Ruby 3.4, No More TypeError with **Nil as It Is Treated as an Empty Hash

Re:Ruby 3.4, No More TypeError with **Nil as It Is Treated a... by pzmarzly on Monday May 13, 2024 @03:02PM (#40346970)

Interestingly, JavaScript has the opposite problem: {...null} is fine (returns {}), but [...null] is not (TypeError: Spread syntax requires ...iterable not be null or undefined).

[ Parent | on HN ]

Re:Ruby 3.4, No More TypeError with **Nil as It Is Treated a... by fwlr on Tuesday May 14, 2024 @01:19AM (#40351868)

I was quite surprised by this, I was certain it wasn’t true, but I checked and it is!

[ Parent | on HN ]

Re:Ruby 3.4, No More TypeError with **Nil as It Is Treated a... by kevincox on Tuesday May 14, 2024 @09:09AM (#40354852)

This tricks me all of the time. Part of me wants for it to work to avoid allocating an empty array object in situations like `[...opts?.foo.map(format)]`. On the other hand I appreciate the stricter type safety.

[ Parent | on HN ]

Re:Ruby 3.4, No More TypeError with **Nil as It Is Treated a... by psychoslave on Friday May 24, 2024 @03:50AM (#40463935)

Thanks for making me aware of this. I guess {...something ?? {} } and [...something ?? [] ] are then appropriate to avoid any trouble for now.

[ Parent | on HN ]