ruvector/crates/ruvector-postgres/sql/graph_examples.sql
rUv 84f8b685c1 feat(postgres): Add 53 SQL function definitions for all advanced modules (#46)
* feat(postgres): Add 7 advanced AI modules to ruvector-postgres

Comprehensive implementation of advanced AI capabilities:

## New Modules (23,541 lines of code)

### 1. Self-Learning / ReasoningBank (`src/learning/`)
- Trajectory tracking for query optimization
- Pattern extraction using K-means clustering
- ReasoningBank for pattern storage and matching
- Adaptive search parameter optimization

### 2. Attention Mechanisms (`src/attention/`)
- Scaled dot-product attention (core)
- Multi-head attention with parallel heads
- Flash Attention v2 (memory-efficient)
- 10 attention types with PostgresEnum support

### 3. GNN Layers (`src/gnn/`)
- Message passing framework
- GCN (Graph Convolutional Network)
- GraphSAGE with mean/max aggregation
- Configurable aggregation methods

### 4. Hyperbolic Embeddings (`src/hyperbolic/`)
- Poincaré ball model
- Lorentz hyperboloid model
- Hyperbolic distance metrics
- Möbius operations

### 5. Sparse Vectors (`src/sparse/`)
- COO format sparse vector type
- Efficient sparse-sparse distance functions
- BM25/SPLADE compatible
- Top-k pruning operations

### 6. Graph Operations & Cypher (`src/graph/`)
- Property graph storage (nodes/edges)
- BFS, DFS, Dijkstra traversal
- Cypher query parser (AST-based)
- Query executor with pattern matching

### 7. Tiny Dancer Routing (`src/routing/`)
- FastGRNN neural network
- Agent registry with capabilities
- Multi-objective routing optimization
- Cost/latency/quality balancing

## Docker Infrastructure
- Dockerfile with pgrx 0.12.6 and PostgreSQL 16
- docker-compose.yml with test runner
- Initialization SQL with test tables
- Shell scripts for dev/test/benchmark

## Feature Flags
- `learning`, `attention`, `gnn`, `hyperbolic`
- `sparse`, `graph`, `routing`
- `ai-complete` and `graph-complete` bundles

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* fix(docker): Copy entire workspace for pgrx build

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* fix(docker): Build standalone crate without workspace

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* docs: Update README to enhance clarity and structure

* fix(postgres): Resolve compilation errors and Docker build issues

- Fix simsimd Option/Result type mismatch in scaled_dot.rs
- Fix f32/f64 type conversions in poincare.rs and lorentz.rs
- Fix AVX512 missing wrapper functions by using AVX2 fallback
- Fix Vec<Vec<f32>> to JsonB for pgrx pg_extern compatibility
- Fix DashMap get() to get_mut() for mutable access
- Fix router.rs dereference for best_score comparison
- Update Dockerfile to copy pre-written SQL file for pgrx
- Simplify init.sql to use correct function names
- Add postgres-cli npm package for CLI tooling

All changes tested successfully in Docker with:
- Extension loads with AVX2 SIMD support (8 floats/op)
- Distance functions verified working
- PostgreSQL 16 container runs successfully

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* feat: Add ruvLLM examples and enhanced postgres-cli

Added from claude/ruvector-lfm2-llm-01YS5Tc7i64PyYCLecT9L1dN branch:
- examples/ruvLLM: Complete LLM inference system with SIMD optimization
  - Pretraining, benchmarking, and optimization system
  - Real SIMD-optimized CPU inference engine
  - Comprehensive SOTA benchmark suite
  - Attention mechanisms, memory management, router

Enhanced postgres-cli with full ruvector-postgres integration:
- Sparse vector operations (BM25, top-k, prune, conversions)
- Hyperbolic geometry (Poincare, Lorentz, Mobius operations)
- Agent routing (Tiny Dancer system)
- Vector quantization (binary, scalar, product)
- Enhanced graph and learning commands

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* fix(postgres-cli): Use native ruvector type instead of pgvector

- Change createVectorTable to use ruvector type (native RuVector extension)
- Add dimensions column for metadata since ruvector is variable-length
- Update index creation to use simple btree (HNSW/IVFFlat TBD)
- Tested against Docker container with ruvector extension

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* feat(postgres): Add 53 SQL function definitions for all advanced modules

Enable all advanced PostgreSQL extension functions by adding their SQL
definitions to the extension file. This exposes all Rust #[pg_extern]
functions to PostgreSQL.

## New SQL Functions (53 total)

### Hyperbolic Geometry (8 functions)
- ruvector_poincare_distance, ruvector_lorentz_distance
- ruvector_mobius_add, ruvector_exp_map, ruvector_log_map
- ruvector_poincare_to_lorentz, ruvector_lorentz_to_poincare
- ruvector_minkowski_dot

### Sparse Vectors (14 functions)
- ruvector_sparse_create, ruvector_sparse_from_dense
- ruvector_sparse_dot, ruvector_sparse_cosine, ruvector_sparse_l2_distance
- ruvector_sparse_add, ruvector_sparse_scale, ruvector_sparse_to_dense
- ruvector_sparse_nnz, ruvector_sparse_dim
- ruvector_bm25_score, ruvector_tf_idf, ruvector_sparse_normalize
- ruvector_sparse_topk

### GNN - Graph Neural Networks (5 functions)
- ruvector_gnn_gcn_layer, ruvector_gnn_graphsage_layer
- ruvector_gnn_gat_layer, ruvector_gnn_message_pass
- ruvector_gnn_aggregate

### Routing/Agents - "Tiny Dancer" (11 functions)
- ruvector_route_query, ruvector_route_with_context
- ruvector_calculate_agent_affinity, ruvector_select_best_agent
- ruvector_multi_agent_route, ruvector_create_agent_embedding
- ruvector_get_routing_stats, ruvector_register_agent
- ruvector_update_agent_performance, ruvector_adaptive_route
- ruvector_fastgrnn_forward

### Learning/ReasoningBank (7 functions)
- ruvector_record_trajectory, ruvector_get_verdict
- ruvector_distill_memory, ruvector_adaptive_search
- ruvector_learning_feedback, ruvector_get_learning_patterns
- ruvector_optimize_search_params

### Graph/Cypher (8 functions)
- ruvector_graph_create_node, ruvector_graph_create_edge
- ruvector_graph_get_neighbors, ruvector_graph_shortest_path
- ruvector_graph_pagerank, ruvector_cypher_query
- ruvector_graph_traverse, ruvector_graph_similarity_search

## CLI Updates
- Enabled hyperbolic geometry commands in postgres-cli
- Added vector distance and normalize commands
- Enhanced client with connection pooling and retry logic

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-12-02 22:49:29 -05:00

327 lines
11 KiB
SQL

-- Graph Operations Examples for ruvector-postgres
-- This file demonstrates the graph database capabilities
-- ============================================================================
-- Basic Graph Operations
-- ============================================================================
-- Create a new graph
SELECT ruvector_create_graph('social_network');
-- List all graphs
SELECT ruvector_list_graphs();
-- ============================================================================
-- Social Network Example
-- ============================================================================
-- Add users
SELECT ruvector_add_node(
'social_network',
ARRAY['Person'],
jsonb_build_object('name', 'Alice', 'age', 30, 'city', 'New York')
) AS alice_id;
SELECT ruvector_add_node(
'social_network',
ARRAY['Person'],
jsonb_build_object('name', 'Bob', 'age', 25, 'city', 'San Francisco')
) AS bob_id;
SELECT ruvector_add_node(
'social_network',
ARRAY['Person'],
jsonb_build_object('name', 'Charlie', 'age', 35, 'city', 'Boston')
) AS charlie_id;
SELECT ruvector_add_node(
'social_network',
ARRAY['Person'],
jsonb_build_object('name', 'Diana', 'age', 28, 'city', 'Seattle')
) AS diana_id;
-- Create friendships
SELECT ruvector_add_edge(
'social_network',
1, 2, -- Alice -> Bob
'FRIENDS',
jsonb_build_object('since', '2020-01-15', 'strength', 0.9)
);
SELECT ruvector_add_edge(
'social_network',
2, 3, -- Bob -> Charlie
'FRIENDS',
jsonb_build_object('since', '2019-06-20', 'strength', 0.8)
);
SELECT ruvector_add_edge(
'social_network',
1, 4, -- Alice -> Diana
'FRIENDS',
jsonb_build_object('since', '2021-03-10', 'strength', 0.7)
);
SELECT ruvector_add_edge(
'social_network',
3, 4, -- Charlie -> Diana
'FRIENDS',
jsonb_build_object('since', '2020-09-05', 'strength', 0.85)
);
-- Get graph statistics
SELECT ruvector_graph_stats('social_network');
-- Find nodes by label
SELECT ruvector_find_nodes_by_label('social_network', 'Person');
-- Get neighbors of Alice (node 1)
SELECT ruvector_get_neighbors('social_network', 1);
-- Find shortest path from Alice to Charlie
SELECT ruvector_shortest_path('social_network', 1, 3, 10);
-- Find weighted shortest path
SELECT ruvector_shortest_path_weighted('social_network', 1, 3, 'strength');
-- ============================================================================
-- Cypher Query Examples
-- ============================================================================
-- Create nodes with Cypher
SELECT ruvector_cypher(
'social_network',
'CREATE (n:Person {name: ''Eve'', age: 27, city: ''Austin''}) RETURN n',
NULL
);
-- Match all persons
SELECT ruvector_cypher(
'social_network',
'MATCH (n:Person) RETURN n.name, n.age',
NULL
);
-- Match with WHERE clause
SELECT ruvector_cypher(
'social_network',
'MATCH (n:Person) WHERE n.age > 28 RETURN n.name, n.age',
NULL
);
-- Parameterized query
SELECT ruvector_cypher(
'social_network',
'MATCH (n:Person) WHERE n.name = $name RETURN n',
jsonb_build_object('name', 'Alice')
);
-- Create relationship with Cypher
SELECT ruvector_cypher(
'social_network',
'CREATE (a:Person {name: ''Frank''})-[:KNOWS {since: 2022}]->(b:Person {name: ''Grace''}) RETURN a, b',
NULL
);
-- ============================================================================
-- Knowledge Graph Example
-- ============================================================================
SELECT ruvector_create_graph('knowledge');
-- Add concepts
SELECT ruvector_cypher(
'knowledge',
'CREATE (ml:Concept {name: ''Machine Learning'', category: ''AI''})
CREATE (nn:Concept {name: ''Neural Networks'', category: ''AI''})
CREATE (dl:Concept {name: ''Deep Learning'', category: ''AI''})
CREATE (cv:Concept {name: ''Computer Vision'', category: ''AI''})
CREATE (nlp:Concept {name: ''Natural Language Processing'', category: ''AI''})
RETURN ml, nn, dl, cv, nlp',
NULL
);
-- Create relationships between concepts
WITH ids AS (
SELECT generate_series(1, 5) AS id
)
SELECT
CASE
WHEN i.id = 1 THEN ruvector_add_edge('knowledge', 1, 2, 'INCLUDES', '{"strength": 0.9}'::jsonb)
WHEN i.id = 2 THEN ruvector_add_edge('knowledge', 2, 3, 'SPECIALIZES_IN', '{"strength": 0.95}'::jsonb)
WHEN i.id = 3 THEN ruvector_add_edge('knowledge', 3, 4, 'APPLIES_TO', '{"strength": 0.85}'::jsonb)
WHEN i.id = 4 THEN ruvector_add_edge('knowledge', 3, 5, 'APPLIES_TO', '{"strength": 0.9}'::jsonb)
END AS edge_id
FROM ids i
WHERE i.id <= 4;
-- Find path from Machine Learning to Computer Vision
SELECT ruvector_shortest_path('knowledge', 1, 4, 10);
-- ============================================================================
-- Recommendation System Example
-- ============================================================================
SELECT ruvector_create_graph('recommendations');
-- Add users and movies
SELECT ruvector_cypher(
'recommendations',
'CREATE (u1:User {name: ''Alice'', preference: ''SciFi''})
CREATE (u2:User {name: ''Bob'', preference: ''Action''})
CREATE (u3:User {name: ''Charlie'', preference: ''SciFi''})
CREATE (m1:Movie {title: ''Inception'', genre: ''SciFi''})
CREATE (m2:Movie {title: ''Interstellar'', genre: ''SciFi''})
CREATE (m3:Movie {title: ''The Matrix'', genre: ''SciFi''})
CREATE (m4:Movie {title: ''Die Hard'', genre: ''Action''})
RETURN u1, u2, u3, m1, m2, m3, m4',
NULL
);
-- Create watch history
SELECT ruvector_add_edge('recommendations', 1, 4, 'WATCHED', '{"rating": 5, "timestamp": "2024-01-15"}'::jsonb);
SELECT ruvector_add_edge('recommendations', 1, 5, 'WATCHED', '{"rating": 4, "timestamp": "2024-01-20"}'::jsonb);
SELECT ruvector_add_edge('recommendations', 2, 7, 'WATCHED', '{"rating": 5, "timestamp": "2024-01-18"}'::jsonb);
SELECT ruvector_add_edge('recommendations', 3, 4, 'WATCHED', '{"rating": 5, "timestamp": "2024-01-22"}'::jsonb);
SELECT ruvector_add_edge('recommendations', 3, 6, 'WATCHED', '{"rating": 4, "timestamp": "2024-01-25"}'::jsonb);
-- Get statistics
SELECT ruvector_graph_stats('recommendations');
-- ============================================================================
-- Organizational Hierarchy Example
-- ============================================================================
SELECT ruvector_create_graph('org_chart');
-- Create organizational structure
SELECT ruvector_cypher(
'org_chart',
'CREATE (ceo:Employee {name: ''Jane Doe'', title: ''CEO'', level: 1})
CREATE (cto:Employee {name: ''John Smith'', title: ''CTO'', level: 2})
CREATE (cfo:Employee {name: ''Emily Brown'', title: ''CFO'', level: 2})
CREATE (dev1:Employee {name: ''Alex Johnson'', title: ''Senior Dev'', level: 3})
CREATE (dev2:Employee {name: ''Sarah Wilson'', title: ''Senior Dev'', level: 3})
CREATE (acc1:Employee {name: ''Michael Davis'', title: ''Accountant'', level: 3})
RETURN ceo, cto, cfo, dev1, dev2, acc1',
NULL
);
-- Create reporting structure
SELECT ruvector_add_edge('org_chart', 2, 1, 'REPORTS_TO', '{}'::jsonb);
SELECT ruvector_add_edge('org_chart', 3, 1, 'REPORTS_TO', '{}'::jsonb);
SELECT ruvector_add_edge('org_chart', 4, 2, 'REPORTS_TO', '{}'::jsonb);
SELECT ruvector_add_edge('org_chart', 5, 2, 'REPORTS_TO', '{}'::jsonb);
SELECT ruvector_add_edge('org_chart', 6, 3, 'REPORTS_TO', '{}'::jsonb);
-- Find all employees reporting to CTO (directly or indirectly)
SELECT ruvector_shortest_path('org_chart', 4, 1, 5); -- Path from dev1 to CEO
SELECT ruvector_shortest_path('org_chart', 5, 1, 5); -- Path from dev2 to CEO
-- ============================================================================
-- Transport Network Example
-- ============================================================================
SELECT ruvector_create_graph('transport');
-- Add cities as nodes
SELECT ruvector_add_node('transport', ARRAY['City'], '{"name": "New York", "population": 8336817}'::jsonb);
SELECT ruvector_add_node('transport', ARRAY['City'], '{"name": "Boston", "population": 692600}'::jsonb);
SELECT ruvector_add_node('transport', ARRAY['City'], '{"name": "Philadelphia", "population": 1584064}'::jsonb);
SELECT ruvector_add_node('transport', ARRAY['City'], '{"name": "Washington DC", "population": 705749}'::jsonb);
-- Add routes with distances
SELECT ruvector_add_edge('transport', 1, 2, 'ROUTE', '{"distance": 215, "mode": "train", "duration": 4.5}'::jsonb);
SELECT ruvector_add_edge('transport', 1, 3, 'ROUTE', '{"distance": 95, "mode": "train", "duration": 1.5}'::jsonb);
SELECT ruvector_add_edge('transport', 3, 4, 'ROUTE', '{"distance": 140, "mode": "train", "duration": 2.5}'::jsonb);
SELECT ruvector_add_edge('transport', 2, 3, 'ROUTE', '{"distance": 310, "mode": "train", "duration": 5.5}'::jsonb);
-- Find shortest route by distance
SELECT ruvector_shortest_path_weighted('transport', 2, 4, 'distance');
-- Find fastest route by duration
SELECT ruvector_shortest_path_weighted('transport', 2, 4, 'duration');
-- ============================================================================
-- Analytics Queries
-- ============================================================================
-- Get all graphs with their statistics
SELECT
name,
(ruvector_graph_stats(name)::jsonb)->>'node_count' AS nodes,
(ruvector_graph_stats(name)::jsonb)->>'edge_count' AS edges
FROM (
SELECT unnest(ruvector_list_graphs()) AS name
) graphs;
-- ============================================================================
-- Cleanup
-- ============================================================================
-- Delete specific graph
-- SELECT ruvector_delete_graph('social_network');
-- Delete all graphs
-- SELECT ruvector_delete_graph(name)
-- FROM unnest(ruvector_list_graphs()) AS name;
-- ============================================================================
-- Performance Testing
-- ============================================================================
-- Create a larger graph for performance testing
SELECT ruvector_create_graph('perf_test');
-- Generate random nodes
DO $$
DECLARE
i INTEGER;
BEGIN
FOR i IN 1..1000 LOOP
PERFORM ruvector_add_node(
'perf_test',
ARRAY['Node'],
jsonb_build_object('id', i, 'value', random() * 100)
);
END LOOP;
END $$;
-- Generate random edges
DO $$
DECLARE
i INTEGER;
source_id INTEGER;
target_id INTEGER;
BEGIN
FOR i IN 1..5000 LOOP
source_id := 1 + floor(random() * 1000)::INTEGER;
target_id := 1 + floor(random() * 1000)::INTEGER;
IF source_id <> target_id THEN
BEGIN
PERFORM ruvector_add_edge(
'perf_test',
source_id,
target_id,
'CONNECTS',
jsonb_build_object('weight', random())
);
EXCEPTION WHEN OTHERS THEN
-- Ignore errors (e.g., duplicate edges)
NULL;
END;
END IF;
END LOOP;
END $$;
-- Check performance stats
SELECT ruvector_graph_stats('perf_test');
-- Test path finding performance
\timing on
SELECT ruvector_shortest_path('perf_test', 1, 500, 20);
SELECT ruvector_shortest_path_weighted('perf_test', 1, 500, 'weight');
\timing off
-- Cleanup performance test
-- SELECT ruvector_delete_graph('perf_test');