pyokera 1.3.0.dev9

Python client for the Okera Data Access Service

Python client for RecordService implementations.

Dependencies

Required:

• Python 3.4+

• six, bit_array, thriftpy2==0.3.12, urllib3

pip3 install six bit_array thriftpy2==0.3.12 urllib3

Optional:

• pandas for conversion to DataFrame objects

Installation

pip3 install pyokera

To verify:

>>> import okera.odas
>>> okera.odas.version()
'##OKERA_RELEASE_VERSION##'

Usage

from okera import context
ctx = context()
with ctx.connect(host='localhost', port=12050) as conn:
    conn.list_databases()
    pd = conn.scan_as_pandas("okera_sample.sample")
    pd

To enable a connection to a server with token-authentication:

from okera import context
ctx = context()
ctx.enable_token_auth(token_str='my-token')
with ctx.connect(host='localhost', port=12050) as conn:
    conn.list_databases()

To enable a connection to a server with kerberos-authentication:

from okera import context
ctx = context()
# Connecting to server principal 'cerebro/service@REALM'
ctx.enable_kerberos('cerebro', host_override='service')
with ctx.connect(host='localhost', port=12050) as conn:
    conn.list_databases()