Opened an interesting ticket: https://dev.clojure.org/jira/browse/CLJS-2485

@mfikes I think I’ve applied all the MapEntry related stuff

@dnolen Thanks! This one was recently discovered, and could be a regression for code like (rseq (first {:a 1})) https://dev.clojure.org/jira/browse/CLJS-2484

k, looking at that one

applied

It is great that the map entry changes end up being more than just for "correctness". Here is the perf snippet from the news article: Finally, using a dedicated map entry type can lead to performance improvements in some code that works with map entries. For example, in :advanced mode, this code [source,clojure]

(simple-benchmark [m (zipmap (range 100) (range))]
  (reduce (fn [a [k v]] (if (even? v) (+ a k) a)) 0 m) 100000)

yeah I suspect as much 🙂

I suppose it is a matter of a dedicated 2-element type being very fast over a generic PersistentVector type that has to delegate to internal arrays and such.

yes