This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2017-07-12
Channels
- # aleph (10)
- # beginners (79)
- # boot (81)
- # chestnut (3)
- # cider (9)
- # cljs-dev (336)
- # cljsrn (17)
- # clojure (121)
- # clojure-boston (1)
- # clojure-italy (4)
- # clojure-nl (1)
- # clojure-russia (218)
- # clojure-spec (32)
- # clojure-uk (98)
- # clojurescript (109)
- # cloverage (1)
- # core-async (5)
- # cursive (17)
- # datascript (15)
- # datomic (38)
- # editors (4)
- # emacs (6)
- # graphql (1)
- # hoplon (140)
- # instaparse (1)
- # jobs (2)
- # klipse (1)
- # leiningen (4)
- # lumo (2)
- # mount (103)
- # off-topic (3)
- # om (8)
- # onyx (19)
- # parinfer (32)
- # pedestal (3)
- # precept (32)
- # re-frame (33)
- # reagent (24)
- # remote-jobs (11)
- # rum (1)
- # spacemacs (1)
- # specter (37)
- # unrepl (4)
- # untangled (43)
- # vim (11)
Has anyone seen issues with :trigger/emit
? I'm running Onyx 0.10.0
, and I see the following exception when I use :trigger/emit
, but not with :trigger/sync
Unfreezable type: class
onyx.refinements$discarding_create_state_update
{:type onyx.refinements$discarding_create_state_update, :as-str "#function[onyx.refinements/discarding-create-state-update]", :original-exception :clojure.lang.ExceptionInfo}
@fellows I haven’t seen that one. Using trigger/emit with the discarding refinement?
that looks pretty similar to what we’re doing in the test https://github.com/onyx-platform/onyx/blob/0.10.x/test/onyx/windowing/emit_aggregate_test.clj#L97
do you think you could try to reproduce it in that test?
@lucasbradstreet -- I'm unable to repro that error running that test in conjunction with my project
and config
settings. But, I can reproduce the error if I start with lein new onyx-app <name>
, change the onyx version to "0.10.0", and add use the following window and trigger settings.
@fellows thanks. If you created a new project would you mind pushing up the fresh project to github? No worries if not, but it'd make my life a little easier
Ah, no worries then. Should be easy enough anyway. Thanks!
that would be perfect
thanks
@fellows oh, it’s because you’re trying to return the window-data arg which is basically a lot of metadata about the event. Unfortunately that record includes a lot of extra stuff that isn’t serializable, such as the refinement fn
No worries, it’d be nice if we checked for that but validating on every submit would cost a lot perf wise.