(ns foo ;;cljs
  (:require ["esrever" :as esrever]))

(let [s "Lorem ipsum 𝌆 dolor sit ameͨ͆t."
      a (clojure.string/reverse s)
      b (esrever/reverse s)]
  (js/console.log s)
  (js/console.log a)
  (js/console.log b)
  (js/console.log (= a b)))

https://dev.clojure.org/jira/browse/CLJS-3073

Use of undeclared Var threedays.core/format

  1  (format "%5d" 3)
      ^---

@m373h4n there is a format in Google closure, but you need to import it

(goog.string.format "%04d" 5)  ;;=> "0005"

misha, dominicm, thank you very much

Warning: React does not recognize the componentDidMount prop on a DOM element.

(defn card
  [item]
  [:div.holder {:component-did-mount (resetMsnryLayout)}
   [:div.card
    [:div.card-image
...

where i should put this :component-did-mount

I have a memory leak in my app, and I wanted to know if anyone has some advice on how to tackle the problem? The leak only appears on Windows

@neo2551 need a lot more information to be helpful - did you confirm via Chrome/Firefox dev tools that you're not leaking there? what version of IE are you talking about? And is the observed behavior in IE a crash?

It happens on Firefox on Windows

But kit Firefox Linux

@dnolen thanks a lot for helping. I will try to do more thorough test with multiple browser first.

What are the common cause of memory leak in cljs? (With reagent/re-frame)

I don't think you can point to a common case

if this is Firefox then you're in luck - it has descent memory profiling tools

if it has a heap snapshot feature that can be very helpful

(er, well actually re: common case, lazy sequences)

that's the first thing to check for - if it's not that then you have to go hunting

Thanks :)

<memory leak in IE 10> = “Just go buy more RAM, thanks”

@neo2551 usual issue is to do some kind of operation that involves an infinite lazy sequence and then realize it - which of course will just eat up memory

but that would affect every browser

But if this is the case, it would consume the memory super fast?

https://developer.mozilla.org/en-US/docs/Tools/Memory

@neo2551 yes it would explode almost immediately, locking up the browser

I would use the above to find the leak

Thanks. I have an app that loops events.

Thanks I will search!

And also thanks a lot for your talks! They are really inspiring

I wish we had the slides for the one when you talk about simpler tools.

@lilactown I constantly get Use of undeclared Var hx.hooks/useEffect When i try to utilize your library

what version are you using?

latest, i think

let me see

lilactown/hx {:mvn/version "LATEST"}

seems weird. does it point to a specific line?

@lilactown https://i.imgur.com/b1VYd0J.png

FYI I’ve never seen that way of referring to maven versions

https://i.imgur.com/lFbNxET.png

I think what’s happening is that it’s not pulling the SNAPSHOT version

so you’re using 0.4.0, which doesn’t have the useAbcXyz named fns

sure, but i have already had it tested iwith what you refer at your site

try "0.5.0-SNAPSHOT"

on it

I haven’t done a proper release of what the docs reflect. sorry about that

no worries

blew up

https://i.imgur.com/UgTnRXC.png

Is it possible to extract ClojureScript compiler warnings as data?

Or turn warnings into errors

anyone know of a good way or workaround to do with-out-str on console.log? Presumably the console entries are stored somewhere, is there some way to access them?

Alternatively, does anyone know what the console uses to display object properties? like some kind of format function I can call on MouseEvent that would return me the string that is displayed in the console with console.log

@ghadi you can hook into warnings via :warning-handlers. I am experimenting with this using https://github.com/emezeske/lein-cljsbuild. Here’s a link to my playground warning-handler: https://github.com/lread/rewrite-cljs-playground/blob/cljc-spike-2/test/rewrite_clj/warning_handler.clj which is specified here: https://github.com/lread/rewrite-cljs-playground/blob/cljc-spike-2/project.clj#L36

oh, useful, thanks!

@lee is that a lein-cljsbuild thing or a clojurescript thing?

I don't see it listed here https://clojurescript.org/reference/compiler-options

## 1.8.40

### Fixes
* CLJS-1603: Only warn for misspelled comp/REPL opts
* :warning-handlers missing for known compiler options
* CLJS-1592: Self-host: Robustness for core tests
* 

It is a clojurescript thing http://cljs.github.io/api/compiler-options/#warning-handlers

strange

lein-cljsbuild made it easy to specify a warning-handler, so I took that route for now.

https://github.com/clojure/clojurescript/blob/master/src/main/clojure/cljs/closure.clj#L206

nice! does :warning-handlers resolve a symbol into a function? or do you need to supply a function from code?

(I can try it)

using something like requiring-resolve would be nice for build tooling

no, that was what I found a bit awkward with it and why I started my experiment with lein-cljsbuild which does resolve for you.

FWIW, here’s where lein-cljsbuild helps out with the setup of warning-handlers: https://github.com/emezeske/lein-cljsbuild/blob/master/plugin/src/leiningen/cljsbuild.clj#L126

I wonder if :watch-fn :watch-error-fn and :warning-handlers could use requiring-resolve

I don't know how that interacts with self-hosting... but it would be nice for build automation

just point to a var already on the classpath

prepl does this aleady with its own version of requiring-resolve https://github.com/clojure/clojure/blob/master/src/clj/clojure/core/server.clj#L263-L270

@ghadi I don't see any issue for self-hosting - and even if there were that would be considered separately

patch welcome for this and I completely agree - would be simple - I think the only consideration here would be making sure this doesn't tie us 1.10 or something like that

ok

are there other callbacks exposed?

more than: https://clojurians.slack.com/archives/C03S1L9DN/p1556216317048400

I only grepped the docs quickly

@ghadi that's probably all that's there at the moment

Is there a guide or tutorial to resolving js promises in cljs? I know that the promise will never reject

@mv i've been successfully using funcool/promesa with js promises for quite a while http://funcool.github.io/promesa/latest/

Yes I saw that, was just curious if there was a stdlib or core way of doing it

If not, thanks!

no stdlib way afaik, other than using the completion callbacks to put the resolution onto a core.async channel

(-> (returns-a-promise-of-a-thing)
    (.then (fn [thing] (do-something-with-thing thing)))

@mv @mccraigmccraig

on the topic of promises — anyone had to implement something to run multiple promises sequentially?

well, sure, if you only want to resolve a single promise - when you want to compose chains of them with sensible error handling you might want a lib

yes @martinklepsch exactly that

FWIW I use promesa heavily, but don't find it generally better than just interop

@martinklepsch not sure I totally understand. do you mean, given a collection of promises, that you run each one after the other, feeding result of one into the next?

we use the promesa impl of the funcool/cats promise monad context a lot - i really like it - the monadic bind syntax makes for very comprehensible code

more like this

a().then(b).then(c).then(function(result) {
   // all are done here
});

I don't actually care about the result of the previous run but yeah, first promise should resolve before second is "started"

@martinklepsch (cats.core/mlet promesa-ctx [a (thing) b (other-thing) c (ya-thing)] (return (x a b c)))

it's hard to know what exactly you want to happen different than writing out that promise chain exactly

oh right, sorry — I want to do this in a map like interface over a sequence of items.

which makes it hard to do it with syntax constructs like mlet etc

fundamentally it's a reduce, right?

so you could do

(defn chainp "Chains a promise-returning function to a promise"
  [p1 f]
  (.then p1 f))

(.then (reduce chainp (js/Promise.resolve) [a b c])
   (fn [result]
     ;; all are done here
    ))

or something like >>= - https://github.com/funcool/cats/blob/master/src/cats/core.cljc#L889

you're right, it's really that simple I guess. 🙂

Argh, how do I set element 0 of a js array again? (set! (nth (.-data obj) 0) "something-else") doesn’t work

Nevermind, it’s aset

Filed https://dev.clojure.org/jira/browse/CLJS-3074 I might not have the time to look at this, so it's up for grabs by any motivated party! @dnolen

I suppose I could take a crack at it. Seems like an easy first issue?

Wait.. wrong room to ask, I’ll head over to #cljs-dev