36
Python Big O: the time complexities of different data structures in Python
(www.pythonmorsels.com)
Welcome to the Python community on the programming.dev Lemmy instance!
Past
November 2023
October 2023
July 2023
August 2023
September 2023
Zero surprises. It's the same as in any other language.
I was a bit surprised that
deque
is implemented as a linked list and not, for example, a ring buffer. It would mean that index reads would be constant time (though insert and delete at an index would be linear time), the opposite of using a linked list.But a ring buffer is a FIFO data structure that can be implemented using linked lists. What ring buffer implementation did you have in mind?
Standard ring buffers or circular buffers are implemented as continuous vectors, with a read and write pointer.
Rust's
VecDeque
is a ring buffer: https://doc.rust-lang.org/std/collections/struct.VecDeque.htmlNo that's a subtly different thing. The storage is a contiguous vector, but because it is a ring buffer there must be one pair of adjacent elements that are not contiguous in memory. That's what that comment is saying.
But because it is only one discontinuity it doesn't affect algorithmic complexity, so it is still O(1) to access elements, unlike a linked list which is O(N).
If you google "circular buffer" there will be loads of diagrams that make it really obvious how it works.