Y'all are surprising me with your programming opinions

Then again, JavaScript or whatever apparently has a Math.max function, while C doesn't

@wallhackio Are you referring to yourself or to me or to us all?

@vaporeon_ I'm referring to anyone who writes code for money instead of writing code for fun

@vaporeon_ this is not a slight on Holly. i hope that is clear. she does code golf so you know she actually likes coding for the hell of it

@wallhackio @vaporeon_ oh yeah all this goes out the window when you're golfing. sometimes it's fun to let loose and write some horseshit

@monorail @vaporeon_ my worst code practices occur when i hack a solution to a challenging data structures/algorithms problem

@wallhackio How can you have bad code practices when there's no such thing as good Java code, it's all terrible

@vaporeon_ I've done DS&A problems in JavaScript and written some nightmarish oneliners doing that

Follow

@vaporeon_ I know how this is gonna go. I will show it to you and then you'll be like, wait this is cool I like it :)

@vaporeon_ okay here's a bad one. i implemented a non-recursive quicksort using the variables gorp, norp, and sorp:

const partition = (arr, start, end, strategy) => {
const q = [[start, end]];
while (q.length > 0) {
const [left, right] = q.pop();

const p = strategy(arr, left, right);
const pVal = arr[p];
swap(arr, p, right);

let gorp = 0;
let norp = 0;
let sorp = 0;

const L = right - left + 1;

for (let i = 0; i < L - 1; ++i) {
const curr = left + gorp + sorp;
if (arr[curr] === pVal) {
swap(arr, left + sorp, curr);
++sorp;
} else if (arr[curr] > pVal) {
swap(arr, curr, right - norp - 1);
++norp;
} else {
++gorp;
}
}

for (let j = 0; j < gorp; ++j) {
swap(arr, left + j, left + sorp + j);
}
swap(arr, left + gorp + sorp, right);

if (gorp > 1) q.push([left, left + gorp - 1]);
if (norp > 1) q.push([left + gorp + sorp + 1, right]);
}
};

@wallhackio I love how my Mastodon posting has made you use gorp, norp, and sorp in actual code

@aescling @vaporeon_ if you are wondering this is using the Dutch National Flag partition, which is the actual real name for it, and I simulated recursion using a stack (a JavaScript array has methods on it that allow it to easily resemble a stack)

I used the ninther strategy for the pivot selection but that was not present in this snippet

@aescling @vaporeon_ oh this was an inefficient implementation of Dutch National Flag too, lol

@wallhackio @aescling @vaporeon_ It's funny how "a JavaScript start has methods on it that allow it to easily resemble a stack" should immediately set off all kinds of red flags for anyone with basic algorithms knowledge, *and yet* because developers regularly use them like that, most JavaScript engines have developed to make those operations fast and otherwise efficient.

@aschmitz @aescling @vaporeon_ I once used a JavaScript array as a queue, using the shift method to pluck elements from the start of the array. Of course, this means iterating over an entire "queue" should be slow, since every call to shift causes every element in the array to be reassigned and reduces the size of the array by 1. I benchmarked my algorithm and it was slower than I hoped, so I immediately worked on the queue as I thought it was a throttle for the algorithm.

I made an object that effectively had a JavaScript array as a private variable and created a method for "removing" the first element from the array. In reality, there was pointer being tracked by the object that simply shifted to the right by 1, simulating the behavior of a queue with an ordinary array. Obviously iterating over this structure should be O(n).

To my surprise, when I benchmarked the algorithm again, it was slower. Evidently, with my first approach, the JIT engine realized I was using an array as a queue and secretly optimized it for me. Without realizing, I tried to outsmart the JIT engine, and lost.

@vaporeon_ here is another hack with a "beautiful" one liner:

Problem Statement
You are given a 0-indexed array of integers nums of length n. You are initially positioned at nums[0].

Each element nums[i] represents the maximum length of a forward jump from index i. In other words, if you are at nums[i], you can jump to any nums[i + j] where:

  • 0 <= j <= nums[i] and
  • i + j < n

Return the minimum number of jumps to reach nums[n - 1]. The test cases are generated such that you can reach nums[n - 1].

Example

  • Input: nums = [2,3,1,1,4]
  • Output: 2
  • Explanation: The minimum number of jumps to reach the last index is 2. Jump 1 step from index 0 to 1, then 3 steps to the last index.

Code

const jump = (nums, i = 0, gorp = {}) => {
if (i === nums.length - 1) return 0;
if (i > nums.length - 1 || nums[i] === 0) return Infinity;
if (gorp[i] !== undefined) return gorp[i];

let min = Infinity;
for (let j = 1, res; j <= nums[i]; ++j) res = 1 + jump(nums, i + j, gorp), min = res < min ? res : min;

return gorp[i] = min;
};

@vaporeon_ oooooh I could have done for (var min = Infinity, j = 1, res; j <= nums[i]; ++j) res = 1 + jump(nums, i + j, gorp), min = res < min ? res : min;

javascript variables declared with var and not block scoped, so min is available outside of the for loop teehee

@wallhackio @vaporeon_ if you look into the semantic diffurences between for with let and var scoped variables, this is not surpurrising; the furmer is surpurrisingly complicated. const allows fur some optimizations over let though iirc

@aescling @vaporeon_ with let the algo ran the test cases ~700ms and with var it shaved ~100ms

Sign in to participate in the conversation
📟🐱 GlitchCat

A small, community‐oriented Mastodon‐compatible Fediverse (GlitchSoc) instance managed as a joint venture between the cat and KIBI families.