Module algorithms 

Graph algorithms operating against the GraphQuery<V> interface.

Ported from Pattern.Graph.Algorithms in the Haskell reference implementation.

All functions are representation-independent: they operate on GraphQuery<V> closures only. The same code works against PatternGraph, in-memory closures, or any future backing store.

Traversal weight semantics

Pass a TraversalWeight<V> to control which edges are traversable and at what cost. Use the canonical functions undirected, directed, or directed_reverse, or supply a custom Rc<dyn Fn(...)>.

An edge with INFINITY cost in a given direction is impassable in that direction.

Functions

all_paths

Enumerate all simple paths from from to to (no repeated nodes).

betweenness_centrality

Betweenness centrality using the Brandes BFS algorithm (unnormalized).

bfs

Breadth-first traversal from start.

connected_components

Partition the graph into connected components.

degree_centrality

Degree centrality for all nodes.

dfs

Depth-first traversal from start.

has_cycle

Returns true if the graph contains a directed cycle.

has_path

Returns true if a path exists from from to to.

is_connected

Returns true if the entire graph is connected under weight.

is_neighbor

Returns true if b is directly reachable from a in one hop under weight.

minimum_spanning_tree

Minimum spanning tree using Kruskal’s algorithm with path-compression union-find.

query_annotations_of

Returns all containers of element that are classified as annotations.

query_co_members

Returns all elements that share container with element, excluding element itself.

query_walks_containing

Returns all containers of element that are classified as walks.

shortest_path

Find the minimum-cost path from from to to using Dijkstra’s algorithm.

topological_sort

Topological sort using iterative DFS post-order with cycle detection.