Fix slowness

[sobrinho]

We don't need to specify either allowing and rejecting.

Fix #127

[gr8bit]

@sobrinho if I understand correctly, we still have a breaking behavior with this PR as @joshmcarthur pointed out in #127, right?

[igorkasyanchuk]

@sobrinho this looks like a good solution. I think it worth to merge.

But can you please check if we need this method content_type_keys

[sobrinho]

Yep, it is a breaking change in terms of not testing the other side of the validator.

[gr8bit]

Yep, it is a breaking change in terms of not testing the other side of the validator.

Ok, but I also think the advantages outweigh that. 👍

[igorkasyanchuk]

@gr8bit from your side do you approve this PR?

[gr8bit]

@gr8bit from your side do you approve this PR?

Yes. 👍

[mvz]

Why exactly is specifying both allowing and rejecting at the same time not allowed anymore? It looks like the rest of the code should handle that just fine.

[sobrinho]

I don't recall why I added that but restoring it works just fine.

Added a new PR: #173

[mvz]

Thanks @sobrinho, I actually added a PR for that as well (#172) 🙂