Fork me on GitHub
#boot
<
2016-09-06
>
richiardiandrea00:09:49

a question about adzerk.env, so in my cljs file I do:

#?(:cljs (env/def
           ASSET_ENDPOINT ""
           API__ENDPOINT ""
           API__WS_ENDPOINT ""))

richiardiandrea00:09:11

will I lose the symbols with :optimizations :advanced ?

richiardiandrea00:09:41

my gut feelings are saying no

onetom02:09:27

@danielsz i saw there was a recent release from system, but the changelog doesn't mention it i would assume 0.3.1-SNAPSHOT became 0.3.1 but then it seems it came after 0.3.2-SNAPSHOT either way, it would be good to mention it in the CHANGELOG im trying it out anyway

flyboarder03:09:18

@richiardiandrea: are you referring to accessing them from JS? It should work with advanced compilation

danielsz03:09:15

@onetom Thank you for pointing it out. I've fixed the CHANGELOG.

danielsz03:09:19

@onetom Latest release is 0.3.1 . There's a 0.3.2-SNAPSHOT available.

onetom03:09:05

quick question. i have a system with a new-web-server component. if starting the system fails i get an error message which shows that the web server was actually started successfully. BUT when i fix the reason of the start failiure i get an "Address already in use" error although I saw the system was stopping:

#'sys/dev:refreshing
Stopping sys/dev
Unloading: (app.profile-test app.users-test app.schema-test sys app.demo boot.maven app.core-test app.excel.data app.excel.sens app.excel.core admin.docs app.profile test-helpers app.schema app.docs app.core app.users)
Reloading: (app.demo sys app.schema-test app.users-test app.profile-test app.users app.core app.docs app.schema test-helpers app.profile admin.docs app.excel.core app.excel.sens app.excel.data app.core-test boot.maven)
...
clojure.lang.ExceptionInfo: Error in component :web in system com.stuartsierra.component.SystemMap calling #'com.stuartsierra.component/start
    data: {:reason
           :com.stuartsierra.component/component-function-threw-exception,
           :function #'com.stuartsierra.component/start,
           :system-key :web,
           :component
           {:options {:port 8200, :join? false},
            :handler
            {:handler
             #object[ring.middleware.cors$wrap_cors$fn__9236 0x76de0359 "ring.middleware.cors$wrap_cors$fn__9236@76de0359"]}},
           :system
           {:db {:uri "datomic:", :conn nil},
            :schema {:db nil},
            :demo-data {:db nil, :schema nil},
            :handler
            {:handler
             #object[ring.middleware.cors$wrap_cors$fn__9236 0x76de0359 "ring.middleware.cors$wrap_cors$fn__9236@76de0359"]},
            :web {:options {:port 8200, :join? false}, :handler nil}}}
    java.net.BindException: Address already in use

onetom03:09:33

it seems the web-server stop code doesn't work correctly then

richiardiandrea03:09:46

@flyboarder: yeah but it works, the problem was elsewhere

onetom03:09:30

@danielsz forgot to mention u for the question above ^^^

danielsz04:09:42

@onetom: Yes, this is a problem we're trying to fix with idempotent start and stops. What particular web component are you using?

onetom04:09:48

im using

(system.components
      [jetty :refer [new-web-server]]
      [datomic :refer [new-datomic-db]])

onetom04:09:21

is there a better option?

onetom04:09:41

@danielsz you mean u would somehow force idempotency? currently the jetty server seems to try being idempotent by doing the dissoc :server... 😕

danielsz04:09:55

@onetom Ah you're using Jetty. This one has idempotent starts and stops. Then it must be something else.

danielsz04:09:32

@onetom jetty is a perfectly good option BTW.

onetom04:09:33

thx. if you haven't seen this issue yet, i will narrow it further down.

danielsz04:09:55

@onetom That would be great. thanks!

mrmcc307:09:39

hello. i’m trying to use boot.core/cp and keep getting

java.lang.IllegalArgumentException: No matching method found: copy
                                            ...
                          boot.tmpdir/add-blob!            tmpdir.clj:   87
Anyone else run into this? seems to work fine when I drop to 2.5.5

alexyakushev07:09:04

Hey everyone. Is there a place to read about how scopes (`provided`, test etc.) work in Boot? AFAIU it is a Maven feature, yet I can't find the explanation how many scopes there are, and what are the rules of applying them.

alexyakushev07:09:39

Also, how can I run something like boot show -d, but exclude things in e.g. test scope?

borkdude09:09:10

is .cljc code automatically reloaded in the clojurescript reload tooling of boot?

juhoteperi09:09:31

Yes, boot-reload only cares about js files written by Cljs

mitchelkuijpers13:09:15

Is there anyone from cljsjs here? I want an updated version of react-select but I saw earlier updates were somehow done by a robot so maybe I should not make a pull request? https://github.com/JedWatson/react-select has a new version o 1.0.0-rc

martinklepsch13:09:30

@mitchelkuijpers hey let's take this to #cljsjs

anmonteiro14:09:15

having the following problem: I’ve got a resources directory that may have a main.js artifact resulting of Figwheel compilation (with lein)

anmonteiro14:09:37

in my Boot-CLJS build I wanna sift that out so it doesn’t interfere with the Boot-cljs build

anmonteiro14:09:16

my sift invocation looks like this:

(sift :include #{#"^devcards\/main.js"} :invert true)

anmonteiro14:09:45

if I run boot that-sift-task-above show -f the main.js file doesn’t appear

anmonteiro14:09:23

however if I run boot that-sift-task-above my-cljs-compilation-task I get the stale main.js and not the one resulting from my Boot-CLJS build

anmonteiro14:09:31

running out of ideas on what to do here

martinklepsch14:09:59

@anmonteiro are you using the watch task as well?

martinklepsch14:09:16

have you reproduced this with and without watch?

anmonteiro14:09:27

let me try without watch

anmonteiro14:09:43

what would be the problem there?

anmonteiro14:09:42

maybe I should sift before the watch task? hrm

martinklepsch14:09:12

I'm just guessing without much thought... 🙂 The watch task stores the fileset and there is some merging going on in cases so could be that this merge op is behaving undexpected

anmonteiro14:09:56

@martinklepsch well I’m sifting the file after the watch task is called, so maybe my mistake

martinklepsch14:09:43

no I think that's probably the right place

martinklepsch14:09:02

any luck without watch?

anmonteiro14:09:34

I had luck without watch, but then it behaved correctly when I readded watch 🙂

anmonteiro14:09:58

something non-deterministic going on here

anmonteiro14:09:08

definitely my favorite type of bug :sarcasm:

anmonteiro14:09:03

oh I also removed ./boot/cache in the process so it might just be that

anmonteiro14:09:16

just realizing I shouldn’t have introduced that variable

martinklepsch14:09:47

not if you meant to track it down 😛 — but it works ¯\(ツ)

anmonteiro14:09:43

also works if I just call sift as the first thing

anmonteiro14:09:13

I don’t expect to have figwheel running in parallel so I can leave it like it is now

martinklepsch14:09:46

if it shows up again might be worth putting some prns inside boot.core/reset-fileset

micha17:09:01

@alexyakushev the scopes are described in the pom reference

micha17:09:48

@mrmcc3 which version of boot demonstrates the issue?

ghadi19:09:32

So last month at TriClojure @stuarthalloway switched clojure versions live in a JVM using boot. How would you do that?

micha19:09:08

@ghadi you would create a new pod with the different version of clojure

micha19:09:26

then you can evaluate expressions in there, start a repl server, etc

martinklepsch19:09:32

hah, that's a nice demo 😄

mrmcc321:09:19

@micha 2.6.0 should I open an issue with repro

micha21:09:56

@mrmcc3 sure, what are you using the cp function for?

mrmcc321:09:56

playing around with striping frontmatter from a file and overwriting the original

micha21:09:18

ah, i would recommend using a different approach

martinklepsch21:09:02

In any case a repro might be good to exclude any unwanted behavior?

micha21:09:04

you are going to need to have a file anyway, so in general i would always just emit the file to a temp dir and use add-resource etc to add it to the fileset

micha21:09:19

yeah the repro will help me fix the bug 🙂

micha21:09:38

the cp function is complicated to use correctly

mrmcc321:09:32

yeah that was my initial approach but wanted to preserve the input/output of the original file.

micha21:09:02

i see, yes

micha21:09:11

that's actually exactly what the cp function is for

micha21:09:33

so disregard my advice, i'll just fix the bug 🙂