Home

Awesome

libserdes

Copyright(C) 2015-2022 Confluent Inc.

libserdes is a schema-based serializer/deserializer C/C++ library with support for Avro and the Confluent Platform Schema Registry.

The library is aimed to be used in the streaming pipeline, e.g. Apache Kafka, to perform data serialization and deserialization with centrally managed schemas.

On the producer side the application only needs to provide the centrally managed schema name or id and libserdes will fetch and load the schema from the remote schema registry. The loaded schema will be used for producer side verification and serialization of Avro objects.

On the consumer side libserdes will automatically extract the schema id of consumed serialized data and fetch the required schemas from the remote schema registry to perform verification and deserialization back to an Avro object.

This functionality is designed to work in close integration with the Kafka consumer or producer, see the examples directory for integrations with the librdkafka client library.

The serialization defaults to Avro but is completely pluggable through the configuration object interface, allowing applications to use libserdes for any serialization type.

libserdes is licensed under Apache License 2.0

Requirements

On Debian/Ubuntu based systems: sudo apt-get install libjansson-dev libcurl-gnutls-dev

NOTE: The avro libraries needs to be installed manually or through Confluent's APT and YUM repositories

Build

./configure
make
sudo make install

Documentation

Full API documentation is available in serdes.h and serdescpp.h

Configuration

libserdes typically only needs to be configured with a list of URLs to the schema registries, all other configuration is optional.