Can Option and Result be used together?

[keiravillekode]

Example A
Suppose we have Lisp/Haskell style lists, and are writing a function to return the tail of the list.

• If the list has one element, we would like to indicate the result is empty, for example by returning none.
• If the list has more than one element, we have a non-empty tail, and we need to return a reference.
• If the list has no elements, we would like to report an error.

Example B
We have a function that constructs binary trees

• An empty tree is a valid result
• A non-empty tree is also a valid result
• If the input is invalid, we want to report an error.

What is the idiomatic way to address such scenarios? Are there examples in the standard library I should look at?

[JalonSolov]

No, they cannot.

There used to only be one type, that incorporated both, but too many people complained that it was too "hard" to distinguish. They actually needed to put an if inside the or block!

So now, they are 2 separate types which cannot be used together.

The simplest way to get the effects of both is to use a Result type, and send back a specific error for empty, etc. Then check for that specific error in the or block of the caller and handle it separately from others.

[Wajinn]

V's documentation gives 4 examples of usage. Perhaps that should be the reference point to start from, concerning possible issues.