54 replies to this topic

[speciesUnknown]

Cornstalks, on 07 February 2012 - 11:33 AM, said:

AltarofScience, on 07 February 2012 - 11:30 AM, said:

I guess I am just a bizarre abberation. I write all my code with every line on the left edge

thing=0;
otherthing=10;
while(thing<otherthing) {
stuff;
stuff;
thing++;
}
more;
code;
all;
on;
the;
left;

spacing things out generally confuses the hell out of me too.

Yup, that's pretty weird. But there's no mistakes in teh codes....

I don't think its that wierd; I've seen code like this in C by a guy who had previously only worked in assembly language for some kind of PIC chips. He was writing very low level code, pretty much treating cc as a code generator. If I encountered code like this for high level operations though, i would very tempted to do a ^a ^k ^f...

[Sik_the_hedgehog]

Test loop for testing the framerate system. Try to find out why the framerate system was always returning 2 frames instead of 1:

for (unsigned i = 0; i <= 300; i++)
   i += update_program();

(update_program returns the number of ellapsed frames)

[Sirisian]

No examples, but that above example reminds me of so many times when using a nested loop that I used the wrong iterator value as an index. Especially with the whole i and j iterator idiom. I remember one time I read over the error for a whole day and it was like my brain had changed a j to an i and I just kept overlooking it. Wasn't going out of order. The numbers in the array were so arbitrary that debugging it didn't make me suspicious at all.

[Eelco]

Sirisian, on 08 February 2012 - 01:10 AM, said:

No examples, but that above example reminds me of so many times when using a nested loop that I used the wrong iterator value as an index. Especially with the whole i and j iterator idiom. I remember one time I read over the error for a whole day and it was like my brain had changed a j to an i and I just kept overlooking it. Wasn't going out of order. The numbers in the array were so arbitrary that debugging it didn't make me suspicious at all.

I think ive grown a seperate brain lobe just to reason about these type of situations. The kind of bugs used to be the bane of my existence as a young coder, but now I dont have to think about it anymore.

[Phobon]

Just happened to me recently:

for(int i; i<N; i++) someArray[i] = i;

Especially fun as it worked fine for days on my debugging machine but failed in obscure ways when run on the production cluster. I'm still wondering why gcc didn't throw a warning here...

[davepermen]

Sirisian, on 08 February 2012 - 01:10 AM, said:

No examples, but that above example reminds me of so many times when using a nested loop that I used the wrong iterator value as an index. Especially with the whole i and j iterator idiom. I remember one time I read over the error for a whole day and it was like my brain had changed a j to an i and I just kept overlooking it. Wasn't going out of order. The numbers in the array were so arbitrary that debugging it didn't make me suspicious at all.

As I often got confused by this i started to name my iterators with real variable names when ever there's more than one loop at a time. This helps me to not just parse-over all i and j and k or x and y and z and think they're the same.

single loops, i typically don't see them anymore (c#, linq), but when i use one, then i use the variable i. So it's clear to me, it's an ordinary, single loop. if it isn't, i change the name.

[jwezorek]

davepermen, on 09 February 2012 - 04:53 AM, said:

single loops, i typically don't see them anymore (c#, linq), but when i use one, then i use the variable i.

This is actually one of the nice side benefits of using std algorithms + lambdas in C++11: you find yourself declaring variables named i much less often.

[froop]

this one got me a few times:

point p = (10, 20);

instead of

point p(10, 20);

or

point p = point(10, 20);

[J-dog]

No code example, but after doing a lot of C++, I switched back to Java for a small project.

I took copy constructors and sending objects as parameters for granted... it became really messy and difficult to track, and I kicked myself when I realised that I was sending object references all over the places, not copies.

I guess I had assumed that java would somehow have taken care of that?!

[Instigator]

//Boolean logic Fail
if (DateTime.Now.DayOfWeek != DayOfWeek.Saturday || DateTime.Now.DayOfWeek != DayOfWeek.Sunday)
			{
				weekday = true
			}
			else
			{
				weekday = false;
			}

[Alpha_ProgDes]

Instigator, on 11 February 2012 - 01:52 PM, said:

//Boolean logic Fail
if (DateTime.Now.DayOfWeek != DayOfWeek.Saturday || DateTime.Now.DayOfWeek != DayOfWeek.Sunday)
			{
				weekday = true
			}
			else
			{
				weekday = false;
			}

I did something very similar with SQL:

select dept, job
from organizations
where (dept = 'Marketing' and job <> 'Sam Wise')
or    (dept = 'Marketing' and job <> 'The Precious')

Note: Shout out to the GD.Net crew who got the keywords to automagically highlight properly without having to declare what language you're using. Impressive!

[Cornstalks]

Alpha_ProgDes, on 11 February 2012 - 03:02 PM, said:

Note: Shout out to the GD.Net crew who got the keywords to automagically highlight properly without having to declare what language you're using. Impressive!

Side note: they use this guy.

[Sik_the_hedgehog]

Instigator, on 11 February 2012 - 01:52 PM, said:

//Boolean logic Fail
if (DateTime.Now.DayOfWeek != DayOfWeek.Saturday || DateTime.Now.DayOfWeek != DayOfWeek.Sunday)
			{
				weekday = true
			}
			else
			{
				weekday = false;
			}

Nonononono, you don't get it, that isn't boolean logic fail, that's company manager boolean logic (especially true in crunch mode!).

[shurcool]

Just today I wrote:

auto TemporaryHeart = MoveHeart(State[8], State[10]);
MoveHeart(TemporaryHeart, State[12]);
MoveHeart(TemporaryHeart, State[3]);
MoveHeart(TemporaryHeart, State[8]);
TemporaryHeart = MoveHeart(State[9], State[11]);
MoveHeart(TemporaryHeart, State[13]);
MoveHeart(TemporaryHeart, State[2]);
MoveHeart(TemporaryHeart, State[9]);

Instead of:

auto TemporaryHeart = MoveHeart(State[8], State[10]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[12]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[3]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[8]);
TemporaryHeart = MoveHeart(State[9], State[11]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[13]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[2]);
TemporaryHeart = MoveHeart(TemporaryHeart, State[9]);

[NickGomes]

I had one mistake that was just so dumb that I overlooked it. The functions worked independently but when i tried using them together...

void function1(Blah whatever)
{

	 //Tons of code
	 for(int i=0; i;whatever.count; i ++)
	 {
	   //tons of code
	   function2(whatever);
	   //tons of code
	 }
	 //more code
}



void function2(Blah& whatever)
{
	 //lots of code that does stuff
	 Blah whatever2 = new Blah();
	 whatever.count ++;
	 //more codes
}

I seem to always mess up iterators in functions.

[Mussi]

Made this mistake a few times, very hard to spot in some situations.

for(int i=0; i < i_max; i++)
{
    for(int j=0; j < j_max; i++)
        array[i][j];
}

[tstrimple]

Sounds like for the most part most people here would benefit tremendously from better tools. I haven't been hung up on anything like this for a couple years now. Check out the video showing how most of these types of problems can be avoided using good tools. It's not going to help most of the logical problems, but it does a fantastic job of helping you avoid silly errors like where a semi-colon is placed.

http://screencast.com/t/U9s4FW1HM9su

[way2lazy2care]

Today I forgot to compile something after making changes that should have totally changed the behavior of something and it took me a few minutes to realize why the behavior wasn't changing at all.

/monday

[rip-off]

way2lazy2care, on 20 February 2012 - 12:33 PM, said:

Today I forgot to compile something after making changes that should have totally changed the behavior of something and it took me a few minutes to realize why the behavior wasn't changing at all.

/monday

This happens to me sometimes when I'm working with our branch and trunk at around the same time. It can drive you crazy!

[Sik_the_hedgehog]

way2lazy2care, on 20 February 2012 - 12:33 PM, said:

Today I forgot to compile something after making changes that should have totally changed the behavior of something and it took me a few minutes to realize why the behavior wasn't changing at all.

/monday

Reminds me of when I made a quick fork of a game to add time attack for a special event. No matter how hard I tried my attempts at resetting the timer would not work no matter what, it was like it was completely ignoring the code I was writing. Turns out I was editing the original file, not the one from the fork *facepalm*