English | 中文

---

@js0.site/proto2js : Convert Protobuf definitions to JavaScript modules

Converts .proto files into ES modules, enabling full type-safety and IntelliSense for gRPC services and messages in modern JavaScript environments.

Table of Contents

- Features
- Usage
- Design
- Technology Stack
- Directory Structure
- A Little History

Features

- Modern JavaScript: Generates ES modules using modern syntax.
- Import Resolution: Automatically parses and merges all imported .proto dependencies into a single definition.
- Service & Message Generation: Creates distinct modules for service definitions and message types (Encoders/Decoders).
- Framework Agnostic: Produces clean, dependency-free modules that can be used with any gRPC library or data serialization framework.
- CLI Interface: Simple and intuitive command-line interface for code generation.

Usage

$3

The primary way to use proto2js is via the protogen command.

Syntax:
``bash
protogen -o [-I ...]
`

| Argument | Description |
| :--- | :--- |
| proto_path | Path to the root .proto file. |

| Option | Alias | Description | Required |
| :--- | :--- | :--- | :--- |
| --out | -o | Output directory for generated JavaScript files. | Yes |
| --include | -I | Directory for resolving .proto imports. Can be specified multiple times. The tool also automatically includes a set of standard Google protobuf types. | No |

$3

Given the following .proto file in test/demo/test.proto:
`protobuf
syntax = "proto3";

package test;

message Echo {
string msg = 1;
}

message Inner {
Echo echo = 1;
}

service Api {
rpc apiTest(Echo) returns (Inner);
}
`

Run the generator:
`bash
protogen test/demo/test.proto -o test/out
`

This will generate the following files:
- test/out/test/Api.js (Service definition)
- test/out/test/EchoD.js (Echo message decoder)
- test/out/test/EchoE.js (Echo message encoder)
- test/out/test/InnerD.js (Inner message decoder)
- test/out/test/InnerE.js (Inner message encoder)

Design

The code generation process follows a clear, multi-stage pipeline:

1. Import Merging (merge.js): The process starts from a single entry .proto file. The tool recursively parses all import statements. It searches for these dependencies within the provided include paths (-I) and merges them into a single, cohesive protocol buffer definition in memory. This creates a virtual .proto file that contains all necessary types.

2. AST Parsing (gen.js): The merged definition is then parsed into an Abstract Syntax Tree (AST) using the proto-parser library.

3. Code Generation (gen.js): The tool traverses the AST.
- For each message, it generates two modules: an E (Encoder) file and a D (Decoder) file. This separation allows for clean, unidirectional data flow.
- For each service, it generates a module that defines the RPC methods, importing the relevant message encoders and decoders.
- Type resolution is handled by findType.js, which correctly maps protobuf types to their corresponding generated JavaScript modules.

4. File Output: The generated JavaScript code is written to the specified output directory, preserving the package structure from the .proto files. The lib.js file orchestrates this entire process.

Technology Stack

- Runtime: Bun
- Dependencies:
- yargs: For command-line argument parsing.
- proto-parser: For converting .proto files into an AST.
- @3-/read, @3-/write: Utility libraries for file system operations.

Directory Structure

`
/
├───src/ # Source code
│ ├───cli.js # Command-line interface logic
│ ├───gen.js # Core code generation from AST
│ ├───lib.js # Main library entry point
│ ├───merge.js # .proto file import resolution and merging
│ └───import/ # Pre-bundled Google protobuf definitions
├───test/ # Test files and demos
│ ├───demo/
│ │ └───test.proto # Example proto file
│ └───out/ # Output of test generation
├───package.json # Project metadata and dependencies
└───README.md # This file
`

A Little History

Protocol Buffers (Protobuf) was developed at Google in the early 2000s as a high-performance, language-neutral, and platform-neutral mechanism for serializing structured data. It was created to be smaller, faster, and simpler than XML, which was the standard for data interchange at the time. After being used internally for many years as the foundation for Google's RPC systems and for data storage, Google open-sourced it in 2008. This project, proto2js, continues that spirit by bringing Protobuf's efficiency and strong typing to the world of modern JavaScript.

---

About

This project is an open-source component of js0.site ⋅ Refactoring the Internet Plan.

We are redefining the development paradigm of the Internet in a componentized way. Welcome to follow us:

* Google Group
* x.com/Js0Site
* js0site.bsky.social

---

@js0.site/proto2js : 将 Protobuf 定义转换为 JavaScript 模块

将 .proto 文件转换为 ES 模块，为现代 JavaScript 环境中的 gRPC 服务和消息提供完整的类型安全和智能提示。

目录

- 功能
- 使用方法
- 设计思路
- 技术栈
- 目录结构
- 相关历史

功能

- 现代 JavaScript: 生成使用现代语法的 ES 模块。
- 导入解析: 自动解析并合并所有导入的 .proto 依赖项，形成单一定义。
- 服务与消息生成: 为服务定义和消息类型（编码器/解码器）创建独立的模块。
- 框架无关: 生成纯净、无依赖的模块，可与任何 gRPC 库或数据序列化框架配合使用。
- CLI 接口: 提供简单直观的命令行工具用于代码生成。

使用方法

$3

proto2js 的主要使用方式是通过 protogen 命令。

语法:
`bash
protogen -o [-I ...]
`

| 参数 | 描述 |
| :--- | :--- |
| proto_path | 根 .proto 文件的路径。 |

| 选项 | 别名 | 描述 | 是否必需 |
| :--- | :--- | :--- | :--- |
| --out | -o | 生成的 JavaScript 文件的输出目录。 | 是 |
| --include | -I | 用于解析 .proto 导入的目录。可多次指定。工具也会自动包含一组标准的 Google Protobuf 类型。 | 否 |

$3

假设在 test/demo/test.proto 中有以下 .proto 文件：
`protobuf
syntax = "proto3";

package test;

message Echo {
string msg = 1;
}

message Inner {
Echo echo = 1;
}

service Api {
rpc apiTest(Echo) returns (Inner);
}
`

运行生成器：
`bash
protogen test/demo/test.proto -o test/out
`

这将生成以下文件：
- test/out/test/Api.js (服务定义)
- test/out/test/EchoD.js (Echo 消息解码器)
- test/out/test/EchoE.js (Echo 消息编码器)
- test/out/test/InnerD.js (Inner 消息解码器)
- test/out/test/InnerE.js (Inner 消息编码器)

设计思路

代码生成过程遵循清晰的多阶段流程：

1. 导入合并 (merge.js): 过程始于单个入口 .proto 文件。工具递归解析所有 import 语句，在提供的包含路径 (-I) 中搜索这些依赖项，并将它们合并成一个内存中的、统一的协议缓冲区定义。这会创建一个包含所有必要类型的虚拟 .proto 文件。

2. AST 解析 (gen.js): 接着，使用 proto-parser 库将合并后的定义解析为抽象语法树 (AST)。

3. 代码生成 (gen.js): 工具遍历 AST。
- 对于每个 message，生成两个模块：一个 E (Encoder) 文件和一个 D (Decoder) 文件。这种分离实现了清晰的单向数据流。
- 对于每个 service，生成一个定义 RPC 方法的模块，并导入相关的消息编码器和解码器。
- 类型解析由 findType.js 处理，它将 Protobuf 类型正确映射到其对应的生成 JavaScript 模块。

4. 文件输出: 生成的 JavaScript 代码被写入指定的输出目录，并保留 .proto 文件中的包结构。整个过程由 lib.js 文件统一调度。

技术栈

- 运行时: Bun
- 依赖项:
- yargs: 用于命令行参数解析。
- proto-parser: 用于将 .proto 文件转换为 AST。
- @3-/read, @3-/write: 用于文件系统操作的实用工具库。

目录结构

`
/
├───src/ # 源代码
│ ├───cli.js # 命令行接口逻辑
│ ├───gen.js # 从 AST 生成代码的核心逻辑
│ ├───lib.js # 库的主入口点
│ ├───merge.js # .proto 文件的导入解析与合并
│ └───import/ # 预置的 Google Protobuf 定义
├───test/ # 测试文件与演示
│ ├───demo/
│ │ └───test.proto # 示例 proto 文件
│ └───out/ # 测试生成结果的输出目录
├───package.json # 项目元数据与依赖
└───README.md # 本文档
`

相关历史

Protocol Buffers (Protobuf) 由 Google 在 21 世纪初开发，作为一种高性能、语言中立、平台中立的结构化数据序列化机制。其创建初衷是为了提供比当时数据交换标准 XML 更小、更快、更简单的替代方案。在作为 Google RPC 系统和数据存储的基础被内部使用了多年之后，Google 于 2008 年将其开源。proto2js` 项目延续了这一精神，将 Protobuf 的高效和强类型特性带入了现代 JavaScript 世界。

---

关于

本项目为 js0.site ⋅ 重构互联网计划 的开源组件。

我们正在以组件化的方式重新定义互联网的开发范式，欢迎关注：

* 谷歌邮件列表
* x.com/Js0Site
* js0site.bsky.social