such union types is also something I really liked when learning elm (https://guide.elm-lang.org/types/custom_types.html). in java, we would model this with an interface and both ResolvableLink and UnresolvableLink implementing it?
Yes, in Java one would use an interface as you suggest. (I also had a comment via email saying the same as you.) The problem then is that both the Java interface solution and the Typescript union type solution lose information -- the type of the item added to the list is lost at runtime and can't easily be recovered by code when it retrieves those items again.
Do you think you could elaborate, e.g. with example(s), on what you mean by 'only run-time checks will tell me that something has broken' and 'I want to be able to see their dependencies locally, without having to execute the code.'
I am asking so that I can better understand what the problem is and also whether the reason why 'run-time' is being mentioned has anything to do with weak/strong typing and/or static/dynamic typing.
In a way @Philip those phrases are my "north star" for this newsletter. I'm writing in order to understand what they mean and to attempt to discover whether they are achievable.
such union types is also something I really liked when learning elm (https://guide.elm-lang.org/types/custom_types.html). in java, we would model this with an interface and both ResolvableLink and UnresolvableLink implementing it?
Yes, in Java one would use an interface as you suggest. (I also had a comment via email saying the same as you.) The problem then is that both the Java interface solution and the Typescript union type solution lose information -- the type of the item added to the list is lost at runtime and can't easily be recovered by code when it retrieves those items again.
I had a go at some Java
```java
import java.util.List;
sealed interface Link { }
record ResolvableLink(String title, String url) implements Link { }
record UnresolvableLink(String title) implements Link { }
class Main {
public static void main(String[] args) {
List<Link> listOfLinks = List.of(
new ResolvableLink("Habitable Code", "https://habitablecode.substack.com/"),
new UnresolvableLink("Free Money")
);
System.out.println("Links=" + listOfLinks);
}
}
```
Here is a Scala union type example:
type Link = UnresolvableLink | ResolvableLink
case class ResolvableLink(title: String, url: String)
case class UnresolvableLink(title: String)
val listOfLinks = List[Link](
ResolvableLink(title = "Habitable Code", url = "https://habitablecode.substack.com/"),
UnresolvableLink(title = "Free Money")
)
println(s"Links=$listOfLinks")
==>
Links=List(ResolvableLink(Habitable Code,https://habitablecode.substack.com/), UnresolvableLink(Free Money))
FWIW, here is a Scala example with a Link ADT (Algebraic Data Type)
type ListOfLinks = List[Link]
enum Link:
case ResolvableLink(title: String, url: String)
case UnresolvableLink(title: String)
import Link._
val listOfLinks = List(
ResolvableLink("Habitable Code", "https://habitablecode.substack.com/"),
UnresolvableLink("Free Money")
)
println(s"Links:$listOfLinks")
FWIW, here is a Scala example using Either:
type Link = Either[UnresolvableLink,ResolvableLink]
case class ResolvableLink(title: String, url: String)
case class UnresolvableLink(title: String)
val listOfLinks = List[Link](
Right(ResolvableLink(title = "Habitable Code", url = "https://habitablecode.substack.com/")),
Left(UnresolvableLink(title = "Free Money"))
)
println(s"Links=$listOfLinks")
Hello Kevin.
Do you think you could elaborate, e.g. with example(s), on what you mean by 'only run-time checks will tell me that something has broken' and 'I want to be able to see their dependencies locally, without having to execute the code.'
I am asking so that I can better understand what the problem is and also whether the reason why 'run-time' is being mentioned has anything to do with weak/strong typing and/or static/dynamic typing.
In a way @Philip those phrases are my "north star" for this newsletter. I'm writing in order to understand what they mean and to attempt to discover whether they are achievable.
I did expend a little more in a later post https://habitablecode.substack.com/p/runtime-and-compile-time, but you're right -- there's a lot more to be said and a lot more exploration needed.
All right, I'll be working my way through the posts and will get to that one.
Thanks.