Module graph 

Graph Data Structures for PageRank Centrality

Efficient graph representation optimized for dependency analysis and PageRank computation. This module implements the core graph structures used in the PageRank centrality algorithm with emphasis on reverse edges for code dependency analysis.

Design Philosophy

• Forward edges: A imports B (A -> B)
• Reverse edges: B is imported by A (B <- A)
• PageRank flows along reverse edges (importance flows to imported files)
• Memory-efficient adjacency list representation for large graphs (10k+ nodes)
• Fast degree queries and statistics calculation

Structs

ConcurrentDependencyGraph

Thread-safe concurrent version of DependencyGraph

DegreeInfo

Degree information for a node

DependencyGraph

Efficient dependency graph representation optimized for PageRank computation Uses integer-based internal representation for massive performance improvements

GraphStatistics

Graph statistics computed lazily and cached

NodeMetadata

Metadata associated with each node in the graph

Enums

TraversalDirection

Direction for graph traversal

Type Aliases

EdgeWeight

Edge weight type (unused in unweighted PageRank, but reserved for extensions)

InternalNodeId

Internal node identifier type for efficient graph operations (usize for array indexing)

NodeId

External node identifier type for the dependency graph (file paths)