d-segmented-hashmap 1.0.4

d-segmented-hashmap

A segmented hash map implementation for D which provides must faster table creation

Implementatio: Almost implements every feature from D AA:

• Stable location so, you can hold up any key/value elocation

Improvements over D AA

• (Better for fire and forget) -> No rehash happens
• Might use less memory, specially for strings as keys since the state is encoded on it
• put is a much faster operation, almost tripling the speed of D AA
• You can pre-initialize its capacity (reserve memory)

Cons:

• A get operation with best case of O(1) and worst case of O(log n). Though it will never be O(n) since no rehash happens
• After removing elements, other segmentations of the map will become unused. True removal can only be achieved when calling .clear()

Usage

import hip.util.shashmap;
HashMap!(string, string) map;
//map.setCapacity(50_000); //You may use that for reserving the size
foreach(i; 0..50000)
{
    map[identifiers[i]] = values[i];
}

foreach(string value; map)
{
    target = value;
}

string* myValue = "someKey" in map;
map.remove("someKey");