| 1 |
<?php
|
| 2 |
|
| 3 |
function leech_news_parse_news_feed(&$xml) {
|
| 4 |
// Some feeds already use CDATA but in "wrong way": http://www.rocketboom.com/vlog/quicktime_daily_enclosures.xml (ie. <description> something <CDATA soemthing else></description>
|
| 5 |
$data = trim(str_replace(array('<![CDATA[', ']]>'), '', $xml));
|
| 6 |
|
| 7 |
// Add CDATA around content which may contain (x)html data, and is not contained in CDATA yet
|
| 8 |
$src = array(
|
| 9 |
'%(<(link|content|content:encoded|description|title|summary|info|tagline|copyright|source|itunes:summary|media:text|text)(?>[^<]*(?<!/)>)(?!<!\[CDATA\[))(.*)(</\2>)%sUS',
|
| 10 |
'%24:(\d\d:\d\d)%' // workaround buggy hour format in feeds
|
| 11 |
/*'%(<(\w+)(?>[^<]*type=")(?:text/html|application/xhtml\+xml|html|xhtml")(?>[^<]*(?<!/)>)(?!<!\[CDATA\[))(.*)(</\2>)%sUS'*/
|
| 12 |
);
|
| 13 |
$dst = array(
|
| 14 |
'$1<![CDATA[$3]]>$4',
|
| 15 |
'00:$1'
|
| 16 |
);
|
| 17 |
$data = preg_replace($src, $dst, $data);
|
| 18 |
|
| 19 |
$xml_parser = new LeechNewsParser();
|
| 20 |
$xml_tree = $xml_parser->Parse($data);
|
| 21 |
return _leech_news_get_feed($xml_tree);
|
| 22 |
}
|
| 23 |
|
| 24 |
|
| 25 |
function _leech_news_get_feed(&$xml_tree) {
|
| 26 |
$feed = new StdClass();
|
| 27 |
|
| 28 |
if ($xml_tree['RSS']) { // RSS 0.91, 0.92, 2.0
|
| 29 |
$feed->format = 'RSS';
|
| 30 |
$feed->root = &$xml_tree['RSS'][0];
|
| 31 |
$feed->channel = &$xml_tree['RSS'][0]['CHANNEL'][0];
|
| 32 |
$items = &$xml_tree['RSS'][0]['CHANNEL'][0]['ITEM'];
|
| 33 |
|
| 34 |
$feed->logo = $feed->channel['IMAGE'][0];
|
| 35 |
$feed->link = $feed->channel['LINK'][0]['VALUE'];
|
| 36 |
$feed->description = $feed->channel['DESCRIPTION'][0]['VALUE'];
|
| 37 |
|
| 38 |
if ($feed->items[0]['GUID'][0]['VALUE']) {
|
| 39 |
$feed->has_guids = 'GUID';
|
| 40 |
}
|
| 41 |
}
|
| 42 |
else if ($xml_tree['RDF:RDF']) {
|
| 43 |
$feed->format = 'RDF';
|
| 44 |
$feed->root = &$xml_tree['RDF:RDF'][0];
|
| 45 |
$feed->channel = &$xml_tree['RDF:RDF'][0]['CHANNEL'][0];
|
| 46 |
$items = &$xml_tree['RDF:RDF'][0]['ITEM'];
|
| 47 |
|
| 48 |
$feed->logo = $feed->root['IMAGE'][0];
|
| 49 |
$feed->link = $feed->channel['LINK'][0]['VALUE'];
|
| 50 |
$feed->description = $feed->channel['DESCRIPTION'][0]['VALUE'];
|
| 51 |
|
| 52 |
if ($feed->items[0]['GUID'][0]['VALUE']) {
|
| 53 |
$feed->has_guids = 'GUID';
|
| 54 |
}
|
| 55 |
}
|
| 56 |
else if ($xml_tree['FEED']) { // Atom 0.3, 1.0
|
| 57 |
$feed->format = 'ATOM';
|
| 58 |
$feed->root = &$xml_tree['FEED'][0];
|
| 59 |
$feed->channel = &$xml_tree['FEED'][0];
|
| 60 |
$items = &$xml_tree['FEED'][0]['ENTRY'];
|
| 61 |
|
| 62 |
$feed->logo = ($feed->channel['LOGO'][0]['VALUE'] ? $feed->channel['LOGO'][0]['VALUE'] : $feed->channel['ICON'][0]['VALUE']);
|
| 63 |
// TODO: remove this Atom hack when we have field mapping or at least specialized parsers in place
|
| 64 |
if (count($feed->channel['LINK']) > 1) {
|
| 65 |
$feed->link = '';
|
| 66 |
foreach ($feed->channel['LINK'] as $l) {
|
| 67 |
if ($l['REL'] == 'alternate') {
|
| 68 |
$feed->link = $l['HREF'];
|
| 69 |
}
|
| 70 |
}
|
| 71 |
}
|
| 72 |
else {
|
| 73 |
$feed->link = $feed->channel['LINK'][0]['HREF'];
|
| 74 |
}
|
| 75 |
$feed->description = $feed->channel['TAGLINE'][0]['VALUE'];
|
| 76 |
|
| 77 |
if ($feed->items[0]['ATOM:ID'][0]['VALUE']) {
|
| 78 |
$feed->has_guids = 'ATOM:ID';
|
| 79 |
}
|
| 80 |
else if ($feed->items[0]['ID'][0]['VALUE']) {
|
| 81 |
$feed->has_guids = 'ID';
|
| 82 |
}
|
| 83 |
}
|
| 84 |
else if ($xml_tree['CHANNEL']) { // RSS 1.1
|
| 85 |
$feed->format = 'RSS';
|
| 86 |
$feed->root = &$xml_tree['CHANNEL'][0];
|
| 87 |
$feed->channel = &$xml_tree['CHANNEL'][0];
|
| 88 |
$items = &$xml_tree['CHANNEL'][0]['ITEMS'][0]['ITEM'];
|
| 89 |
|
| 90 |
$feed->logo = $feed->channel['IMAGE'][0];
|
| 91 |
$feed->link = $feed->channel['LINK'][0]['VALUE'];
|
| 92 |
$feed->description = $feed->channel['DESCRIPTION'][0]['VALUE'];
|
| 93 |
|
| 94 |
if ($feed->items[0]['GUID'][0]['VALUE']) {
|
| 95 |
$feed->has_guids = 'GUID';
|
| 96 |
}
|
| 97 |
}
|
| 98 |
else {
|
| 99 |
// unsupported format
|
| 100 |
$feed->format = '[unknown]';
|
| 101 |
$feed->root = array();
|
| 102 |
$feed->channel = array();
|
| 103 |
$items = array();
|
| 104 |
|
| 105 |
$feed->logo = '';
|
| 106 |
$feed->link = '';
|
| 107 |
$feed->description = '';
|
| 108 |
|
| 109 |
$feed->has_guids = FALSE;
|
| 110 |
return $feed;
|
| 111 |
}
|
| 112 |
|
| 113 |
// Now handle image/logo tags which are parents of other tags with url and stuff
|
| 114 |
if (is_array($feed->logo)) {
|
| 115 |
if ($feed->logo['URL']) {
|
| 116 |
$feed->logo = $feed->logo['URL'][0]['VALUE'];
|
| 117 |
}
|
| 118 |
else {
|
| 119 |
// unsupported
|
| 120 |
$feed->logo = '';
|
| 121 |
}
|
| 122 |
}
|
| 123 |
|
| 124 |
// Try to find author of feed
|
| 125 |
if ($feed->channel['AUTHOR'][0]['VALUE']) {
|
| 126 |
$feed->author = $channel['AUTHOR'][0]['VALUE'];
|
| 127 |
}
|
| 128 |
if ($feed->channel['AUTHOR'][0]['NAME'][0]['VALUE']) {
|
| 129 |
$feed->author = $feed->channel['AUTHOR'][0]['NAME'][0]['VALUE'];
|
| 130 |
}
|
| 131 |
else if ($feed->channel['DC:CREATOR']) {
|
| 132 |
$feed->author = $feed->channel['DC:CREATOR'][0]['VALUE'];
|
| 133 |
}
|
| 134 |
else {
|
| 135 |
$feed->author = '';
|
| 136 |
}
|
| 137 |
|
| 138 |
// TODO: find nicer way for handling namespaces ;)
|
| 139 |
if ($items[0]['PUBDATE']) $feed->has_dates = 'PUBDATE'; // RSS 2.0
|
| 140 |
else if ($items[0]['DC:DATE']) $feed->has_dates = 'DC:DATE'; // Dublin core
|
| 141 |
else if ($items[0]['DATE']) $feed->has_dates = 'DATE'; // Dublin core
|
| 142 |
else if ($items[0]['DCTERMS:ISSUED']) $feed->has_dates = 'DCTERMS:ISSUED'; // Dublin core
|
| 143 |
else if ($items[0]['ISSUED']) $feed->has_dates = 'ISSUED'; // Dublin core
|
| 144 |
else if ($items[0]['DCTERMS:CREATED']) $feed->has_dates = 'DCTERMS:CREATED'; // Dublin core
|
| 145 |
else if ($items[0]['CREATED']) $feed->has_dates = 'CREATED'; // Dublin core
|
| 146 |
else if ($items[0]['DCTERMS:MODIFIED']) $feed->has_dates = 'DCTERMS:MODIFIED'; // Dublin core
|
| 147 |
else if ($items[0]['MODIFIED']) $feed->has_dates = 'MODIFIED'; // Dublin core
|
| 148 |
else if ($items[0]['ATOM:UPDATED']) $feed->has_dates = 'ATOM:UPDATED'; // Atom
|
| 149 |
else if ($items[0]['UPDATED']) $feed->has_dates = 'UPDATED'; // Atom
|
| 150 |
else $feed->has_dates = NULL;
|
| 151 |
|
| 152 |
// Set default value
|
| 153 |
$feed->has_unique_links = TRUE;
|
| 154 |
$feed->items = array();
|
| 155 |
$temp = array();
|
| 156 |
for ($index = count($items) - 1; $index >= 0; $index--) {
|
| 157 |
$feed->items[] = _leech_news_get_item($items[$index], $feed);
|
| 158 |
|
| 159 |
// If there was no guid try to use link as guid.
|
| 160 |
if (!$feed->has_guids && is_array($temp)) {
|
| 161 |
$link = &$feed->items[count($feed->items)-1]->link;
|
| 162 |
// Check if link is unique
|
| 163 |
if (!isset($temp[$link])) {
|
| 164 |
$temp[$link] = 1;
|
| 165 |
}
|
| 166 |
else {
|
| 167 |
$feed->has_unique_links = FALSE;
|
| 168 |
unset($temp);
|
| 169 |
}
|
| 170 |
}
|
| 171 |
}
|
| 172 |
|
| 173 |
return $feed;
|
| 174 |
}
|
| 175 |
|
| 176 |
function _leech_news_get_item(&$data, &$feed) {
|
| 177 |
$item = new StdClass();
|
| 178 |
|
| 179 |
// Description field is needed early for case when no title is specified
|
| 180 |
if ($data['DESCRIPTION']) { // RSS 0.91, 0.92, 1.0, 1.1, 2.0
|
| 181 |
$item->body = $data['DESCRIPTION'][0]['VALUE'];
|
| 182 |
}
|
| 183 |
else if ($data['SUMMARY']) { // Atom 0.3, 1.0
|
| 184 |
$item->body = $data['SUMMARY'][0]['VALUE'];
|
| 185 |
}
|
| 186 |
|
| 187 |
$temp = '';
|
| 188 |
if ($data['CONTENT']) { // Atom 0.3, 1.0
|
| 189 |
$temp = &$data['CONTENT'][0]['VALUE'];
|
| 190 |
}
|
| 191 |
else if ($data['CONTENT:ENCODED']) { // Don't know where it came from but it can be found in RSS 2.0 feeds
|
| 192 |
$temp = &$data['CONTENT:ENCODED'][0]['VALUE'];
|
| 193 |
}
|
| 194 |
if (strlen($item->body) < strlen($temp)) {
|
| 195 |
if ($item->body) {
|
| 196 |
$item->teaser = $item->body;
|
| 197 |
}
|
| 198 |
$item->body = $temp;
|
| 199 |
}
|
| 200 |
|
| 201 |
/*
|
| 202 |
** Resolve the item's title. If no title is found, we use
|
| 203 |
** up to 40 characters of the description ending at a word
|
| 204 |
** boundary but not splitting potential entities.
|
| 205 |
*/
|
| 206 |
if (!($item->title = $data['TITLE'][0]['VALUE'])) {
|
| 207 |
$item->title = preg_replace('/^(.*)[^\w;&].*?$/', "\\1", truncate_utf8($item->body, 40));
|
| 208 |
}
|
| 209 |
|
| 210 |
// If title was "escaped" then it may still contain entities, becuase each & from entity was also escaped to & before
|
| 211 |
// TODO: the same for content?
|
| 212 |
if ($data['TITLE'][0]['MODE'] == 'escaped') {
|
| 213 |
$item->title = parse_entities($item->title);
|
| 214 |
}
|
| 215 |
$item->title = strip_tags($item->title);
|
| 216 |
|
| 217 |
/*
|
| 218 |
** Resolve the items link.
|
| 219 |
*/
|
| 220 |
if ($data['LINK']) {
|
| 221 |
// TODO: remove this Atom hack when we have field mapping or at least specialized parsers in place
|
| 222 |
if (count($data['LINK']) > 1) {
|
| 223 |
$item->link = $feed->link;
|
| 224 |
foreach ($data['LINK'] as $temp) {
|
| 225 |
if ($temp['REL'] == 'alternate') {
|
| 226 |
$item->link = $temp['HREF'];
|
| 227 |
}
|
| 228 |
}
|
| 229 |
}
|
| 230 |
else {
|
| 231 |
$item->link = ($data['LINK'][0]['HREF'] ? $data['LINK'][0]['HREF'] : $data['LINK'][0]['VALUE']);
|
| 232 |
}
|
| 233 |
}
|
| 234 |
elseif ($data['GUID'] && (strncmp($data['GUID'][0]['VALUE'], 'http://', 7) == 0) && $data['GUID'][0]['ISPERMALINK'] != 'false') {
|
| 235 |
$item->link = $data['GUID'][0]['VALUE'];
|
| 236 |
}
|
| 237 |
else {
|
| 238 |
$item->link = $feed->link;
|
| 239 |
}
|
| 240 |
|
| 241 |
// Try to "sniff" real link from feeds like news.google.com which "hide" real link behind own url
|
| 242 |
if (strpos($item->link, 'http://news.google.com/news/url?') === 0) {
|
| 243 |
if (preg_match('/\&url=(.*)\&/U', $item->link, $matches) && $matches[1]) {
|
| 244 |
$item->link = rawurldecode($matches[1]);
|
| 245 |
}
|
| 246 |
}
|
| 247 |
else if (preg_match('/^\w+:\/\/(?:\w+\.|\.)*yahoo.com\/dailynews\/rss\/.*\*(.*)/', $item->link, $matches)) {
|
| 248 |
$item->link = rawurldecode($matches[1]);
|
| 249 |
}
|
| 250 |
|
| 251 |
/*
|
| 252 |
** Resolve the items source.
|
| 253 |
*/
|
| 254 |
// RSS 2.0 description of SOURCE is a bit different from ATOM and DC.
|
| 255 |
// It says link should point to XML data of source (so i guess to feed/channel??),
|
| 256 |
// while ATOM and DC say it just points to original data (and from examples on web
|
| 257 |
// it looks like it means link to original article on site, not in RSS/ATOM format).
|
| 258 |
if ($data['SOURCE'][0]['VALUE'] && $data['SOURCE'][0]['URL']) { // RSS 2.0
|
| 259 |
$item->source_title = $data['SOURCE'][0]['VALUE'];
|
| 260 |
$item->source_xml = $data['SOURCE'][0]['URL'];
|
| 261 |
}
|
| 262 |
if ($data['DC:SOURCE'][0]['VALUE'] || (!$item->source_xml && $data['SOURCE'][0]['VALUE'])) { // Dublin core
|
| 263 |
$item->source_link = $data['DC:SOURCE'][0]['VALUE'];
|
| 264 |
}
|
| 265 |
else if ($data['SOURCE'] || $data['ATOM:SOURCE']) { // ATOM 1.0
|
| 266 |
if ($data['SOURCE'][0]['TITLE']) $item->source_title = $data['SOURCE'][0]['TITLE'][0]['VALUE'];
|
| 267 |
else if ($data['SOURCE'][0]['ATOM:TITLE']) $item->source_title = $data['SOURCE'][0]['ATOM:TITLE'][0]['VALUE'];
|
| 268 |
if ($data['SOURCE'][0]['LINK']) $item->source_link = $data['SOURCE'][0]['LINK'][0]['VALUE'];
|
| 269 |
else if ($data['SOURCE'][0]['ATOM:LINK']) $item->source_link = $data['SOURCE'][0]['ATOM:LINK'][0]['VALUE'];
|
| 270 |
}
|
| 271 |
|
| 272 |
if (!$item->source_title) {
|
| 273 |
$item->source_title = '';
|
| 274 |
}
|
| 275 |
if (!$item->source_link) {
|
| 276 |
$item->source_link = '';
|
| 277 |
}
|
| 278 |
if (!$item->source_xml) {
|
| 279 |
$item->source_xml = '';
|
| 280 |
}
|
| 281 |
|
| 282 |
if ($feed->has_dates) {
|
| 283 |
$item->date = strtotime($data[$feed->has_dates][0]['VALUE']); // strtotime() returns -1 on failure
|
| 284 |
if ($item->date < 0) {
|
| 285 |
$item->date = parse_w3cdtf($data[$feed->has_dates][0]['VALUE']); // also returns -1 on failure
|
| 286 |
if ($item->date < 0) {
|
| 287 |
$item->date = time(); // better than nothing
|
| 288 |
}
|
| 289 |
}
|
| 290 |
}
|
| 291 |
else {
|
| 292 |
$item->date = time();
|
| 293 |
}
|
| 294 |
|
| 295 |
// Try to use RSS:GUID/ATOM:ID as unique identifier
|
| 296 |
if ($data['GUID'][0]['VALUE']) { // RSS 2.0
|
| 297 |
$item->guid = $data['GUID'][0]['VALUE'];
|
| 298 |
}
|
| 299 |
else if ($data['ATOM:ID'][0]['VALUE']) { // ATOM 0.3, 1.0
|
| 300 |
$item->guid = $data['ATOM:ID'][0]['VALUE'];
|
| 301 |
}
|
| 302 |
else if ($data['ID'][0]['VALUE']) { // ATOM 0.3, 1.0
|
| 303 |
$item->guid = $data['ID'][0]['VALUE'];
|
| 304 |
}
|
| 305 |
else {
|
| 306 |
$feed->has_guids = FALSE;
|
| 307 |
}
|
| 308 |
// TODO: is there anyway to check if DC:IDENTIFIER is unique?
|
| 309 |
// http://dublincore.org/documents/usageguide/elements.shtml says it can be non-unique so useles for us :(
|
| 310 |
|
| 311 |
// Make relative URLs to be global
|
| 312 |
$base = '';
|
| 313 |
if ($item->link) {
|
| 314 |
$q = strpos($item->link, '?');
|
| 315 |
if ($q > 0) {
|
| 316 |
$q = strrpos(substr($item->link, 0, $q), '/');
|
| 317 |
}
|
| 318 |
else {
|
| 319 |
$q = strrpos($item->link, '/');
|
| 320 |
}
|
| 321 |
$base = substr($item->link, 0, $q);
|
| 322 |
}
|
| 323 |
$item->body = parse_relative_urls($item->body, $base);
|
| 324 |
$item->teaser = parse_relative_urls($item->teaser, $base);
|
| 325 |
|
| 326 |
return $item;
|
| 327 |
}
|
| 328 |
|
| 329 |
/**
|
| 330 |
* Private function;
|
| 331 |
* Parse the W3C date/time format, a subset of ISO 8601. PHP date parsing
|
| 332 |
* functions do not handle this format.
|
| 333 |
* See http://www.w3.org/TR/NOTE-datetime for more information.
|
| 334 |
* Origionally from MagpieRSS (http://magpierss.sourceforge.net/).
|
| 335 |
*
|
| 336 |
* @param $date_str A string with a potentially W3C DTF date.
|
| 337 |
* @return A timestamp if parsed successfully or -1 if not.
|
| 338 |
*/
|
| 339 |
function parse_w3cdtf($date_str) {
|
| 340 |
if (preg_match('/(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2})(:(\d{2}))?(?:([-+])(\d{2}):?(\d{2})|(Z))?/', $date_str, $match)) {
|
| 341 |
list($year, $month, $day, $hours, $minutes, $seconds) = array($match[1], $match[2], $match[3], $match[4], $match[5], $match[6]);
|
| 342 |
// calc epoch for current date assuming GMT
|
| 343 |
$epoch = gmmktime($hours, $minutes, $seconds, $month, $day, $year);
|
| 344 |
if ($match[10] != 'Z') { // Z is zulu time, aka GMT
|
| 345 |
list($tz_mod, $tz_hour, $tz_min) = array($match[8], $match[9], $match[10]);
|
| 346 |
// zero out the variables
|
| 347 |
if (!$tz_hour) {
|
| 348 |
$tz_hour = 0;
|
| 349 |
}
|
| 350 |
if (!$tz_min) {
|
| 351 |
$tz_min = 0;
|
| 352 |
}
|
| 353 |
$offset_secs = (($tz_hour * 60) + $tz_min) * 60;
|
| 354 |
// is timezone ahead of GMT? then subtract offset
|
| 355 |
if ($tz_mod == '+') {
|
| 356 |
$offset_secs *= -1;
|
| 357 |
}
|
| 358 |
$epoch += $offset_secs;
|
| 359 |
}
|
| 360 |
return $epoch;
|
| 361 |
}
|
| 362 |
else {
|
| 363 |
return -1;
|
| 364 |
}
|
| 365 |
}
|
| 366 |
|
| 367 |
/**
|
| 368 |
* Private function;
|
| 369 |
* from: http://pl2.php.net/manual/en/function.html-entity-decode.php#51055
|
| 370 |
* Used as callback function for preg_replace_all() to decode numeric entities to UTF-8 chars
|
| 371 |
*
|
| 372 |
* @param $ord Number
|
| 373 |
* @return UTF-8 string
|
| 374 |
*/
|
| 375 |
function _parse_num_entity($ord) {
|
| 376 |
$ord = $ord[1];
|
| 377 |
if (preg_match('/^x([0-9a-f]+)$/i', $ord, $match)) {
|
| 378 |
$ord = hexdec($match[1]);
|
| 379 |
}
|
| 380 |
else {
|
| 381 |
$ord = intval($ord);
|
| 382 |
}
|
| 383 |
|
| 384 |
$no_bytes = 0;
|
| 385 |
$byte = array();
|
| 386 |
|
| 387 |
if ($ord == 128) {
|
| 388 |
return chr(226).chr(130).chr(172);
|
| 389 |
}
|
| 390 |
else if($ord == 129) {
|
| 391 |
return chr(239).chr(191).chr(189);
|
| 392 |
}
|
| 393 |
else if($ord == 130) {
|
| 394 |
return chr(226).chr(128).chr(154);
|
| 395 |
}
|
| 396 |
else if($ord == 131) {
|
| 397 |
return chr(198).chr(146);
|
| 398 |
}
|
| 399 |
else if($ord == 132) {
|
| 400 |
return chr(226).chr(128).chr(158);
|
| 401 |
}
|
| 402 |
else if($ord == 133) {
|
| 403 |
return chr(226).chr(128).chr(166);
|
| 404 |
}
|
| 405 |
else if($ord == 134) {
|
| 406 |
return chr(226).chr(128).chr(160);
|
| 407 |
}
|
| 408 |
else if($ord == 135) {
|
| 409 |
return chr(226).chr(128).chr(161);
|
| 410 |
}
|
| 411 |
else if($ord == 136) {
|
| 412 |
return chr(203).chr(134);
|
| 413 |
}
|
| 414 |
else if($ord == 137) {
|
| 415 |
return chr(226).chr(128).chr(176);
|
| 416 |
}
|
| 417 |
else if($ord == 138) {
|
| 418 |
return chr(197).chr(160);
|
| 419 |
}
|
| 420 |
else if($ord == 139) {
|
| 421 |
return chr(226).chr(128).chr(185);
|
| 422 |
}
|
| 423 |
else if($ord == 140) {
|
| 424 |
return chr(197).chr(146);
|
| 425 |
}
|
| 426 |
else if($ord == 141) {
|
| 427 |
return chr(239).chr(191).chr(189);
|
| 428 |
}
|
| 429 |
else if($ord == 142) {
|
| 430 |
return chr(197).chr(189);
|
| 431 |
}
|
| 432 |
else if($ord == 143) {
|
| 433 |
return chr(239).chr(191).chr(189);
|
| 434 |
}
|
| 435 |
else if($ord == 144) {
|
| 436 |
return chr(239).chr(191).chr(189);
|
| 437 |
}
|
| 438 |
else if($ord == 145) {
|
| 439 |
return chr(226).chr(128).chr(152);
|
| 440 |
}
|
| 441 |
else if($ord == 146) {
|
| 442 |
return chr(226).chr(128).chr(153);
|
| 443 |
}
|
| 444 |
else if($ord == 147) {
|
| 445 |
return chr(226).chr(128).chr(156);
|
| 446 |
}
|
| 447 |
else if($ord == 148) {
|
| 448 |
return chr(226).chr(128).chr(157);
|
| 449 |
}
|
| 450 |
else if($ord == 149) {
|
| 451 |
return chr(226).chr(128).chr(162);
|
| 452 |
}
|
| 453 |
else if($ord == 150) {
|
| 454 |
return chr(226).chr(128).chr(147);
|
| 455 |
}
|
| 456 |
else if($ord == 151) {
|
| 457 |
return chr(226).chr(128).chr(148);
|
| 458 |
}
|
| 459 |
else if($ord == 152) {
|
| 460 |
return chr(203).chr(156);
|
| 461 |
}
|
| 462 |
else if($ord == 153) {
|
| 463 |
return chr(226).chr(132).chr(162);
|
| 464 |
}
|
| 465 |
else if($ord == 154) {
|
| 466 |
return chr(197).chr(161);
|
| 467 |
}
|
| 468 |
else if($ord == 155) {
|
| 469 |
return chr(226).chr(128).chr(186);
|
| 470 |
}
|
| 471 |
else if($ord == 156) {
|
| 472 |
return chr(197).chr(147);
|
| 473 |
}
|
| 474 |
else if($ord == 157) {
|
| 475 |
return chr(239).chr(191).chr(189);
|
| 476 |
}
|
| 477 |
else if($ord == 158) {
|
| 478 |
return chr(197).chr(190);
|
| 479 |
}
|
| 480 |
else if($ord == 159) {
|
| 481 |
return chr(197).chr(184);
|
| 482 |
}
|
| 483 |
else if($ord == 160) {
|
| 484 |
return chr(194).chr(160);
|
| 485 |
}
|
| 486 |
|
| 487 |
if ($ord < 128) {
|
| 488 |
return chr($ord);
|
| 489 |
}
|
| 490 |
else if ($ord < 2048) {
|
| 491 |
$no_bytes = 2;
|
| 492 |
}
|
| 493 |
else if ($ord < 65536) {
|
| 494 |
$no_bytes = 3;
|
| 495 |
}
|
| 496 |
else if ($ord < 1114112) {
|
| 497 |
$no_bytes = 4;
|
| 498 |
}
|
| 499 |
else {
|
| 500 |
return;
|
| 501 |
}
|
| 502 |
|
| 503 |
switch ($no_bytes) {
|
| 504 |
case 2:
|
| 505 |
$prefix = array(31, 192);
|
| 506 |
break;
|
| 507 |
|
| 508 |
case 3:
|
| 509 |
$prefix = array(15, 224);
|
| 510 |
break;
|
| 511 |
|
| 512 |
case 4:
|
| 513 |
$prefix = array(7, 240);
|
| 514 |
break;
|
| 515 |
}
|
| 516 |
|
| 517 |
for ($i = 0; $i < $no_bytes; $i++) {
|
| 518 |
$byte[$no_bytes - $i - 1] = (($ord & (63 * pow(2, 6 * $i))) / pow(2, 6 * $i)) & 63 | 128;
|
| 519 |
}
|
| 520 |
|
| 521 |
$byte[0] = ($byte[0] & $prefix[0]) | $prefix[1];
|
| 522 |
|
| 523 |
$ret = '';
|
| 524 |
for ($i = 0; $i < $no_bytes; $i++) {
|
| 525 |
$ret .= chr($byte[$i]);
|
| 526 |
}
|
| 527 |
|
| 528 |
return $ret;
|
| 529 |
}
|
| 530 |
|
| 531 |
/**
|
| 532 |
* Private function; Convert named entities to UTF-8 characters
|
| 533 |
* from: http://pl2.php.net/manual/en/function.html-entity-decode.php#51722
|
| 534 |
*/
|
| 535 |
function _parse_name_entities(&$data) {
|
| 536 |
static $ttr;
|
| 537 |
if (!$ttr) {
|
| 538 |
$trans_tbl = get_html_translation_table(HTML_ENTITIES);
|
| 539 |
foreach ($trans_tbl as $k => $v) {
|
| 540 |
$ttr[$v] = utf8_encode($k);
|
| 541 |
}
|
| 542 |
$ttr['''] = "'";
|
| 543 |
}
|
| 544 |
return strtr($data, $ttr);
|
| 545 |
}
|
| 546 |
|
| 547 |
/**
|
| 548 |
* Private function; Convert all entities to UTF-8 characters
|
| 549 |
*/
|
| 550 |
function parse_entities(&$data) {
|
| 551 |
$result = _parse_name_entities($data);
|
| 552 |
return preg_replace_callback('/&#([0-9a-fx]+);/mi', '_parse_num_entity', $result);
|
| 553 |
}
|
| 554 |
|
| 555 |
|
| 556 |
/**
|
| 557 |
* Private function; Convert relative URLs
|
| 558 |
*/
|
| 559 |
function parse_relative_urls(&$data, $base_url) {
|
| 560 |
$src = '%( href| src)="(?!\w+://)/?([^"]*)"%';
|
| 561 |
$dst = '$1="'. trim($base_url, '/') .'/$2"';
|
| 562 |
return preg_replace($src, $dst, $data);
|
| 563 |
}
|
| 564 |
|
| 565 |
class LeechNewsParser {
|
| 566 |
var $xml_parser;
|
| 567 |
|
| 568 |
var $xml_tree;
|
| 569 |
var $xml_paths;
|
| 570 |
var $xml_path_cur;
|
| 571 |
var $xml_timer;
|
| 572 |
|
| 573 |
var $_start;
|
| 574 |
var $_end;
|
| 575 |
|
| 576 |
function LeechNewsParser() {
|
| 577 |
$this->xml_parser = NULL;
|
| 578 |
$this->xml_tree = array();
|
| 579 |
$this->xml_paths[] = &$this->xml_tree;
|
| 580 |
$this->xml_path_cur = 0;
|
| 581 |
$this->xml_timer = 0;
|
| 582 |
$this->_start = 0;
|
| 583 |
$this->_end = 0;
|
| 584 |
}
|
| 585 |
|
| 586 |
function Parse(&$data) {
|
| 587 |
$this->xml_tree = array();
|
| 588 |
$this->xml_paths[] = &$this->xml_tree;
|
| 589 |
$this->xml_path_cur = 0;
|
| 590 |
|
| 591 |
$this->_start = microtime();
|
| 592 |
|
| 593 |
$this->xml_parser = drupal_xml_parser_create($data);
|
| 594 |
if ($this->xml_parser == NULL) {
|
| 595 |
return $this->xml_tree;
|
| 596 |
}
|
| 597 |
|
| 598 |
xml_set_object($this->xml_parser, $this);
|
| 599 |
xml_set_element_handler($this->xml_parser, '_element_start', '_element_end');
|
| 600 |
xml_set_character_data_handler($this->xml_parser, '_element_data');
|
| 601 |
xml_parser_set_option($this->xml_parser, XML_OPTION_CASE_FOLDING, 1);
|
| 602 |
xml_parser_set_option($this->xml_parser, XML_OPTION_SKIP_WHITE, 1);
|
| 603 |
if (!xml_parse($this->xml_parser, $data, 1)) {
|
| 604 |
$this->xml_tree['parser_error'] = xml_error_string(xml_get_error_code($this->xml_parser));
|
| 605 |
$this->xml_tree['parser_line'] = xml_get_current_line_number($this->xml_parser);
|
| 606 |
}
|
| 607 |
else {
|
| 608 |
unset($this->xml_tree['parser_error']);
|
| 609 |
unset($this->xml_tree['parser_line']);
|
| 610 |
}
|
| 611 |
xml_parser_free($this->xml_parser);
|
| 612 |
|
| 613 |
$this->_end = microtime();
|
| 614 |
list($sec, $usec) = explode(' ', $this->_start);
|
| 615 |
$this->_start = $sec + $usec;
|
| 616 |
list($sec, $usec) = explode(' ', $this->_end);
|
| 617 |
$this->xml_tree['parser_time'] = $this->xml_timer = ($sec + $usec) - $this->_start;
|
| 618 |
|
| 619 |
return $this->xml_tree;
|
| 620 |
}
|
| 621 |
|
| 622 |
function _element_start($parser, $name, $attributes) {
|
| 623 |
$temp = &$this->xml_paths[$this->xml_path_cur++];
|
| 624 |
$temp[$name][] = $attributes;
|
| 625 |
$this->xml_paths[$this->xml_path_cur] = &$temp[$name][count($temp[$name])-1];
|
| 626 |
}
|
| 627 |
|
| 628 |
function _element_end($parser, $name) {
|
| 629 |
$temp = &$this->xml_paths[$this->xml_path_cur];
|
| 630 |
array_pop($this->xml_paths);
|
| 631 |
$this->xml_path_cur--;
|
| 632 |
if (isset($temp['VALUE'])) {
|
| 633 |
$temp['VALUE'] = trim(parse_entities($temp['VALUE']));
|
| 634 |
}
|
| 635 |
}
|
| 636 |
|
| 637 |
function _element_data($parser, $data) {
|
| 638 |
$temp = trim($data);
|
| 639 |
if (strlen($temp) > 0) {
|
| 640 |
$temp = &$this->xml_paths[$this->xml_path_cur];
|
| 641 |
$temp['VALUE'] .= $data;
|
| 642 |
}
|
| 643 |
}
|
| 644 |
}
|
| 645 |
|
| 646 |
?>
|
Parent Directory
| 
Revision Log
| 
Revision Graph