ImageProxy: avoid oversight of wrapping an entire HTML document around image #212
Conversation
…d image In reply to <FreshRSS#206 (comment)>. With special thanks to this comment: <https://www.php.net/manual/en/domdocument.savehtml.php#119767>
|
Hm, I was just reading more and realized this may not be ideal, though at first glance the output looked okay:
|
|
I have not looked precisely, but a typical way is to |
|
That's the first thing I tried, as But I see what you're referring to is actually a for loop appending to a string. I had considered that, but it felt hacky. |
|
I think it makes a difference in behaviour to pass the node inside the |
|
And no, without loop. See e.g. |
|
Ah, in that case presumably it should simply be something like: $output = $dom->getElementsByTagName('body')->item(0);
return $dom->saveHTML($output);I'll test that tomorrow (if I find the time). PS Here are the other results. I hadn't notice the third was slightly different, whoops. |
|
For reference, this is the idea I played with first yesterday: $body = $doc->getElementsByTagName('body')->item(0);
$output = '';
foreach ($body->childNodes as $node) {
$output .= $doc->saveHTML($node);
}
return $output;The sample uses regex to effect the desired result. I will do so as well to limit new string creation. |
With the method in FreshRSS#212 undesired transformations could sometimes be effected.
* ImageProxy: better DOMDocument HTML handling With the method in #212 undesired transformations could sometimes be effected. * Update xExtension-ImageProxy/extension.php Co-authored-by: Alexandre Alapetite <[email protected]> * Bump version to 0.7.1 --------- Co-authored-by: Alexandre Alapetite <[email protected]>
In reply to #206 (comment).
With special thanks to this comment: https://www.php.net/manual/en/domdocument.savehtml.php#119767