Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Deon]

Thanks for reply Starver! And don't forget that I use C#, so no &/* magic for me, just straight references!

[AlStar]

This is why it's always good policy to lead with the constant when not doing assignment.

Don't check "if(x==1)", check "if(1==x)" instead and it will catch a lot of those mistakes that might otherwise slip by.

At least, that's how I've always done it: If Desired Value is the same as Given Value, then...

For most C++ compilers there are optional warnings you can enable to make it shout angry words at you for doing assignment in a conditional. In 9.998 out of 10 cases it's unintentional so I always enable it and elevate to an error, along with the one for "not all control paths returns a value" and a few others.

Randomly curious - what case would that .002% usage look like?

I'm trying to think of a situation where you'd want to do that, and failing.

[olemars]

It saves you a line of code sometimes, so its intentional use is usually either a case of laziness or inexperence. The lazy wants to type as little as possible, and the inexperienced thinks there is some sort of achievement in writing short code.

I've seen it intentionally used once at work (a case of laziness) since it triggered the warning I had elevated, but then in a while loop like

Code: [Select]

while(x = getNextSomething())
{
     do shit to x in the safe knowledge that it is not NULL
}

You can do that without triggering the warning/error by adding a test, ie

Code: [Select]

while( (x = getNextSomething()) != 0 )although that adds the new pitfall of forgetting one set of parantheses.

[Starver]

Thanks for reply Starver! And don't forget that I use C#, so no &/* magic for me, just straight references!

[edit: Whoops, first meant to say "Well, I hope you understood it, at least..." (with the emphasis on me having written something that was generally understandable!), before going on to waffle with the following diversion...]


I need to get into C#.  I know most other variations on C (from K&R onwards) but, as you can see, I'm currently on a bit of a Perl-binge for prototyping[1], and tend to use a form of it even in what should be Pseudocode, and I really need to force myself back to the bleeding-edge again, and last time I touched it I got distracted by some other shiny-shiny language or other...

And while I can't believe that you can't do something very similar in C#, I'm pretty sure that it's not best practice to overly rely on it.  (Probably akin to littering the code with GOTOs... )  And the type-freedom of Perl does tend to spoil me...

Code: (Example) [Select]

my $variable;
$variable = "Hello World"; print "$variable\n";
$variable = [split(//,$variable)]; print join(",",@$variable)."\n";
my $temp=$variable; $variable =+{}; map {$$variable{$_}++} @$temp;
print join("\t",map {"$_: ".$$variable{$_} } sort {$$variable{$b} <=> $$variable{$a}} keys %$variable);
#And that's not even using coderefs and globrefs!


Hello World
H,e,l,l,o, ,W,o,r,l,d
l: 3    o: 2    e: 1    H: 1    r: 1    W: 1     : 1    d: 1

(But as for readability... I can do much better than that... )



[1] In the more mundane "build it quickly, to test", sense, rather than the usual procedural-definition sense!

[Virex]

(But as for readability... I can do much better than that... )

I'd hope so. To me that almost looks like a piece of assembly code that's been stripped of comments...

[Araph]

if(x = 1) means TRUE, because any value that is not 0 is also TRUE and variable x has from now on the value 1.
if(x == 1) actually tests whether x has the value 1.

Oh, WOW. How did I miss that?! Everywhere else, I didn't mess up on that, and then I somehow managed to screw it up here.

Thank you again for all the help! I'm so excited that I can finally move along with this project!

EDIT: Accidental-psyche! I have a new question: is there a reason that, according to the Firebug console, the variable xOne would go '1, 0.9, 0.8, 0.7000001'? It should only be decreasing by .1 each iteration. Also, looping that function seems to gradually freeze Firefox.

[Normandy]

@Araph:
Floating point numbers are subject to rounding errors, and since they're not stored in base-10, not all simple decimal fractions will necessarily be represented well in floating point.

[Deon]

snip

You can use *var and &var in C#, but this triggers a lot of warnings and you have to enable the "allow unsafe code" check for the Visual Studio to compile it. C# has a lot of inbuilt memory management. For example, you don't have to clear memory, and that's great for me (because all of my more or less long programs in C++ always had memory leaks).

[Araph]

@Araph:
Floating point numbers are subject to rounding errors, and since they're not stored in base-10, not all simple decimal fractions will necessarily be represented well in floating point.

Oh, okay. I switched the decimals out for whole numbers and changed the size of the boxes to compensate, and now that works. Is Firefox freezing just from the looping function? If so, is it possible to fix that?

[olemars]

@Araph:
Floating point numbers are subject to rounding errors, and since they're not stored in base-10, not all simple decimal fractions will necessarily be represented well in floating point.

Oh, okay. I switched the decimals out for whole numbers and changed the size of the boxes to compensate, and now that works. Is Firefox freezing just from the looping function? If so, is it possible to fix that?

I've never done much with javascript, but I suspect that each time setInterval is called you're adding another callback to the function, building up until they bog down execution and freeze up firefox.

[Siquo]

No, that should not be it, since there is no real loop. Could you post the entire thing (html+js) in a spoiler so I can see?

[olemars]

No, that should not be it, since there is no real loop. Could you post the entire thing (html+js) in a spoiler so I can see?

Assuming this is reasonably close to the current code he's using, setInterval will be called a number of times depending on the value of xFour.
setInterval isn't singleshot, so each time that function is called a new recurring timer callback will be added with a 10ms timeout. They'll never stop, and if it is pass by value the number of them will grow exponentially.

setTimeout is singleshot, which I think is what's wanted here.

[Siquo]

setTimeout is singleshot, which I think is what's wanted here.

#$%^& damn did I mess up. So, to fix: use setTimeout not setInterval. Sorry 

[Araph]

setTimeout is singleshot, which I think is what's wanted here.

Hooray! It works now! Thanks for pointing that out!

setTimeout is singleshot, which I think is what's wanted here.

#$%^& damn did I mess up. So, to fix: use setTimeout not setInterval. Sorry 

No need to be sorry, you've been an amazing help!

[Deon]

I wanted to ask, what is Console.SetBufferSize(x, y) for?