all 89 comments

sorted by: hot top controversial new old
[–] 86 points 11 months ago (3 children)

The advantage of that last approach is that it has side effects and cannot therefore be optimized out by the compiler.

  • source
  • hideshow 3 child comments
  • [–] 13 points 11 months ago (2 children)

    That’s only one advantage. In theory it does not necessarily terminate, so that’s another one.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 58 points 11 months ago (4 children)
    function myFunction() {
      try {
        x = new Random().nextInt();
        if (x != 10) {
         throw "not 10";
        }
        else {
          return (10)
        }
        catch(err) {
          myFunction()
        }
      }
    }
    
    x = myFunction()
    

    Commit notes: Added error handling

  • source
  • hideshow 4 child comments
  • [–] 39 points 11 months ago (8 children)

    For a time on Reddit (some years ago when I still used it) there was a trend of finding the worst way of implementing is_even(x: int) -> bool. My contribution to that was a function that ran Ackerman(x,x) flipping a Boolean at every iteration, and check if it was true or false at the end.

    It works btw, I will find the proof later

  • source
  • hideshow 8 child comments
  • [–] 13 points 11 months ago (7 children)

    I would love to see the implementaion.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 18 points 11 months ago* (6 children)

    The implementation is not very exciting, I capture a variable in python. It could have been done more cleanly.

    1000041934

    The proof is this. But, I could have made mistakes, it was many years ago.

    1000041935

    Note that in python you'll never be able to run is_even(5) the stack cannot handle it

    Edit: daaaamn, that variable is ugly as hell. I would never do things like that now.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 1 point 11 months ago (2 children)

    It never occurred to me that you could assign fields to a function. I mean, it totally makes sense considering that functions are objects in Python. It just never occurred to me that this is a thing one can do. Crazy.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 11 months ago (1 child)

    Please don't do that, I was stupid when I wrote that. But still, in very dynamic languages like python or js everything is an object, including functions, so you can just do object stuff on them.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 11 months ago

    I wasn't going to, and after I saw it it totally makes sense that it's possible, it just never occurred to me.

    I guess this could be used like static variables inside functions in c. So scope-limited global variables. Not a good design choice in most cases.

  • source
  • parent
  • [–] 1 point 11 months ago (2 children)

    That's , uh...

    Yeah. Cooler than anything I could've achieved for purposefully bad is_even

    My first idea of a purposefully bad is_even is this:

    def is_even(i):
        return True if i == 0 else not is_even(abs(i)-1)
    

    But I'm sure I could come up with worse given enough time.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 11 months ago* (last edited 11 months ago) (1 child)

    That's also slower than most of the stuff you could come up with, it is so slow that there is no hyperoperation fast enough to describe it. There were other approaches that were almost worse though, like "the function is a switch-case that returns false by default. As complaint tickets are opened, more cases get added to the switch-case"

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 11 months ago

    the function is a switch-case that returns false by default. As complaint tickets are opened, more cases get added to the switch-case”

    Oh if that is acceptable, then my secondary idea of using an API call for this should work too. I thought that it would have to be guaranteed to be correct (as long as you don't reach a stack overflow or something)

  • source
  • parent
  • [–] 29 points 11 months ago (7 children)

    The compiler will optimize it anyway. /s

  • source
  • hideshow 7 child comments
  • [–] 16 points 11 months ago* (1 child)

    You jest, but you aren't wrong. At least if we are talking about C, C++ or Rust. https://godbolt.org/z/oPPfdfcf5

    .NET compiler is weak when it comes to optimizing your code; I assume Go's is as bad.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 4 points 11 months ago

    Technically yes... But I think he was more making the excuse for the gore "from the goresmith's perspective."

    And I'm not sure if the compiler in any language would change a random check function... The others are a possibility.

  • source
  • parent
  • [–] 6 points 11 months ago (4 children)

    Not sure about the last one though. The other two are trivial to optimize away.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 3 points 11 months ago* (3 children)

    An infinite loop canot be ruled out in the last case, so a compiler couldn't optimize this away without potentially changing the program behavior.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 11 months ago (2 children)

    Infinite loops are often weird though. They could be seen as undefined behavior and the compiler may do whatever it feels like.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 11 months ago (1 child)

    How could an infinite loop be considered UB?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 11 months ago

    Even though this isn't C, but if we take from the C11 draft §6.8.5 point 6 (https://www.open-std.org/jtc1/sc22/wg14/www/docs/n1570.pdf):

    An iteration statement whose controlling expression is not a constant expression, that performs no input/output operations, does not access volatile objects, and performs no synchronization or atomic operations in its body, controlling expression, or (in the case of a for statement) its expression-3, may be assumed by the implementation to terminate

    "new Random().nextInt()" might perform I/O though so it could still be defined behavior. Or the compiler does not assume this assumption.

    But an aggressive compiler could realize the loop would not terminate if x does not become 10 so x must be 10 because the loop can be assumed to terminate.

  • source
  • parent
  • [–] 25 points 11 months ago (6 children)

    That's not even enough to get you a job these days.
    You now have to use:

    do {
        x = reinterpret_cast<int>(AI::Instance().ask("Do Something. Anything. Be efficient and productive. Use 10 tokens."));
    } while (x != 10);
    
  • source
  • hideshow 6 child comments
  • [–] 8 points 11 months ago (2 children)

    This isn't just a function, it's a bold restatement of what it means to write code — a symphony of characters, questioning the very nature of the cutting edge language models that I want to beat with hammers.

  • source
  • parent
  • hideshow 2 child comments
  • [+] 6 points 11 months ago (2 children)
  • [–] 2 points 11 months ago (1 child)
  • [–] 2 points 11 months ago*

    First, we'll deep dive into "What is a variable?", then together we'll examine "Who sets a variable?", "What is an LLM?" and finally, "Who would set a variable without using an LLM?"

    You'll be a coding pro in no time!

    How does that sound?

    (I felt gross writing this lmao)

  • source
  • parent
  • [–] 24 points 11 months ago (8 children)

    How about

    x=x-x

    x++

    x++

    x++

    x++

    x++

    x++

    x++

    x++

    x++

    x++

  • source
  • hideshow 8 child comments
  • [–] 8 points 11 months ago*

    Freshman year of college doing assembly programming, I spent a while figuring out a "programmic" way to solve a problem, trying to wrangle labels and gotos. My friend came in with essentially this but as lookup table. It blew my mind.

    It was then that I learned to trade space for complexity.

  • source
  • parent
  • [–] 6 points 11 months ago (3 children)

    This is actually a valid brainf*ck program, but it results in 19, not 10.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 6 points 11 months ago (2 children)

    How so? It’s only 10 increments

  • source
  • parent
  • hideshow 2 child comments
  • [–] 4 points 11 months ago

    Because the only brainfuck instructions in your comment where a - which decrements and 20 +, each of which increments.
    Mine echos the first two characters from stdin, because of the commas and dots.

  • source
  • parent
  • [–] 1 point 11 months ago (2 children)

    Make sure you initialize x with x=x/x-x/x for better precision

  • source
  • parent
  • hideshow 2 child comments
  • [–] 14 points 11 months ago (9 children)

    What is the value of x in the Good example before the loop?

  • source
  • hideshow 9 child comments
  • [–] 14 points 11 months ago (6 children)

    An int. Value doesn't matter because it's overwritten.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 3 points 11 months ago (5 children)

    Not in this case. First, i is declared and assigned a value of 0. Next, x is declared and assigned a value of -i or -0. On the first loop iteration, i will decrement to -1, perform the conditional check, then execute the loop body which will assign x to -i or -(-1) or positive 1, and so on.

    The only time a variable is created without a value is if you declare one without assigning a value like with

    [int]i;

  • source
  • parent
  • hideshow 5 child comments
  • [–] 5 points 11 months ago (4 children)

    I know. OP asked what x was before the loop, and I just said it's an int. The int can be any value because as you pointed out it will be set to 0 in the first loop iteration.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 2 points 11 months ago (3 children)

    Shit, you're right. x is declared inside the loop, so it doesn't exist until the loop begins execution.

    Technically, I suppose you could say the compiler will allocate memory for x without assigning a value before the loop is executed and... I'm understanding what you mean now, I think.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 11 months ago (2 children)

    The code seems to be C-style language with curly braces and types in front for variable declarations, probably java. This means the variable must be declared of screen before the loop or it would not compile. It could have a previous value or be uninitialized, but that does not affect the end result.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 11 months ago

    Yeah, it does look like C now that I think about it. You're right about the end result too. I believe C# will let you do inline declaration and assignment like that, so maybe that's what we're looking at? Been a while, could be wrong

  • source
  • parent
  • [–] 1 point 11 months ago

    I read in on C but it's also true for JavaScript. The code implies that x was declared as an int sometime previously, or if JavaScript, just an object if not assigned a value giving it a type.

  • source
  • parent
  • [–] 1 point 11 months ago

    If this is JavaScript, it would have a value of -0, which is actually valid and works the same as normal zero thanks to type coercion. I think the only difference is some methods that detect if a number is negative will return true instead of false, but otherwise, JS treats -0 the same way as 0

  • source
  • parent
  • [–] 13 points 11 months ago (1 child)

    It would get you promoted at Twitter, where lines of code is the productivity metric.

  • source
  • hideshow 1 child comment
  • [–] 10 points 11 months ago (1 child)

    Probably Microsoft: You’re hired! Go work on GitHub

    Context

  • source
  • hideshow 1 child comment
  • [–] 21 points 11 months ago (1 child)

    I once was helping to organize the testing of town-level algorithmic competition for school students.

    The competition had one entry level issue that was basically solvable by reading the question properly, recognising that it's just multiplication of two numbers, and writing the simplest app ever.

    And there was one student who passed the automatic tests. We had to read the code too for the protocol, just to make sure there was no cheating.

    We looked in the code. What? Why? It had two nested for loops and a++ inside. When we understood what's going on we couldn't stop laughing for like solid ten minutes.

  • source
  • hideshow 1 child comment
  • [–] 8 points 11 months ago (2 children)

    Oddly enough, out of all of these the one the compiler has the best chance of optimizing out is the last one

  • source
  • hideshow 2 child comments
  • [–] 2 points 11 months ago

    What?

    First one is optimized obvious.

    Second one optimizes to x = 10 via constant propagation.

    Third one first unrolls the loop, propagates constants including booleans, and then eliminates dead code to arrive at x = 10.

    The last one cannot be optimized as “new” created objects that get used, nextInt() changes the state of those objects, and the global state of the random number system is impacted.

  • source
  • parent
  • [–] 4 points 11 months ago

    Now write a function to unroll the while loop to "optimize it for the compiler"

  • source
  • [–] 4 points 11 months ago
     x = (function(){
    return 10
    })
    

    Or something like that

  • source
  • [–] 3 points 11 months ago*

    Something like

    int *a = new int(10)
    
    Int*b = null
    
    While *b !=10 { b = rand(); a=new int(10)}
    
    Return *b
    

    I haven't coded recently enough in c/c++ to remember syntax but the concept might work eventually if you're lucky and have enough memory... Might need a time variant seed on the rand()...

  • source
  • [–] 3 points 11 months ago (3 children)
  • [–] 2 points 11 months ago (16 children)

    x = -i;

    Do many languages let you do that? When it's in front of a variable I would've expected it to be a subtraction operator only and you would need to do x = -1 * i;

  • source
  • hideshow 16 child comments
  • [–] 2 points 11 months ago (1 child)

    Personally I would expect it to behave the same in front of a numeric literal and in front of a variable. I do think most languages do that, but I haven't actually tested that many and could br wrong.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 11 months ago (4 children)

    I just tested it in PowerShell. Works fine

    $i = 1
    $x = -$i
    $x
    

    Outputs -1

  • source
  • parent
  • hideshow 4 child comments
  • [–] 2 points 11 months ago (3 children)

    Works fine in any language I ever used.

    I'm honestly quite surprised that this very basic language feature is even a matter of discussion here.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 2 points 11 months ago (2 children)

    It certainly makes me question a lot of things. This sub somehow manages to both feed my impostor syndrome and makong me feel like a genius programmer depending on the thread.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 11 months ago* (1 child)

    Totally, yes. I guess there's a ton of non-programmers and total beginners in this community.

    But sometimes there are some crazy good programmers here as well.

    What's really weird though is that I got two downvotes a bit further up for claiming that unary minus is a standard language feature.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 11 months ago* (last edited 11 months ago) (2 children)

    Why would they not let you do that? I honestly don't know a single language that wouldn't let you do that. Same as basic math notation allows you to do that.

    x = -i

    is a totally valid mathematical equation.

    For the downvoters: Find me a single language that supports operators but doesn't have an unary minus operator

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 11 months ago (1 child)

    It's a valid mathematical notation, sure. But there is an implicit understanding that the - in this case is making a number negative rather than subtracting (or, an implicit subtraction from 0).

    With the way negative numbers generally work in binary there would be much different ones and zeroes stored behind the scenes, so handling that would have to be pretty intentional.

    That said, I did just try it in Java because that's what I work in normally and I swear I had a gotcha with that. But it worked fine as far as I can tell.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 11 months ago (4 children)

    But of course -- It's just flipping around the -= operator!

  • source
  • parent
  • hideshow 4 child comments
  • [–] 3 points 11 months ago (2 children)

    Nope, it is not.

    x = 5
    i = 2
    x -= i // x => 3
    

    while

    x = 5
    i = 2
    x = -i // x => -2
    

    x=-i is the unary minus operator which negates the value right of it. It doesn't matter if that value is a literal (-3), a variable (-i) or a function (-f()).

    x-=i is short for x = x-i, and here it's a binary subtraction, so x is set to the result of i subtracted from x.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 11 months ago (1 child)

    I need to append /s to my future silly replies I think... that said, I'll never pooh-pooh a well thought response, so thanks for the nice write-up!

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 11 months ago*

    Thanks, I totally missed your sarcasm :)

    There's a couple people in this threat who seem to actually think that x = -i is some weird magic instead of a standard feature that's present in every major programming language.

  • source
  • parent
  • [–] 2 points 11 months ago

    bad but more chaotic

    you want to be punk rock not blues or jazz

  • source
  • [–] 1 point 11 months ago

    x = printf("one + nine");

  • source
  • [–] 1 point 11 months ago

    Can someone make it an async function?

  • source
  • [–] 1 point 11 months ago

    it takes two instructions to materialize a constant in risc-v. X64 has LEA.

    Risc-v is better!

  • source