This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2017-11-13
Channels
- # beginners (71)
- # boot (61)
- # clara (49)
- # cljs-dev (9)
- # cljsjs (2)
- # cljsrn (5)
- # clojure (55)
- # clojure-android (1)
- # clojure-italy (4)
- # clojure-spec (39)
- # clojure-uk (56)
- # clojurescript (69)
- # cursive (5)
- # data-science (1)
- # defnpodcast (6)
- # devcards (1)
- # duct (12)
- # figwheel (3)
- # fulcro (18)
- # leiningen (35)
- # lumo (19)
- # midje (1)
- # off-topic (22)
- # om (3)
- # onyx (23)
- # portkey (3)
- # re-frame (20)
- # reagent (23)
- # ring-swagger (6)
- # shadow-cljs (119)
- # specter (7)
- # unrepl (25)
@hkjels they are... why do you ask?
Hey folks, i'm investigating options for organizing state in my client app. I want DB-like experience, but existing solutions are still not convincing, including DataScript. How do you think, is it viable to use just normalized maps and query/update them with Specter? Is it easy to emulate joins with Specter?
@ul joins are not in the scope of specter, but otherwise specter is very helpful for managing state
since it's so elegant to handle data structures of arbitrary complexity, it gives you the freedom to model your data in the best possible way
I see at least two pros on that way: 1. My data will be highly explorable outside main query mechanics (because it will stay just Clojure datastructures) 2. It will be friendly to reactive approach I use: Specter queries will fit perfectly into reactive formulas, and with proper data organization will trigger not much redundant recalculations.