การกำหนดสคีมา Protobuf

Welcome! In this lesson, we'll dive into Protocol Buffers, often called Protobuf. It's a key technology for gRPC.

• Protobuf is a language-neutral, platform-neutral, extensible mechanism for serializing structured data. Think of it as a highly efficient way to define and exchange data.
• It's like JSON or XML, but smaller, faster, and simpler.
• gRPC uses Protobuf to define the service interface and the structure of the payload messages.

Protobuf schemas are defined in special files ending with the .proto extension. These files act as contracts for your data.

• Every .proto file starts with a syntax declaration, usually syntax = "proto3";. This tells the Protobuf compiler which version of the syntax to use.
• It's crucial for forward and backward compatibility.

In Protobuf, a message is a structured record of information. It's similar to a class in object-oriented programming or a struct in C.

You define a message using the message keyword, followed by its name and a body containing its fields.

Inside a message, you define fields, each with a specific data type and a unique field number.

• Data Type: Specifies the type of data (e.g., string, int32, bool).
• Field Name: A unique identifier for the field within the message.
• Field Number: A unique, positive integer tag (from 1 to 2²⁹-1). These numbers are critical for identifying fields in the binary format and must remain stable for compatibility.

Protobuf supports a range of scalar data types. Here are some of the most common ones:

• int32, int64: Integer numbers.
• float, double: Floating-point numbers.
• bool: Boolean (true/false).
• string: UTF-8 encoded text.
• bytes: Raw byte sequences.

Choose the type that best fits your data to optimize storage and transmission.

Let's define a Person message with a few common fields. Notice the unique field numbers assigned to each field.

These numbers are like unique IDs for your data pieces. If you change a field's name, its number must stay the same for older systems to understand it.

Sometimes you need a field to have a value from a predefined list of options. This is where enums come in handy.

• Enums define a set of named integer constants.
• The first value in an enum must be 0, as it's the default value when a field isn't set.

Now, let's create a User message and include our UserRole enum as a field. This ensures that the user's role can only be one of the defined values.

It provides strong typing and prevents invalid data from being sent.

For more complex data, you can nest messages within other messages. This helps organize your schema and represent hierarchical data naturally.

Think of it like having an object inside another object in programming languages.

Consider the following Protobuf message definition:

Great job! You've learned the fundamentals of defining Protobuf schemas:

• .proto Files: The contract for your data.
• Messages: Structured data definitions, similar to classes.
• Fields: Each has a type, name, and unique field number.
• Scalar Types: Common types like string, int32, bool.
• Enums: For predefined sets of values, starting with 0.
• Nesting: Organizing complex data by embedding messages.

Next, we'll see how to turn these definitions into code!