mutate one array, instead of pushing sorted items in a new array
Otherwise it can be shown as an example of a binary algorithm, but actually it's so strongly slow that even doesn't make any sense. Just because ... has O(n) where n is .length and you repeat it again and again. Whereas mutating an existing array you will just sometimes shift elements.
Array push has O(1), but sometimes it costs O(n). Becase it's like a rubber. It has some extra space at the end and when this space ran out it just copies everything to a new array with bigger extra space.
In case when you need to create a new sorted array you can just do it at the very beginning. I mean create a clone-array and then mutate it.
I used this example to teach the idea of quicksorting and recursion. This is not a real-world example. By the way, you can easily get rid of shift() using first item as an index arr[0] and start iterating by staring from index 1. And I don't think ... gives you any side effects.
But you might be right about push(). If you truly improve the performance of your sort, you should really focus on choosing better pivots I picked the first for simplicity. Maybe try using Median of three(First,Middle,Last) or maybe median of three random elements to avoid O(n2).
So anyway, I'm open for improvements to ease out the learning curve of this algorithm.
... doesn't give you any side-effect, you're right. It's FP pure. But it is copying the whole array every time :) That's where the problem is.
If you truly improve the performance of your sort, you should really focus on choosing better pivots I picked the first for simplicity
I think you may even set pivot equal 0 each iteration and even then it would be faster than version with ... :)
... ruins this algorithm. Most of algorithms have to be impure to be useful.
So you can dramatically improve it by getting rid of excess copies. I think it'd be better even in educational point of view
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
Unfortunately in real world you also need:
...Otherwise it can be shown as an example of a binary algorithm, but actually it's so strongly slow that even doesn't make any sense. Just because ... has O(n) where n is .length and you repeat it again and again. Whereas mutating an existing array you will just sometimes shift elements.
Array push has O(1), but sometimes it costs O(n). Becase it's like a rubber. It has some extra space at the end and when this space ran out it just copies everything to a new array with bigger extra space.
In case when you need to create a new sorted array you can just do it at the very beginning. I mean create a clone-array and then mutate it.
I used this example to teach the idea of quicksorting and recursion. This is not a real-world example. By the way, you can easily get rid of
shift()using first item as an indexarr[0]and start iterating by staring from index 1. And I don't think...gives you any side effects.But you might be right about
push(). If you truly improve the performance of your sort, you should really focus on choosing better pivots I picked the first for simplicity. Maybe try using Median of three(First,Middle,Last) or maybe median of three random elements to avoid O(n2).So anyway, I'm open for improvements to ease out the learning curve of this algorithm.
...doesn't give you any side-effect, you're right. It's FP pure. But it is copying the whole array every time :) That's where the problem is.I think you may even set pivot equal 0 each iteration and even then it would be faster than version with
...:)...ruins this algorithm. Most of algorithms have to be impure to be useful.So you can dramatically improve it by getting rid of excess copies. I think it'd be better even in educational point of view