Skip to main content

Developing a query module in Python


Developing a module

In this guide, we'll create a random walk algorithm. Position yourself in the MAGE repository you cloned earlier. Specifically, go in the python sub-directory and create a new file called Next, import mgp, Memgraph's internal data structure module. Among others, it contains definitions for Vertex and Node data structures.


Install the mgp Python module so your editor can use typing annotations properly and suggest methods and classes it contains. You can install the module by running pip install mgp.

Here's the code for the random walk algorithm:

import mgp
import random

def get(
start: mgp.Vertex,
length: int = 10,
) -> mgp.Record(path=mgp.Path):
"""Generates a random path of length `length` or less starting
from the `start` vertex.

:param mgp.Vertex start: The starting node of the walk.
:param int length: The number of edges to traverse.
:return: Random path.
:rtype: mgp.Record(mgp.Path)
path = mgp.Path(start)
vertex = start
for _ in range(length):
edge = random.choice(list(vertex.out_edges))
vertex = edge.to_vertex
except IndexError:

return mgp.Record(path=path)

The get_path is decorated with the @mgp.read_proc decorator, which tells Memgraph it's a read procedure, meaning it won't change the graph. The path is created from the start node, and edges are appended to it iteratively.


Start Memgraph and MAGE, and copy the module you developed into the /usr/lib/memgraph/query_modules.

Instructions for a local Memgraph installation (Debian/Ubuntu):

sudo systemctl start memgraph
cp python/ /usr/lib/memgraph/query_modules/

Instructions for a docker Memgraph instance:

docker run --rm -d --name memgraph -p 7687:7687 memgraph/mage
docker cp python/ memgraph:/usr/lib/memgraph/query_modules/

A more advanced approach is to use docker volumes. That will allow you to have query modules synchronized between your mage repository and your docker container.

Then, use a Memgraph client like MemgraphLab or mgconsole to load the newly added function.

CALL mg.load('random_walk')

Lastly, run a query and test your module:

MERGE (start:Node {id: 0})-[:RELATION]->(mid:Node {id: 1})-[:RELATION]->(end:Node {id: 2})
CALL random_walk.get(start, 2) YIELD path


Test decoupled parts of your code that don't depend on Memgraph like you would in any other setting. E2e (end to end) tests, on the other hand, depend on internal Memgraph data structures, like nodes and edges. After running Memgraph, we need to prepare the testing environment on the host machine. Position yourself in the mage directory you cloned from GitHub. The expected folder structure for each module is the following:

└── e2e
└── random_walk_test
└── test_base
β”œβ”€β”€ input.cyp
└── test.yml

input.cyp represents a Cypher script for entering the data into the database. To simplify this tutorial, we'll leave the database empty. test.yml specifies which test query should be run by the database, and what should be the result or exception. Create the files following the aforementioned directory structure.




query: >
MATCH (start:Node {id: 0})
CALL random_walk.get(start, 2) YIELD path

output: []

Lastly, run the e2e tests with python:

python test_e2e