infong/php-readability
1
0
Fork 0
mirror of https://github.com/j0k3r/php-readability.git synced 2026-09-27 06:26:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Do not remove nofollow links

Browse Source 
Most the time, they can be usefull.
At least, it'll be a link to something unrelated. But we won't lose a link inside the content.

Also, adding some extra space.
This commit is contained in:
Jeremy Benoist
2015-09-22 19:25:57 +02:00
parent 1830dc45d4
commit b77876b30a
3 changed files with 108 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/JSLikeHTMLElement.php
+4
View File
@@ -50,9 +50,11 @@ class JSLikeHTMLElement extends \DOMElement
for ($x = $this->childNodes->length - 1; $x >= 0; --$x) {
$this->removeChild($this->childNodes->item($x));
}
// $value holds our new inner HTML
if ($value != '') {
$f = $this->ownerDocument->createDocumentFragment();
// appendXML() expects well-formed markup (XHTML)
$result = @$f->appendXML($value); // @ to suppress PHP warnings
if ($result) {
@@ -63,12 +65,14 @@ class JSLikeHTMLElement extends \DOMElement
// $value is probably ill-formed
$f = new \DOMDocument();
$value = mb_convert_encoding($value, 'HTML-ENTITIES', 'UTF-8');
// Using <htmlfragment> will generate a warning, but so will bad HTML
// (and by this point, bad HTML is what we've got).
// We use it (and suppress the warning) because an HTML fragment will
// be wrapped around <html><body> tags which we don't really want to keep.
// Note: despite the warning, if loadHTML succeeds it will return true.
$result = @$f->loadHTML('<htmlfragment>'.$value.'</htmlfragment>');
if ($result) {
$import = $f->getElementsByTagName('htmlfragment')->item(0);
foreach ($import->childNodes as $child) {