bobluppes/graaf

Summary PR #448 adds pagerank under include/graaflib/algorithm/ranking/, but no corresponding benchmark was added to the Graaf_perf executable. Every other algorithm category under perf/graaflib/ (traversal, shortest-path, MST, coloring, clique detection) has a matching benchmark; ranking/ is currently the only category without one. Scope Add perf/graaflib/pagerank_benchmark.cpp, following the structure of the existing benchmarks in that directory (e.g. bron_kerbosch_benchmark.cpp, prim_benchmark.cpp). It's picked up automatically by the file(GLOB PERF_SOURCES "graaflib/*.cpp" "graaflib/*/*.cpp") glob in perf/CMakeLists.txt, so no build-file changes should be needed. This is scoped to the Graaf-only perf/graaflib executable. It does not include adding a comparison benchmark to perf/boost/ — Boost's PageRank lives in the Parallel/Distributed BGL rather than the sequential BGL core, so a fair one-to-one comparison there is a separate, larger discussion. Suggestions Benchmark on synthe

Summary Graaf's vertex-properties module exposes only raw degree counts. None of the standard centrality measures (betweenness, closeness, eigenvector) are implemented. Current state include/graaflib/properties/vertex_properties.h currently exposes exactly three functions: vertex_degree, vertex_outdegree, vertex_indegree. A search across include/graaflib/ for "centrality" (betweenness/closeness/eigenvector) returns no results. Why this matters Degree is the simplest possible notion of "importance" in a graph, but it's rarely sufficient on its own — betweenness centrality (how often a vertex lies on shortest paths between other vertices) and closeness centrality (how close a vertex is to all others) are the standard next tier of analysis for questions like "which node is a bottleneck" or "which node can reach the rest of the network fastest," and they come up constantly in social-network analysis, infrastructure/routing analysis, and organizational-graph analysis. Their absence means

Summary Graaf has no PageRank implementation, or any other iterative graph-ranking algorithm. Current state A search across include/graaflib/ for "pagerank"/"page_rank" returns no results, and there is no algorithm/ranking/ (or similar) category. There is a related historical request, #90 "[ALGO] Page Rank", which was closed without being implemented (closed due to inactivity, not as "won't do"). Why this matters PageRank (and the broader family of iterative eigenvector-based ranking algorithms it represents) is one of the most widely recognized graph algorithms outside of pure computer science — used for web ranking, recommendation systems, citation analysis, and influence scoring in social/organizational networks. Its absence is a notable gap for anyone evaluating the library specifically for graph-analytics use cases (as opposed to pathfinding/traversal use cases, which are already well covered). Suggested resolution Implement PageRank (the standard damping-factor, iterative-unti

Summary Graaf implements two minimum-spanning-tree algorithms (Kruskal, Prim) but not a third standard one: Borůvka's algorithm. Current state include/graaflib/algorithm/minimum_spanning_tree/ contains only kruskal.h/.tpp and prim.h/.tpp. There is a related historical request, #113 "[ALGO] Borůvka's algorithm", which was closed without being implemented (closed due to inactivity, not as "won't do"). Why this matters Borůvka's algorithm is one of the three classical MST algorithms taught and used alongside Kruskal's and Prim's, and it has a property neither of the other two has out of the box: it parallelizes naturally, since each iteration finds the minimum outgoing edge for every component simultaneously. For a library whose own MST category already covers two of the three standard approaches, omitting the third — particularly the one most relevant to parallel/large-scale processing — is a visible completeness gap for anyone comparing MST algorithm coverage against alternatives

Summary Graaf only implements DFS-based topological sorting. Kahn's algorithm (the BFS/in-degree-based alternative) is not implemented. Current state include/graaflib/algorithm/topological_sorting/ contains only dfs_topological_sorting.h/.tpp. There is a related historical request, #114 "[ALGO] Kahn's Algorithm", which was closed without being implemented (closed due to inactivity, not as "won't do"). Why this matters Kahn's algorithm is the standard alternative to DFS-based topological sorting, and it has properties the DFS version doesn't: it naturally detects cycles as a side effect (if not all vertices are output, the graph has a cycle), it processes vertices in in-degree order rather than via recursion (avoiding recursion-depth concerns on very large or deep graphs), and it more naturally supports "level" or "layer" based processing where all vertices at the same dependency depth are grouped together. Offering only one of the two standard approaches leaves a well-known, commonly r