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

@ -329,7 +329,7 @@ class Handler_Public extends Handler {
if (!$og_image) {
$tmpdoc = new DOMDocument();
if (@$tmpdoc->loadHTML(mb_substr($content, 0, 131070))) {
if (@$tmpdoc->loadHTML('<?xml encoding="UTF-8">' . mb_substr($content, 0, 131070))) {
$tmpxpath = new DOMXPath($tmpdoc);
$imgs = $tmpxpath->query("//img");

View File

@ -1200,12 +1200,8 @@ class RSSUtils {
static function cache_media($html, $site_url) {
libxml_use_internal_errors(true);
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
$doc = new DOMDocument();
$doc->loadHTML($charset_hack . $html);
$doc->loadHTML('<?xml encoding="UTF-8">' . $html);
$xpath = new DOMXPath($doc);
$entries = $xpath->query('(//img[@src])|(//video/source[@src])|(//audio/source[@src])');

View File

@ -562,7 +562,7 @@
libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadHTML($html);
$doc->loadHTML('<?xml encoding="UTF-8">' . $html);
$xpath = new DOMXPath($doc);
$base = $xpath->query('/html/head/base[@href]');
@ -1518,14 +1518,10 @@
// plugins work on original source URLs used before caching
function rewrite_cached_urls($str) {
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
$res = trim($str); if (!$res) return '';
$doc = new DOMDocument();
$doc->loadHTML($charset_hack . $res);
$doc->loadHTML('<?xml encoding="UTF-8">' . $res);
$xpath = new DOMXPath($doc);
$entries = $xpath->query('(//img[@src]|//picture/source[@src]|//video[@poster]|//video/source[@src]|//audio/source[@src])');
@ -1580,16 +1576,10 @@
$res = trim($str); if (!$res) return '';
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
$res = trim($res); if (!$res) return '';
libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadHTML($charset_hack . $res);
$doc->loadHTML('<?xml encoding="UTF-8">' . $res);
$xpath = new DOMXPath($doc);
$rewrite_base_url = $site_url ? $site_url : get_self_url_prefix();
@ -2115,7 +2105,7 @@
libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadHTML($content);
$doc->loadHTML('<?xml encoding="UTF-8">' . $content);
$xpath = new DOMXPath($doc);
$entries = $xpath->query('/html/head/link[@rel="alternate" and '.
'(contains(@type,"rss") or contains(@type,"atom"))]|/html/head/link[@rel="feed"]');
@ -2136,7 +2126,7 @@
}
function is_html($content) {
return preg_match("/<html|DOCTYPE html/i", substr($content, 0, 100)) !== 0;
return preg_match("/<html|DOCTYPE html/i", substr($content, 0, 8192)) !== 0;
}
function url_is_html($url, $login = false, $pass = false) {

View File

@ -19,11 +19,7 @@ class Af_Fsckportal extends Plugin {
$doc = new DOMDocument();
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
@$doc->loadHTML($charset_hack . $article["content"]);
@$doc->loadHTML('<?xml encoding="UTF-8">' . $article["content"]);
if ($doc) {
$xpath = new DOMXPath($doc);

View File

@ -172,14 +172,10 @@ class Af_Readability extends Plugin {
if (!$tmpdoc->loadHTML($tmp))
return false;
// this is the worst hack yet :(
if (strtolower($tmpdoc->encoding) != 'utf-8') {
$tmpxpath = new DOMXPath($tmpdoc);
foreach ($tmpxpath->query("//meta") as $elem) {
$elem->parentNode->removeChild($elem);
}
$tmp = $tmpdoc->saveHTML();
$tmp = preg_replace("/<meta.*?charset.*?\/>/i", "", $tmp);
$tmp = mb_convert_encoding($tmp, 'utf-8', $tmpdoc->encoding);
}
try {
@ -210,7 +206,6 @@ class Af_Readability extends Plugin {
} catch (Exception $e) {
return false;
}
}
return false;

View File

@ -25,12 +25,8 @@ class Af_Tumblr_1280 extends Plugin {
if (!function_exists("curl_init") || ini_get("open_basedir"))
return $article;
$charset_hack = '<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
</head>';
$doc = new DOMDocument();
$doc->loadHTML($charset_hack . $article["content"]);
$doc->loadHTML('<?xml encoding="UTF-8">' . $article["content"]);
$found = false;
@ -92,4 +88,4 @@ class Af_Tumblr_1280 extends Plugin {
return 2;
}
}
}

View File

@ -155,7 +155,7 @@ class Af_Zz_ImgProxy extends Plugin {
$proxy_all = $this->host->get($this, "proxy_all");
$doc = new DOMDocument();
if (@$doc->loadHTML($article["content"])) {
if (@$doc->loadHTML('<?xml encoding="UTF-8">' . $article["content"])) {
$xpath = new DOMXPath($doc);
$imgs = $xpath->query("//img[@src]");

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])');