Remove Twingly::URL::Hasher

[jage]

We had a discussion about the Twingly::URL::Hasher discussions today. What we want to do is to remove the helper functions:

• Twingly::URL::Hasher.taskdb_hash(url)
• Twingly::URL::Hasher.blogstream_hash(url)
• Twingly::URL::Hasher.documentdb_hash(url)
• Twingly::URL::Hasher.autopingdb_hash(url)

Since we only use the hash with Twingly::URL objects we should introduce these functions as instance methods to Twingly::URL instead. I would prefer if we allowed one method (documentdb_hash) to be the default go to hash algorithm and call it to_hash as well, in the long run we should avoid using different hash implementations.

[dentarg]

I would prefer if we allowed one method (documentdb_hash) to be the default go to hash algorithm and call it to_hash as well

Don't think to_hash is suitable name in Ruby land :-) Otherwise I'm not against the idea of a default hash

[walro]

Since we only use the hash with Twingly::URL objects

Actually, we only use these with string objects. For instance: https://github.com/twingly/zambezi/blob/7f0a7a5cfdf074e495d54aeb387f79b0e0d2c7a5/lib/converter.rb#L85 (follow backwards to see original passed values).

I think we could stuff the methods into the utilities class instead, but other than that I don't mind how things currently look.

I don't mind the idea of putting the hasher methods on the URL objects, my only objection is that Zambezi would have to parse every string (i.e. convert to a Twingly::URL) before it can hash it, I know it's a slow process :)