Py2Neo - how to get around iterating the record object returned from cypher queries to access the query data

Question

I use py2neo to connect to Neo4j.

E.g I have a query that gives this result (I can actually be anything: numbers, strings....)

number      |   first name   | last name
--------------------------------------
1           |   Anne         | Hathaway
2           |   Tom          | Cruise
3      ....

Whenever I execute the cypher query from python (using the graph.cypher.execute or graph.cypher.stream commands), I get a Record object returned and I need to iterate through the record object to access the individual elements to access them and store them in for example a numpy matrix.

Is there a way to have the result from a cypher query stored immediately in a list,numpy matrix,.... in python?

Basically: how can I skip the record object that gets returned and avoid unnecessary manipulations to prepare the data for further calculations?

Answer 1

What you get back from cypher.execute() is a RecordList containing Record objects.

Both RecordList and Record behave like iterables, i.e. you can iterate with for x in recordlist through all records and with for x in record through all return values of a single record.

ResultList has an attribute records which returns the list of Record objects. You can pass this to any constructor that accepts a list of lists, such as Pandas DataFrame or numpy matrix:

import py2neo
import pandas as pd
import numpy as np

result = graph.cypher.execute(myquery)

# Pandas DataFrame
df = pd.DataFrame(result.records, columns=result.columns)

# numpy matrix
matrix = np.matrix(result.records)

This works of course only if you return data types which can be stored in a DataFrame or matrix.