semantics of assigning to a?.x

[sigmundch]

What should we expect with the ?. operator when used with assignments:

exp1?.x = exp2

My current expectation is that this should be equivalent to:

((a, b) => a == null ? null : a.x = b) (exp1, exp2);

(both expressions are evaluated once, exp2 is evaluated even when exp1 is null)

[nex3]

Since set x is just a method, it should presumably do the same thing as exp1?.x(exp2) with regards to whether exp2 is evaluated or not. I'm not sure what that same thing is, though.

[gbracha]

Current spec says:

Evaluation of an assignment a of the form e1?.v = e2 is equivalent to the evaluation of the expression ((x) => x == null?null : x.v = e2)(e1). The static type of a is the static type of e2. Let T be the static type of e1 and let y be a fresh variable of type T. Exactly the same static warnings that would be
caused by y.v = e2 are also generated in the case of e1?.v = e2.

In other words, e2 is not evaluated if e1 is null. This is consistent with our treatment of method arguments in general. Basically, if the target is null, nothing happens. Very similar to writing

(a = e1) != null ? a.x = e2 : null

which is presumably what you would have had to write before this worthy construct was introduced.

[nex3]

👍

[sigmundch]

great, thanks for the clarification!