1
0
mirror of https://tt-rss.org/git/tt-rss.git synced 2024-06-24 11:56:36 +02:00

af_readability: add a workaround for meta charset html pages

This commit is contained in:
Andrew Dolgov 2015-07-07 10:15:08 +03:00
parent 6322fc6870
commit b7d1306b19

View File

@ -101,6 +101,19 @@ class Af_Readability extends Plugin {
$tmp = fetch_file_contents($article["link"]);
if ($tmp) {
$tmpdoc = new DOMDocument("1.0", "UTF-8");
$tmpdoc->loadHTML($tmp);
if ($tmpdoc->encoding != 'UTF-8') {
$tmpxpath = new DOMXPath($tmpdoc);
foreach ($tmpxpath->query("//meta") as $elem) {
$elem->parentNode->removeChild($elem);
}
$tmp = $tmpdoc->saveHTML();
}
$r = new Readability($tmp, $article["link"]);
if ($r->init()) {