When drawing a certain state, wich has among others a grid of 40 by 50, will displaying it as a canvas give much better performance then creating the HTML with rows, as I do now? Since I only want to show some things on the grid it seems to make more sense. Probably I will need to detect/set the size of the canvas, and lose a bit in responsiveness

I already found some awnsers and it seems to offer much better performance, and not being that much more difficult then what I’m using now. It would also seem to be fairly well possible to make it responsive at the same time

Check out react-virtualized

thanks, I will look into that, although as its for a game, it doesn’t seem a good fit

that's probably true

https://twitter.com/roman01la/status/832165875797082112

Hi, what approach similar to: https://github.com/exponent/react-native-invertible-scroll-view would you take on web?

@roman01la nice is that in Canary or is that already in Chrome now?

@dnolen in Canary, not enabled by default

nice

I'm trying to access *data-readers* in cljs, (specifically lumo), I saw a reference to *cljs-data-readers* in an issue, but this also doesn't resolve. Does this var not exist in cljs for any particular reason?

i have a function that returns a list with two elements and another function that takes two arguments. i would like to apply the list's elements as arguments to the second function? google results that i found for apply are about macros and Java interop, couldn't see an obvious solution

use clojure.core/apply

thanks

it was too obvious :))

@dominicm: I still haven't got around to implementing proper tagged literal support in Lumo. Can you open an issue if there isn't one already, please?

Done!

why not compile (.-x y) to y["x"] instead of y.x.

@doglooksgood then there would be no way to have closure advanced compilation rename property names

understood. but this case is quite few, I think.

when I want to get property from some already defined struct, such as js/Event, I should always use aget?

use aget if it is not a standard name closure would know about, or that you don't have an extern for

by the way, i think it can make a big difference, and this is why closure gives the smallest size in most tests

also, there are some macros and libraries floating around to help with this, for example: https://github.com/reagent-project/reagent/blob/master/src/reagent/interop.clj#L26

Is there an equivalent of *file* in cljs?

@isak I haven't use closure compiler before I knew clojurescript. how can I get the list of those names that closure compiler can recognise?

@doglooksgood AFAIK, it knows about the ones you can find here: https://github.com/google/closure-compiler/tree/master/externs

also here is a nice place to try things out in advanced mode: https://closure-compiler.appspot.com/home

cool, it's helpful

@matthavener there isn’t currently

Where's the best place to see a list of known issues?

I think 1.9.473 has an issue with desctructured bindings on :keys.

@phil this is the CLJS issue tracker: http://dev.clojure.org/jira/browse/CLJS

I don’t think anything like what you’re describing has been reported. Do you have a minimal failing example?

@phil the first thing to do is just report what you are seeing here or in the IRC channel or mailing list

then if someone confirms the problem - make an issue

Ok, thanks.

@anmonteiro doesn't have the same issue, so my build might be a mess.

that conclusion sounds unlikely - I don’t see how your build could affect destructuring

(let [{:keys [:city :something]} {:city "London"}] city) => nil

That's what I'm seeing.

@phil is this something you’re seeing under advanced or something?

Ah, yes, advanced.

and you did a clean build? are you using code splitting modules? etc

Just trying in a completely fresh build (I'd tried lein clean)

I'm coming back to cs after a long break.

lein clean doesn’t do what you think

unless you specified :clean-targets

your bug is most likely a dirty build

like a bad constants table or something

@phil are you using keywords in your :keys vector? should be a vector of symbols

@jr it works regardless