Not a bug Problem with swear filter masking part of longer words

[Paul B]

I wrote <s n i g g e r> in another thread but without the spaces and it was filtered; e.g. <snigger>

Now I fully understand why that word is in the swear filter but if it's going to catch out longer words which utlise part of the forbidden word then that's a problem.

I presume the famous town on the east coast of the UK is also going to be filtered then ; Scunthorpe.

Indeed it is.

 

[Elizabeth]

That's part of the problem with a swear filter. IT'S not human and can not tell when a word is not a swear word . A lot of innocent words get caught.  

 

[Paul B]

I think it's just how the swear filter is configured.
If wildcards are used then it will catch everything.

I can post both of the masked words on my forum with no problem but the forbidden words are still caught.

 

[Elizabeth]

In my old forum, the average age was 40+, so I turned off the swear filter. Didn't need it. 

 

[Mike]

That particular word is setup to be masked where ever it occurs, even as a substring. That allows it to catch other forms of it, but yes it does mean it can catch some false positives. It's not really a bug (and the censor is empty by default), but it's down to how we want to configure things here.

 

[Paul B]

Thanks Mike.
I thought that might be the case so thanks for confirming it.

It should be easy then to configure exact words so longer strings are allowed.

I have to say, configuring the swear filter is one of my least favourite tasks but it's something that needs to be done sadly.

 

[James]

I take it we can configure it to allow longer words that contain the characters without actually having to manually write exceptions?

Test to see if BBCode bypasses the filter: Scuntorphe. Is there no workaround for this yet?

 

[Chromaniac]

http://en.wikipedia.org/wiki/****** (this is a test)

on the first time i pressed edit, the link showed fine in the edit field. but it does not now. its converted into the censored version.

i posted the wikipedia page for the word mentioned above.

 

[Brandon_R]

If someone wants to bypass the filter system, they will. xenForo just needs to get the basic implementation solid IMHO.

 

[James]

Yes indeed and using BBCode is the simplest way to bypass the filter.
Scunthorpe (empty plain bbcodes)

 

[Paul B]

Whilst it's true that people who are determined to bypass the swear filter will do so, we don't want to be filtering normal words as that will just cause confusion and then force people to be creative to be able to post words such as s n i g g e r.

 

[erich37]

is it possible to exclude certain words from the swear-filter ?
I would like to have a certain swear-word being allowed to use for my website.

 

[Paul B]

is it possible to exclude certain words from the swear-filter ?
I would like to have a certain swear-word being allowed to use for my website.

The swear filter is empty by default.

 

[Chromaniac]

i would have other uses for it. like preventing people from posting links to spammy/rival sites.

 

[Shamil]

i would have other uses for it. like preventing people from posting links to spammy/rival sites.

that's what happened on the first vBulletin board I properly frequented, to which I then became an "honorary" moderator.

 

[Erik]

Would it be possible to get regex support for the swear filter?

 

[lazer]

is it possible to exclude certain words from the swear-filter ?

Yea, it would be pretty bad if you were the admin for the biggest forum all about Scunthorpe!

 

[Erik]

Also, this should be marked as [Not a bug].

 

[Paul M]

What would be nice in any censor system is an exceptions list - so you could censor c***, but add scunthorpe as an exception.