mogge... more Fulcro today

mogguh

Mornin'... more Code review today

How’s Fulcro?

challenging

is it worth the complexity?

Not 100% convinced yet to be honest. It does solve some really complex problems and it is very clever at doing that. It also includes everything including the kitchen sink.

So can't adopt it by using little pieces (as far as I can tell). So it has a rather steep learning curve IMHO.

yeah. it's an all or nothing thing isn't it

I've been pondering about babashka, whether I should leave JDBC+Postgres support in (which I included last week), since it might be going to "everything including the kitchen sink" direction. Just interacting with psql and the mysql CLI by shelling out is already quite powerful for small sql things and it doesn't put extra features into bb...

did you get (m)any request for SQL support?

not really, so that's why I may re-consider. most people just shell out to their respective sql CLIs

it was more like: wow, this works with GraalVM... super cool!

I'll release it behind a feature flag in the next release probably, so people who desperately need it can compile babashka for themselves

Fulcro uses panthom, right?

I’ve been wanting to take a look at fulcro. But so far I’ve been really happy with just using Re-frame + some clojure web stack

Fulcro and Pathom work really wel together. You can use the separately if you want to though.

So they tend to get used together a lot

Would you use Panthom without Fulcro?

Probably not.

but as far as I know you can.

I'm currently building a commercial product using Fulcro, Pathom, and Datomic. I inherited a legacy code base that used rum, minimally, and was more of a proof-of-concept than something I'd be comfortable with in production. So we stopped and asked ourselves if we should continue on as-is or pivot. We pivoted:

(tip: when uploading a chunk of text as a file, it becomes collapsible)

Yes, you can use Fulcro without Pathom. If you're using Datomic on the backend it may not matter much if you use it or not since Fulcro has its origins in om.next which means Fulcro's protocol is a lot like Datomic's pull-syntax. Using Pathom with Datomic provides marginal benefit, but if you're using SQL on the backend then Pathom would be super helpful in parsing the client request so it can be easily turned into SQL

Our rationale for switching to Fulcro

This https://gitlab.com/tvaughan/fulcro-template is a Fulcro app I built for myself to learn and evaluate Fulcro.

cool... I'll have a look at that