Google2Pandas

Google2Pandas may eventually be a set of tools that will allow for easy querying of various google database products (Analytics, etc.) with the results returned as pandas.DataFrame objects (http://pandas.pydata.org/).

At this point, only queries to Google Analytics and the Multi-Channel Funnels Reporting via the core reporting API are supported.

Nomenclature

Suggested usage:

from google2pandas import *

Quick Setup

Install the latest version via pip:

sudo pip install Google2Pandas

or install the latest development version via:

sudo pip install git+https://github.com/panalysis/Google2Pandas

You will first need to enable the Analytics API, in particular you will need to follow Step 1 here.

Place the client_secrets_v3.json file in your dist-packages/google2pandas/ directory, and you're ready to go! Note that if this package has been installed system-wide (default), you will likely need to adjust the permissions/ownership of client_secrets_v3.json as well as the created analytics.dat token file. In particular, if you wish to create a system-wide token file (by default the class looks in /path/to/your/dist-packages/google2pandas/analytics.dat) you will likely need to instantiate the GoogleAnalyticsQuery class specifying a local location for the token file, and manually relocate it later.

Alternatively, store your credentials anywhere you like and simply pass a pointer to client_secrets_v3.json and analytics.dat when instantiating the class.

Quick Demo

from google2pandas import *

query = {\
    'ids'           : <valid_ids>,
    'metrics'       : 'pageviews',
    'dimensions'    : ['date', 'pagePath', 'browser'],
    'filters'       : ['pagePath=~iPhone', 'and', 'browser=~Firefox'],
    'start_date'    : '8daysAgo',
    'max_results'   : 10}

conn = GoogleAnalyticsQuery(
        token_file_name='my_analytics.dat',
	secrets='my_client_secrets_v3.json')
df, metadata = conn.execute_query(**query)

New and Improved (more of a work in progess really)

Support has now been added for the GA Reporting API V4 as suggested in issue #21 via the GoogleAnalyticsQueryV4 class. The support is rather rough for now, the primary reason being that since I'm not working with GA much at all these days I do not have the time to fully learn the features present in the new API.

For now, what this means is that there is zero parsing of the queries provided, it's down to the user to structure them correctly. As well, no guarantees are provided as to the ability to of the resp2frame method to convert the JSON object from GA to a pandas.DataFrame object in a manner that is generically robust. The as_dict keyword argument causes the restructuring step to be skipped; if you find room for improvements please do not hesitate to make a PR with your suggestions!

To use this module, one needs to follow the new setup process to enable acces. No more analytics.dat file, instead one needs to simply add the generated email address to the GA view you wish to access.

I also suggest naming the client_secrets file to something that indicates it is for the V4 API, as it is quite a different thing than the V3 version (default behaviour is to look for client_secrets_v4.json in dist-packages/google2pandas/).

Quick Demo

from google2pandas import *

query = {
    'reportRequests': [{
        'viewId' : <valid_ids>,

        'dateRanges': [{
            'startDate' : '8daysAgo',
            'endDate'   : 'today'}],

        'dimensions' : [
            {'name' : 'ga:date'}, 
            {'name' : 'ga:pagePath'},
            {'name' : 'ga:browser'}],

        'metrics'   : [
            {'expression' : 'ga:pageviews'}],

        'dimensionFilterClauses' : [{
            'operator' : 'AND',
            'filters'  : [
                {'dimensionName' : 'ga:browser',
                 'operator' : 'REGEXP',
                 'expressions' : ['Firefox']},

                {'dimensionName' : 'ga:pagePath',
                 'operator' : 'REGEXP',
                 'expressions' : ['iPhone']}]
        }]
    }]
}

# Assume we have placed our client_secrets_v4.json file in the current
# working directory.

conn = GoogleAnalyticsQueryV4(secrets='my_client_secrets_v4.json')
df = conn.execute_query(query)