[dcl.init.string] p1 Successive characters initializing array elements make nosense after P2314

[xmh0511]

Successive characters of the value of the string-literal initialize the elements of the array.

Since P2314 introduces a clear concept about "code unit", which is an integer value of a character type that is the type of the element of the character arrays. So, this outdated wording should be improved

Successive code unit values of the string-literal initialize the elements of the array.

[jensmaurer]

The problem is that [lex.string] only creates a sequence of code unit values if we initialize a string literal object, but that doesn't appear in [dcl.init.string].

See the pull request.

[xmh0511]

@jensmaurer I think the following modification may be clearer.

Each element of the array is copy-initialized from the corresponding code unit value in the sequence of code unit values represented by the string-literal.

This can stress that every element of the array is a container that saves the code unit value. [lex.charset] p5 implies this point but it is not exposed in the initialization of the character array.

A code unit is an integer value of character type ([basic.fundamental]).

The character type is exactly the type of the element of the array.