Clean text string

[zzz]

I has a user that posted a message but with the following formatting:

[QUOTE]

Message here...
[/QUOTE]

Because the top "line" was wider than the area where it was meant to appear it messed up my page formatting, which made me think that I need to parse the text string through a function that will detect things like this and remove them. Because I do not know what next can they use , i.e. I think I want to make a check to make sure that it is just alpha-numeral values and possibly if the same character is repeated, say, more then 5 times -- remove it.

I think I need some sort of regex. Not sure how to go about it though...

[NogDog]

You could try:

if(preg_match('/\S{20,}/', $text))
{
   // too long
}

Where you could change 20 to whatever the number of characters in one word is that you want to trigger an error.