investigate using gradle instead of sbt at some point

[outkaj]

No description provided.

[adridadou]

Let's open a discussion about it here. What's the reason to move to gradle instead of sbt ? What do we want to gain from doing this ?

[outkaj]

We just touched on it briefly in our discussion today. It's more a longer term open question - moving to use mroth's sbt builder image is in our immediate roadmap. We were wondering if it could be helpful since some of our projects use multiple languages and also avoid some of the slowness of sbt. @fforbeck raised Gradle and can maybe add more points here.

[fforbeck]

As I mentioned in the call it was just an idea that could help us to save some building time. There is a nice post about that from LinkedIn engineering team:

• https://engineering.linkedin.com/blog/2018/07/how-we-improved-build-time-by-400-percent

Overall it seems to be faster for building Play java apps of different sizes. The key points they noticed were: Productivity boost, Faster builds, Maintainability, Scalability (they were talking about hundreds of microservices and several daily deployments).

There is a Scala plugin for Gradle, perhaps we could look more into that in the future. It might not be suitable for our needs atm, but just wanted to point that out.

[adridadou]

For play apps, we should use that: https://docs.gradle.org/current/userguide/play_plugin.html

I like the idea of investigating different building tools and see which one suits us the best.

I've heard of bazel too but not sure if it fits our needs either