_domain(gchar *p, gchar **b, gchar **e)

102 {

103 #ifdef PARSE_TEST

104 g_print("read_domain: %s\n", p);

105 #endif

106 *b = p;

107 if(*p != '['){

108 while(isalnum(*p) || (*p == '-') || (*p == '.'))

109 p++;

110 }else{

111 p++;

112 while(isalpha(*p) || (*p == '.'))

113 p++;

114 if(*p != ']'){

115 parse_error =

116 g_strdup_printf("']' expected at end of literal address %s", *b);

117 return FALSE;

118 }

119 p++;

120 }

121 *e = p;

122 return TRUE;

123 }

124

125 gboolean parse_address_rfc822(gchar *string,

126 gchar **local_begin, gchar **local_end,

127 gchar **domain_begin, gchar **domain_end,

128 gchar **address_end)

129 {

130 gint angle_brackets = 0;

131

132 gchar *p = string;

133 gchar *b, *e;

134

135 *local_begin = *local_end = NULL;

136 *domain_begin = *domain_end = NULL;

137

138 /* might be some memory left from previous call: */

139 if(parse_error != NULL){

140 g_free(parse_error);

141 parse_error = NULL;

142 }

143

144 /* leading spaces and angle brackets */

145 while(*p && (isspace(*p) || (*p == '<'))){

146 if(*p == '<')

147 angle_brackets++;

148 p++;

149 }

150

151 if(*p){

152 while(TRUE){

153 if(read_word_with_dots(p, &b, &e)){

154 p = e;

155 #ifdef PARSE_TEST

156 g_print("after read_word_with_dots: %s\n", p);

157 #endif

158 /* eat white spaces and comments */

159 while((*p && (isspace(*p))) || (*p == '(')){

160 if(*p == '('){

161 if(!(p = skip_comment(p))){

162 parse_error =

163 g_strdup("missing right bracket ')'");

164 return FALSE;

165 }

166 }else

167 p++;

168 }

169 /* we now have a non-space char that is not

170 the beginning of a comment */

171

172 if(*p == '@'){

173 /* the last word was the local_part

174 of an addr-spec */

175 *local_begin = b;

176 *local_end = e;

177 #ifdef PARSE_TEST

178 g_print("found local part: %s\n", *local_begin);

179 #endif

180 if(*p == '@'){

181 p++; /* skip @ */

182 /* now the domain */

183 if(read_domain(p, &b, &e)){

184 p = e;

185 *domain_begin = b;

186 *domain_end = e;

187 }

188 else

189 return FALSE;

190 }else{

191 /* unqualified? */

192 *domain_begin = *domain_end = NULL;

193 }

194 break;

195 }else if(*p == '<'){

196 /* addr-spec follows */

197 while(isspace(*p) || (*p == '<')){

198 if(*p == '<')

199 angle_brackets++;

200 p++;

201 }

202 if(read_word_with_dots(p, &b, &e)){

203 p = e;

204 *local_begin = b;

205 *local_end = e;

206 #ifdef PARSE_TEST

207 g_print("found local part: %s\n", *local_begin);

208 #endif

209 }else

210 return FALSE;

211 if(*p == '@'){

212 p++;

213 if(read_domain(p, &b, &e)){

214 p = e;

215 *domain_begin = b;

216 *domain_end = e;

217 }else

218 return FALSE;

219 }else{

220 /* may be unqualified address */

221 *domain_begin = *domain_end = NULL;

222 }

223 break;

224 }else if(!*p || *p == '>'){

225 *local_begin = b;

226 *local_end = e;

227 #ifdef PARSE_TEST

228 g_print("found local part: %s\n", *local_begin);

229 #endif

230 *domain_begin = *domain_end = NULL;

231 break;

232 }else if(strchr(specials, *p) || iscntrl(*p) || isspace(*p)){

233 parse_error = g_strdup_printf("unexpected character: %c", *p);

234 return FALSE;

235 }

236 }else

237 return FALSE;

238 }

239 /* trailing spaces and angle brackets */

240 #ifdef PARSE_TEST

241 g_print("down counting trailing '>'\n");

242 #endif

243 while(*p && (isspace(*p) || (*p == '>'))){

244 if(*p == '>')

245 angle_brackets--;

246 p++;

247 }

248

249 *address_end = p;

250

251 if(angle_brackets != 0){

252 if(angle_brackets > 0)

253 parse_error = g_strdup("missing '>' at end of string");

254 else

255 parse_error = g_strdup("superfluous '>' at end of string");

256 return FALSE;

257 }else{

258 /* we successfully parsed the address */

259 return TRUE;

260 }

261 /* we never get here */

262 }

263 return FALSE;

264 }

265

266 gboolean parse_address_rfc821(gchar *string,

267 gchar **local_begin, gchar **local_end,

268 gchar **domain_begin, gchar **domain_end,

269 gchar **address_end)

270 {

271 gint angle_brackets = 0;

272

273 gchar *p = string;

274 gchar *b, *e;

275

276 *local_begin = *local_end = NULL;

277 *domain_begin = *domain_end = NULL;

278

279 /* might be some memory left from previous call: */

280 if(parse_error != NULL){

281 g_free(parse_error);

282 parse_error = NULL;

283 }

284

285 /* leading spaces and angle brackets */

286 while(*p && (isspace(*p) || (*p == '<'))){

287 if(*p == '<')

288 angle_brackets++;

289 p++;

290 }

291

292 if(*p){

293 while(TRUE){

294 if(read_word_with_dots(p, &b, &e)){

295 p = e;

296 #ifdef PARSE_TEST

297 g_print("after read_word_with_dots: %s\n", p);

298 #endif

299 *local_begin = b;

300 *local_end = e;

301 #ifdef PARSE_TEST

302 g_print("found local part: %s\n", *local_begin);

303 g_print("local_end = %s\n", *local_end);

304 #endif

305 if(!(*p) || isspace(*p) || (*p == '>')){

306 /* unqualified ?*/

307 domain_begin = domain_end = NULL;

308 break;

309 }else if(*p == '@'){

310 p++;

311 if(read_domain(p, &b, &e)){

312 p = e;

313 *domain_begin = b;

314 *domain_end = e;

315 }

316 break;

317 }else{

318 parse_error =

319 g_strdup_printf("unexpected character after local part '%c'",*p);

320 return FALSE;

321 }

322 } else

323 return FALSE;

324 }

325

326 /* trailing spaces and angle brackets */

327 #ifdef PARSE_TEST

328 g_print("down counting trailing '>'\n");

329 #endif

330 while(*p && (isspace(*p) || (*p == '>'))){

331 if(*p == '>')

332 angle_brackets--;

333 p++;

334 }

335 *address_end = p;

336

337 if(angle_brackets != 0){

338 if(angle_brackets > 0)

339 parse_error = g_strdup("missing '>' at end of string");

340 else

341 parse_error = g_strdup("superfluous '>' at end of string");

342 return FALSE;

343 }else{

344 /* we successfully parsed the address */

345 return TRUE;

346 }

347 /* we never get here */

348 }

349 return FALSE;

350 }

351

352 /*

353 allocate address, reading from string.

354 On failure, returns NULL.

355 after call, end contatins a pointer to the end of the parsed string

356 end may be NULL, if we are not interested.

357

358 parses both rfc 821 and rfc 822 addresses, depending on flag is_rfc821

359 */

360

361 address *_create_address(gchar *string, gchar **end, gboolean is_rfc821)

362 {

363 gchar *loc_beg, *loc_end;

364 gchar *dom_beg, *dom_end;

365 gchar *addr_end;

366

367 if (string && (string[0] == 0)) {

368 address *addr = g_malloc(sizeof(address));

369 addr->address = g_strdup("");

370 addr->local_part = g_strdup("");

371 addr->domain = g_strdup(""); /* 'NULL' address (failure notice),

372 "" makes sure it will not be qualified with a hostname */

373 return addr;

374 }

375

376 if(is_rfc821 ?

377 parse_address_rfc821(string,

378 &loc_beg, &loc_end, &dom_beg, &dom_end, &addr_end) :

379 parse_address_rfc822(string,

380 &loc_beg, &loc_end, &dom_beg, &dom_end, &addr_end)){

381 address *addr = g_malloc(sizeof(address));

382 gchar *p = addr_end;

383

384

385 memset(addr, 0, sizeof(address));

386

387 if(loc_beg[0] == '|'){

388 parse_error = g_strdup("no pipe allowed for RFC 822/821 address");

389 return NULL;

390 }

391

392 while(*p && (*p != ',')) p++;

393 addr->address = g_strndup(string, p - string);

394

395 addr->local_part = g_strndup(loc_beg, loc_end - loc_beg);

396

397 #ifdef PARSE_TEST

398 g_print("addr->local_part = %s\n", addr->local_part);

399 #endif

400

401 if(dom_beg != NULL){

402 addr->domain = g_strndup(dom_beg, dom_end - dom_beg);

403 }else{

404 if(addr->local_part[0] == 0)

405 addr->domain = g_strdup(""); /* 'NULL' address (failure notice),

406 "" makes sure it will not be qualified with a hostname */

407 else

408 addr->domain = NULL;

409 }

410

411 if(end != NULL)

412 *end = p;

413

414 #ifndef PARSE_TEST

415 addr_unmark_delivered(addr);

416 #endif

417

418 return addr;

419 }

420 return NULL;

421 }

422

423 address *create_address_rfc822(gchar *string, gchar **end){

424 return _create_address(string, end, FALSE);

425 }

426

427 address *create_address_rfc821(gchar *string, gchar **end){

428 return _create_address(string, end, TRUE);

429 }

430

431 GList *addr_list_append_rfc822(GList *addr_list, gchar *string, gchar *domain)

432 {

433 gchar *p = string;

434 gchar *end;

435

436 while(*p){

437 address *addr = _create_address(p, &end, FALSE);

438 if(addr){

439 if(domain)

440 if(addr->domain == NULL)

441 addr->domain = g_strdup(domain);

442

443 addr_list = g_list_append(addr_list, addr);

444 p = end;

445 }else

446 break;

447 while(*p == ',' || isspace(*p)) p++;

448 }

449 return addr_list;

450 }