Glitch on forums

Every time I search for a post, then click on the link to jump to that post it just jumps to the top of the thread. This has been happening recently and is quite annoying. Same thing when I actually make a post of my own. Anyone else experiencing the same problem?

I’ve also been having the same problem, and it happens on every browser I use.

Yeah. I even tried clearing my cache and it didn’t work. :confused: It’s annoying.

I take it you guys are on Google Chrome? There is indeed a bug that causes the page to refresh if there’s a hash in the URL. (The hash is what makes your browser automatically scroll to the post you’re looking for.)

Thank you for your post. We have been investigating this issue and hope to have a resolution soon.

> I take it you guys are on Google Chrome? There is indeed a bug that causes the page to refresh if there’s a hash in the URL. (The hash is what makes your browser automatically scroll to the post you’re looking for.)

Yes but this is only recent. It never used to do this before.

> Thank you for your post. We have been investigating this issue and hope to have a resolution soon.

Thanks.

> > I take it you guys are on Google Chrome? There is indeed a bug that causes the page to refresh if there’s a hash in the URL. (The hash is what makes your browser automatically scroll to the post you’re looking for.)
>
> Yes but this is only recent. It never used to do this before.

I didn’t mean to make it sound like this bug has been around for ages.

A community member alerted me to it on Twitter a few days ago, posted in the Bugs sticky, and noted that he was on Google Chrome. I was simply asking to clarify that you’re on the same browser he was on, and that this isn’t a separate bug (or the same bug triggering under unexpected circumstances).

Side note: I just tested it and it seems like the bug only occurs if there is a hash in the URL and there is no “find=lastpost” parameter in the query string. Odd.

Examining the behavior of a few tested URLs, it seems like there is some JavaScript that listens for a “find=lastpost” parameter in the query string. If the parameter is detected, a hash is added, to force scrolling to the last post on the page. Perhaps this code is malfunctioning in a Chrome-specific way?

Google Chrome’s development tools trace the forced reload back to jquery.address-1.4.js. (Specifically, it’s directing my attention to line 1820 of the tested thread, and line 1820 includes the linked code file.)

That file defines a set of functions for working with URLs. I don’t know if that file is solely responsible for the page load, or if one of its functions is being called from elsewhere and that’s causing the reload. But that file is definitely the “end of the line” for this bug.

EDIT: I found the problem. Check out lines 367 and 368 of that code file:

> } else if (!_supported && _hrefHash() !== ‘’) {
> _l.replace(_l.href.substr(0, _l.href.indexOf(’#’)));

If something isn’t supported and there is a hash in the URL, then replace* the current location* with the same URL minus its hash.

Wow, you’re good at code. xD So is there anyway to fix it?

> I take it you guys are on Google Chrome? There is indeed a bug that causes the page to refresh if there’s a hash in the URL. (The hash is what makes your browser automatically scroll to the post you’re looking for.)

I get the same problem and I’m using my iPhone.

Hm, odd. The problem is no longer occurring on Firefox.

> > I take it you guys are on Google Chrome? There is indeed a bug that causes the page to refresh if there’s a hash in the URL. (The hash is what makes your browser automatically scroll to the post you’re looking for.)
>
> I get the same problem and I’m using my iPhone.

Okay, so that narrows it down to something within WebKit, then…

> Wow, you’re good at code. xD So is there anyway to fix it?

We can’t. 343i probably can.

The trick is, I don’t know why they added the problematic file. The code inside looks like a library: it’s not code that exists for its own sake; it’s one piece of code that gets used by other pieces of code located elsewhere. However, I can’t seem to find those other pieces of code (maybe there aren’t any at all), so the million-dollar question at this point is, “What is 343i using the problem file for?”

Depending on that, they might have to edit the file, edit the code that uses the file, or just completely remove the file and whatever features rely on it.

@crazybydefault: As far as I’m aware, Firefox has never been affected. You’re probably right about it being WebKit, so desktop Safari should be able to repro this bug.

Did someone fix it?! Because I’m no longer having the issue. Thanks to whoever did. :slight_smile:

> Did someone fix it?! Because I’m no longer having the issue. Thanks to whoever did. :slight_smile:

Yep! Thank the Web Team! Take a look at this thread! :slight_smile: