FlatBuffers
26.6K ·GitHub
What is this?
FlatBuffers is a cross-platform serialization library built for maximum memory efficiency: you read serialized data directly without a parsing/unpacking step, and it keeps forwards/backwards compatibility.
FlatBuffers is a cross-platform serialization library built for maximum memory efficiency: you read serialized data directly without a parsing/unpacking step, and it keeps forwards/backwards compatibility.So is it a library I link into my code, or a separate tool?Both halves: runtime libraries you link into your app, plus the flatc compiler that takes your schema and generates language-specific accessors and builder code from a .fbs file.
Both halves: runtime libraries you link into your app, plus the flatc compiler that takes your schema and generates language-specific accessors and builder code from a .fbs file.What's the workflow to get my data into a FlatBuffer?Build flatc with cmake, define your schema in a .fbs file, run flatc to generate code for your language, serialize with FlatBufferBuilder, then read the buffer back with the generated accessors.
Build flatc with cmake, define your schema in a .fbs file, run flatc to generate code for your language, serialize with FlatBufferBuilder, then read the buffer back with the generated accessors.Which languages can I use it from?Code generation and runtimes for C, C++, C#, Dart, Go, Java, JavaScript, Kotlin, Lobster, Lua, PHP, Python, Rust, Swift, TypeScript and Nim, shipped via npm, PyPI, NuGet, Maven, crates.io, pub.dev and more.
Code generation and runtimes for C, C++, C#, Dart, Go, Java, JavaScript, Kotlin, Lobster, Lua, PHP, Python, Rust, Swift, TypeScript and Nim, shipped via npm, PyPI, NuGet, Maven, crates.io, pub.dev and more.Are there any gotchas around versioning when I add it as a dependency?Yes — releases are dated instead of following SemVer, so don't expect 1.2.3-style version numbers when pinning it in your package manager.
Yes — releases are dated instead of following SemVer, so don't expect 1.2.3-style version numbers when pinning it in your package manager.Does it work on Windows and macOS, or only Linux?All of them — Windows, macOS, Linux and Android are explicitly supported, plus any other OS with a recent C++ compiler (C++ 11 or newer).
All of them — Windows, macOS, Linux and Android are explicitly supported, plus any other OS with a recent C++ compiler (C++ 11 or newer).Is it free to use, even commercially?Yes, it's licensed under Apache License 2.0, a permissive open-source license with no paid tier.
Yes, it's licensed under Apache License 2.0, a permissive open-source license with no paid tier.Can I read a buffer written by another language, or an older schema version?Yes. The reader doesn't need the same language or schema version — FlatBuffers keeps data readable across languages and schema versions; the README shows Rust reading a buffer that C++ wrote.
Yes. The reader doesn't need the same language or schema version — FlatBuffers keeps data readable across languages and schema versions; the README shows Rust reading a buffer that C++ wrote.Why pick it over a format I have to parse, like JSON?The core pitch: you access serialized data directly without parsing or unpacking it first, which is where the memory efficiency comes from, and cross-version compatibility is built in.
The core pitch: you access serialized data directly without parsing or unpacking it first, which is where the memory efficiency comes from, and cross-version compatibility is built in.What are similar projects?protobuf, protobuf.js, arrow, thrift, grpc, hello-algo, CtCI-6th-Edition, protobuf
protobuf, protobuf.js, arrow, thrift, grpc, hello-algo, CtCI-6th-Edition, protobufWhich topics is it in?zero-copy, json-parser, protocol-buffers, protobuf, serialization, schema, rpc, c-sharp, dart, c-plus-plus, grpc, lua
zero-copy, json-parser, protocol-buffers, protobuf, serialization, schema, rpc, c-sharp, dart, c-plus-plus, grpc, lua