Redash Refresh API usage example with parameters Raw

refresh.py

import os
import requests
import time
from pprint import pprint


def poll_job(s, redash_url, job):
    # TODO: add timeout
    while job['status'] not in (3,4):
        response = s.get('{}/api/jobs/{}'.format(redash_url, job['id']))
        job = response.json()['job']
        time.sleep(1)

    if job['status'] == 3:
        return job['query_result_id']
    
    return None

def get_fresh_query_result(redash_url, query_id, api_key, parameters):
    s = requests.Session()
    s.headers.update({'Authorization': 'Key {}'.format(api_key)})

    body = {
        "parameters": parameters
    }
    response = s.post('{}/api/queries/{}/results'.format(redash_url, query_id), json=body)

    if response.status_code != 200:
        raise Exception('Refresh failed.')

    result_id = poll_job(s, redash_url, response.json()['job'])

    if result_id:
        response = s.get('{}/api/queries/{}/results/{}.json'.format(redash_url, query_id, result_id))
        if response.status_code != 200:
            raise Exception('Failed getting results.')
    else:
        raise Exception('Query execution failed.')

    return response.json()['query_result']['data']['rows']

if __name__ == '__main__':
    params = {'param': 1243}
    query_id = 1234
    # Need to use a *user API key* here (and not a query API key).
    api_key = '...'
    pprint(get_fresh_query_result('https://app.redash.io/acme', query_id, api_key, params))

Well this works just fine with 2 year old installation of redash, which is rather unexpected for some random gist on github.
Thanks!

@czen this "random gist on github" was written by the founder of Redash 😀

Nowadays you can also use redash-toolbelt to do this and many other operations as well. It's the officially maintained Redash API Python client.

i wonder, would there happen to be a way to have the query results stored as an excel file / csv / tsv instead of a json? would certainly make it easier for mass consumption.

i wonder, would there happen to be a way to have the query results stored as an excel file / csv / tsv instead of a json? would certainly make it easier for mass consumption.

CSV, TSV, and Excel versions of the results are already available through the API.

Are there any plans to add the ability to retrieve results from a parameterized query with a query api token instead of a user api token? My understanding is that currently this is not possible (as denoted by line 45 in the sample code above).