การสร้างเส้นทางแบบสหพันธ์

In a Micro Frontend (MFE) architecture, federated routing means each individual MFE is responsible for managing its own specific routes. Instead of one central application dictating all routes, each MFE declares its own pathways.

This approach gives teams more autonomy, allowing them to develop and deploy routing logic independently without tightly coupling to a main host application's routing system.

Federated routing offers several key advantages:

• Autonomy: MFE teams control their own navigation.
• Independent Deployment: Changes to an MFE's routes don't require redeploying the entire host app.
• Scalability: Easier to manage routing logic as the number of MFEs grows.
• Technology Agnostic: Different MFEs can use different routing libraries.

The fundamental idea behind federated routing is that each Micro Frontend publishes its own set of routes. Think of it like a mini-application within the larger system, responsible for its internal navigation structure.

The host application then acts as an aggregator, discovering and integrating these MFE-owned routes into a single, cohesive routing experience for the user.

Here's how a Micro Frontend (let's call it "MFE A") might define its internal routes. These are specific paths that MFE A is responsible for rendering.

We're using a simple array structure to represent routes, where each route has a path and a component or view it should render.

Similarly, another Micro Frontend ("MFE B") will define its own set of routes. Notice how its paths are distinct, often prefixed to avoid conflicts, ensuring clear ownership.

This separation allows MFE A and MFE B to evolve their navigation independently.

The host application's role is to act as the central orchestrator. It needs to discover and import the route definitions provided by each Micro Frontend.

Using Module Federation, these route definitions can be exposed by remote MFEs and consumed by the host, effectively building a global route map from distributed sources.

The host application then takes all the aggregated routes and integrates them into its own global router instance. This example simulates how a host app might load and process routes from different MFEs to handle navigation.

Try changing the currentPath variable in the code and run it!

Once the host's global router has integrated all federated routes, navigation between different Micro Frontends becomes seamless. From the user's perspective, it feels like a single, unified application.

When a user clicks a link to /app-b/settings, the global router identifies that this path belongs to MFE B and delegates the rendering or activation of MFE B accordingly.

With multiple MFEs defining routes, conflicts can arise if two MFEs define the same path (e.g., both have a /dashboard route).

Common strategies to prevent this include:

• Namespacing: Prefixing MFE routes (e.g., /mfea/dashboard, /mfeb/dashboard).
• Unique Paths: Ensuring each MFE's routes are inherently unique.
• Host Override: Allowing the host to prioritize or remap conflicting routes.

Imagine you have a host application integrating routes from two Micro Frontends, MFE-A and MFE-B. MFE-A defines /products and /products/:id. MFE-B defines /cart and /checkout.

Which of the following best describes the host application's role in this federated routing setup?

You've learned about federated routing, where each Micro Frontend owns and defines its own routes. We saw how a host application aggregates these routes to create a unified navigation experience.

• Each MFE declares its own routes.
• Host app collects and integrates MFE routes.
• Ensures MFE autonomy and independent deployment.
• Namespacing helps prevent conflicts.

This approach is crucial for large-scale MFE systems, balancing independence with a consistent user experience.