Time for an adventure then I guess haha

yeah too many factors to make a generic setup

I did some benchmarks recently using the js-framework-benchmark https://github.com/thheller/js-framework-shadow-grove

but thats a rather specific benchmark

some older benchmarks I used the npm benchmark package https://github.com/thheller/shadow-experiments/blob/master/src/test/shadow/experiments/grove/bench_db.cljs

works ok but not great

Yea there is a cascade of dependencies and system requirement 😅

Do you know if this is the right way to go for testing libraries @U05224H0W? Or am I going down the wrong path here? https://clojurescript.org/community/running-tests

I don't understand the question

testing what?

I’m trying to figure out the best way to benchmark Datahike across different JS VM’s. I also guess this would not work with IndexedDB and only work for in-memory so I would need to actually have the headless browsers. So perhaps I am going down the wrong path.

what are different JS VMs for you? you mean just different browsers or actually different lower level JS VMs without being hooked up to a browser, so jsc directly instead of safari?

I mean if you intend this to run in the browser then I'd be testing the browsers?

We do intend to offer node support etc. Despite the value prop in our opinion not being that high compared to using it on JVM. So I was investigating the most general testing and then going to work to more specific. From my basic tests at the REPL the in-memory is pretty similar to IndexedDB so I was thinking that this would give us the best view into our performance. We could then more accurately identify where the performance problems are.

Arguably some people might want to do some IOT stuff using Datahike from JS.

Or even Cljs