obm_confluent_blog VS Protobuf

So what is actually the Schema Registry? And how does it help to make sense of binary data? In essence Schema Registry is an application with some Rest endpoints, from which schema's can be registered and retrieved. It used to only support Apache Avro. Later support for Protobuf and JSON Schema was added. Part of the same Github project, and what makes Schema Registry easy to use, is a Collection of Java classes that are used to go from bytes to typed objects and vice versa. There are several classes that support Kafka Streams and ksqlDB next to the more low level Kafka [Producer(https://kafka.apache.org/documentation/#producerapi) and Consumer clients. There are more advanced use cases, but basically you supply the url for the Schema Registry, and the library will handle the rest. For producing data this will mean optionally register a new schema too get the correct id. The consumer will use the encoded id to fetch the schema used to produce the data. It can also be used with other frameworks like Spring Cloud Stream. For example in the Kotlin Command Handler by using the SpecificAvroSerde class. You might need to set additional properties to get this working.

It's because they changed the versioning format: https://github.com/protocolbuffers/protobuf/releases?page=5

But I suppose old version still receive bugfixes.

For at least 4 years protobuf has had decent support for self-describing messages (very similar to avro) as well as reflection

https://github.com/protocolbuffers/protobuf/blob/main/src/go...

Xgooglers trying to make do on the cheap will just create a Union of all their messages and include the message def in a self-describing message pattern. Super-sensitive network I/O can elide the message def (empty buffer) and any for RecordIO clone well file compression takes care of the definition.

Definitely useful to be able to dig out old defs but protobuf maintainers have surprisingly added useful features so you don’t have to.

Bonus points tho for extracting the protobuf defs that e.g. Apple bakes into their binaries.

gRPC Gateway is a protoc plugin that reads gRPC service definitions and generates a reverse proxy server that translates a RESTful JSON API into gRPC.

To use more recent versions of protoc in future applications, you can download them from the Protobuf repository.

Use the Protobuf CLI protoc and the plugin protoc-gen-tonic.

> didn’t find any standard for separating protobuf messages

The fact that protobufs are not self-delimiting is an endless source of frustration, but I know of 2 standards:

- SerializeDelimited* is part of the protobuf library: https://github.com/protocolbuffers/protobuf/blob/main/src/go...

- Riegeli is "a file format for storing a sequence of string records, typically serialized protocol buffers. It supports dense compression, fast decoding, seeking, detection and optional skipping of data corruption, filtering of proto message fields for even faster decoding, and parallel encoding": https://github.com/google/riegeli

It looks like it is in fact universal. Just glancing at the code here, it looks like the tool searches any arbitrary file for bytes that look like encoded protobuf descriptors, specifically looking for bytes that are plausibly the beginning of a FileDescriptorProto message defined here:

https://github.com/protocolbuffers/protobuf/blob/main/src/go...

This takes advantage of the fact that such descriptors are commonly compiled into programs that use protobuf. The descriptors are usually embedded as constant byte arrays. That said, not all protobuf implementations embed the descriptors and those that do often have an option to inhibit such embedding (at the expense of losing some dynamic introspection features).

Not much. You can check the source code in https://github.com/protocolbuffers/protobuf. For example, for serializing a boolean in C#: https://github.com/protocolbuffers/protobuf/blob/main/csharp/src/Google.Protobuf/WritingPrimitives.cs#L165. Strings and objects are a bit more complicated, but it is all about turning the data into its byte representation.