On 7/19/17 11:47 AM, Jonathan Marler wrote:
On Wednesday, 19 July 2017 at 15:39:02 UTC, Steven Schveighoffer wrote:
On 7/19/17 9:30 AM, sontung wrote:
So I was thinking of a way of extending if statements that have declarations. The following being as example of the current use of if statements with declarations:

     if(int* weDontPollute = someFunc())
     {
          // use weDontPollute
     }

That's great and all, but it only works by checking if the variable evaluates to true or false. Which is fine for a pointer but otherwise useless for anything else, like integers where zero is usually valid input (index for array). So currently the only way to do something like this in the language, that i've found, is to use a for statement.

     for(int i = someFunc(); i >= 0;)
     {
         // use i

         break;
     }

Not that ideal to use a for statement. It makes it hard to read and if the break condition isn't there it might very well be an infinite loop. So I was thinking of some sort of syntax like this:

     if(int i = someFunc(); i >= 0)
     {
         // use i
     }
Thoughts on this sort of feature?


I really like the idea. Only thing I don't like is the possibility for abuse/confusion/errors:

if(int i = someFunc(); j >= 0) // typo, or weird relationship, or just intentional obfuscation?

It reminds me a bit of why we got rid of the comma operator.

This is why I've liked suggestions in the past like:

if((int i = foo()) >= 0)

That is, you want to use 'if' on an expression while saving the expression, but the if is only looking at a property of that expression.

Note this makes if(arr) (the correct meaning, that is ;) much more palatable:

if((auto x = getArray()).length)

Don't get me wrong, if this syntax is what gets this idea in, I'm fine with it.

One possibility is to require usage of the declared variable in the condition.


I respectfully disagree with this. I recall many times where I want to declare variables that should be limited to the scope of the conditional block but aren't used in the condition itself, i.e

{
     auto a = ...;
     auto b = ...;
     while(something)
     {
         // use a and b
     }
}

This is different. The variable exist outside the scope of the loop. This is more like a for-loop. Arguably, for-loops are better suited for this, and already support it:

for(auto a = ..., b = ...; something;)

An if statement runs once. There isn't an "exists for all the loops" for an if statement. So it's clean and neat to declare them in one place or the other. Only if you need to use it for the condition does the declaration have to be in the condition expression.

But arguably, for loops can be (and have been) abused, so this isn't exactly new territory. Like I said, I'm OK with the proposal if that's what gets it done.

-Steve

Reply via email to