Jamie Gaskins

Thanks for the report. It looks like this may have been broken somewhere in the transition from `Neo4j.map_node` to `include Neo4j::Serializable::Node` because unions deserialize into `Neo4j::Node` first to get the...

Interesting. I was actually going to alias `NATS::Headers` to `HTTP::Headers`, but for some reason I thought NATS message headers didn't allow duplicates.

@samueleaton Awesome, thanks! I'm a little overburdened this week, but hopefully can get to these next week.

[NARRATOR]: He did not get to these the following week

Looks like [NATS does support duplicate headers](https://github.com/nats-io/nats.go/blob/b795e86201ac8846ae7836720d02502395ba1762/nats.go#L3272-L3305), but they're case-sensitive, at least in the Golang client. Using `HTTP::Headers` in Crystal would have different semantics since it treats headers as case-insensitive....

Making the naming (especially for this very specific thing) more consistent is something I've been meaning to do, tbh. I don't even remember where I got `body` from because I...

> its common for methods in crystal to use `raw` for the, _eh-hem_, more raw representation. how about `#raw_data` is the bytes, and `#data` does a lazy stringify. I really...

@samueleaton If you want to change these up to be `data : Bytes` and `data_string : String`, I'll go ahead and merge.

Yeah, I think what I was going for here was similar to the Postgres-style URI where the path points to the DB name. I wouldn't mind adding support for a...

Disconnection handling is the part of the codebase I have the least confidence in despite also [testing it](https://github.com/jgaskins/nats/blob/4533a8cdc37c27045b4c535a8a93ec32d7266894/spec/nats_spec.cr#L165-L193) since closing the socket like I did in the spec sends a...