MATLAB already has a built in class to handle Hash Map/Tables using containers.Map(). However, support for differing data types is limited (optional arguments exist to support differing data types. Some data types are not supported at all (structs, function handles, etc).

This implementation gives you more of a modern implementation of hash maps, allowing mixing of data types, functions/structs/classes as map keys, and concise language to improve code readability.

HashMap will auto rehash as new data gets added. Rehash determination is based on number of entries in the map and current number of buckets allocated to the internal array.

A psudo seperate chaining takes place when hash collisions occur (psudo as in not using a more traditional linked list as storage).

Look up - Best/Average case O(1), worst case O(N) [when hash function produces key collisions] Space - O(N) array increases linearly with number of dictionary entries

Double click on the toolbox file (HashMap.mtlbx) or drag the file into the MATLAB command window. This will install the class globally within MATLAB.

• get(key) -> get value associated with key
• set(key,value) -> set new element in map by key and value
• delete(key) -> delete entry by key
• has(key) -> Returns true/false based on whether the map contains the key
• keys() -> returns all keys of map
• values() -> returns all values of map
• size - current size of the map (number of keys)

map = HashMap(); map.size

size = 0

map = HashMap({'a','b','c'},{1,2,3}); map.size

3

map = HashMap({'a','b','c'},{1,2,3}); map.size

3

map.set('d','4'); map.size

4

map.has('d')

true

map.get('d')

4

map.delete('a'); map.has('a') map.size

false, 3

map.keys()

'c' 'b' 'd'

map.values()

3, 2, 4