[Feature Idea] Default to wildcard if too many merged imports

If I have a number of qualified imports sharing a package name, such as below:

import cats.data.AndThen
import cats.data.Binested
import cats.data.BinestedBitraverse 
import cats.data.Kleisli
import cats.data.Reader
import cats.data.OptionT
import cats.data.State
import cats.data.StateT

The optimizer under the Merge configuration (rightly) groups them all together:

import cats.data.{AndThen, Binested, BinestedBitraverse, Kleisli, OptionT, Reader, State, StateT}

However at some point the number of things imported from the same package can be so large that the merge grouping will go over the column limit. Of course, Scalafmt can catch this and turn it into this:

import cats.data.{
  AndThen,
  Binested,
  BinestedBitraverse,
  Kleisli,
  Reader,
  OptionT,
  State,
  StateT
}

But why not have an option to simplify to a wildcard like below? It’s much cleaner, albeit less precise w.r.t scope.

import cats.data._

So the feature idea I’m proposing is to add a configuration to the groupedImports rule such as SmartMerge which will default to a wildcard if the merged group goes over the standard column limit (100 chars?).

[liancheng]

BTW, the reason why the groupedImports option defaults to Explode is that having each import selector on their own line as a fully-qualified import is friendlier to version control and text searching. I personally prefer not grouping them together. It also avoids exceeding the column limit.

[nasadorian]

Awesome, thanks for the update @liancheng! Gonna take it for a spin in one of my repos.