Protovalidate is the semantic validation library for Protobuf. It provides standard annotations to validate common rules on messages and fields, as well as the ability to use CEL to write custom rules. It's the next generation of protoc-gen-validate.
With Protovalidate, you can annotate your Protobuf messages with both standard and custom validation rules:
syntax="proto3";
packageacme.user.v1;
import"buf/validate/validate.proto";
messageUser {
stringid=1 [(buf.validate.field).string.uuid = true];
uint32age=2 [(buf.validate.field).uint32.lte = 150]; // We can only hope.stringemail=3 [(buf.validate.field).string.email = true];
stringfirst_name=4 [(buf.validate.field).string.max_len = 64];
stringlast_name=5 [(buf.validate.field).string.max_len = 64];
option(buf.validate.message).cel= {
id: "first_name_requires_last_name"message: "last_name must be present if first_name is present"expression: "!has(this.first_name) || has(this.last_name)"
};
}Once you've added protovalidate to your project, validation is idiomatic Python:
try:
protovalidate.validate(message)
exceptprotovalidate.ValidationErrorase:
# Handle failure.Tip
The easiest way to get started with Protovalidate for RPC APIs are the quickstarts in Buf's documentation. There's one available for Python and gRPC.
To install the package, use pip:
pip install protovalidateComprehensive documentation for Protovalidate is available at protovalidate.com.
Highlights for Python developers include:
- The developer quickstart
- A comprehensive RPC quickstart for Python and gRPC
- A migration guide for protoc-gen-validate users
Protovalidate isn't just for Python! You might be interested in sibling repositories for other languages:
protovalidate-go(Go)protovalidate-java(Java)protovalidate-cc(C++)protovalidate-es(TypeScript and JavaScript)
Additionally, protovalidate's core repository provides:
- Protovalidate's Protobuf API
- Conformance testing utilities for acceptance testing of
protovalidateimplementations
We include a set of benchmarks to verify different types of validation operations based on pytest-benchmark.
You can run them with
$ uv run poe benchThis persists the runs results. It is often helpful to compare benchmark results between runs, for example before and after an optimization. To compare persisted results, use
$ uv run poe bench-compareWe genuinely appreciate any help! If you'd like to contribute, check out these resources:
- Contributing Guidelines: Guidelines to make your contribution process straightforward and meaningful
- Conformance testing utilities: Utilities providing acceptance testing of
protovalidateimplementations
Offered under the Apache 2 license.