ternary conditional operator with statement support

[MisterKidX]

in C#, the ternary conditional operator, AKA the ?: operator, is used to assign values based on a conditional. You can symbolize its intent as follows:

variable = conditional ? yes expression : no expression

What I would love for this syntax to offer, is statement support, in which I don't have to declare a variable and assign it the value of the ternary expression. Goes something like this:

conditional ? Foo() : Bar();

Where Foo() and Bar() can be void methods.

void Foo() { }
void Bar() { }

// this is allowed
bool conditional = true;
int x = conditional ? 1 : 0;

// this is not allowed, what I'm suggesting
conditional ? Foo() : Bar();

// the latter would make this obsolete
// braces added for extra drama
if (conditional)
{
    Foo();
}
else
{
    Bar();
}

[spydacarnage]

You're comparing the ternary to the longest form of the if statement.

You could just use

if (condition) Foo(); else Bar();

and then that isn't much longer, and it's a lot easier to read...

[DavidArno]

I 100% disagree with your assertion that if (condition) Foo(); else Bar(); is easier to read than condition ? Foo() : Bar();

[DavidArno]

There was talk around four years ago of allowing switch statements to use the switch expression => style syntax (#3038), allowing statement blocks in switch expressions (#3037) and just allowing a switch expression to be a statement expression (#2632 and #2860), but these all seemed to have died a death as - as far as I can tell - there appears little interest amongst the language team in implementing any of them. They would typically allowed things like:

conditional switch { 
    true => Foo(), 
    false => Bar() 
};

Since there seems little chance of even this style of "void in expressions" happening, I think the chances of the team embracing statements in the ternary condition operator are pretty much non-existent.

[jnm2]

Duplicate of #6013.

[MisterKidX]

Thanks, haven't found it when I searched

[DavidArno]

Interesting how that original discussion received lots of upvotes, whereas this time around it's just downvotes.

[jnm2]

@MisterKidX No worries. Until GitHub throws some AI at this problem, this happens all the time.

[CyrusNajmabadi]

The numbers are so small, it doesn't seem statistically significant.

Since there seems little chance of even this style of "void in expressions" happening

Not sure where that idea is coming from. There's a wg looking at exactly this issue.

[DavidArno]

Not sure where that idea is coming from. There's a wg looking at exactly this issue.

Excellent news. I look forward to reading the results from that working group.