Hacker News Reader

Comment by vletal | Hacker News Reader

Exciting! Especially I like the named attribute matching [1] and dict matching [2]. Scala does not have that. If you have a case (data) class with 13 parameters, you need to match on all of them, instead of doing `MyClass(param=value)`. Similarly you can not easily pattern match on `Map` (`dict`).

One thing which I am missing though is matching on f-strings. Such a powerful feature in Scala.

[1] https://www.python.org/dev/peps/pep-0636/#adding-a-ui-matchi...

[2] https://www.python.org/dev/peps/pep-0636/#going-to-the-cloud...

View on HN

I don’t understand what you are saying? In scala you do not have to match on all parameters. You can use the underscore operator to skip the other attributes.

Also you can match a map since every map is just a sequence of Tuples in scala. However why would you ever do this? It is really impractical for large sequences and if I have only a very limited number of attributes I would not use a map I guess.

> You can use the underscore operator to skip the other attributes

And have to add an underscore in the right position when you ever add an attribute? With partial matching like described, you don't have to change the pattern when adding fields to the data structure. I guess the GP was referring to this.

Well that is a little bit different I guess. You probably could do some unapply shenanigans to achieve the same result but that off the top of my head I would not know how to do that.

Well sure, wildcards still count as matching. The ability to use named params instead could clean some codebases a bit.

Also I get that there are reasons for the Map matching not being present in Scala. All I am saying is that it's cool that it's in python.

You can use guards to achieve a lesser version of this in Scala, e.g.:

  case x:YourCaseClass if(x.oneField=="value") => ...

Reply on news.ycombinator.com