" : " and " ? " in java

irishguy

private static int max( int lhs, int rhs )
{
return lhs > rhs ? lhs : rhs;
}


can some expalin what the " ? " and " : " mean in the above code please

timeout

its an if else situation.

if (lhs > rhs){
            return lhs;
          } else{
            return rhs;
          }

its just hander for simple comparison expressions without the need of all the extra text and brackets associated with the if else. So the ? represents what to do if the expression is true and : if its false.

Timeout

irishguy

timeout wrote:

its an if else situation.

if (lhs > rhs){
            return lhs;
          } else{
            return rhs;
          }

its just hander for simple comparison expressions without the need of all the extra text and brackets associated with the if else. So the ? represents what to do if the expression is true and : if its false.

Timeout

Cool thanks, never seen it before

Tobias Greeshman

It's called the Ternary operator, it originates from C.

http://www.devdaily.com/java/edu/pj/pj010018/

Skud

In performance tests on a Sun Solaris workstation, the ternary operator was four times faster than the Math.min() method when comparing float variables, and twice as fast when comparing integer variables. In similar cases in your code, I recommend testing to determine if performance improvements can be achieved. As the old saying goes, your performance may vary.

... You learn something new everyday. That obviously doesnt correspond to unix or windows systems exactly... So it's better to use this?

is_that_so

TBH , don't use it with anything more complex than your example. It leads to very terse code that is extremely hard to maintain and you may not even remember what you were doing in six months time. :rolleyes: If you do have to use it do so sparingly. It's a tradeoff between efficiency and maintainability.
Nothing wrong with if .....else

BrokenArrows

cool never saw that before.

Hobbes

is_that_so wrote:

TBH , don't use it with anything more complex than your example.

I agree, although I did get something like this in the exam.

answer = lhs++ > rhs-- ? lhs++ : rhs++;

You had to list the values, but in certain cases the numbers won't increment/decrement.

Mutant_Fruit

answer = lhs++ > rhs-- ? lhs++ : rhs++;

Converts to...

if(lhs++ > rhs--)
{
    lhs++;
}
else
{
    rhs++;
}

Doesn't it? Also, i assume that if it evaluates true... lhs will end up being 2 bigger, and rhs will be one smaller.

If it evaluates false, lhs ends up one bigger, and rhs ends up with its initial value? At least their all post increments and not a mixture of pre- and post-increments, otherwise it'd be even more confusing :P

satchmo

Yeah that's right, and answer would be assigned inside there, ending up with the relevant value before the increment.

Interesting question, what exam did you get that in Hobbes?

Nalz

irishguy wrote:

Cool thanks, never seen it before

I honestly think C should be taught (in programming courses) before java. it would make life a lot better for future porgammers...but that argument is for another day i suppose

is_that_so

Trilla wrote:

I honestly think C should be taught (in programming courses) before java. it would make life a lot better for future porgammers...but that argument is for another day i suppose

Or put you off.:p

Hobbes

satchmo wrote:

Interesting question, what exam did you get that in Hobbes?

SCJP1.4

a lot of the questions are trick questions.

timeout

you could get smacked down my sun for saying that having signed their disclaimer in all Doubt they'll follow through since its 1.5 now

rsynnott

is_that_so wrote:

Or put you off.:p

Anyone put off by C shouldn't be a programmer anyway.

Hobbes

timeout wrote:

you could get smacked down my sun for saying that having signed their disclaimer in all Doubt they'll follow through since its 1.5 now

hardly. it is not an exact copy of the question, because tbh I can't remember the exact question but you do get ones along that lines.

You would learn more from jchq.net tbh.

Nalz

rsynnott wrote:

Anyone put off by C shouldn't be a programmer anyway.

very very true

Mutant_Fruit

You're not a true programmer til you've malloced and realloced!

Still... i got through 18 months of C programming in Electronic Engineering without having to use malloc. I used it for the first time last week because i felt i wasn't a real programmer yet. Now i am

rsynnott

Mutant_Fruit wrote:

You're not a true programmer til you've malloced and realloced!

Still... i got through 18 months of C programming in Electronic Engineering without having to use malloc. I used it for the first time last week because i felt i wasn't a real programmer yet. Now i am

Next step: LISP!

PhantomBeaker

Lisp or scheme (currently hacking around in scheme in uni)

Actually with C, you can do some incredibly fun hacks, because of the way it evaluates true/false... i.e. false is 0, anything else is true.

So to test for odd/even numbers
(i % 2)? something : something else.

One hack I was particularly proud of (because there's one lecturer we have and he seems to promote the use of hackish/obscure code while learning C++) involved something like

cout << ((++on % 2) ? "Light is on" : "Light is off");

So hacky! So painfully, painfully hacky! However, being a masochist, I found it strangely fun!

aoife

rsynnott

Tsk

http://www.ioccc.org/

PhantomBeaker

Yeah, I'm not that much of a masochist... yet.

I just find that I've written it to be just enough of a pain to rewrite if I ever need to.

But yeah, I just like the fun that can be had with c's interesting syntax... and ternary operators are definately one of 'em.