"""Generate comprehensive knowledge graph statistics."""
import asyncio
import sys
from pathlib import Path
from collections import Counter
from datetime import datetime
sys.path.insert(0, str(Path(__file__).parent.parent))
from mcp_server.mcp_tools import query_decisions
async def get_graph_statistics():
print("="*60)
print("FAULKNER DB - KNOWLEDGE GRAPH STATISTICS")
print("="*60)
# Query all nodes
print("\nš Querying knowledge base...")
nodes = await query_decisions(query="")
print(f"ā
Loaded {len(nodes)} total nodes\n")
# Node type breakdown
node_types = Counter(node.get('type', 'unknown') for node in nodes)
print("š Node Types:")
for ntype, count in node_types.most_common():
print(f" {ntype}: {count}")
# Relationship statistics
total_edges = 0
nodes_with_edges = 0
relationship_types = Counter()
for node in nodes:
edges = node.get('edges', [])
if edges:
nodes_with_edges += 1
total_edges += len(edges)
for edge in edges:
relationship_types[edge.get('type', 'unknown')] += 1
connectivity_pct = (nodes_with_edges / len(nodes) * 100) if nodes else 0
print(f"\nš Relationships:")
print(f" Total edges: {total_edges:,}")
print(f" Connected nodes: {nodes_with_edges:,} ({connectivity_pct:.1f}%)")
print(f" Isolated nodes: {len(nodes) - nodes_with_edges:,}")
print(f" Avg edges/node: {total_edges / len(nodes):.1f}" if nodes else " N/A")
print(f"\nš Relationship Types:")
for rtype, count in relationship_types.most_common():
print(f" {rtype}: {count:,}")
# Temporal analysis (if timestamps available)
dated_nodes = [n for n in nodes if n.get('timestamp')]
if dated_nodes:
dates = [datetime.fromisoformat(n['timestamp'].replace('Z', '+00:00')) for n in dated_nodes]
earliest = min(dates)
latest = max(dates)
print(f"\nš
Temporal Range:")
print(f" Earliest: {earliest.strftime('%Y-%m-%d')}")
print(f" Latest: {latest.strftime('%Y-%m-%d')}")
print(f" Span: {(latest - earliest).days} days")
# Top keywords (if available)
all_keywords = []
for node in nodes:
keywords = node.get('keywords', [])
all_keywords.extend(keywords)
if all_keywords:
keyword_counts = Counter(all_keywords)
print(f"\nš·ļø Top Keywords:")
for keyword, count in keyword_counts.most_common(10):
print(f" {keyword}: {count}")
# Quality metrics
nodes_with_rationale = sum(1 for n in nodes if n.get('rationale'))
nodes_with_alternatives = sum(1 for n in nodes if n.get('alternatives'))
print(f"\nāØ Data Quality:")
print(f" With rationale: {nodes_with_rationale}/{len(nodes)} ({nodes_with_rationale/len(nodes)*100:.1f}%)")
print(f" With alternatives: {nodes_with_alternatives}/{len(nodes)} ({nodes_with_alternatives/len(nodes)*100:.1f}%)")
print("\n" + "="*60)
print("STATISTICS COMPLETE")
print("="*60)
if __name__ == "__main__":
asyncio.run(get_graph_statistics())
