Using Python to query GraphQL

Using Python to query GraphQL

Situation: We need to query GraphQL, in JSON format, to convert to dataframe.

There are several ways to go about this. The simplest way is to use the requests library to make HTTP requests to the API endpoint, then the json library to convert those requests into JSON format:

import requests
import json
import pandas as pd

def run_query(q):
    request = requests.post('https://api-endpoint'
                            '',
                            json={'query': query})
    if request.status_code == 200:
        return request.json()
    else:
        raise Exception('Query failed. return code is {}.     {}'.format(
            request.status_code, query))

# basic query first
query = """
{
    transferBanks(first: 1000, orderBy: timestamp, orderDirection: asc, subgraphError: allow) {
    id
    from_address
    to_address
    amount
    amount_display
    timestamp
    timestamp_display
  }
}
"""

# returns JSON
result = run_query(query)

An alternative way is to use the gql and the gql.transport.aiohttp libraries:

from gql import gql, Client
from gql.transport.aiohttp import AIOHTTPTransport

# Select transport with url endpoint
transport = AIOHTTPTransport(
    url="https://api.studio.thegraph.com/query/1121/bankv1/v0.0.5")

# create GraphQL client using defined transport
client = Client(transport=transport, fetch_schema_from_transport=True)

# GraphQL query
query = gql("""
{
    transferBanks(first: 1000, where: {timestamp_gte: "1635403557"}, orderBy: timestamp, orderDirection: asc, subgraphError: allow) {
    id
    from_address
    to_address
    amount
    amount_display
    timestamp
    timestamp_display
    }
}
""")

# run query on transport
result = client.execute(query)
print(result)

For more content on data science, R, and Python find me on Twitter.