Fork me on GitHub
#om
<
2016-03-30
>
threepointone03:03:04

What are you not able to do on the server side? I don't understand the 'up in the air' comment.

jimmy03:03:41

@isak: it's good to know that someone already have some work on .net 😄

seanirby04:03:43

what version of om are you all using?

petterik05:03:36

Alpha 31 currently. Updating to 32 some time this week

tomjack06:03:54

is there any hope of using react-motion's components directly?

tomjack06:03:16

react-motion will call your child-fn from RAF, so need to manually convey bindings for the factories in child-fn. after that, so far so good..

marianoguerra13:03:19

hi, is there an official api to get the state history in om.next (alpha33-SNAPSHOT)?

marianoguerra13:03:40

I'm showing om.next at a local react.js meetup, I already have the export/import state like circleci

marianoguerra13:03:01

but I would like to enable a time traveling widget with the last N values of the state atom

seanirby13:03:55

@marianoguerra: I've never used it yet but it looks like om.next/from-history will do what you need

marianoguerra13:03:42

@seanirby: thing is I need to get the uuid from somewhere simple_smile

marianoguerra13:03:10

if can I subscribe to the thing that logs in the console to collect the uuids then I'm done

marianoguerra13:03:46

the alternative is to call app-staet and add-watch

anmonteiro13:03:03

and find something that solves your problem

marianoguerra13:03:27

I think I will call add-watch on the atom returned from app-state and see if it works

marianoguerra13:03:50

I think the nice official way would be to be able to subscribe to the transactions and move the logging as a subscriber of that API

seanirby13:03:09

hey @anmonteiro, i noticed when bumping my version from alpha30 to alpha32 that :normalize true had no effect when I passed an atom to the reconciler. do you know if this is a regression or is the reconciler expecting a map for normalization to occur?

marianoguerra13:03:27

add-watch works like a charm simple_smile

anmonteiro13:03:20

I'll try it out later and let you know if there's a regression

anmonteiro16:03:48

@dnolen: there's no symbol-identical? in Clojure

dnolen16:03:10

oh …. sorry right

dnolen16:03:15

PR said that. OK.

tomjack20:03:19

I was also tripped up by 6243eb499b

tomjack20:03:19

My fix was to pass a map, then (defonce app-state (om/app-state app-reconciler)) to get ahold of the atom

tomjack20:03:41

I was confused, but the new behavior makes sense to me now

seanirby20:03:28

tomjack: I haven't needed to pull the app-state out of the reconciler yet. I've just been using @reconciler. What's the difference?

tomjack20:03:59

I mean if you had been def'ing an atom and passing that in

tomjack20:03:15

Then instead you can pass the map, but still get the atom after

tomjack20:03:28

Otherwise just pass a map :)

tomjack20:03:30

I like to have the atom so I can swap/reset for development purposes

anmonteiro21:03:54

@dnolen: just noticed we have dispatch repeated in 3 places: om.next.impl.parser, om.next and om.next.server

dnolen21:03:40

I think just punted because it’s not very important and I didn’t want to think about where to put it

dnolen21:03:00

it’s also annoying to have to import it use it

anmonteiro21:03:00

not sure if it's too much of a problem, since it's a very simple thing

dnolen21:03:18

so I copied it into each namespace that somebody might use by itself so to speak

anmonteiro21:03:06

yea.. we could probably do (def dispatch om.next.impl.parser/dispatch) on the others if we want to avoid the repetition

anmonteiro21:03:23

since parser can be imported in both clj and cljs

dnolen21:03:34

yeah that pattern is just to be avoided

dnolen21:03:41

aliasing prevents dead code elimination

anmonteiro21:03:46

because of advanced optimizations?

anmonteiro21:03:27

alright, just something I noticed, not much we can do about it then