The current output format makes each item readable by itself, but it causes repetition of the field names which is unneeded in some situations. I'd find it useful to be able to have each row (both header and item) represented as an array.
The headers would be at index 0, and then the rows would proceed on from that.
Is this something you would be interested in adding to the project? If you don't want to code yourself I could write a pull request,