The current choices mirror what's in the official Scala Toolkit (sttp, upickle, oslib). I also added Decline :)

But ... it's all bikesheddable :)

• scala/toolkit#14

Anything we wanna do prior to a first release? :)

Here's the one posed by Virtus Lab:

Imagine you have a common business problem to solve. You need to write a program that fetches text from a web service and then creates a file in the local filesystem with content and path based on the downloaded information.

Scala Toolkit is getting a bunch of tutorials / examples, we can cheat by copying those and then slapping on a parTraverse 😝

• scalacenter/docs.scala-lang#8

A common complaint is about the enormous amount of imports it takes to do stuff. e.g. http4s/http4s#6696

What if ... we just exported a bunch of useful stuff under a single package?

package typelevel.toolkit

export cats.*
export cats.effect.*
export fs2.*
export org.http4s.*
export org.http4s.ember.client.EmberClientBuilder
// etc.

Not gonna lie, it's playing with fire a bit. Probably instead of wildcards we should be thoughtful about the specific things we export.

But then:

import typelevel.toolkit.*

and you are off to the races!

Scala Toolkit supports 2.12. All of our libraries do so far, so it's not any extra effort to do it here.

But, why. Here you go @SethTisue 😜

It will be helpful to have tests in CI that check that the toolkit is able to compile a few test scripts.

I'm not sure we can achieve something like this, but we can try. WDYT @armanbilge?

I ran the example code from the README and it doesn't print anything:

➜  toolkit git:(main) ✗ scala-cli toolkit.sc
Compiling project (Scala 3.2.2, JVM)
Compiled project (Scala 3.2.2, JVM)
➜  toolkit git:(main) ✗ 

I'm not sure if this is because of scala-cli or Cats Effect.

This will be a huge win if we can land it.

• typelevel/cats-effect#3149

I've tried using this toolkit for my Scala Native project, but it fails with

scala.MatchError: 16 (of class java.lang.Integer)

for both scala-cli and sbt on all platforms:

• https://github.com/andrzejressel/typelevel-toolkit-native/actions/runs/5272627423
• https://github.com/andrzejressel/typelevel-toolkit-native/actions/runs/5272627424

• scala/toolkit#37

Languages like python provide really fast and simple ways to perform most operations, which lends itself very well to scripting.
I think that Scala fails at this, while it's better than Java it's still not great. I don't think this is a problem for most situations.

I do however think it would be very nice if toolkit could provide default implementations of some common tasks in a referential transparent and resource-safe way. An example would be reading and writing files.

I'm pretty new to using CE so I ended up writing this, which I'm not even sure is the correct way to do this.

def fileWriter(path: String): Resource[IO, FileWriter] = 
    Resource.make(IO.blocking(new FileWriter(new File(name))))(file => IO.blocking(file.close()))
def openFile(path: String) : Resource[IO, Source] = 
    Resource.make(IO.blocking(Source.fromFile(path)))(file => IO.blocking(file.close()))

def writeToFile(path: String, content: String) =
    fileWriter(path).use(writer => IO.blocking(writer.write(content)))

if instead similar operations were already implemented for me I think that would be really great. It would at least make it a lot easier to convince other people to use this :)

All of this might obviously be outside of the scope of toolkit as it's a collection of libraries, but it would certainly make it easier to pick up for people who just want something to write small tools with.