This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
2019-07-23
Channels
- # announcements (2)
- # beginners (165)
- # boot (11)
- # cider (11)
- # clj-kondo (7)
- # cljdoc (1)
- # cljsrn (5)
- # clojure (120)
- # clojure-dev (21)
- # clojure-europe (3)
- # clojure-france (1)
- # clojure-italy (62)
- # clojure-nl (8)
- # clojure-spec (26)
- # clojure-uk (40)
- # clojuredesign-podcast (1)
- # clojurescript (3)
- # cursive (2)
- # data-science (2)
- # datomic (10)
- # emacs (2)
- # figwheel-main (1)
- # fulcro (17)
- # graphql (5)
- # hoplon (5)
- # jackdaw (15)
- # jobs (2)
- # juxt (1)
- # luminus (5)
- # off-topic (1)
- # onyx (11)
- # pathom (4)
- # pedestal (1)
- # re-frame (4)
- # reagent (11)
- # reitit (1)
- # remote-jobs (5)
- # shadow-cljs (48)
- # spacemacs (2)
- # specter (4)
- # sql (24)
- # tools-deps (25)
- # vim (82)
Is there a way to control HTTP timeouts or retries with Maven dependencies when using tools-deps? I must have a shakey connection because I’m getting HTTP errors when performing clojure -Stree
.
With enough retries I can get them all, but this isn’t working very well my build process. Any suggestions?
no support for that, sorry
I'm not even sure what Maven offers in the apis for that, but nothing is surfaced through tools.deps
Ok, thanks for the response! Google tells me that there’s some Maven HTTP configurations through XML, but I have very little knowledge in this area. Most of my Maven interactions have been through Clojure tooling. 🙂
XML in pom.xml or in settings.xml?
In settings.xml. If I understand correctly? https://maven.apache.org/guides/mini/guide-http-settings.html#Connection_Timeouts
yes, that's settings.xml, but this is for wagons, which is the "old stuff". tools.deps uses the newer Maven resolver with transports instead
I doubt anything we're doing in tools.deps is looking at it right now (although honestly I'm not certain)
Ah, I see. Thanks. I figured I was not getting something because I don’t know the ecosystem very well.
we're using org.eclipse.aether.transport.http.HttpTransporter, which does seem to have some sort of configuration facility for http parameters. how to actually get that configured would take some spelunking
I kinda see how it could be done
anyhow, if you want to file a ticket to request this, you could do so at https://clojure.atlassian.net/servicedesk/customer/portal/1 - this is the first time I've seen someone ask for it, so pretty far down in the queue
:thumbsup: Thanks. That’s understandable. Dependency pulls have worked well for me too, until my summer relocation to a cabin in the woods. Has been great for hammocking, now just trying to get those dependencies. 🙂
you just need to mirror maven central and clojars before heading to the woods
Today I've found https://clojars.org/luchiniatwork/cambada which has been forked by https://github.com/mikeananev/cambada . Are you using any of those ? How are you doing aot in tools-deps. The fork is 1.0.4 , they are realeasing more often but the issue list is disabled.
I’m using Cambada in my project to do AOT and create an überjar. It’s been working well for me.
some things i've heard of include: https://github.com/EwenG/badigeon https://github.com/juxt/pack.alpha
the clj script I have (version 1.9.0.375 I guess? from the -Sdescribe, not sure how to track that to a tools.dep version) doesn't seem to handle '-i' in what I thought was the correct way. my understanding is -i means "load this file and give me a repl", and what happens is it loads the file and then exits
You need -r
to get a REPL if you specify any of the main opts I think.
(and it's still true on the latest version of the clj script I'm fairly sure)
Confirmed:
(! 693)-> clj -Sdescribe
{:version "1.10.1.466"
...
Tue Jul 23 16:08:19
(sean)-(jobs:0)-(~/clojure)
(! 694)-> clj -i script.clj
Hello from script
Tue Jul 23 16:08:29
(sean)-(jobs:0)-(~/clojure)
(! 695)-> clj -i script.clj -r
Clojure 1.10.1
Hello from script
user=>
ah, I wonder if that has always been the case and I just forgot about it because of how little I use -i
That has been the same forever