Fork me on GitHub
Steve Upstill00:02:12

@gonewest818 Yep, that's me. Not too many of us Upstills around.


Very nice to meet you!

Steve Upstill00:02:17

Back at you! I don't often run into Companion users anymore--though, admittedly, it's been a while since I've been to SIGGRAPH.


I'm writing a library that deals with geometry and I did it all with points as maps with :x :y keywords. this has some advantages as other types can use point functions but the library works and I'm refactoring it and wondering how strongly I should feel about doing something like deftype or defrecord and how to think about this vs other sins I may be comitting


@awerner32 How strongly do you feel about performance vs convenience?


Any behaviour difference between put! and >! in core.async other than put! can be attached with a callback and >! must be called inside a go block ?


@funyako.funyao156 callback will called when you put! something on channel(channel can have fixed size) and >! will stop execution of go block until something can be added


@funyako.funyao156 >! ends up calling put! so the differences are fairly minimal


>! is basically (call-cc (fn [k] (put! c val k))) if Clojure had cal-cc

Markus Åkerlund Larsson14:02:30

Anyone familiar with cljs-http? I'm trying to use it, but no matter what method or url I use I end up getting a {:status 0, :success false}.


You can try debugging options provided by clj-http -

Markus Åkerlund Larsson15:02:18

I'm not sure the libraries work the same way. Either way it doesn't seem to do much.


my bad, sorry. I miss s in library name

Markus Åkerlund Larsson16:02:14

Seems it's not a library issue at all but to do with same-origin policy stuff in the browser.


are you seeing browser console messages about CORS headers?

Markus Åkerlund Larsson17:02:01

Yep. Something about * while 'include'

Markus Åkerlund Larsson17:02:06

I only realized I can actually look at the browser console about half an hour ago, so that's the level I'm at when it comes to the web


i was baffled by that status 0 return too


I’m wondering how good or bad idea is to use alter-var-root function to “decorate” my functions with metrics logic . I mean , if enabled ( collecting some metrics ) I will replace the root-var of some functions with my metric logic and all the code remains the same


I am new to spec. Is there a way to spec a function that expects an Atom?


(def arg1 (atom ""))

(defn some-func 

(some-func arg1)


@tkjone you could spec that it is an (partial instance? clojure.lang.Atom). Unless you want to spec the contents of the atom. Then, don’t. 🙂 Write a function that takes data, returns data; and another that derefs/swaps the atom to provide that data.


I have 2 vectors and I need to calculate their mutual difference using


the problem I have is that I want to define equality for the elements of those arrays in a custom way


any suggestions?