Clojurians
# om

This page is not created by, affiliated with, or supported by Slack Technologies, Inc.

danielstockton 00:02:19

in an om context or the word itself?

danielstockton 00:03:30

novelty - "the quality of being new, original, or unusual."
in om, its something new from the server that you want to merge into your app state

haywood 03:30:28

wooord, nice thanks man

haywood 03:31:47

I've spent the weekend just modeling my backend with datomic so the pull queries against the data match the data on the client. now trying to figure out how to wire the om.next client with the server

haywood 03:32:04

do I need to use transit?

haywood 03:32:57

fyi not really looking for answers... just happy making noise

haywood 03:33:05

I'll get it eventually

danielstockton 10:07:09

transit is a good choice

danielstockton 10:07:23

especially since you're dealing with clojure on both ends

amann 15:39:08

For posterity’s sake on (https://clojurians.slack.com/archives/om/p1479413311007034):

I posted here a little while back about seeing really high CPU util from FireFox and that I had traced it down to some GC and Render stuff going on. What made this unique was that Chrome had no such problems. I was able to narrow down the errant code later that day, but am still on the hunt for the Chrome vs Firefox question.

Turned out, in the CircleCI Om code that we had a little something like this going on:

(reify om/IWillUpdate (will-update [_ something _] (when (something-booleany? something) (om/set-state! owner [:this-is-cool-yeah?] "woah boy!"))))

Changing this to first check that the owner’s state hasn’t been set dropped CPU util down to comparable between Chrome and FireFox. Changing this entire thing to be under IWillReceiveProps was an even cleaner solution (and ultimately where this should have been to begin with). Why this is a different issue between the two browsers, I do not know, but this is what I’m aiming to figure out.

For those interested in the actual code, have fun! https://github.com/circleci/frontend/commit/c787a719bc73ba9b62f414e71bd276194a637ee2

fatihict 21:02:17

Hi folks, I (a ClojureScript noob) am struggling with the following:

I want to use this library: [cljsjs/react-grid-layout "0.13.7-0"] and have included it in my project and imported in my namespace. I have the following code to create a ReactGridLayout component: (def GridLayout (js/React.createFactory js/ReactGridLayout)) but I get errors in my console saying that ReactGridLayout is not defined... Can anyone point me in the right direction to solve this problem. I have no clue how to solve this.. :confounded:

haywood 21:04:25

I don't think it's in the js/ namespace

fatihict 21:10:23

:thinking_face: Thanks for your response. Do you know in which namespace I can look for it?

haywood 21:12:24

maybe try (:require [cljsjs.react-grid-layout :as grid]) then grid/ReactGridLayout

haywood 21:16:41

that require assumes it's inside the ns macro, in which case you do not need to quote the vector

jr 21:19:33

you need to require cljsjs.react-grid-layout for it to be available as a global

fatihict 21:23:53

Thanks for the tips. I have required react-grid-layout and also tried grid/ReactGridLayout instead of js/ReactGridLayout, but that also didn't work for me.

haywood 21:24:31

you could just add the script tag for the lib in your browser for now

haywood 21:24:39

and access it via js/ReactGridLayout

fatihict 21:26:36

The library is being included <script type="text/javascript" src="http://localhost:8080/js/out/goog/../react-grid-layout.inc.js"></script>, but somehow ReactGridLayout is still not defined..

haywood 21:29:27

gonna load up a test chestnut app

haywood 21:29:28

and see what's up

fatihict 21:29:43

I was using om version 0.8.8, I am now updating to the bleeding edge version... Maybe that will solve my issue..

fatihict 21:30:06

Oh, and it works now :slightly_smiling_face:

fatihict 21:30:21

Thanks @haywood for thinking with me!

haywood 21:30:40

:smile:

haywood 21:50:50

ah, so it does get thrown on the js/ namespace

fatihict 22:04:43

Yes, it does :slightly_smiling_face:

mlimotte 23:48:16

I'm working with a toy om.next app to learn about it. I have a page with a couple of divs which I want to make into Tabs with goog.ui.TabPane.
I was only able to get this to work by adding the root component and then creating the Tab:
(om/add-root! reconciler RootView (gdom/getElement "app")) (TabPane. (gdom/getElement "Root")) ; "Root" refers the rendered RootView I.e. The TabPane constructor is called outside of any component's (render) method. How can I do this inside a Component?
Can I pass a component as the arg to TabPane? Or how else do I identify the container to create a TabPane from, since (gdom/getElement "Root") seems to be nil until after the render is done.

mlimotte 23:49:07

I thought about doing the (TabPane. ...) in the RootView components' componentDidMount() method. Not sure if that will work or is the recommended approach. I'm going to try it?

haywood 23:50:44

you could create a React wrapper around the tab pane, which would mostly consist of lifecycle method manipulation.(TabPane. React.findDOMNode(this))

mlimotte 23:51:01

componentDidMount works! But is that the recommended approach. Seems non-obvious to me, but I'm also new to front-end development, as well as Om dev.

haywood 23:51:24

that's what you gotta do when you're bringing jquery plugins into react

haywood 23:51:29

or other non-react stuff

mlimotte 23:51:56

@haywood can you point me at an example?

haywood 23:54:35

hm, doing some googling

haywood 23:56:14

you can create the react wrapper with om or just dirty js interop

haywood 23:56:22

om has all the lifecycle interfaces