1
0
mirror of https://git.tt-rss.org/git/tt-rss.git synced 2025-12-13 18:06:02 +00:00

domdocument: remove old meta charset unicode hacks, replace with shorter xml preamble utf8 hack (on loadhtml where it makes sense)

af_readability: better (?) charset hack for non-unicode pages
This commit is contained in:
Andrew Dolgov
2019-03-21 21:08:02 +03:00
parent 3bd3324e5a
commit 671f4cee65
8 changed files with 15 additions and 46 deletions

View File

@@ -190,12 +190,8 @@ class Cache_Starred_Images extends Plugin implements IHandler {
return;
}
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
$doc = new DOMDocument();
$doc->loadHTML($charset_hack . $content);
$doc->loadHTML('<?xml encoding="UTF-8">' . $content);
$xpath = new DOMXPath($doc);
$entries = $xpath->query('(//img[@src])|(//video/source[@src])');