Connect FilePulse is a multipurpose, scalable and reliable, Kafka Connector that makes it easy to parse, transform and stream any file, in any format, into Apache Kafka™. It provides capabilities for reading files from: local-filesystem, Amazon S3, Azure Storage and Google Cloud Storage.
In organizations, data is frequently exported, shared and integrated from legacy systems through the use of files in a wide variety of formats (e.g. CSV, XML, JSON, Avro, etc.). Dealing with all of these formats can quickly become a real challenge for enterprise that usually end up with a complex and hard to maintain data integration mess.
A modern approach consists in building a scalable data streaming platform as a central nervous system to decouple applications from each other. Apache Kafka™ is one of the most widely used technologies to build such a system. The Apache Kafka project packs with Kafka Connect a distributed, fault tolerant and scalable framework for connecting Kafka with external systems.
The Connect File Pulse project aims to provide an easy-to-use solution, based on Kafka Connect, for streaming any type of data file with the Apache Kafka™ platform.
Some of the features of Connect File Pulse are inspired by the ingestion capabilities of Elasticsearch and Logstash.
Connect FilePulse provides a set of built-in features for streaming files from multiple filesystems into Kafka. This includes, among other things:
- Support for recursive scanning of local directories.
- Support for reading files from Amazon S3, Azure Storage and Google Cloud Storage.
- Support multiple input file formats (e.g: CSV, JSON, AVRO, XML).
- Support for Grok expressions.
- Parsing and transforming data using built-in or custom processing filters.
- Error handler definition
- Monitoring files while they are being written into Kafka
- Support pluggable strategies to clean up completed files
- Etc.
You think this project can help you or your team to ingest data into Kafka ? Please 🌟 this repository to support us!
The best way to learn Kafka Connect File Pulse is to follow the step by step Getting Started.
If you want to read about using Connect File Pulse, the full documentation can be found here
File Pulse is also available on Docker Hub 🐳
https://hub.docker.com/r/streamthoughts/kafka-connect-file-pulse:latestAny feedback, bug reports and PRs are greatly appreciated! See our guideline.
-
Source Code: https://github.com/streamthoughts/kafka-connect-file-pulse
-
Issue Tracker: https://github.com/streamthoughts/kafka-connect-file-pulse/issues
-
Documentation: https://streamthoughts.github.io/kafka-connect-file-pulse/
-
Releases: https://github.com/streamthoughts/kafka-connect-file-pulse/releases
This code base is available under the Apache License, version 2.