#!/usr/bin/env python3

"""

Query script for kg-3 (llm-graph-builder) Knowledge Graph.

Connects to Neo4j via Bolt driver. Runs analytics and traversal queries.

"""

import os

from pathlib import Path

from neo4j import GraphDatabase

# Load .env from project dir or workspace root

for env_path in [Path(__file__).parent / ".env", Path("/workspace/.env")]:

if env_path.exists():

for line in env_path.read_text().splitlines():

line = line.strip()

if line and not line.startswith("#") and "=" in line:

k, v = line.split("=", 1)

os.environ.setdefault(k.strip(), v.strip())

NEO4J_URI = os.environ.get("NEO4J_URI", "bolt://host.docker.internal:7690")

NEO4J_USER = os.environ.get("NEO4J_USERNAME", "neo4j")

NEO4J_PASS = os.environ.get("NEO4J_PASSWORD", "kg-eval-password")

def run_query(session, title, cypher):

"""Run a query and print results."""

print(f"\n{'='*60}")

print(f" {title}")

print(f"{'='*60}")

try:

result = session.run(cypher)

records = list(result)

if not records:

print(" (no results)")

return

keys = records[0].keys()

# Column widths

widths = {k: len(k) for k in keys}

rows = []

for rec in records:

row = {k: str(rec[k])[:50] for k in keys}

for k in keys:

widths[k] = max(widths[k], len(row[k]))

rows.append(row)

# Print

header = " | ".join(k.ljust(widths[k]) for k in keys)

sep = "-+-".join("-" * widths[k] for k in keys)

print(f" {header}")

print(f" {sep}")

for row in rows:

print(f" {' | '.join(row[k].ljust(widths[k]) for k in keys)}")

except Exception as e:

print(f" ERROR: {e}")

def main():

driver = GraphDatabase.driver(NEO4J_URI, auth=(NEO4J_USER, NEO4J_PASS))

with driver.session() as s:

run_query(s, "GRAPH OVERVIEW",

"MATCH (n) WITH labels(n)[0] AS label, count(*) AS cnt "

"RETURN label, cnt ORDER BY cnt DESC")

run_query(s, "RELATIONSHIP COUNTS",

"MATCH ()-[r]->() RETURN type(r) AS rel_type, count(r) AS cnt "

"ORDER BY cnt DESC")

run_query(s, "TOP PEOPLE",

"MATCH (p:Person)-[r]-() "

"RETURN p.id AS person, count(r) AS connections "

"ORDER BY connections DESC LIMIT 15")

run_query(s, "CLIENTS",

"MATCH (c:Client) OPTIONAL MATCH (c)<-[r]-() "

"RETURN c.id AS client, count(r) AS rels "

"ORDER BY rels DESC LIMIT 15")

run_query(s, "ACTION ITEMS WITH ASSIGNEES",

"MATCH (a:ActionItem)-[:ASSIGNED_TO]->(p:Person) "

"RETURN a.id AS action_item, p.id AS assigned_to LIMIT 15")

run_query(s, "TOPICS DISCUSSED IN MEETINGS",

"MATCH (m:Meeting)-[:DISCUSSED]->(t:Topic) "

"RETURN m.id AS meeting, t.id AS topic LIMIT 15")

run_query(s, "MULTI-HOP: Person -> Meeting -> Topic",

"MATCH (p:Person)-[:ATTENDED]->(m:Meeting)-[:DISCUSSED]->(t:Topic) "

"RETURN p.id AS person, m.id AS meeting, t.id AS topic LIMIT 10")

run_query(s, "DOCUMENTS AND ENTITY COUNTS",

"MATCH (d:Document)-[:MENTIONS]->(e) "

"RETURN d.id AS document, count(e) AS entities "

"ORDER BY entities DESC LIMIT 10")

driver.close()

print(f"\n{'='*60}")

print(" QUERY COMPLETE")

print(f"{'='*60}\n")

if __name__ == "__main__":

main()