I'd appreciate a pointer to the Standard paragraph stating that the following code is well-formed:
int main() {
int var = 0;
return [] { return sizeof(var); }(); // var is not captured!
}
Similar example appears e.g. in § 8.4.5.2, but I could not find any verbal description of it.
It's specified in terms of when an entity must be captured, rather than in terms of when it may not.
[expr.prim.lambda.capture] (with some omissions)
8 ... If a lambda-expression or an instantiation of the function call operator template of a generic lambda odr-uses this or a variable with automatic storage duration from its reaching scope, that entity shall be captured by the lambda-expression. ...
[ Example:
void f1(int i) { int const N = 20; auto m1 = [=]{ int const M = 30; auto m2 = [i]{ int x[N][M]; // OK: N and M are not odr-used // ... }; }; // ... }— end example ]
The key in your code sample is that var is not odr-used, because it is an unevaluated operand. As such, it does not need to be captured.