| 1 |
/* |
| 2 |
* This file is part of Sargasso, http://zamez.org/sargasso |
| 3 |
* Licensed under the GNU General Public License, |
| 4 |
* http://www.opensource.org/licenses/gpl-license |
| 5 |
* Copyright 2006 James Bursa <james@zamez.org> |
| 6 |
*/ |
| 7 |
|
| 8 |
#include <assert.h> |
| 9 |
#include <ctype.h> |
| 10 |
#include <errno.h> |
| 11 |
#include <stdio.h> |
| 12 |
#include <stdlib.h> |
| 13 |
#include <string.h> |
| 14 |
#include <libxml/debugXML.h> |
| 15 |
#include "feed.h" |
| 16 |
|
| 17 |
|
| 18 |
unsigned int feed_count = 0; |
| 19 |
struct feed *feeds = 0; |
| 20 |
bool feed_work_needed = false; |
| 21 |
const char *feed_error = 0; |
| 22 |
static const char *feed_status_name[] = { "NEW", "FETCHING", "OK", "ERROR" }; |
| 23 |
static CURLM *curl_multi_handle; |
| 24 |
static unsigned int fetching = 0; |
| 25 |
#define MAX_FETCHES 3 |
| 26 |
|
| 27 |
|
| 28 |
static void feed_set_status(struct feed *feed, feed_status status); |
| 29 |
static void feed_work_feed(struct feed *feed); |
| 30 |
static void feed_create_fetch(struct feed *feed); |
| 31 |
static void feed_start_fetch(struct feed *feed); |
| 32 |
static size_t feed_header_callback(void *ptr, size_t size, size_t nmemb, |
| 33 |
void *stream); |
| 34 |
static size_t feed_write_callback(void *ptr, size_t size, size_t nmemb, |
| 35 |
void *stream); |
| 36 |
static void feed_fetched(struct feed *feed, CURLcode result); |
| 37 |
static void feed_parse(struct feed *feed); |
| 38 |
static void feed_parse_item(struct feed *feed, xmlNode *node); |
| 39 |
static void feed_free_item(struct feed *feed, unsigned int i); |
| 40 |
static void feed_clean_text(xmlChar *text); |
| 41 |
|
| 42 |
|
| 43 |
/** |
| 44 |
* Initialise the feed module. |
| 45 |
*/ |
| 46 |
|
| 47 |
bool feed_init(void) |
| 48 |
{ |
| 49 |
CURLcode code; |
| 50 |
|
| 51 |
code = curl_global_init(CURL_GLOBAL_ALL); |
| 52 |
if (code != CURLE_OK) { |
| 53 |
feed_error = curl_easy_strerror(code); |
| 54 |
return false; |
| 55 |
} |
| 56 |
|
| 57 |
curl_multi_handle = curl_multi_init(); |
| 58 |
if (!curl_multi_handle) { |
| 59 |
feed_error = "Failed to initialise curl"; |
| 60 |
return false; |
| 61 |
} |
| 62 |
|
| 63 |
xmlInitParser(); |
| 64 |
|
| 65 |
return true; |
| 66 |
} |
| 67 |
|
| 68 |
|
| 69 |
/** |
| 70 |
* Quit the feed module. |
| 71 |
*/ |
| 72 |
|
| 73 |
void feed_quit(void) |
| 74 |
{ |
| 75 |
while (feed_count) |
| 76 |
feed_remove(0); |
| 77 |
free(feeds); |
| 78 |
feeds = 0; |
| 79 |
|
| 80 |
xmlCleanupParser(); |
| 81 |
|
| 82 |
curl_multi_cleanup(curl_multi_handle); |
| 83 |
} |
| 84 |
|
| 85 |
|
| 86 |
/** |
| 87 |
* Add a new feed. |
| 88 |
*/ |
| 89 |
|
| 90 |
bool feed_add(const char *url) |
| 91 |
{ |
| 92 |
struct feed *feeds1; |
| 93 |
struct feed *feed; |
| 94 |
char *url1; |
| 95 |
unsigned int i; |
| 96 |
|
| 97 |
assert(url); |
| 98 |
|
| 99 |
feeds1 = realloc(feeds, sizeof *feed * (feed_count + 1)); |
| 100 |
if (!feeds1) { |
| 101 |
feed_error = "Out of memory"; |
| 102 |
return false; |
| 103 |
} |
| 104 |
feeds = feeds1; |
| 105 |
|
| 106 |
url1 = strdup(url); |
| 107 |
if (!url1) { |
| 108 |
feed_error = "Out of memory"; |
| 109 |
free(url1); |
| 110 |
return false; |
| 111 |
} |
| 112 |
|
| 113 |
feed = &feeds[feed_count]; |
| 114 |
feed->url = url1; |
| 115 |
feed->status = FEED_NEW; |
| 116 |
feed->error = 0; |
| 117 |
feed->status_line = 0; |
| 118 |
feed->etag = 0; |
| 119 |
feed->redirect = 0; |
| 120 |
feed->data = 0; |
| 121 |
feed->data_size = 0; |
| 122 |
feed->title = 0; |
| 123 |
feed->description = 0; |
| 124 |
feed->link = 0; |
| 125 |
feed->copyright = 0; |
| 126 |
feed->pub_date = 0; |
| 127 |
feed->category = 0; |
| 128 |
for (i = 0; i != FEED_MAX_ITEMS; i++) { |
| 129 |
feed->item[i].title = 0; |
| 130 |
feed->item[i].description = 0; |
| 131 |
feed->item[i].link = 0; |
| 132 |
feed->item[i].author = 0; |
| 133 |
feed->item[i].pub_date = 0; |
| 134 |
feed->item[i].category = 0; |
| 135 |
feed->item[i].guid = 0; |
| 136 |
feed->item[i].new_item = false; |
| 137 |
} |
| 138 |
feed->item_count = 0; |
| 139 |
|
| 140 |
feed_count++; |
| 141 |
feed_work_needed = true; |
| 142 |
|
| 143 |
printf("added feed %s\n", url); |
| 144 |
return true; |
| 145 |
} |
| 146 |
|
| 147 |
|
| 148 |
/** |
| 149 |
* Remove a feed. |
| 150 |
*/ |
| 151 |
|
| 152 |
bool feed_remove(unsigned int i) |
| 153 |
{ |
| 154 |
unsigned int j; |
| 155 |
|
| 156 |
assert(i < feed_count); |
| 157 |
|
| 158 |
if (feeds[i].status == FEED_FETCHING) { |
| 159 |
curl_multi_remove_handle(curl_multi_handle, feeds[i].curl); |
| 160 |
curl_easy_cleanup(feeds[i].curl); |
| 161 |
curl_slist_free_all(feeds[i].headers); |
| 162 |
feeds[i].headers = 0; |
| 163 |
} |
| 164 |
|
| 165 |
for (j = 0; j != feeds[i].item_count; j++) |
| 166 |
feed_free_item(&feeds[i], j); |
| 167 |
|
| 168 |
free(feeds[i].url); |
| 169 |
free(feeds[i].status_line); |
| 170 |
free(feeds[i].etag); |
| 171 |
free(feeds[i].redirect); |
| 172 |
free(feeds[i].data); |
| 173 |
if (feeds[i].title) |
| 174 |
xmlFree(feeds[i].title); |
| 175 |
if (feeds[i].description) |
| 176 |
xmlFree(feeds[i].description); |
| 177 |
if (feeds[i].link) |
| 178 |
xmlFree(feeds[i].link); |
| 179 |
if (feeds[i].copyright) |
| 180 |
xmlFree(feeds[i].copyright); |
| 181 |
if (feeds[i].pub_date) |
| 182 |
xmlFree(feeds[i].pub_date); |
| 183 |
if (feeds[i].category) |
| 184 |
xmlFree(feeds[i].category); |
| 185 |
|
| 186 |
if (i != feed_count - 1) |
| 187 |
memmove(&feeds[i], &feeds[i + 1], |
| 188 |
(sizeof feeds[0]) * (feed_count - i - 1)); |
| 189 |
|
| 190 |
feed_count--; |
| 191 |
|
| 192 |
return true; |
| 193 |
} |
| 194 |
|
| 195 |
|
| 196 |
/** |
| 197 |
* Set the status of a feed. |
| 198 |
*/ |
| 199 |
|
| 200 |
void feed_set_status(struct feed *feed, feed_status status) |
| 201 |
{ |
| 202 |
printf("status %s %s => %s\n", feed->url, |
| 203 |
feed_status_name[feed->status], |
| 204 |
feed_status_name[status]); |
| 205 |
feed->status = status; |
| 206 |
feed->updated = true; |
| 207 |
} |
| 208 |
|
| 209 |
|
| 210 |
/** |
| 211 |
* Do some work on the feeds. |
| 212 |
*/ |
| 213 |
|
| 214 |
bool feed_work(void) |
| 215 |
{ |
| 216 |
unsigned int i; |
| 217 |
int running; |
| 218 |
int queue; |
| 219 |
CURLMsg *msg; |
| 220 |
struct feed *feed; |
| 221 |
|
| 222 |
for (i = 0; i != feed_count; i++) { |
| 223 |
feeds[i].updated = false; |
| 224 |
feed_work_feed(&feeds[i]); |
| 225 |
} |
| 226 |
|
| 227 |
feed_work_needed = false; |
| 228 |
|
| 229 |
while (curl_multi_perform(curl_multi_handle, &running) == |
| 230 |
CURLM_CALL_MULTI_PERFORM) |
| 231 |
continue; |
| 232 |
|
| 233 |
if ((msg = curl_multi_info_read(curl_multi_handle, &queue))) { |
| 234 |
if (msg->msg == CURLMSG_DONE) { |
| 235 |
curl_easy_getinfo(msg->easy_handle, CURLINFO_PRIVATE, |
| 236 |
&feed); |
| 237 |
feed_fetched(feed, msg->data.result); |
| 238 |
} |
| 239 |
} |
| 240 |
|
| 241 |
for (i = 0; i != feed_count; i++) |
| 242 |
if (feeds[i].status == FEED_NEW || |
| 243 |
feeds[i].status == FEED_FETCHING || |
| 244 |
feeds[i].status == FEED_UPDATE) |
| 245 |
feed_work_needed = true; |
| 246 |
|
| 247 |
for (i = 0; i != feed_count; i++) |
| 248 |
if (feeds[i].updated) |
| 249 |
return true; |
| 250 |
return false; |
| 251 |
} |
| 252 |
|
| 253 |
|
| 254 |
/** |
| 255 |
* Do some work on a feed. |
| 256 |
*/ |
| 257 |
|
| 258 |
void feed_work_feed(struct feed *feed) |
| 259 |
{ |
| 260 |
assert(feed); |
| 261 |
|
| 262 |
if ((feed->status == FEED_NEW || feed->status == FEED_UPDATE) && |
| 263 |
fetching < MAX_FETCHES) { |
| 264 |
feed_create_fetch(feed); |
| 265 |
if (feed->status != FEED_ERROR) |
| 266 |
feed_start_fetch(feed); |
| 267 |
} |
| 268 |
} |
| 269 |
|
| 270 |
|
| 271 |
/** |
| 272 |
* Create a fetch for a feed. |
| 273 |
*/ |
| 274 |
|
| 275 |
void feed_create_fetch(struct feed *feed) |
| 276 |
{ |
| 277 |
CURL *curl; |
| 278 |
struct curl_slist *headers = 0; |
| 279 |
struct curl_slist *headers2 = 0; |
| 280 |
|
| 281 |
curl = curl_easy_init(); |
| 282 |
if (!curl) { |
| 283 |
feed_set_status(feed, FEED_ERROR); |
| 284 |
feed->error = "Failed to create curl session"; |
| 285 |
return; |
| 286 |
} |
| 287 |
|
| 288 |
headers2 = curl_slist_append(headers, "Accept: " |
| 289 |
"application/rss+xml, application/xml, text/xml"); |
| 290 |
if (!headers2) { |
| 291 |
curl_easy_cleanup(feed); |
| 292 |
curl_slist_free_all(headers); |
| 293 |
feed_set_status(feed, FEED_ERROR); |
| 294 |
feed->error = "Out of memory"; |
| 295 |
return; |
| 296 |
} |
| 297 |
headers = headers2; |
| 298 |
|
| 299 |
if (feed->etag) { |
| 300 |
size_t n = 20 + strlen(feed->etag); |
| 301 |
char if_none_match[n]; |
| 302 |
snprintf(if_none_match, n, "If-None-Match: %s", feed->etag); |
| 303 |
headers2 = curl_slist_append(headers, if_none_match); |
| 304 |
if (!headers2) { |
| 305 |
curl_easy_cleanup(curl); |
| 306 |
curl_slist_free_all(headers); |
| 307 |
feed_set_status(feed, FEED_ERROR); |
| 308 |
feed->error = "Out of memory"; |
| 309 |
return; |
| 310 |
} |
| 311 |
headers = headers2; |
| 312 |
} |
| 313 |
|
| 314 |
curl_easy_setopt(curl, CURLOPT_VERBOSE, 1L); |
| 315 |
curl_easy_setopt(curl, CURLOPT_URL, feed->url); |
| 316 |
curl_easy_setopt(curl, CURLOPT_PRIVATE, feed); |
| 317 |
curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, feed_write_callback); |
| 318 |
curl_easy_setopt(curl, CURLOPT_WRITEDATA, feed); |
| 319 |
curl_easy_setopt(curl, CURLOPT_HEADERFUNCTION, feed_header_callback); |
| 320 |
curl_easy_setopt(curl, CURLOPT_HEADERDATA, feed); |
| 321 |
curl_easy_setopt(curl, CURLOPT_USERAGENT, |
| 322 |
"Sargasso (http://zamez.org/sargasso)"); |
| 323 |
curl_easy_setopt(curl, CURLOPT_ENCODING, "gzip"); |
| 324 |
curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers); |
| 325 |
curl_easy_setopt(curl, CURLOPT_LOW_SPEED_LIMIT, 1L); |
| 326 |
curl_easy_setopt(curl, CURLOPT_LOW_SPEED_TIME, 60L); |
| 327 |
curl_easy_setopt(curl, CURLOPT_NOSIGNAL, 1L); |
| 328 |
curl_easy_setopt(curl, CURLOPT_CONNECTTIMEOUT, 30L); |
| 329 |
|
| 330 |
feed->curl = curl; |
| 331 |
feed->headers = headers; |
| 332 |
feed->redirect_count = 0; |
| 333 |
} |
| 334 |
|
| 335 |
|
| 336 |
/** |
| 337 |
* Start fetching a feed. |
| 338 |
*/ |
| 339 |
|
| 340 |
void feed_start_fetch(struct feed *feed) |
| 341 |
{ |
| 342 |
CURLMcode mcode; |
| 343 |
|
| 344 |
mcode = curl_multi_add_handle(curl_multi_handle, feed->curl); |
| 345 |
if (mcode != CURLM_OK) { |
| 346 |
feed_set_status(feed, FEED_ERROR); |
| 347 |
feed->error = curl_multi_strerror(mcode); |
| 348 |
curl_easy_cleanup(feed->curl); |
| 349 |
curl_slist_free_all(feed->headers); |
| 350 |
feed->curl = 0; |
| 351 |
feed->headers = 0; |
| 352 |
return; |
| 353 |
} |
| 354 |
|
| 355 |
free(feed->status_line); |
| 356 |
feed->status_line = 0; |
| 357 |
feed->data_size = 0; |
| 358 |
feed->error = 0; |
| 359 |
feed_set_status(feed, FEED_FETCHING); |
| 360 |
|
| 361 |
printf("fetching feed %s\n", feed->url); |
| 362 |
|
| 363 |
fetching++; |
| 364 |
feed_work_needed = true; |
| 365 |
} |
| 366 |
|
| 367 |
|
| 368 |
/** |
| 369 |
* Callback for receiving headers for a feed. |
| 370 |
*/ |
| 371 |
|
| 372 |
size_t feed_header_callback(void *ptr, size_t size, size_t nmemb, |
| 373 |
void *stream) |
| 374 |
{ |
| 375 |
struct feed *feed = (struct feed *) stream; |
| 376 |
size_t n = size * nmemb; |
| 377 |
char header[n + 1]; |
| 378 |
char *value; |
| 379 |
|
| 380 |
strncpy(header, ptr, n); |
| 381 |
header[n] = 0; |
| 382 |
while (0 < n && header[n - 1] <= 32) |
| 383 |
header[--n] = 0; |
| 384 |
|
| 385 |
if (!feed->status_line) { |
| 386 |
feed->status_line = strdup(header); |
| 387 |
return size * nmemb; |
| 388 |
} |
| 389 |
|
| 390 |
value = strchr(header, ':'); |
| 391 |
if (!value) |
| 392 |
return size * nmemb; |
| 393 |
*value = 0; |
| 394 |
value++; |
| 395 |
while (isspace(*value)) |
| 396 |
value++; |
| 397 |
|
| 398 |
if (strcasecmp(header, "ETag") == 0 && value[0] == '"') { |
| 399 |
free(feed->etag); |
| 400 |
feed->etag = strdup(value); |
| 401 |
} else if (strcasecmp(header, "Location") == 0) { |
| 402 |
free(feed->redirect); |
| 403 |
feed->redirect = strdup(value); |
| 404 |
} |
| 405 |
|
| 406 |
return size * nmemb; |
| 407 |
} |
| 408 |
|
| 409 |
|
| 410 |
/** |
| 411 |
* Callback for receiving data for a feed. |
| 412 |
*/ |
| 413 |
|
| 414 |
size_t feed_write_callback(void *ptr, size_t size, size_t nmemb, |
| 415 |
void *stream) |
| 416 |
{ |
| 417 |
struct feed *feed = (struct feed *) stream; |
| 418 |
char *data; |
| 419 |
|
| 420 |
printf("received %u for %s\n", size * nmemb, feed->url); |
| 421 |
|
| 422 |
data = realloc(feed->data, feed->data_size + size * nmemb); |
| 423 |
if (!data) { |
| 424 |
feed_set_status(feed, FEED_ERROR); |
| 425 |
feed->error = "Out of memory"; |
| 426 |
return 0; |
| 427 |
} |
| 428 |
|
| 429 |
memcpy(data + feed->data_size, ptr, size * nmemb); |
| 430 |
feed->data = data; |
| 431 |
feed->data_size += size * nmemb; |
| 432 |
|
| 433 |
return size * nmemb; |
| 434 |
} |
| 435 |
|
| 436 |
|
| 437 |
/** |
| 438 |
* Process a complete feed fetch. |
| 439 |
*/ |
| 440 |
|
| 441 |
void feed_fetched(struct feed *feed, CURLcode result) |
| 442 |
{ |
| 443 |
long http_code; |
| 444 |
|
| 445 |
printf("finished %s with result %i %s\n", |
| 446 |
feed->url, result, curl_easy_strerror(result)); |
| 447 |
|
| 448 |
fetching--; |
| 449 |
|
| 450 |
if (result == CURLE_OK) { |
| 451 |
curl_easy_getinfo(feed->curl, CURLINFO_RESPONSE_CODE, |
| 452 |
&http_code); |
| 453 |
printf("HTTP code %li\n", http_code); |
| 454 |
if (http_code == 0 || http_code == 200 /* OK */) { |
| 455 |
feed_parse(feed); |
| 456 |
} else if (http_code == 300 /* Multiple Choices */ || |
| 457 |
http_code == 301 /* Moved Permanently */ || |
| 458 |
http_code == 302 /* Found */ || |
| 459 |
http_code == 303 /* See Other */ || |
| 460 |
http_code == 307 /* Temporary Redirect */) { |
| 461 |
if (feed->redirect_count++ == 5) { |
| 462 |
feed_set_status(feed, FEED_ERROR); |
| 463 |
feed->error = "Too many redirects."; |
| 464 |
} else if (feed->redirect) { |
| 465 |
curl_multi_remove_handle(curl_multi_handle, |
| 466 |
feed->curl); |
| 467 |
curl_easy_setopt(feed->curl, CURLOPT_URL, |
| 468 |
feed->redirect); |
| 469 |
feed_start_fetch(feed); |
| 470 |
if (http_code == 301 /* Moved Permanently */) { |
| 471 |
free(feed->url); |
| 472 |
feed->url = feed->redirect; |
| 473 |
feed->redirect = 0; |
| 474 |
} |
| 475 |
return; |
| 476 |
} else { |
| 477 |
feed_set_status(feed, FEED_ERROR); |
| 478 |
feed->error = "Invalid redirect."; |
| 479 |
} |
| 480 |
} else if (http_code == 304 /* Not Modified */) { |
| 481 |
feed_set_status(feed, FEED_OK); |
| 482 |
} else { |
| 483 |
feed_set_status(feed, FEED_ERROR); |
| 484 |
if (feed->status_line) |
| 485 |
feed->error = feed->status_line; |
| 486 |
else |
| 487 |
feed->error = "Response not understood."; |
| 488 |
} |
| 489 |
|
| 490 |
} else { |
| 491 |
feed_set_status(feed, FEED_ERROR); |
| 492 |
if (!feed->error) |
| 493 |
feed->error = curl_easy_strerror(result); |
| 494 |
} |
| 495 |
|
| 496 |
curl_multi_remove_handle(curl_multi_handle, feed->curl); |
| 497 |
curl_easy_cleanup(feed->curl); |
| 498 |
feed->curl = 0; |
| 499 |
curl_slist_free_all(feed->headers); |
| 500 |
feed->headers = 0; |
| 501 |
free(feed->data); |
| 502 |
feed->data = 0; |
| 503 |
} |
| 504 |
|
| 505 |
|
| 506 |
/** |
| 507 |
* Parse a feed's XML. |
| 508 |
*/ |
| 509 |
|
| 510 |
void feed_parse(struct feed *feed) |
| 511 |
{ |
| 512 |
xmlDoc *doc; |
| 513 |
xmlNode *rss; |
| 514 |
xmlNode *channel; |
| 515 |
xmlNode *node; |
| 516 |
|
| 517 |
assert(feed); |
| 518 |
assert(feed->status == FEED_FETCHING); |
| 519 |
|
| 520 |
doc = xmlReadMemory(feed->data, feed->data_size, feed->url, 0, 0); |
| 521 |
if (!doc) { |
| 522 |
feed_set_status(feed, FEED_ERROR); |
| 523 |
feed->error = "failed to parse XML"; |
| 524 |
return; |
| 525 |
} |
| 526 |
//xmlDebugDumpDocument(stdout, doc); |
| 527 |
|
| 528 |
free(feed->data); |
| 529 |
feed->data = 0; |
| 530 |
|
| 531 |
for (rss = doc->children; rss; rss = rss->next) |
| 532 |
if (rss->type == XML_ELEMENT_NODE && !strcmp(rss->name, "rss")) |
| 533 |
break; |
| 534 |
if (!rss) { |
| 535 |
feed_set_status(feed, FEED_ERROR); |
| 536 |
feed->error = "rss element not found"; |
| 537 |
xmlFreeDoc(doc); |
| 538 |
return; |
| 539 |
} |
| 540 |
|
| 541 |
for (channel = rss->children; channel; channel = channel->next) |
| 542 |
if (channel->type == XML_ELEMENT_NODE && |
| 543 |
!strcmp(channel->name, "channel")) |
| 544 |
break; |
| 545 |
if (!channel) { |
| 546 |
feed_set_status(feed, FEED_ERROR); |
| 547 |
feed->error = "channel element not found"; |
| 548 |
xmlFreeDoc(doc); |
| 549 |
return; |
| 550 |
} |
| 551 |
|
| 552 |
for (node = channel->last; node; node = node->prev) { |
| 553 |
if (node->type != XML_ELEMENT_NODE) |
| 554 |
continue; |
| 555 |
|
| 556 |
if (!strcmp(node->name, "title")) { |
| 557 |
if (feed->title) |
| 558 |
xmlFree(feed->title); |
| 559 |
feed->title = xmlNodeGetContent(node); |
| 560 |
|
| 561 |
} else if (!strcmp(node->name, "description")) { |
| 562 |
if (feed->description) |
| 563 |
xmlFree(feed->description); |
| 564 |
feed->description = xmlNodeGetContent(node); |
| 565 |
|
| 566 |
} else if (!strcmp(node->name, "link")) { |
| 567 |
if (feed->link) |
| 568 |
xmlFree(feed->link); |
| 569 |
feed->link = xmlNodeGetContent(node); |
| 570 |
|
| 571 |
} else if (!strcmp(node->name, "copyright")) { |
| 572 |
if (feed->copyright) |
| 573 |
xmlFree(feed->copyright); |
| 574 |
feed->copyright = xmlNodeGetContent(node); |
| 575 |
|
| 576 |
} else if (!strcmp(node->name, "pubDate")) { |
| 577 |
if (feed->pub_date) |
| 578 |
xmlFree(feed->pub_date); |
| 579 |
feed->pub_date = xmlNodeGetContent(node); |
| 580 |
|
| 581 |
} else if (!strcmp(node->name, "category")) { |
| 582 |
if (feed->category) |
| 583 |
xmlFree(feed->category); |
| 584 |
feed->category = xmlNodeGetContent(node); |
| 585 |
|
| 586 |
} else if (!strcmp(node->name, "item")) { |
| 587 |
feed_parse_item(feed, node); |
| 588 |
} |
| 589 |
} |
| 590 |
|
| 591 |
xmlFreeDoc(doc); |
| 592 |
|
| 593 |
feed_clean_text(feed->title); |
| 594 |
feed_clean_text(feed->description); |
| 595 |
feed_clean_text(feed->link); |
| 596 |
feed_clean_text(feed->copyright); |
| 597 |
feed_clean_text(feed->category); |
| 598 |
|
| 599 |
feed_set_status(feed, FEED_OK); |
| 600 |
} |
| 601 |
|
| 602 |
|
| 603 |
void feed_parse_item(struct feed *feed, xmlNode *node) |
| 604 |
{ |
| 605 |
xmlNode *child; |
| 606 |
xmlChar *title = 0; |
| 607 |
xmlChar *description = 0; |
| 608 |
xmlChar *link = 0; |
| 609 |
xmlChar *author = 0; |
| 610 |
xmlChar *pub_date = 0; |
| 611 |
xmlChar *category = 0; |
| 612 |
xmlChar *guid = 0; |
| 613 |
unsigned int i; |
| 614 |
|
| 615 |
for (child = node->children; child; child = child->next) { |
| 616 |
if (child->type != XML_ELEMENT_NODE) |
| 617 |
continue; |
| 618 |
if (!strcmp(child->name, "title")) |
| 619 |
title = xmlNodeGetContent(child); |
| 620 |
else if (!strcmp(child->name, "description")) |
| 621 |
description = xmlNodeGetContent(child); |
| 622 |
else if (!strcmp(child->name, "link")) |
| 623 |
link = xmlNodeGetContent(child); |
| 624 |
else if (!strcmp(child->name, "author")) |
| 625 |
author = xmlNodeGetContent(child); |
| 626 |
else if (!strcmp(child->name, "pubDate")) |
| 627 |
pub_date = xmlNodeGetContent(child); |
| 628 |
else if (!strcmp(child->name, "category")) |
| 629 |
category = xmlNodeGetContent(child); |
| 630 |
else if (!strcmp(child->name, "guid")) |
| 631 |
guid = xmlNodeGetContent(child); |
| 632 |
} |
| 633 |
|
| 634 |
feed_clean_text(title); |
| 635 |
feed_clean_text(description); |
| 636 |
feed_clean_text(link); |
| 637 |
feed_clean_text(author); |
| 638 |
feed_clean_text(category); |
| 639 |
feed_clean_text(guid); |
| 640 |
|
| 641 |
for (i = 0; i != feed->item_count; i++) { |
| 642 |
if (guid) { |
| 643 |
if (feed->item[i].guid && |
| 644 |
!strcmp(feed->item[i].guid, guid)) |
| 645 |
break; |
| 646 |
} else if (link) { |
| 647 |
if (feed->item[i].link && |
| 648 |
!strcmp(feed->item[i].link, link)) |
| 649 |
break; |
| 650 |
} |
| 651 |
} |
| 652 |
if (i != feed->item_count) { |
| 653 |
/* old item */ |
| 654 |
feed_free_item(feed, i); |
| 655 |
feed->item[i].title = title; |
| 656 |
feed->item[i].description = description; |
| 657 |
feed->item[i].link = link; |
| 658 |
feed->item[i].author = author; |
| 659 |
feed->item[i].pub_date = pub_date; |
| 660 |
feed->item[i].category = category; |
| 661 |
feed->item[i].guid = guid; |
| 662 |
|
| 663 |
} else { |
| 664 |
/* new item */ |
| 665 |
if (feed->item_count == FEED_MAX_ITEMS) |
| 666 |
feed_free_item(feed, FEED_MAX_ITEMS - 1); |
| 667 |
memmove(feed->item + 1, feed->item, |
| 668 |
sizeof *feed->item * (FEED_MAX_ITEMS - 1)); |
| 669 |
feed->item[0].title = title; |
| 670 |
feed->item[0].description = description; |
| 671 |
feed->item[0].link = link; |
| 672 |
feed->item[0].author = author; |
| 673 |
feed->item[0].pub_date = pub_date; |
| 674 |
feed->item[0].category = category; |
| 675 |
feed->item[0].guid = guid; |
| 676 |
feed->item[0].new_item = true; |
| 677 |
if (feed->item_count != FEED_MAX_ITEMS) |
| 678 |
feed->item_count++; |
| 679 |
} |
| 680 |
} |
| 681 |
|
| 682 |
|
| 683 |
void feed_free_item(struct feed *feed, unsigned int i) |
| 684 |
{ |
| 685 |
if (feed->item[i].title) |
| 686 |
xmlFree(feed->item[i].title); |
| 687 |
if (feed->item[i].description) |
| 688 |
xmlFree(feed->item[i].description); |
| 689 |
if (feed->item[i].link) |
| 690 |
xmlFree(feed->item[i].link); |
| 691 |
if (feed->item[i].author) |
| 692 |
xmlFree(feed->item[i].author); |
| 693 |
if (feed->item[i].pub_date) |
| 694 |
xmlFree(feed->item[i].pub_date); |
| 695 |
if (feed->item[i].category) |
| 696 |
xmlFree(feed->item[i].category); |
| 697 |
if (feed->item[i].guid) |
| 698 |
xmlFree(feed->item[i].guid); |
| 699 |
} |
| 700 |
|
| 701 |
|
| 702 |
void feed_clean_text(xmlChar *text) |
| 703 |
{ |
| 704 |
xmlChar *s, *d; |
| 705 |
|
| 706 |
if (!text) |
| 707 |
return; |
| 708 |
|
| 709 |
s = d = text; |
| 710 |
while (*s) { |
| 711 |
char *gt; |
| 712 |
if (*s == '<' && (gt = strchr(s, '>'))) { |
| 713 |
if (s[1] == '/' && s[2] == 't' && s[3] == 'd') |
| 714 |
*d++ = ' ', *d++ = '|', *d++ = ' '; |
| 715 |
else if (s[1] == 'b' && s[2] == 'r') |
| 716 |
*d++ = 0xe2, *d++ = 0x80, *d++ = 0x94; |
| 717 |
s = gt + 1; |
| 718 |
} else if (*s == '&') { |
| 719 |
if (s[1] == '#' && s[2] == '3' && s[3] == '9' && |
| 720 |
s[4] == ';') |
| 721 |
*d++ = '\'', s += 5; |
| 722 |
else if (s[1] == 'n' && s[2] == 'b' && s[3] == 's' && |
| 723 |
s[4] == 'p' && s[5] == ';') |
| 724 |
*d++ = ' ', s += 6; |
| 725 |
else if (s[1] == 'q' && s[2] == 'u' && s[3] == 'o' && |
| 726 |
s[4] == 't' && s[5] == ';') |
| 727 |
*d++ = '"', s += 6; |
| 728 |
else if (s[1] == 'a' && s[2] == 'm' && s[3] == 'p' && |
| 729 |
s[4] == ';') |
| 730 |
*d++ = '&', s += 5; |
| 731 |
else if (s[1] == 'c' && s[2] == 'o' && s[3] == 'p' && |
| 732 |
s[4] == 'y' && s[5] == ';') |
| 733 |
*d++ = 0xc2, *d++ = 0xa9, s += 6; |
| 734 |
else |
| 735 |
*d++ = *s++; |
| 736 |
} else |
| 737 |
*d++ = *s++; |
| 738 |
} |
| 739 |
*d = 0; |
| 740 |
|
| 741 |
/* collapse whitespace */ |
| 742 |
s = d = text; |
| 743 |
while (*s == '\t' || *s == '\r' || *s == '\n' || *s == ' ') |
| 744 |
s++; |
| 745 |
while (*s) { |
| 746 |
while (*s && !(*s == '\t' || *s == '\r' || *s == '\n' || |
| 747 |
*s == ' ')) |
| 748 |
*d++ = *s++; |
| 749 |
if (*s) |
| 750 |
*d++ = ' '; |
| 751 |
while (*s && (*s == '\t' || *s == '\r' || *s == '\n' || |
| 752 |
*s == ' ')) |
| 753 |
s++; |
| 754 |
} |
| 755 |
*d = 0; |
| 756 |
} |
| 757 |
|
| 758 |
|
| 759 |
/** |
| 760 |
* Start updating all feeds. |
| 761 |
*/ |
| 762 |
|
| 763 |
void feed_update(void) |
| 764 |
{ |
| 765 |
unsigned int i; |
| 766 |
|
| 767 |
for (i = 0; i != feed_count; i++) |
| 768 |
feeds[i].status = FEED_UPDATE; |
| 769 |
|
| 770 |
feed_work_needed = true; |
| 771 |
} |
| 772 |
|
| 773 |
|
| 774 |
/** |
| 775 |
* Load list of feeds. |
| 776 |
*/ |
| 777 |
|
| 778 |
bool feed_list_load(const char *path) |
| 779 |
{ |
| 780 |
FILE *stream; |
| 781 |
char url[4000]; |
| 782 |
|
| 783 |
stream = fopen(path, "r"); |
| 784 |
if (!stream) { |
| 785 |
feed_error = strerror(errno); |
| 786 |
return false; |
| 787 |
} |
| 788 |
|
| 789 |
while (!feof(stream)) { |
| 790 |
url[0] = 0; |
| 791 |
fgets(url, sizeof url, stream); |
| 792 |
if (url[0] == 0 || url[0] == '\n') |
| 793 |
continue; |
| 794 |
url[strlen(url) - 1] = 0; |
| 795 |
if (!feed_add(url)) { |
| 796 |
fclose(stream); |
| 797 |
return false; |
| 798 |
} |
| 799 |
} |
| 800 |
|
| 801 |
if (fclose(stream)) { |
| 802 |
feed_error = strerror(errno); |
| 803 |
return false; |
| 804 |
} |
| 805 |
|
| 806 |
return true; |
| 807 |
} |
| 808 |
|
| 809 |
|
| 810 |
/** |
| 811 |
* Save list of feeds. |
| 812 |
*/ |
| 813 |
|
| 814 |
bool feed_list_save(const char *path) |
| 815 |
{ |
| 816 |
FILE *stream; |
| 817 |
unsigned int i; |
| 818 |
|
| 819 |
stream = fopen(path, "w"); |
| 820 |
if (!stream) { |
| 821 |
feed_error = strerror(errno); |
| 822 |
return false; |
| 823 |
} |
| 824 |
|
| 825 |
for (i = 0; i != feed_count; i++) { |
| 826 |
fputs(feeds[i].url, stream); |
| 827 |
fputc('\n', stream); |
| 828 |
} |
| 829 |
|
| 830 |
if (fclose(stream)) { |
| 831 |
feed_error = strerror(errno); |
| 832 |
return false; |
| 833 |
} |
| 834 |
|
| 835 |
return true; |
| 836 |
} |
| 837 |
|
| 838 |
|
| 839 |
/** |
| 840 |
* Output a feed. |
| 841 |
*/ |
| 842 |
|
| 843 |
void feed_print(struct feed *feed) |
| 844 |
{ |
| 845 |
unsigned int i; |
| 846 |
|
| 847 |
assert(feed); |
| 848 |
|
| 849 |
printf("URL: %s\n", feed->url); |
| 850 |
printf("Status: %s\n", feed_status_name[feed->status]); |
| 851 |
|
| 852 |
if (feed->status == FEED_OK) { |
| 853 |
printf("Title: %s\n", feed->title); |
| 854 |
printf("Description: %s\n", feed->description); |
| 855 |
printf("Link: %s\n", feed->link); |
| 856 |
printf("Copyright: %s\n", feed->copyright); |
| 857 |
printf("Publ'n date: %s\n", feed->pub_date); |
| 858 |
printf("Category: %s\n", feed->category); |
| 859 |
for (i = 0; i != feed->item_count; i++) { |
| 860 |
printf(" Title: %s\n", feed->item[i].title); |
| 861 |
printf(" Description: %s\n", |
| 862 |
feed->item[i].description); |
| 863 |
printf(" Link: %s\n", feed->item[i].link); |
| 864 |
printf(" Author: %s\n", feed->item[i].author); |
| 865 |
printf(" Publ'n date: %s\n", feed->item[i].pub_date); |
| 866 |
printf(" Category: %s\n", feed->item[i].category); |
| 867 |
printf(" GUID: %s\n", feed->item[i].guid); |
| 868 |
printf(" New item: %s\n", feed->item[i].new_item ? |
| 869 |
"yes" : "no"); |
| 870 |
} |
| 871 |
} else if (feed->status == FEED_ERROR) { |
| 872 |
printf("Error: %s\n", feed->error); |
| 873 |
} |
| 874 |
} |
Parent Directory
| 
Revision Log