Improved hash handling

[miuirussia]

Description of problem in #31

[natefaubion]

Can you clarify the intent of the implementations so we can discuss whether or not it makes sense overall? If I understand correctly, this is ultimately checking the hash as part of end. I would argue that this is not correct, semantically. Hash is similar to query parameters, which end does not check either, and I think that's the right thing to do. In my opinion, backtracking on a hash is dubious as a design constraint, but if one insists, I think it would be better to have a different combinator similar to to end for the hash specifically.

[miuirussia]

Can you clarify the intent of the implementations so we can discuss whether or not it makes sense overall?

It needs for correctly match path, as described in #31

If I understand correctly, this is ultimately checking the hash as part of end

Agreed, we probably need something like strictEnd that doesn't ignore the rest of the parsing