Goto's?

[wildfrog]

The other code constructs (like break and switch and continue) do not suffer from the "comes from" problem. You know exactly where the code is coming from, and its exact state when it arrives.

Well, actually when you got multiple continues/break within a loop you may lose track.

The major problem with 'goto' is IMO what TheCPUWizard mentioned. That you can easily jump over variable declaration/definition/initialization without ever being notified (not even a warning) and that may lead to strange results.

I see the "comes from" problem all the time, and to me it just another issue that I have to deal with when debugging.

int i = // some value
while (i > 0)
{
    // do something...
}

// at this point i =< 0. How can I know if the above loop was ever entered? 
// And if it was, how many iterations were there?

- petter

[S_M_A]

I sure don't regret risking that war! This was interesting reading.

I have to admit that I have never given much thought of how and why to get rid of break & continue in loops. Maybe I have seen them for so long that I'm just take their neccessity (and how is that spelled...) for being a law of nature but now I see your points.

[SuperKoko]

That you can easily jump over variable declaration/definition/initialization without ever being notified (not even a warning) and that may lead to strange results.

Your compiler is buggy:

#include <iostream>

int main() {
  goto next;
  int x=42; /* initializer or constructor */
  next: std::cout << x;
}

Is ill-formed code.
A diagnostic message is required.
Most compilers generate a fatal diagnostic message (i.e. an error), which is a good thing.

[TheCPUWizard]

Guess no one is going to tell me what the "Ultimate Question" was??

[reference to Hitchhikers Guide to the Galaxy....]

The "Ultimate Anwer" we know to be 42, so my code posted above, should give the question:

"f(42)" = ????

[MikeAThon]

Guess no one is going to tell me what the "Ultimate Question" was??

[reference to Hitchhikers Guide to the Galaxy....]

The "Ultimate Anwer" we know to be 42, so my code posted above, should give the question:

D@mn!! If it wasn't for all those gotos, we would have been able to figure it out!!

Mike

[exterminator]

See this - http://groups.google.co.in/group/com...c70a74a925f12/

[ChaosTheEternal]

Guess no one is going to tell me what the "Ultimate Question" was??

If it weren't for the division by 0 and the missing label, then maybe you'd be able to figure it out.

Of course, that's assuming there's nothing fundamentally wrong with the Universe.

[ahoodin]

42:goto question

[TheCPUWizard]

If it weren't for the division by 0 and the missing label, then maybe you'd be able to figure it out.

Of course, that's assuming there's nothing fundamentally wrong with the Universe.

OK, i fixed the missing label (d)

But I dont see a divide by 0...

[ChaosTheEternal]

But I dont see a divide by 0...

//...
     goto b;
e:
     if (x % 0)
        goto b;
//...

Dev-C++, using the g++ compiler, gives a warning about it being a Divide by 0.
VS2005 gives the warning "potential mod by 0", but since it is Modulus division, it's still dividing by 0.

[TheCPUWizard]

Dev-C++, using the g++ compiler, gives a warning about it being a Divide by 0.
VS2005 gives the warning "potential mod by 0", but since it is Modulus division, it's still dividing by 0.

Told you "I didnt see it", did not say it didnt exist.

Yes "(x%0)" is mathematically undefined, I actually meant to use the much more useful "(x%1)"... (going back to edit again)....

[ChaosTheEternal]

Told you "I didnt see it", did not say it didnt exist.

I had to type something, and I hate just saying "Here".

[S_M_A]

I sense that this thread is going from "potential goto war" -> "it's friday"...

[TheCPUWizard]

I sense that this thread is going from "potential goto war" -> "it's friday"...

Nope..after midnight, now saturday!!!!

[wildfrog]

Yep, it's Saturday 'morning' (strangely enough I can feel that in my bones, muscles, limbs and head...).

Your compiler is buggy:

You're right. I'm using VC++ 2005.

I thought that TheCPUWizards code contained 'jumps over variable initializations', but when I look at the code now, I connot find it.. maybe it's been edited

- petter