GLEAM_FSHARP_COMPARISON.md

Yog: Gleam vs F# Implementation Comparison

This document compares the Gleam and F# implementations of the Yog graph algorithm library.

Quick Summary

Aspect	Gleam	F#
Repository	code-shoily/yog	code-shoily/yog-fsharp
Language	Gleam (BEAM/Erlang VM)	F# (.NET)
Package	hex.pm/packages/yog	nuget.org/packages/Yog.FSharp
Documentation	HexDocs	GitHub Pages
Status	Stable, Production Ready	0.5.0 Pre-release
Total Algorithms	60+	50+
Lines of Code	~13,000	~8,500

Core Data Structures

Feature	Gleam	F#	Notes
Graph<'n, 'e>	✅	✅	Directed/Undirected with generic node/edge data
MultiGraph	✅	✅	Parallel edges between nodes
DAG (Directed Acyclic Graph)	✅	✅	Type-safe wrapper with cycle prevention
Disjoint Set (Union-Find)	✅	✅	Path compression and union by rank

Pathfinding Algorithms

Algorithm	Gleam	F#	Complexity
Dijkstra	✅	✅	O((V+E) log V)
A*	✅	✅	O((V+E) log V)
Bellman-Ford	✅	✅	O(VE)
Floyd-Warshall	✅	✅	O(V³)
Distance Matrix	✅	✅	All-pairs distances
Implicit Pathfinding	✅	✅	State-space search

Status: ✅ Feature parity - All algorithms present in both

Graph Traversal

Algorithm	Gleam	F#	Notes
BFS	✅	✅	Breadth-first search
DFS	✅	✅	Depth-first search
Early Termination	✅	✅	Stop on goal found
Implicit Traversal	✅	✅	On-demand graph exploration
Topological Sort	✅	✅	Kahn's algorithm
Lexicographical Topo Sort	✅	✅	Stable ordering

Status: ✅ Feature parity

Flow & Optimization

Algorithm	Gleam	F#	Status
Edmonds-Karp (Max Flow)	✅	✅	Both fully functional
Min Cut from Max Flow	✅	✅	Both fully functional
Stoer-Wagner (Global Min Cut)	✅	✅	Both fully functional
Network Simplex (Min Cost Flow)	✅ ✅	⚠️ ❌	Gleam: Complete (930 LOC), F#: Incomplete (349 LOC)

Status: ✅ Gleam has complete, production-ready Network Simplex

Network Simplex Details

Component	Gleam	F#
Initial state setup	✅	✅
Demand validation	✅	✅
Entering edge selection	✅	✅
find_cycle	✅	❌
find_leaving_edge	✅	❌
augment_flow	✅	❌
Tree updates	✅	❌
Potential updates	✅	❌
Tests pass	✅ All tests	❌ Fails all tests

Gleam Advantage: Full Network Simplex implementation with all pivot logic, extensively tested.

Graph Properties & Analysis

Feature	Gleam	F#	Notes
Connectivity	✅	✅
- Bridges	✅	✅	Tarjan's algorithm
- Articulation Points	✅	✅	Tarjan's algorithm
- Strong Components (SCC)	✅	✅	Tarjan's & Kosaraju's
Cyclicity	✅	✅	Cycle detection
Eulerian Paths/Circuits	✅	✅	Hierholzer's algorithm
Bipartite Graphs	✅	✅	Detection & max matching
Stable Marriage	✅	✅	Gale-Shapley algorithm
Cliques	✅	✅	Bron-Kerbosch algorithm

Status: ✅ Feature parity

Centrality Measures

Measure	Gleam	F#	Notes
Degree Centrality	✅	✅
Betweenness Centrality	✅	✅	Int & Float variants
Closeness Centrality	✅	✅	Int & Float variants
Harmonic Centrality	✅	✅	Int & Float variants
PageRank	✅	✅	Iterative algorithm
Eigenvector Centrality	✅	✅	Power iteration
Katz Centrality	✅	✅
Alpha Centrality	✅	✅

Status: ✅ Feature parity - All 8 centrality measures in both

Community Detection

Algorithm	Gleam	F#	Notes
Louvain	✅	❌	Fast modularity optimization
Leiden	✅	❌	Quality guaranteed partitions
Label Propagation	✅	❌	Near-linear time scaling
Girvan-Newman	✅	❌	Hierarchical edge betweenness
** Walktrap**	✅	❌	Random walk distances
Infomap	✅	❌	Information-theoretic flow
Clique Percolation	✅	❌	Overlapping communities
Local Community	✅	❌	Massive graphs, seed expansion
Fluid Communities	✅	❌	Exact k partitions, fast
Metrics & Modularity	✅	❌	Quality evaluation

Status: ✅ Gleam exclusive feature

Minimum Spanning Trees

Algorithm	Gleam	F#	Notes
Kruskal's MST	✅	✅	O(E log E)
Prim's MST	✅	✅	O(E log V)

Status: ✅ Feature parity

Graph Generators

Classic Deterministic Graphs

Generator	Gleam	F#	Description
Complete (K_n)	✅	✅	Every node connected
Cycle (C_n)	✅	✅	Ring structure
Path (P_n)	✅	✅	Linear chain
Star (S_n)	✅	✅	Hub with spokes
Wheel (W_n)	✅	✅	Cycle + center hub
Grid 2D	✅	✅	Rectangular lattice
Binary Tree	✅	✅	Complete binary tree
Complete Bipartite	✅	✅	K_{m,n}
Petersen Graph	✅	✅	Famous 10-node graph
Empty Graph	✅	✅	Isolated nodes

Status: ✅ Feature parity - All 10 classic generators

Random Network Models

Generator	Gleam	F#	Description
Erdős-Rényi G(n,p)	✅	✅	Edge probability p
Erdős-Rényi G(n,m)	✅	✅	Exactly m edges
Barabási-Albert	✅	✅	Scale-free networks
Watts-Strogatz	✅	✅	Small-world networks
Random Trees	✅	✅	Uniformly random spanning trees

Status: ✅ Feature parity - All 5 random generators

Graph Builders

Builder	Gleam	F#	Use Case
Labeled Builder	✅	✅	Use custom labels instead of IDs
Live Builder	✅	✅	Interactive construction
Grid Builder	✅	✅	Lattice/grid graphs

Status: ✅ Feature parity

Graph Transformations

Operation	Gleam	F#	Notes
Transpose	✅	✅	O(1) edge reversal
Map Nodes	✅	✅	Transform node data
Map Edges	✅	✅	Transform edge data
Filter Nodes	✅	✅	Remove nodes by predicate
Filter Edges	✅	✅	Remove edges by predicate
Subgraph	✅	✅	Extract by node set
Merge	✅	✅	Combine graphs
Contract Edges	✅	❌	Gleam only - Merge nodes

Status: ✅ Gleam has edge contraction

Visualization & I/O

Format	Gleam	F#	Purpose
DOT (Graphviz)	✅	✅	Professional visualization
JSON	Extenstion Package	✅	Web APIs, data interchange
Mermaid	✅	✅	Markdown diagrams
GraphML	🔶 Planned	✅	XML format for Gephi, yEd, Cytoscape
GDF	🔶 Planned	✅	Gephi lightweight format

Status: ⚠️ F# has 2 additional export formats (planned for Gleam)

Future Enhancement: GraphML and GDF export would be valuable additions to Gleam's I/O capabilities for better integration with tools like Gephi and yEd.

DAG-Specific Algorithms

Feature	Gleam	F#	Notes
Type-safe DAG wrapper	✅	✅	Prevents cycles at compile time
Longest Path	✅	✅	Critical path analysis
Topological Sort	✅	✅	Guaranteed success on DAG
Transitive Closure	✅	✅	Reachability matrix
Transitive Reduction	✅	✅	Minimal equivalent DAG

Status: ✅ Feature parity

MultiGraph Support

Feature	Gleam	F#	Notes
Parallel Edges	✅	✅	Multiple edges between nodes
Edge IDs	✅	✅	Unique identification
Eulerian for MultiGraphs	✅	✅	Specialized implementation
MultiGraph Traversal	✅	✅	BFS/DFS with edge IDs

Status: ✅ Feature parity

Performance Optimizations

Feature	Gleam	F#
Pairing Heap (Priority Queue)	✅	❌
Two-List Queue (BFS)	✅	❌
Mutable Arrays for Hot Paths	Limited	✅
O(1) Transpose	✅	✅

Gleam Advantage: Custom data structures (Pairing Heap, Two-List Queue) optimized for graph algorithms.

Testing & Quality

Aspect	Gleam	F#
Unit Tests	✅ Extensive	✅ Extensive
Property-Based Tests	✅ qcheck	✅ FsCheck
Example Count	25+	37+
Documentation Coverage	✅ Complete	✅ Complete
CI/CD	✅ GitHub Actions	✅ GitHub Actions

Platform & Ecosystem

Aspect	Gleam	F#
Runtime	BEAM/Erlang VM	.NET CLR
Target Platforms	Erlang, JavaScript	Windows, Linux, macOS
Concurrency Model	Actor model (OTP)	async/await, Tasks
Package Manager	Hex	NuGet
Interactive	Gleam REPL	F# Interactive (FSI)

Unique Features

Gleam Only

• ✅ Community Detection Suite - 10 algorithms including Louvain, Leiden, Infomap, Fluid
• ✅ Complete Network Simplex - Full min cost flow implementation (930 LOC)
• ✅ Edge Contraction - Graph transformation
• ✅ Pairing Heap - Custom priority queue for pathfinding
• ✅ Two-List Queue - Optimized BFS queue
• ✅ Production Ready - Stable 5.0.0 release

F# Only

• ✅ GraphML Export/Import - XML graph format for Gephi, yEd, Cytoscape
• ✅ GDF Export - Gephi lightweight format
• ✅ More Examples (37 vs 25)
• ✅ .NET Integration - Seamless with C#/VB.NET

Migration Guide

F# → Gleam

Straightforward migration, but note:

• ✅ Network Simplex works correctly in Gleam
• ✅ All core algorithms present and tested
• ✅ GraphML/GDF features are in an extension package yog-io

Gleam →

Mostly straightforward, but watch for:

• ⚠️ Network Simplex is incomplete in F# (use Gleam or wait for update)
• ✅ All other algorithms are functionally equivalent
• ✅ F# has additional export formats (GraphML, GDF)

Version History

Version	Gleam	F#
Latest	5.0.0	0.5.0
First Release	2024	2025
Stability	Stable	Pre-release

Recommendations

Choose Gleam If

• ✅ Building BEAM/Erlang applications or microservices
• ✅ Need min cost flow (Network Simplex) in production
• ✅ Want battle-tested, stable code (0.6.0+)
• ✅ Prefer functional programming on Erlang VM
• ✅ Building fault-tolerant, distributed systems
• ✅ Using Phoenix, Nerves, or other BEAM ecosystem tools

Choose F# If

• ✅ Working in .NET ecosystem
• ✅ Need GraphML/GDF export for Gephi/yEd integration
• ✅ Want seamless C# interop
• ✅ Can wait for Network Simplex completion or use alternatives
• ✅ Prefer statically typed .NET with excellent tooling
• ✅ Building desktop or Windows-centric applications

Roadmap

Gleam Planned

• GraphML Export/Import - Add XML-based graph format support
• GDF Export - Add Gephi lightweight format
• More examples to match F# (target: 35+)
• Performance benchmarks vs F#
• Additional random graph models

F# Planned

• Complete Network Simplex - Port full pivot logic from Gleam
• Edge contraction transformation
• Custom data structures (Pairing Heap, Two-List Queue)

Both

• Additional centrality measures
• Graph isomorphism detection
• Graph coloring algorithms

Contributing

Both implementations welcome contributions!

• Gleam: github.com/code-shoily/yog
• F#: github.com/code-shoily/yog-fsharp

Summary

Both implementations are high-quality, feature-rich graph libraries with excellent documentation and test coverage.

Algorithm Coverage: ~98% feature parity Quality: Both production-ready (Gleam has complete Network Simplex) Documentation: Excellent in both Community: Active maintenance in both

Key Differentiators

Gleam Strengths:

• ✅ Community Detection Suite (10 algorithms)
• ✅ Complete Network Simplex (production-ready min cost flow)
• ✅ Edge contraction
• ✅ Battle-tested on BEAM VM
• ✅ Stable 5.0.0 release
• ✅ Custom optimized data structures

F# Strengths:

• ✅ GraphML/GDF export formats
• ✅ More examples (37 vs 25)
• ✅ .NET ecosystem integration

Recommended Use:

• Production min cost flow problems: Choose Gleam
• Gephi/yEd visualization workflows: Choose F# (or wait for Gleam support)
• Platform-specific: BEAM → Gleam, .NET → F#

---

Last Updated: March 2025 Gleam Version: 5.0.0 F# Version: 0.5.0