freesixty

Freesixty

A simple Google Analytics API data extraction.

Installation

pip install freesixty

Access credentials

To set up access to your Google Analytics follow first step of these instructions. Store them in your local machine and enter their path into KEY_FILE_LOCATION variable.

Get data

import freesixty

KEY_FILE_LOCATION = './client_secrets.json'
VIEW_ID = 'XXXXXXX'

query = {
    'reportRequests': [
    {
        'viewId': VIEW_ID,
        'dateRanges': [{'startDate': '2009-01-01', 'endDate': '2019-01-05'}],
        'metrics': [{'expression': 'ga:sessions'}],
        'dimensions': [{'name': 'ga:country', 'name': 'ga:date'}]
    }]
}

analytics = freesixty.initialize_analyticsreporting(KEY_FILE_LOCATION)
result, is_data_golden = freesixty.execute_query(analytics, query)

On the other hand if we want to store resulting data to a desired URI.

import freesixty

KEY_FILE_LOCATION = './client_secrets.json'
VIEW_ID = 'XXXXXXX'
folder_uri = 'file:///tmp/example/folder'

query = {
    'reportRequests': [
    {
        'viewId': VIEW_ID,
        'dateRanges': [{'startDate': '2009-01-01', 'endDate': '2019-01-05'}],
        'metrics': [{'expression': 'ga:sessions'}],
        'dimensions': [{'name': 'ga:country', 'name': 'ga:date'}]
    }]
}

analytics = freesixty.initialize_analyticsreporting(KEY_FILE_LOCATION)
freesixty.store_query(analytics, query, folder_uri)

Getting more data

In case a query would return over 100k rows of data it will fail. We can get around it by splitting the date range into smaller chunks:

queries = freesixty.split_query(query=query, start_date='2019-01-01', end_date='2019-02-01', freq='D')

for query in queries:
    freesixty.store_query(analytics, query, folder_uri)

Useful links

• Try out queries
• Compose queries

TODO:

• More complete tests

:cake: