Python SDK
The official CoordiNode Python SDK provides sync and async gRPC clients.
bash
pip install coordinodeQuick Start
python
from coordinode import CoordinodeClient
client = CoordinodeClient("localhost:7080")
# Execute a Cypher query
results = client.cypher("MATCH (n:Person) RETURN n.name LIMIT 10")
for row in results:
print(row["n.name"])
# Vector search
hits = client.vector_search(
label="Document",
property="embedding",
vector=[0.1, 0.2, ...], # 384-dim
top_k=5,
)Source Location Tracking
Enable debug_source_tracking in development to let the Query Advisor map slow-query suggestions to exact call sites in your application.
python
from coordinode import CoordinodeClient
client = CoordinodeClient(
"localhost:7080",
debug_source_tracking=True,
app_name="my-service",
app_version="1.2.3",
)
# Every query call automatically attaches the call site to the request.
# The server will record:
# x-source-file: "api/handlers/feed.py"
# x-source-line: "47"
# x-source-app: "my-service"
results = client.cypher("MATCH (u:User)-[:FOLLOWS]->(f) RETURN f LIMIT 10")Source tracking uses Python's inspect.stack() to read the immediate caller's file and line number. The overhead is negligible and is zero when debug_source_tracking=False.
Recommendation: Enable in development and staging; disable in production.
See Also
- TypeScript SDK — Node.js gRPC client
- LlamaIndex integration — PropertyGraphIndex with CoordiNode backend
- LangChain integration — GraphCypherQAChain
- GitHub: structured-world/coordinode-python
