CODEWITHSUNDEEP
c++
jtbandes
jtbandes

Reputation: 118691

What is expr[{ ... }] for?

This page on cppreference says:

Built-in subscript operator
The subscript operator expressions have the form

expr1 [ expr2 ] (1)
expr1 [ { expr, ... } ] (2) (C++11)

…
2) The form with brace-enclosed list inside the square brackets is only used to call an overloaded operator[]

Other than this, there's no example or explanation of what it means to use braces inside operator[].

On godbolt.org I verified that, given S s;, s[{3}] can be used to call a custom S::operator[](int), while if we had S s[5]; then s[{3}] would be invalid. However, it doesn't seem to work for multiple arguments i.e. s[{3, 4}] (Clang's error mentions the argument is treated as an initializer list).

So...what's the point of this feature? When would I want to use [{}] instead of just plain []?

Upvotes: 4

Views: 839

Answers (1)

StoryTeller - Unslander Monica
StoryTeller - Unslander Monica

Reputation: 170123

Its purpose is to allow initialization of any argument type from a brace-init-list. It's added in C++11 to go along with the new initialization syntax, where braces were allowed as initializers for all types (as opposed to just for aggregates).

An example

std::map<std::pair<int, int>, double> m {
 // content
};

// later

m[{1, 2}] = 2; // {1, 2} initializes a std::pair which is then used for sub-scripting.

Since a { ... } is not an expression under the C++ grammar, we can't use {} in the production [ expr2 ]. An explicit grammar rule was needed.

Upvotes: 6

Related Questions