Singleton monoid instances

[TimWSpence]

Prompted by http4s/http4s#6712, which locally caches a particular Monoid[List[*]] instance - the issue is a generic one so it seems better to apply this optimisation here.

There are other cases where this kind of change could be made too but I wanted to get feedback before doing any more work on it.

Carries on the work done in #4309. Many thanks to @bplommer for doing most of the work!

[satorg]

Perhaps, in the light of Scala3 and future migrations we should avoid using private [this] modifier:
https://docs.scala-lang.org/scala3/reference/dropped-features/this-qualifier.html

I think just private would be private enough :)

[armanbilge]

Just for the record, here's the difference.

//> using scala "2.13.8"

object Foo {
  private val bar: String = "bar"
  private[this] val baz: String = "baz"

  def publicBar = bar
  def publicBaz = baz
}

/*
 * Decompiled with CFR 0.152.
 */
public final class Foo$ {
    public static final Foo$ MODULE$ = new Foo$();
    private static final String bar = "bar";
    private static final String baz = "baz";

    private String bar() {
        return bar;
    }

    public String publicBar() {
        return this.bar();
    }

    public String publicBaz() {
        return baz;
    }

    private Foo$() {
    }
}

[satorg]

Yeah, optimizations and portability are often a trade off :)

[TimWSpence]

So what is the conclusion here sorry? In general we seem to use private[this] absolutely everywhere so personally I don't see the harm in adding one more to the mountain but am happy to go with whatever the concensus is!

[bplommer]

Yeah, when scala 3 actually deprecates/removes it we can replace private[this] with private but there seems no harm using it in the meantime

[satorg]

I'm totally fine with private[this] too, I don't think it is that important for now.

---

I personally prefer to avoid using private [this] on project that are supposed to be migrated to Scala3 eventually. But since the majority of projects are still using Scala2, they may benefit from such optimization.

Btw, I've just checked the @armanbilge's example with Scala 3.2.0 (just out of curiosity :) ) and looks like Scala3 does a great job on optimizing private fields indeed:

/*
 * Decompiled with CFR 0.152.
 * 
 * Could not load the following classes:
 *  scala.runtime.ModuleSerializationProxy
 */
import java.io.Serializable;
import scala.runtime.ModuleSerializationProxy;

public final class Foo$
implements Serializable {
    private static final String bar;
    private static final String baz;
    public static final Foo$ MODULE$;

    private Foo$() {
    }

    static {
        MODULE$ = new Foo$();
        bar = "bar";
        baz = "baz";
    }

    private Object writeReplace() {
        return new ModuleSerializationProxy(Foo$.class);
    }

    public String publicBar() {
        return bar;
    }

    public String publicBaz() {
        return baz;
    }
}

[armanbilge]

I personally prefer to avoid using private [this] on project that are supposed to be migrated to Scala3 eventually.

I am 💯 % confident that find/replace private[this] to private will be the fastest and easiest part of Scala 3 migration in Cats :)

[satorg]

Thank you!