Module tree 

Tree genomes for genetic programming

This module provides tree-based genomes for symbolic regression and genetic programming applications.

Deep trees and stack safety

Every traversal in this module uses an explicit work stack instead of recursion, so no operation overflows the call stack even for pathologically deep trees. This covers the read-side traversals (TreeNode::depth, TreeNode::size, TreeGenome::evaluate, and the position collectors backing TreeNode::positions/TreeNode::terminal_positions/ TreeNode::function_positions), the crate::operators point-mutation traversal, AND teardown: TreeNode implements a stack-safe Drop (EV-60) that frees an arbitrarily deep tree iteratively, so even dropping a deep tree implicitly (never calling TreeGenome::dismantle) cannot overflow the stack. The Drop impl moves each node’s children out with mem::take — which is permitted under Drop, unlike moving a field out by value (E0509) — so the operator layer must likewise use mem::take/in-place mutation rather than by-value destructuring of an owned TreeNode. TreeGenome::dismantle and drop_node_iteratively remain as explicit, self-documenting entry points but are no longer required for correctness.

Structs

TreeGenome

Tree genome for genetic programming

Enums

ArithmeticFunction

Standard arithmetic functions for symbolic regression

ArithmeticTerminal

Standard arithmetic terminals for symbolic regression

TreeNode

A node in a GP tree

Traits

Function

Trait for function nodes in GP trees

Terminal

Trait for terminal nodes in GP trees

TreeGenomeType

Trait for tree genome types (marker trait for operators)

Functions

drop_node_iteratively

Free a tree node and all of its descendants without deep recursion.