As mentioned in previous chapters, a route just describes when and how to call the function we passed in.

In the previous chapters we focused on how it calls our function, now we will talk more about when a route is matched.

Remember when you see route, it's imported from spirit-router from our very first example.

const route = require("spirit-router")

Creating Routes

Routes are always grouped with define, even if we only have 1 route:

  route.get("/", [], some_function)

This specifies the route will match and call our some_function with no arguments []. It will match when the incoming request is a GET / request. That is, the request is a GET method, for path /.

Since there are no arguments, we can optionally omit it and write the above example as:

  route.get("/", some_function)

Routes are tried in the order they first appear, and are considered matched when the incoming request's method and path matches the routes. Once this happens, the function for the route will be called with any arguments defined.

However, the router will only consider the routing is over if the function of the route returns a value that isn't undefined.

If undefined is returned, the router will keep trying until a route finally has a response.

// #=> GET /

  // matches but returns undefined, the router continues
  route.get("/", () => undefined),
  // matches and returns a value, routing ends
  route.get("/", () => "Hello")


spirit-router exports common methods besides just get.

It also exports: put, post, delete, head, options, patch:


If we wanted a route to match any http method, we use any:

  route.any("/test", ...)

Which would would match any http request for path "/test".

For custom http methods that aren't pre-defined we can use method:

  route.method("custom", "/", ...)

Which would match for a http request with CUSTOM method for path "/".


The path of a route doesn't have to always be a string.

They can be a string pattern, or regexp. (They work exactly the same way as Express and other web libraries.)

const greet = (name) => {
  return "Hello, " + name

  route.get("/:name", ["name"], greet)
// #=> GET /bob
// { status: 200, 
//   headers: { "Content-Type": "text/html; charset=utf-8" }, 
//   body: "Hello, bob" }

As shown in the comment, the result is "Hello, bob" when we make a GET /bob request.

It's able to pass in name because of dependency injection (discussed also in Request).

When the route matches, it'll see "/:name" means to hold on to the value in the path. The ["name"] signals greet depends on the value of "name" before we can call it. So "name" is looked up on the request and greet is called with it.

We can also use regexp itself, or regexp characters such as "*" (which matches any path):

const inspect = (url) => {
  return "You made a request to: " + url

  route.get("*", ["url"], inspect)
// #=> GET /test-test
// { status: 200, 
//   headers: { "Content-Type": "text/html; charset=utf-8" }, 
//   body: "You made a request to: /test-test" }

results matching ""

    No results matching ""