This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2016-04-26
Channels
- # admin-announcements (4)
- # beginners (3)
- # boot (78)
- # cider (13)
- # cljs-dev (29)
- # cljs-edn (8)
- # cljsjs (11)
- # cljsrn (15)
- # clojure (81)
- # clojure-beijing (2)
- # clojure-belgium (3)
- # clojure-canada (1)
- # clojure-dusseldorf (8)
- # clojure-greece (6)
- # clojure-russia (40)
- # clojure-sg (1)
- # clojure-uk (59)
- # clojurebridge (1)
- # clojurescript (101)
- # core-logic (1)
- # cursive (3)
- # data-science (1)
- # datomic (60)
- # emacs (4)
- # error-message-catalog (12)
- # funcool (1)
- # hoplon (60)
- # jobs (1)
- # jobs-discuss (40)
- # leiningen (5)
- # liberator (1)
- # mount (22)
- # off-topic (8)
- # om (16)
- # onyx (53)
- # re-frame (11)
- # reagent (2)
- # specter (4)
- # testing (18)
- # untangled (51)
Before I open an issue about it, has anyone gotten the :source-map-timestamp
compiler option to work with boot-cljs?
@domkm I have it true yes, what problem do you have?
@richiardiandrea: Does it modify either the js file name or the source mapping url comment?
lemme check
@domkm: I am trying to format it in a decent way
what is the name of the key in the json?
Oh I thought it was a JSON file
@domkm: I don't see the timestamp either
only "version": 3
@richiardiandrea: Thanks
So, I’m pretty new to Boot (and Clojure/Script) in general, but something I noticed when experimenting a bit… is it really desirable to have the target
task overwrite the specified/default “target” directory by default? Seems a little risky since a value of .
(or worse, a directory above that) will get 100% replaced when the Boot task is run.
Or really to have it overwrite any directory at or above the current directory. Would a separate task to “copy” or otherwise move the correct build artifacts after target
has been called make more sense?
Might not be any different I suppose since it’s still explicit just with target
… anyway, just a thought
@maxp: I think there’s no way to omit alpha/beta as these are handled as “regular releases” by maven/aether
@rain_omen: good thought, maybe a warning if the —dir
passed to target
is .
or similar would help?
Warning probably isn’t enough. Guess it should be fatal. In the past I haven’t heard of people erasing their projects accidentally but this might a good addition anyways
@martinklepsch: Yeah, I would think fatal would be better. If only because the place a warning might come up (I’d imagine within the task itself) would be too late before that task already cleaned the target dir.
Hey, working with boot cljs repl. For some reason, when I enter a Namespace with (in-ns) it does not seem to require all the dependencies.
cljs.user=> (in-ns 'booty.core)
nil
booty.core=> (om/from-history reconciler "882b71b2-1584-4b86-95ac-5fe87abea1a1")
WARNING: No such namespace: om, could not locate om.cljs, om.cljc, or Closure namespace "" at line 1 <cljs repl>
WARNING: Use of undeclared Var om/from-history at line 1 <cljs repl>
WARNING: Use of undeclared Var booty.core/reconciler at line 1 <cljs repl>
If I do this, it works:
booty.core=> (ns booty.core
#_=> (:require-macros [cljs.core.async.macros :refer [go]])
#_=> (:require [goog.dom :as gdom]
#_=> [cljs.core.async :as async :refer [<! >! put! chan]]
#_=> [clojure.string :as string]
#_=> [om.next :as om :refer-macros [defui]]
#_=> [om.dom :as dom])
#_=> (:import [goog Uri]
#_=> [ Jsonp]))
booty.core=> (om/from-history reconciler #uuid "07f06ad2-1964-4cd7-8345-0d86fbf007ac")
WARNING: Use of undeclared Var booty.core/reconciler at line 1 <cljs repl>
{:om.next/queries {#object[booty.core.AutoCompleter] {:params {:query "dimite"}}}, :search/results #js ["Dimitar Berbatov" "Dimitri Payet" "Dimitrije Ljotić" "Dimitri Ashkenazy" "Dimitrie D. Pătrășcanu" "Dimitrije Bužarovski" "Dimitrios Papanikolaou" "Dimitrios Kiousopoulos" "Dimitrios Konstantopoulos" "Dimitris Diamantidis"], :om.next/tables #{}}
@dimiter: i'm not very familiar with the cljs repl, but in clojure you usually need to (require 'some-ns)
before you (in-ns 'some-ns)
in-ns
will create an empty namespace if one has not already been created/populated
has anyone run into this issue before with boot-http? it just started and I dont know what I changed. .. 😞
java.io.FileNotFoundException: Could not locate ring/middleware/reload__init.class or ring/middleware/reload.clj on classpath.
i’m not even using boot-http in the task I’m running, but boot still complains with this and throws
is this coming out of boot-http for sure?
are you guys using a docker workflow? if yes, do you map a local project folder to be visible in the container and launch boot repl
inside it?
@richiardiandrea: that's how i've done it... and sometimes i map m2 in the container to the host's m2
thus caching maven stuff
great, I saw someone actually pushing it to the limit and have the editor run from the container as well
but I think it is not strictly necessary
X11 gets really tricky w/ docker in my experience
yep, there was a talk
at the last conj
so I saw there is a boot docker image somewhere right? I am about to bootstrap a project and I might take that path (and maybe deploy with kubernetes...alternatives?)
I mean, at the end of the day the whole point is to deploy the docker container somewhere...consistently and to scale up if necessary...
we use docker and docker-compose to dev on things but are kind of moving away from it, at least on AWS
thanks @alandipert you saved me some clicks 😄
can I ask you which direction you are taking from here (I am curious)
well we are on AWS, and so experimented with both docker and war deploys on elastic beanstalk
i guess it's really EB we're moving away from
(I am asking because I remember micha telling me that AWS is a pain to scale ... if my memory does not fail)
ah yes I remember that
maybe he was talking about elastic beanstalk
yes I think so
AWS itself, esp. the core services like EC2/S3 are rock solid pieces to build on
have you considered Google Cloud?
I am investigating what is the best way to deploy docker containers directly as well
You guys seem clever enough to follow your footprints 😄 😄
ahah we're just trying to get by without messing up too much
while keeping our jobs and some semblance of our reputations
ahahahhaha
but yeah re: deploys we basically have our own conventions and "infrastructure" to do it
our own machine images, conventions for config files/logging
shell scripts, this is why micha made jt, to process AWS api responses
yeah I can imagine...
jt
is awesome
meaning really awesome
not "not everything is awesome"
in any case, I like your attitude to a lot of things, at ClojureWest Daniel's talk was really good, and exactly your style, fun and informative
oh were you at clojure/west?
very cool, bummed i couldn't make this one. maybe the next
the best for me was to meet people in person, especially people like bbatsov 😄
yeah meeting people for real is so much fun
definitely
and yeah, daniel's talk was really great, just caught it on youtube. altho i must say his style and awesomeness are all his own, i can take no credit
other than having the good taste to be his friend
he has such an awesome design sense also, something lacking in almsot everything i do lol
yeah but this is what I mean, you've kind of happen to be a good team overall
most things i make look visually like an academic homepage from 1996
ahahah