Hi everyone.

I am working on a personal project where I need to represent a horizontal tree in a visual way for a user to interact with it, like shown at FIG. 1.

FIG. 1

IMG_2128.jpg2798×2568 1.14 MB

I have thought about representing it as a list of lists as it usually goes, but I have this constraints:

• I need to increase the depth of the tree at user request
• The content of the new nodes (at depth expansion) depends solely on the leafs of the tree

To do this, I need to know the content of each leaf every time I want to increase its depth, and transversing the lists every time seems unnecessary and expensive. With other programming languages I could store the memory reference of each leaf at creation time, but with elixir this is not possible.

It came to my mind the possibility to make every node a process, having a “tree manager” that stores the PID of the root and the PIDs of each leaf. Leading to interact directly with the leafs at depth expansion without interacting with the rest of the nodes. But my lack of experience doesn’t let me know if this is unnecessary memory usage, seeing that a process uses 338 words when spawned, including a heap of 233 words compared with other data structures.

The tree is not very wide, but can increase its depth at user request. Maybe I can set a limit to change the root of the tree for one closer to the leafs every time the user passes it.

What do you guys think?