Codeberg is giving me headache on PM

[nicaragua]

Tried in 32.1.1 and 32.1.0 x86 on W10. Fresh profile, `palemoon --no-remote --profile 123qwe`.

1. Try opening a single tab, e.g. https://codeberg.org/hiiamboris/red-spa ... plates.red
Let it painfully load, then press Ctrl+F to try and find anything. 10-15 second lags on every action.
2. Try opening 5-10 tabs on the same page. You can kill the browser then, it's no use anymore.

I realize this is likely some half assed coding on their side. But I migrated to Codeberg because GH was super slow. Now they change sides. Dunno how one can bring Gitea fork to such condition.

Anyway, it seems more or less okay in Brave, so either there's some conflict with PM, or it's just too old for such crappy sites.

Any workaround known or comes to mind? I can tell it's lousy CSS - if I turn off CSS (in not fresh instance), it's fast again (but I can't make heads or tails of the page =)

Code: Select all

Content Security Policy: The page’s settings observed the loading of a resource at self (“script-src https://codeberg.org https://*.codeberg.org https://codeberg.org”). A CSP report is being sent. Source: 
	window.addEventListener('error', funct....  
templates.red:23
Content Security Policy: The page’s settings observed the loading of a resource at self (“script-src https://codeberg.org https://*.codeberg.org https://codeberg.org”). A CSP report is being sent. Source: onerror attribute on SCRIPT element.  
templates.red
Content Security Policy: The page’s settings observed the loading of a resource at self (“script-src https://codeberg.org https://*.codeberg.org https://codeberg.org”). A CSP report is being sent. Source: call to eval() or related function blocked by CSP.  
index.js:19

[Moonchild]

This seems to be related to their scripting responding to viewport size changes (it changes when you open the search bar, for example). wouldn't surprise me if it is the infamous mutation observer problem where they call the mutation observer in their own observer callback causing a recursive mutation event storm.

I've not had time to dig into their web code, though, so it's a wild guess.