Please check the build logs for more information.
See Builds for ideas on how to fix a failed build, or Metadata for how to configure docs.rs builds.
If you believe this is docs.rs' fault, open an issue.
Behavior trees are a common means of composing behavior in game AI. They are related to state machines, and in many cases can be reduced to one, but are conceptually easier to reason about, especially when the complexity gets larger, and a state machine's state flows start to resemble goto spaghetti.
In this library, behavior trees are implemented in a very generalized manner, making them very versatile. At each step, a behavior tree node steps some, and stop at either an exit point, which abstracts over that node's terminal states, or a decision point, which abstracts over that node's nonterminal states. At a terminal state, only a transition to an entirely new behavior tree node is possible. However, at a decision point, the parent behavior tree node can either decide to step the behavior tree node as normal, or cause a transition to an entirely new behavior tree node, which abandons the original child node. The parent can also themselves transition to an exit point, which necessarily causes their children to be halted and dropped.
Behavior trees here are also implemented in a zero-cost manner. The behavior tree is not an actual structure in memory, but a logical structure, which when translated to code, reduces to a type whose state transitions are very similar to the state machine one would write by hand, but without the tedium or the copypaste errors. Only the memory needed to hold the state of the active nodes is used, and the conceptual tree-walk is translated to something more like a state machine transition in code, especially if optimizations are turned on.