@akiroz When I try (:body response), I get nothing

*printing that to the console prints nothing

does it print nil or not print anything at all?

i.e. (.log js/console (str "Body Response: " (:body response))) prints Body Response: to console, and nothing else

use clojure's printing fns instead

Ah, I can use that in the browser?

yep, just (prn response)

oh wow

oh yeah, you need to set (enable-console-print!) first

yeah that's the HTML stir I want

(I am seeing the HTML str in the console now, after doing (prn "response was: " response))

👍

It seems that, using console-log or Clojure's prn, I can see (in the browser console) the html str I want

But the response itself seems to be goog.net.XhrIo object

And (:body response) evals to null

getResponseBody and getResponseText appear to be methods of the goog.net.XhrIo's parent class, but calling them on response also just produces 'null'

all you polymer geeks who are dying to help out: see the stuff in index.html and src/shell (except the driver.js bit). driver in ES6, implementation delegated to clojurescript.

at https://github.com/miraj-project/psk2

👋 I'm setting up a simple form with reagent and I can't figure out how to correctly set a default-value on the input. The default-value comes from an api call whose results are eventually sent into this EntryEditor function. Any thoughts on how I can get this to work?

@chris_ care to share the code you're using to make the request?

Got it! I found what I was looking for in the todomvc example for anyone interested: https://github.com/tastejs/todomvc/blob/gh-pages/examples/reagent/src/cljs/todomvc/components/todo_edit.cljs

the example from https://clojurescript.org/news/2017-07-12-clojurescript-is-not-an-island-integrating-node-modules should work by having 1.9.854 in the classpath, right?

@bensu I guess you might be missing :install-deps true in your compiler options

we ended up changing the default behavior last minute and the example should account for it, I’ll edit it

let me know if that doesn’t fix it

https://github.com/clojure/clojurescript-site/pull/128/

Thanks!!!

It worked. Do you mind if I correct the post?

It took me 1 hour to ask the question 🙂

@bensu I submitted the PR above to correct the post

sorry for the waste of time

btw, we don’t install deps by default because of this reason: you may choose to install deps yourself (e.g. with yarn, and not even pass :npm-deps)

and we’ll know that you’re requiring a node module 🙂

as long as they’re installed before CLJS compilation begins, the compiler can figure it out

no need to be sorry! this feature is great. thank you for your help

hey guys, I'm trying out the new :npm-deps and stumbled upon an java.lang.RuntimeException : INTERNAL COMPILER ERROR. at this JS code:

node_modules/dom-helpers/transition/properties.js:17:4
animationEnd = void 0;
node_modules/dom-helpers/transition/properties.js:15:0
var prefix = void 0,

Stacktrace:

java.lang.NullPointerException: null
 at com.google.javascript.rhino.Node.replaceWith (Node.java:872)
    com.google.javascript.jscomp.ProcessCommonJSModules$RewriteModule.updateNameReference (ProcessCommonJSModules.java:1075)
    com.google.javascript.jscomp.ProcessCommonJSModules$RewriteModule.maybeUpdateName (ProcessCommonJSModules.java:954)
    com.google.javascript.jscomp.ProcessCommonJSModules$RewriteModule.visit (ProcessCommonJSModules.java:712)
    com.google.javascript.jscomp.NodeTraversal.traverseBranch (NodeTraversal.java:730)
    com.google.javascript.jscomp.NodeTraversal.traverseChildren (NodeTraversal.java:802)
    com.google.javascript.jscomp.NodeTraversal.traverseBranch (NodeTraversal.java:726)
    com.google.javascript.jscomp.NodeTraversal.traverseChildren (NodeTraversal.java:802)
    com.google.javascript.jscomp.NodeTraversal.handleScript (NodeTraversal.java:680)
    com.google.javascript.jscomp.NodeTraversal.traverseBranch (NodeTraversal.java:705)
    com.google.javascript.jscomp.NodeTraversal.traverse (NodeTraversal.java:307)
    com.google.javascript.jscomp.NodeTraversal.traverseEs6 (NodeTraversal.java:647)
    com.google.javascript.jscomp.ProcessCommonJSModules.process (ProcessCommonJSModules.java:134)
    com.google.javascript.jscomp.Compiler.processAMDAndCommonJSModules (Compiler.java:2106)
    com.google.javascript.jscomp.Compiler.parseInputs (Compiler.java:1792)
    com.google.javascript.jscomp.Compiler.parse (Compiler.java:1007)

@akiroz probably https://github.com/google/closure-compiler/pull/2579

will be in the next version of Closure which we’ll hopefully be able to squeeze into the next release of ClojureScript

@anmonteiro awesome, thanks! 🙂

How do I require the spec.gen namespace in ClojureScript now that we have cljs.spec.alpha? I had [cljs.spec.impl.gen :as gen] earlier.

cljs.spec.gen.alpha

Hmm, that's what I tried without success...

@curlyfry won’t work if you dont have test.check as a dep in your project

@dnolen Sorry about that, I had something else wrong! I think I also got a bit confused by what has been fixed in a later cljs version than what I'm using: CLJS-2242: Lots of undeclared Var warns in cljs.spec.gen.alpha

Can I safely use 1.9.854 btw? I saw it referenced as a 'release candidate'

yes - it should work, there are some minor things we need to fix, but I think no deal breakers

Ok, thanks!

Hi there, following my questions of yesterday, I have a dozen definitions/modules, I load one during the macro expansion phase. I expect the output javascript to NOT include the other modules (compilation :advanced)

which is wrong apparently, am I missing something?

what you’re saying doesn’t really make sense to me

“load during macro expansion phase”, what does that mean?

@dnolen I have a macro that reads an env variable, and require the corresponding module

but there’s no such thing as requiring a module that way, I’m not sure how you tried to accomplish that

but it’s not recommended at all

is this completely wrong: https://clojurians.slack.com/archives/C03S1L9DN/p1501417046949044 ?

I don’t know - but dynamic require is simply not possible

require must be top level only

and it must appear at the top of your source file only

if anything you’re trying doesn’t take these assumptions into account - it’s not going to work

I'm afraid we're not talking about the same thing

I rely on the fact that the macro expansion happens in clojure

macroexpansion doesn’t matter

so in "clojure mode" I read the env variable, require the module and load a variable from it, to produce a simple (def preloaded {:my-def [{:a 1} {:b 2} ...]})

if you mean you’re using the macro splice stuff in, then that’s a different thing

still I would rewind - it’s very hard for me to understand what you are trying to accomplish

I have a evaluator that process a bot definition and user input and produces an output

my evaluator AND the bot definitions (a dozen) are written in clojure

I compile this to js to use it in a google cloud function

I want to avoid loading a dozen definitions every time my cloud function starts. I'm hoping to get a quick win by just picking ONE definition at compile time, that'd produce a smaller js and I'd use it without changing the rest of my code.

So I try the macro route, since I know that's eval'd during the clojure phase, I wrote a macro that loads a module and pull a variable from it. Instead of loading 12 modules, I have a single (def preload {:bot-001 CONTENT-OF-MODULE-DEFINITION})

but why do you need to bother with all this? advanced compilation includes whole program tree shaking / dead code elimination

@lsenta you are bitten by the fact that CLJS includes ALL files it found in your :source-paths in the :advanced compilation

that is no longer true if you use :modules though which only includes what is reached by your :entries

maybe it makes sense to make a :module per “bot”? not sure what you are building

@lsenta maybe :closure-defines could help? see https://cljs.github.io/api/cljs.core/goog-define

@dnolen that's what I'm trying to get, I load a single module so that dead code elimination gets rid of the unused 11 modules

For this I use an env variable to decide at compile time

but @thheller do you have a reference for this? it sounds like dead code elimination would work only for imported libraries then

Same issue with modules & closure defines, the cond over the goog-define value is optimized (I copied the example with identical? in https://cljs.github.io/api/cljs.core/goog-define)

but all the modules are included in the javascript output. Even the one that are never accessed.

@lsenta that will only happen if you have side effects in your leaf entries

if someone is never accessed it will be DCE’ed, if it’s not being DCE’ed something is wrong with the code

Hi -- I’m trying to migrate from consuming an external bundle to consuming node_modules or :npm-deps directly, using cljs master and closure-compiler master. Files using require('assert') fail with JSC_JS_MODULE_LOAD_WARNING. Failed to load module "assert". Do I need to pass an option to clojurescript related to the node API? Alternatively, adding the commonjs assert package to package.json or :npm-deps doesn’t remove the error -- if using this package is the right solution, should I be adding something to my namespace as well?

@wcohen that needs more investigation, can you please open a JIRA ticket?

I think I know what’s wrong but need to have a more in-depth look later

Issue is here: https://dev.clojure.org/jira/browse/CLJS-2290. This is my first bug report, so please let me know if additional or different information would be useful

thanks

Using cljs :infer-externs true I get WARNING - name goog is not defined in the externs coming from the prefix goog.string; and some longer paths under goog.string there should be no reason to create externs for goog closure stuff, so this seems harmless. However, what is causing it? in v1.9.671 I tried searching and best I found was a similar question asked on slack quite a while back.

didn’t see it in jira, unless my searches are failing

oh, also I don’t believe I Have any sort of type hints involving goog.string

@mikerod not enough information to understand what you are saying - come up with something minimal and report an issue in JIRA

@dnolen yeah, not sure how to recreate it so was first checking if this is just a “known thing”. I’ll see about coming up with something that reproduces it

never heard of anything like that

ok, good to know

is there a clojurescript equivalent to alter-var-root? trying to figure how to get https://github.com/bhb/expound printer to be pervasic in a cljs repl

oh nvm, set! was working I was just being thrown by the return value function spraying js all over my repl 😆

any way to stop the repl from trying to print fns?