oksanaromaniv

@johnny5 , you’re right. There was another redirect setup that caused this loop. Thanks for the help!

Yes, I’m not sure why it goes to 404 on the website. This page: https://www.mein-adventskalender.de/ideen/fuellen-erwachsene-2/gibt-es-den-weihnachtsmann/ doesn’t exist, but this: https://www.mein-adventskalender.de/ideen/fuellen-erwachsene/gibt-es-den-weihnachtsmann/ does, but redirects back to https://www.mein-adventskalender.de/ideen/fuellen-erwachsene-2/gibt-es-den-weihnachtsmann/

When I check in redirects checker both links produce “Too many redirects” for me. I also disabled the regex redirect at the moment.

@johnny5 , It’s https://www.mein-adventskalender.de/ideen/fuellen-erwachsene-2/gibt-es-den-weihnachtsmann/ . When you check this in the https://www.redirect-checker.org/index.php tool, you can see the redirects. The URL should be: https://www.mein-adventskalender.de/ideen/fuellen-erwachsene/gibt-es-den-weihnachtsmann/

@johnny5 , ok, thank you, will start hunting this one ??

Hello @johnny5 ,

This problem could be caused by regex-only redirects?

Thanks!

Hello @jorgefilipecosta ,

Didn’t think about the REST API approach, that could work nicely.

Thanks,
Oksana

Great! I hope that priority “10” will work well ??

So was the lyte JS (and CSS) missing when you used autoptimize_html_after_minify?

On the working page the styles (general and background) were printed inline. Not sure about the JS though. On the not working page, there were no inline styles for the lyte-wrapper.

and when talking about “on the draft page”, are you also referring to the solution with autoptimize_html_after_minify?

Yes.

Hey, managed to get it working with the next code:

remove_filter( 'the_content', 'lyte_parse', 4 );
add_filter( 'the_content', 'lyte_parse', 10 );

Looks like the “do_blocks” is added with priority 9, so moving the parser helped in my case ??

Hello @optimizingmatters ,

With the autoptimize_html_after_minify it parsed the content inside reusable blocks, but didn’t attach the event listeners: the background image and “click” event were not added as on other pages. What is strange that on the draft page it picked content from the reusable block and added events correctly, but on the /aktuell/ page not ??

You’re right. The links to videos are in the reusable blocks that are saved as blocks with “ref” attribute. I will try the workaround, thanks! Though I thought that the plugin is hooking into the_content() after all the blocks are parsed so it should have access to the full HTML, shouldn’t it?

@optimizingmatters , you’re right. No errors are logged or tracked with QM. The filter fixed the issue for me. I do suspect that that’s most probably connected to the Cloudways files serving. Thanks!

I wasn’t successful in loggin any errors. I can’t find anything in the apache errors log or PHP error log. The thumbnails worked correctly on Sireground on the staging website, but the live website is hosted by Cloudways. They use Ngnix to serve static files and Apache to serve dynamic content with Varnish in between. Can we log anything to the Query Monitor plugin? Maybe the error logs on the server don’t work correctly.

Hello @optimizingmatters and @jhmattern, can confirm on my side that the pages work correctly ??

The strange thing is that if I add the random URL param at the beginning of the URL (as a first param), the link doesn’t work again. But can confirm it works with the param appended to the URL at the end. This leads me to think that maybe it’s connected to the string parsing logic?