Clojurians Log v2

Hi plexus and greater Lambda Island team. I love kaocha. We've been using it for almost 2 years now. I just started playing around with kaocha-cucumber and was able to get things to work from the CLI. I was wondering if there is a way to test/run a feature from the REPL like a Cider focused test?

Hi @amarpotghan! There's the kaocha.repl namespace which does exactly that.

I didn't say you could only do small bugs. Everything is an incremental improvement at this point.

I dove back in once again and now:

➜ clojure -A:test -m kaocha.runner
SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
SLF4J: Defaulting to no-operation (NOP) logger implementation
SLF4J: See <http://www.slf4j.org/codes.html#StaticLoggerBinder> for further details.
[((..)(...)(.............)(......)(F.)(F...)(.........)(.....F)(.....)(.......)(..)(...)(..)(....)(..)(.............)(FF..FF)(.........)(...)(FFF..)(..)(..)(...............)(........................)(...........)(....)(..........................................................................................)(.)(........)(.)(..)(....)(....)(FFFFFFFF)(FF....)(....)(.)(......)(..)(..)(.............)(.)(..)(FF)(......)(.......)(...........)(FF.....FF...)(..)(.)()(...)(..................................................)(....)(.......)(.....)()(....................)(...)(..)(..)(..)(..)(F.FFFF)(....)(..)(.)(.....)(.....................................................)(....)(..)(.......)()()(....)(...........)(...........)(.......)(...)(...)(......)(..)(...........)(...............)(.)(.....)(..)(..)(F..........................................................................F)(...)(........)(F)(...)(.)(..)(....)(.....................)(...)(.........)(.)(.......................................................................)(.)(...)(......)(..........)(...)(..)(....)(..)(..............................)(.......)(.................)(....)(.........)(..)(.......)(......)(......)(....)(.)(.)(....)(..)(................................)(....)(..)(..)(..........)(....)(..)(..)(.....F)(..)(.................................)(.FF.....F........................)(..)(......)(................)(..)(....)(..)(...)(...)(...)(.)(...)(......)(F)(.)(..)(..)(....)(F.)(...)(.)(.......)(..)(........)(.)(..)(...............)(............)(..)(..)(.....)(..)(.F.F..)(........)(.)(..)(..)(..)(..........................................)(...)(..)(.....)(.)(............)(..)()(.)(.)(..........)(.)(...)(..)(..)(....)(....))]
Randomized with --seed 355021920

Is there an easy way to use chui-ui with kaocha that isn’t baked into the shadow build loop?

For reference here is the node runner I am using:

(ns kaocha-node-runner
  (:require [lambdaisland.chui.runner :as runner]
            [lambdaisland.chui.test-data :as test-data]
            [lambdaisland.glogi :as log]))
            
(log/set-levels
 '{:glogi/root :debug
   lambdaisland :all
   lambdaisland.chui.interceptor :error})

(test-data/capture-test-data!)

(defn start []
  ;; for dev, enable this to update the UI on hot reload
  ;; (ui/render! (.getElementById js/document "chui-container"))
  (runner/run-tests))

(defn stop [done]
  (runner/terminate! done))

(defn ^:export init []
  (start))

Which is basically a copy and pasted version of the browser runner

I am guessing the error

Test var xxx/yyy is known to the compiler, but was not defined at runtime. Did the file load correctly?

OHHHHH I see what is happening, we have integration tests

and those are defn’s gated behind a when

Oh definitely didn’t read into what you had said that I have to only work on specific issues. I would rather just take something small to get used to the codebase ATM

hah, glad I added a check for that, before this would have given a completely impenetrable failure deep inside chui

> Is there an easy way to use chui-ui with kaocha that isn’t baked into the shadow build loop? can you elaborate? I'm not sure what this means. Chui-ui is not tied to Kaocha, you can include it in any build, shadow or not

Ideal world is the chui-ui is something is run at the command line or included with funnel

Which I am sure is possible at the moment but it's just a little confusing how to tell chui to run, get the tests from a remote client via funnel, and then run those tests.

I guess I expect chui-ui to run like the dashboard for shadow?

If that makes it any clearer

yeah chui as a UI for tests that run elsewhere is definitely something interesting to try out at some point, cause then it could also act for instance as a UI for Clojure tests run by Kaocha. The UI itself is independent enough to do that without changes, but you need something like chui-remote to pump the results into chui. Currently chui-remote only does the opposite, it runs tests on demand and sends the results out from chui to somewhere else (in our case kaocha-cljs2)

lots of use cases to consider and I can't do them all at the same time. My current priority is to make sure kaocha-cljs2 works for vanilla cljs, figwheel, and shadow, on browsers and node, and to have examples and docs for these 6 cases.

I have a limited amount of funding from Nextjournal but that's really only for the shadow-cljs use case, and doesn't cover half of the work put in so far... I think once kaocha-cljs2 is ready for general consumption I'm going to shift gears a bit and focus more on outreach to actually get a bit of a community around this stuff. We're really going to need more contributors to keep this sustainable.

I will definitely see what I can do about getting a some funding from work. No promises but if we do get things integrated then we definitely have a reason to support.

🙏:skin-tone-2: thank you! we're building up a small reserve at https://opencollective.com/lambda-island , which I actually mostly intend to use to help support other contributors.

@plexus did you want me to package up that node runner and provide a PR?