https://github.com/bufferapp/kiner

Python AWS Kinesis Producer with error handling and thread support.
https://github.com/bufferapp/kiner

aws kinesis-producer

Last synced: 9 months ago
JSON representation

Python AWS Kinesis Producer with error handling and thread support.

• Host: GitHub
• URL: https://github.com/bufferapp/kiner
• Owner: bufferapp
• License: mit
• Created: 2017-07-20T15:32:29.000Z (over 8 years ago)
• Default Branch: main
• Last Pushed: 2022-10-05T23:12:32.000Z (over 3 years ago)
• Last Synced: 2025-04-15T06:55:25.793Z (9 months ago)
• Topics: aws, kinesis-producer
• Language: Python
• Homepage:
• Size: 37.1 KB
• Stars: 45
• Watchers: 5
• Forks: 20
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          


    





    A simple Python AWS Kinesis Producer.



     

[![Build Status](https://travis-ci.org/bufferapp/kiner.svg?branch=master)](https://travis-ci.org/bufferapp/kiner)

[![PyPI version](https://badge.fury.io/py/kiner.svg)](https://badge.fury.io/py/kiner)

[![License](https://img.shields.io/github/license/mashape/apistatus.svg)](LICENSE)

### Features

- Error handling and retrying with exponential backoff

- Automatic batching and flush callbacks

- Threaded execution

Inspired by the AWS blog post [Implementing Efficient and Reliable Producers with the Amazon Kinesis Producer Library](https://aws.amazon.com/blogs/big-data/implementing-efficient-and-reliable-producers-with-the-amazon-kinesis-producer-library/).

## Installation

You can use `pip` to install Kiner.

```bash

pip install kiner

```

## Usage

To use Kiner, you'll need to have AWS authentication credentials configured

as stated in the [`boto3` documentation](https://boto3.readthedocs.io/en/latest/guide/quickstart.html#configuration)

```python

from kiner.producer import KinesisProducer

p = KinesisProducer('stream-name', batch_size=500, max_retries=5, threads=10)

for i in range(10000):

    p.put_record(i)

p.close()

```

To be notified when data is flushed to AWS Kinesis, provide a flush_callback

```python

from uuid import uuid4

from kiner.producer import KinesisProducer

def on_flush(count, last_flushed_at, Data=b'', PartitionKey='', Metadata=()):

    print(f"""

        Flushed {count} messages at timestamp {last_flushed_at}

        Last message was {Metadata['id']} paritioned by {PartitionKey} ({len(Data)} bytes)

    """)

p = KinesisProducer('stream-name', flush_callback=on_flush)

for i in range(10000):

    p.put_record(i, metadata={'id': uuid4()}, partition_key=f"{i % 2}")

p.close()

```

## Contributions

- Logo design by [@area55git](https://github.com/area55git)