ENH: add google bigquery interface

[linwoodc3]

Use [pandas.io.gbq](http://pandas.pydata.org/pandas-docs/stable/generated/pandas.read_gbq.html#pandas.read_gbq)

First you will need to pip install: pip install --upgrade google-api-python-client
Here is a working query:

# load keys;  requires you to be registered
keys = json.load(open('/Users/linwood/Desktop/keysforapps/apikeys.txt'))

# setup google creds; not sure if this is required yet; but you need to do it once to authorize the api from your python ecosystem
from apiclient.discovery import build
service = build('bigquery', 'v2', developerKey=keys['google']['apikey']+"2")

# load query in proper SQL syntax as string
from pandas.io import gbq
q="""
SELECT MonthYear,count(*)c,count(IF(Actor1Code LIKE 'MUS',1,null)) c_up
FROM [gdelt-bq.full.events] WHERE EventRootCode = '19'
GROUP BY MonthYear ORDER BY MonthYear;"""


# run the query
df = gbq.read_gbq(q, project_id=<projectid>)
:[out]
Requesting query... ok.
Query running...
Query done.
Cache hit.

Retrieving results...
Got 461 rows.

Total time taken 0.75 s.
Finished at 2017-05-30 10:26:21