Problems with unicode and URL tags

[Glow]

This is how it should be :

http://www.vbulletin.com/forum/showthread.php?t=286979

This is the reality :

http://www.idnforums.com/forums/showthread.php?p=117528#post117528

This is the input

ASCII  [ URL="http://www.test.com"]test[/ URL ]
punycode + ascii [ URL="http://www.xn--ee3a.com"]test[/ URL ]
unicode + ascii  [ URL=http://www.试.com]test[/URL]
unicode + ascii  [ URL="http://www.试.com"]test[/URL]
punycode url + unicode [URL="http://www.xn--ee3a.com"]试[/ URL ]
unicode all [ URL="http://www.试.com"]试[/ URL ]

It anyone familiar with this problem ?

[Jake Bunce]

Interesting. I have never seen that problem before. I don't know what would be causing that, but this thread has some stuff you can try:

http://www.vbulletin.com/forum/showthread.php?t=259250

[Glow]

Thanks Jake,

I followed up on your adviced and checked CP but all your tips seem to be implemented already.

Anyone else with ideas on this ?

[Glow]

We turned off all plug-ins and post-caching, the problem persists .

Restating the problem :
After any non-ascii character all bbcodes break down.

[Glow]

Also in PM's and signatures..

[Jake Bunce]

Unfortunately I am out of ideas. That thread has all of the explanations I know of for character problems like this.

[Glow]

I'll let you know if we find a solution Jake, thanks for the assistance.

[Glow]

In the mean time, Jake, can you ask around in the support crew ?

[Jake Bunce]

If you can't fix it then we can take a look if you submit a trouble ticket with your forum and server logins:

http://members.vbulletin.com/membersupport_contactform.php

[gamma]

Hi Jake, I'm working on this with Glow.

Before we get to the last stage of providing log-in info, in summary here's what we've diagnosed:

When the VB software is building the HTML to display a post, as soon as it encounters a non-ASCII character within the text stored in the "post" database record, any further bbcodes within the post text are not converted to HTML. bbcode prior to the non-ASCII character are converted fine.

Also, if the user has a non-ASCII character in their signature the exact same thing happens. bbcodes within the signature prior to the non-ASCII character are converted correctly to HTML. bbcodes after the non-ASCII character are left as square bracketed bbcodes.

Same thing with PM's.

This points towards the bbcode parser aborting when it hits a non-ASCII character.

This occurs with default templates and plug-in's switched off via the "define('DISABLE_HOOKS', true);" hard-switch in config.php.

What could be causing the parser to stop converting bbcode to HTML when it hits a non-ASCII character?

[Jake Bunce]

Probably something to do with the character encoding which is why I linked that thread before. That is the only possible explanation I can think of.