Finally I decided implement the most common collections that I myself usually need. All implementations support "objects". APIs are not necessarily consistent. Implementations are not necessarily efficient for large lists.
- Arraylist: a preallocated fixed size list
- Linkedlist: good old double linked linkedlist which can be used as a stack &| queue
- Set: based on the linkedlist implementation (i.e. no hashmap)
- Dict: will be based on the set implementation (i.e. no hashmap)
- "Proper" unit tests: currently there are a few test functions that can be called in main
Set and Dict should ideally use hashmaps but that will complicate things a lot. For the time being this is/will be a library for small/embedded use-cases where sizes shouldnt grow too big.