Open eval-on-point opened 2 weeks ago
Many thanks for your suggestion! I think it is a reasonable approach, though it hurts a bit to put something in addition to true
and false
an option for a key ending in ?
. Still, I don't really think we need yet another key-value pair here, so I'll see what I can to do to make this work. I'll let you know when I've figured it out and finished it.
That is an interesting philosophical point and is possibly evidence that ?
should be reserved only for boolean-valued predicates. Check out https://github.com/bbatsov/clojure-style-guide/issues/182#issuecomment-516303299 for an interesting perspective on this. I am not exactly sure where I land.
From the changelog:
I was discussing the
lift-ns?
option with my team this morning and we all (N=3) agreed that we would prefer namespaces to be lifted if there were more than one key in a map, but not to lift it if there were only one. We thought it would be slick iflift-ns?
could optionally take in integer argument instead oftrue
. This would set the minimum number of map elements required beforezprint
performs a namespace lift in a backwards compatible way.