Introduction

A simple reallocating vector library for nim.

But wait a second, there is already a seq for holding dynamic arrays, why should I use this stupid vector?

The reasons are:

• --gc=none
  • seqs are managed in memory through the gc
• Performance:
  • You can allocate the vector with a big size to ensure low allocation count and use it as a simple memory pool

The vector's buffer extends when you add an item using push or extend by a power of 2 when it runs out of memory.

For example:

var vec = initVector[int]()  # vec's buffer size is 4 (int.sizeof())

vec.extend(@[1, 2, 3, 4, 5])  # vec's buffer size is 32
# 32 is the lowest power of 2 that is bigger than 20 (5 * int.sizeof())

Notes

The buffer never resizes down (shrinks).

If you really want to shrink your vector, you can allocate a new one using the old one:

var vec = initVector[uint8](1000)  # vec's buffer size is 1024
vec.clear()  # still 1024
var newVec = initVector(vec)  # newVec's buffer size is now float.sizeof()