ksqldb 1.0.1

A simple Python wrapper for the KSQLdb REST API

KSQLdb for Python

Project Overview

A Python wrapper for the KSQLdb REST API

Installation

pip install ksqldb

Client configuration

from ksqldb import KSQLdbClient

client = KSQLdbClient('http://localhost:8088')

It can also be configured with Basic authentication:

from ksqldb import KSQLdbClient

client = KSQLdbClient('http://localhost:8088', api_key="<KEY>", api_secret="<SECRET>")

Usage

client.get_properties()

Returns KSQLdb properties

client.get_properties()

client.ksql(ksql_string, stream_properties=None)

For a full list of statement options make sure to check KSQLdb API: /ksql endpoint

Examples of executing KSQL statements:

client.ksql("show topics;")
client.ksql("show streams;")
client.ksql("show tables;")
client.ksql("describe <STREAM_NAME> extended;")

client.query_sync(query_string, stream_properties=None, timeout=10)

Runs a query synchronously. Can't be used with EMIT CHANGES queries.

Examples of executing KSQL queries:

client.query_sync("select * from STREAM_NAME;")

# To get data from beginning of stream use: 
client.query_sync("select * from STREAM_NAME;", stream_properties={"ksql.streams.auto.offset.reset": "earliest"})

client.query_async(query_string, stream_properties=None, timeout=10)

Runs a query asynchronously.

To test this in python shell can use python -m asyncio Examples of executing KSQL async queries:

async for x in client.query_async("select * from STREAM_NAME emit changes;", timeout=None):
    print(x)
    
# To get data from beginning of stream use: 
async for x in client.query_async("select * from STREAM_NAME emit changes;", stream_properties={"ksql.streams.auto.offset.reset": "earliest"}, timeout=None):
    print(x)

close_query(query_id)

Usually you don't need to close a sync query, but should be done for async ones.

client.close_query("QUERY_ID")

inserts_stream(stream_name, rows)

Inserts data into a stream.

rows = [
    {
        "col1" : "val1",
        "col2": 2.3,
        "col3": True
    },
    {
        "col1" : "val1",
        "col2": 2.3,
        "col3": True
    },
]
client.inserts_stream("STREAM_NAME", rows)