@darwin: greetings, so this error is a chrome-related, right?

Custom Formatter Failed: ReferenceError: bindRemoteObject is not defined

Has anyone encountered this problem? I have an input element that reads values from an atom (the 'db' in re-frame) and updates the atom upon :on-change with the entered value. This works with IMEs that map one keystroke to one character, but when several keystrokes are required to compose the desired character (e.g. Japanese), the on-change event is picked up and the value is updated before the desired character is composed. This seems to not be a problem with pure react with some experimenting, but do not have any clue. Would appreciate any tips/pointers. Thanks!

@iku000888: you can try combine change event with compositionend event https://developer.mozilla.org/en-US/docs/Web/Events/compositionend

@misha: https://github.com/binaryage/cljs-devtools/issues/14

@delaguardo: Thanks for the response! Taking a look at it

@delaguardo: I tried specifying :compositined, but looks like the event is not being picked up. Looking for other variations.

Where would I report a cljs bug (I read the wiki, but was not closer)?

@iku000888: maybe :on-compositionend ?

@hlolli: mailing list? I thought you could report on jira but looks like you need to sign up

@hlolli: sorry what is the bug in your example?

looking at the source code of cljs.reader/numeric? then it should basically be the same as goog.string/numeric. But irrelevant of that fact, the console print prints QMARK not defined something. Are you not able to reproduce the error message?

@hlolli: well, yeah, but it goes away if I actually require cljs.reader

ok, then its not a bug. But what made me almost crazy last week, was that using figwheel did not produce the bug (maybe cljs.reader is loaded with figwheel?), so everytime I did advance compilation it popped up, took me ages to find the cause. Feel sorry for other developers if they have to go trough the same 🙂

glad you got it working in the end

yes, I guess every developer knows that feeling 🙂

@hlolli: ouch

what, if any, is the cljs equivalent to clojure.core/ns-resolve?

hmm; seems like dynamically referencing vars in cljs is a little more arcane

@shader you are looking at clojurescript self host territory there

I wrote an (hopefully interesting) post about the topic here: http://lambdax.io/blog/posts/2016-03-22-replumb-ast.html

Clojurescript basically does not thread namespaces in the same way as clojure

“Bootstrapped ClojureScript and Planck” presented at Seajure meetup last night at Amperity: https://youtu.be/HnQ89r_dKEM

Hi all, does anyone know how to deal with getting read-string to properly parse the new namespaced map literals? i.e. (read-string “#:a{:b1 :c1}”)

When I try this both in the repl and in dev, I get:

Could not find tag parser for :a in ("inst" "uuid" "queue" "js")

nvm, I see that it isn’t yet supported: http://dev.clojure.org/jira/browse/CLJS-1706

I've just spend several hours trying to get :foreign-libs working and it just wont' work. The file isn't included in the compiled javascript and no error is given.

not fun

Is there any way to reconstruct a literal, so that (foo 1.0) => “1.0” and (foo 1) => “1”? I’m doing a compilation step to glsl, and the fact that I have to write my numbers as strings is annoying me.

@mfikes: cool presentation so far. haven’t watched it fully yet, but one of the questions they asked: https://closure-compiler.appspot.com/home

there’s a closure compiler web service that processes JS for you

ahh. Cool :)

there’s also a REST API for that too 🙂