Overzealous filters

[Nateowami]

The default filters block https://facebook.github.io/. The filter responsible is *facebook.* I think there ought to be a way to match the main domain and all sub domains without blocking domains that end with the same name, but I can't think of what the right syntax would be. Perhaps something like *.?facebook.* Seems that could be problematic.

For this particular rule, would there by any problem changing it to *facebook.com? That would still block domains that end with facebook.com of course. Are there any Facebook TLDs that need to be blocked, other than .com, that would make this not work?

[JoshTheDerf]

I believe there was another facebook TLD, and yes, we should probably just block TLDs in the default filters. Making the Regex more complicated is something that should be avoided.

[Nateowami]

In other words, we should have two separate filters, *facebook.com and
*facebook.some_other_tld. Is that what you're saying?

I agree making the regex more complicated is not a good idea. Perhaps we
could make it that *.facebook.com also matches facebook.com?

On 06/30/2016 09:45 AM, Joshua Bemenderfer wrote:

I believe there was another facebook TLD, and yes, we should probably
just block TLDs in the default filters. Making the Regex more
complicated is something that should be avoided.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#16 (comment),
or mute the thread
https://github.com/notifications/unsubscribe/AF2zJksinYbNDlXBQt2Jt7EJHwXta-dVks5qQx_FgaJpZM4JBtJJ.

[JoshTheDerf]

*facebook.com would match that, right?

And yes. I think facebook's like buttons or ad thingies are on a different TLD. Still can't find it though. :/

[Nateowami]

Yes, it would match both facebook.com and something.facebook.com, but also somethingelsefacebook.com.

Looks like it's connect.facebook.net that serves the like buttons. I can split it into two rules.

[JoshTheDerf]

Ah, that was it.