This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2021-05-13
Channels
- # announcements (8)
- # architecture (11)
- # babashka (159)
- # beginners (112)
- # biff (3)
- # chlorine-clover (4)
- # cider (10)
- # clj-kondo (51)
- # cljs-dev (43)
- # cljsrn (10)
- # clojure (45)
- # clojure-bay-area (5)
- # clojure-europe (11)
- # clojure-france (4)
- # clojure-italy (4)
- # clojure-nl (2)
- # clojure-norway (1)
- # clojure-sweden (1)
- # clojure-uk (8)
- # clojurescript (75)
- # code-reviews (1)
- # community-development (2)
- # conjure (88)
- # cryogen (5)
- # data-science (1)
- # datomic (3)
- # dirac (2)
- # fulcro (4)
- # helix (1)
- # jackdaw (5)
- # kaocha (5)
- # leiningen (2)
- # lsp (49)
- # malli (9)
- # mid-cities-meetup (1)
- # off-topic (8)
- # pathom (3)
- # polylith (19)
- # re-frame (6)
- # releases (3)
- # rewrite-clj (1)
- # shadow-cljs (98)
- # spacemacs (2)
- # tools-deps (6)
- # vim (4)
- # xtdb (6)
so this can be true: (and (= x x2) (not= (crux/entity node x) (crux/entity node x2))
thanks to metadata.. bug?
in particular this really caught me off guard:
(crux/submit-tx node [[:crux.tx/put {:crux.db/id (with-meta {:a 1} {:meta 1})}]]) (crux/entity (crux/db node) {:a 1}) ; => nil
particularly because their (hash ...)
are the same, so crux's own hashing is breaking clojure semantics (and fixing it at this point would mean a breaking change)
Hey @U797MAJ8M - thanks for raising this - it would indeed be a breaking change. I've got some notes on this, will clean them up and publish them into an issue