Repository files navigation

Get your app up andrunning withMarko!

• 🚀 Fastest way to build a Marko app
• 💖 Scales from zero configuration
• ⚡️ Pages live-reload as you make changes
• 📁 Directory-based routes, layouts, and middleware
• 🖌️ TypeScript powered editor support
• 🧬Designed with web standards to run anywhere

And when you build your production-ready app:

• 🔥 Blazing fast server-side rendering
• 🌊 Streams content to users ASAP
• 📦 Partial hydration & automatic code splitting
• 🚢 Deploy to multiple platforms

WarningThis project is in BETA - use at your own peril, but please do provide helpful feedback.

npm install @marko/run

marko-run makes it easy to get started without little to no config. The package ships with a default Vite config and node-based adapter.

To get started from a template:

1. npm init marko -- -t basic
2. cd ./<PROJECT_NAME>
3. npm run dev

Or manually create a project:

1. Install@marko/run
2. Create filesrc/routes/+page.marko
3. Runnpm exec marko-run

Finally openhttp://localhost:3000 🚀

dev - Start a development server in watch mode

> npmexec marko-run

or (with explicit sub command)

> npmexec marko-run dev

build - Create a production build

> npmexec marko-run build

preview - Create a production build and start the preview server

> npmexec marko-run preview

The plugin looks for route files in the configuredroutes directory. By default, that’s./src/routes, relative to the Vite config file.

To change what directory routes are found in:

// vite.config.tsimport{defineConfig}from"vite";importmarkofrom"@marko/run/vite";exportdefaultdefineConfig({plugins:[marko({routesDir:'src/pages'// Use `./src/pages` (relative to this file) as the routes directory})]})

The router only recognizes certain filenames which are all prefixed with+ (Why?). The following filenames will be discovered in any directory inside your application’sroutes directory.

These files establish a route at the current directory path which will be served forGET requests with the HTML content of the page. Only one page may exist for any served path.

These files provide alayout component, which will wrap all nested layouts and pages.

Layouts are like any other Marko component, with no extra constraints. Each layout receives the request, path params, URL, and route metadata as input, as well as arenderBody which refers to the nested page that is being rendered.

<main>  <h1>My Products</h1>  <${input.renderBody}/> // render the page or layout here</main>

These files establish a route at the current directory path which can handle requests forGET,POST,PUT, andDELETE HTTP methods.

Typically, these will be.js or.ts files depending on your project. Like pages, only one handler may exist for any served path. A handler should export functions

These files are like layouts, but for handlers. Middleware files are called before handlers and let you perform arbitrary work before and after.

Note: Unlike handlers, middleware run for all HTTP methods.

These files represent static metadata to attach to the route. This metadata will be automatically provided on the routecontext when invoking a route.

In addition to the files above which can be defined in any directory under theroutes directory, some special files can only be defined at its top level.

These special pages are subject to a root layout file (pages/+layout.marko in the default configuration).

This special page responds to any request where:

• TheAccept request header includestext/html
• And no other handler or page rendered the request

Responses with this page will have a404 status code.

This special page responds to any request where:

• TheAccept request header includestext/html
• And an uncaught error occurs while serving the request

Responses with this page will have a500 status code.

Given the following routes directory structure

routes/  about/    +handler.js    +layout.marko    +middleware.js    +page.marko  +layout.marko  +middleware.js  +page.marko

When the path/about is requested, the routable files execute in the following order:

1. Middlewares from root-most to leaf-most
2. Handler
3. Layouts from root-most to leaf-most
4. Page

sequenceDiagram    participant MW1 as routes/+middleware.js    participant MW2 as routes/about/+middleware.js    participant H as routes/about/+handler.js    participant L1 as routes/+layout.marko    participant L2 as routes/about/+layout.marko    participant P as routes/about/+page.marko    Note over L1,P: Combined at build-time as a single component    MW1->>MW2: next()    MW2->>H: next()    H->>L1: next()    L1->L2: ${input.renderBody}    L2->P: ${input.renderBody}    L1-->>H: Stream Response    H-->>MW2: Response    MW2-->>MW1: Response

Within theroutes directory, the directory structure determines the path from which the route is served. There are four types of directory names:static,pathless,dynamic, andcatch-all.

1. Static directories - The most common type, and the default behavior. Each static directory contributes its name as a segment in the route's served path, like a traditional fileserver. Unless a directory name matches the requirements for one of the below types, it is seen as a static directory.

   Examples:

   /foo/users/projects

2. Pathless directories - These directories donot contribute their name to the route's served path. Directory names that start with an underscore (_) will be ignored when parsing the route.

   Examples:

   /_users/_public

3. Dynamic directories - These directories introduce a dynamic parameter to the route's served path and will match any value at that segment. Any directory name that starts with a single dollar sign ($) will be a dynamic directory, and the remaining directory name will be the parameter at runtime. If the directory name is exactly$, the parameter will not be captured but it will be matched.

   Examples:

   /$id/$name/$

4. Catch-all directories - These directories are similar to dynamic directories and introduce a dynamic parameter, but instead of matching a single path segment, they match to the end of the path. Any directory that starts with two dollar signs ($$) will be a catch-all directory, and the remaining directory name will be the parameter at runtime. In the case of a directory named$$, the parameter name will not be captured but it will match. Catch-all directories can be used to make404 Not Found routes at any level, including the root.

   Because catch-all directories match any path segment and consume the rest of the path, you cannot nest route files in them and no further directories will be traversed.

   Examples:

   /$$all/$$rest/$$

Flat routes let you define paths without needing additional folders. Instead the folder structure can be defined either in the file or folder name. This allows you to decouple your routes from your folder structure or co-locate them as needed. To define a flat route, use periods (.) to deliniate each path segment. This behaves exacly like creating a new folder and each segment will be parsed using the rules described above for static, dynamic and pathless routes.

Flat routes syntax can be used for both directories and routable files (eg. pages, handlers, middleware, etc.). For these files, anything preceeding the plus (+) will be treated as the flat route.

For example to define a page at/projects/$projectId/members with a root layout and a project layout:

Without flat routes you would have a file structure like:

routes/  projects/    $projectId/      $members/        +page.marko      +layout.marko  +layout.marko

With flat routes move the path defined by the folders into the files and separate with a period

routes/  +layout.marko  projects+layout.marko  projects.$projectId.members+page.marko

Additionally, you can continue to organize files under folders to decrease duplication and use flat route syntax in the folder name

routes/  projects.$projectId/    +layout.marko    members+page.marko  +layout.marko

Finally, flat routes and routes defined with directories are all treated equally and merged together. For example this page will have layout

routes/  projects/    $projectId/      +layout.marko  projects.$projectId+page.marko

Along with descibing multiple segements, flat route syntax supports defining routes that match more than one path and segments that are optional. To describe a route that matches multiple paths, use a comma (,) and define each route.

For example the following page matches/projects/$projectId/members and/projects/$projectId/people

routes/  projects.$projectId.members,projects.$projectId.people+page.marko

This file name is a bit long so you might do something like this

routes/  projects.$projectId    members,people+page.marko

We can simplify this by introducing another concept:grouping. Groups allows you to define segments within a flat route that match multiple sub-paths by surrounding them with parentheses (( and)). For the example, this means you can do the following:

routes/  projects.$projectId.(members,people)+page.marko

This is a simple example of grouping but you can nest groups and make them as complicated as you want.

The last concept isoptionallity. By introducing an empty segment or pathless segment along with another value you can make that segment optional. For example, If we want a page that matches/projects and/projects/home, you can create a flat route that optionally matcheshome

routes/  projects.(home,)+page.marko

or

routes/  projects.(home,_pathless)+page.marko

While both of these create a route which matches the paths, they have slightly different semantics. Using a pathless segment is the same as creating a pathless folder which allows you to isolate middleware and loayouts. Using an empty segement is the same as defining a file at the current location.

This package’s Vite plugin discovers your route files, generates the routing code, and registers the@marko/vite plugin to compile your.marko files.

// vite.config.tsimport{defineConfig}from"vite";importmarkofrom"@marko/run/vite";// Import the Vite pluginexportdefaultdefineConfig({plugins:[marko()]// Register the Vite plugin})

Adapters provide the means to change the development, build, and preview process to fit different deployment platforms and runtimes while allowing authors to write idiomatic code.

Specify your adapter in the Vite config when registering the@marko/run plugin

// vite.config.tsimport{defineConfig}from"vite";importmarkofrom"@marko/run/vite";importnetlifyfrom"@marko/run-adapter-netlify"// Import the adapterexportdefaultdefineConfig({plugins:[marko({adapter:netlify({edge:true})// Configure and apply the adapter})]})

• @marko/run-adapter-node
• @marko/run-adapter-netlify
• @marko/run-adapter-static

Generally, when using an adapter, this runtime will be abstracted away.

import*asRunfrom '@marko/run/router`;

asyncfunctionfetch<T>(request:Request,platform:T)=>Promise<Response|void>;

This asynchronous function takes aWHATWGRequest object and an object containing any platform-specific data you may want access to, and returns any of

• aWHATWGResponse object (generated from executing any matched route files)
• undefined (if the request was not explicitly handled)
• a404 status code response (if no route matches the requested path)
• a500 status code response (if an error occurs)

Express example:

importexpressfrom"express";import*asRunfrom"@marko/run/router";express().use(async(req,res,next)=>{constrequest=// ...code to create a WHATWG Request from `req`constresponse=awaitRun.fetch(request,{      req,      res});if(response){// ...code to apply a response to `res`}else{next();}}).listen(3000);

In some cases, you might want more control over when route matching and invocation (creating a response) occur. For instance, you may have middleware in your server which needs to know if there is a matched route. The runtime provides these additional methods:

interfaceRoute{params:Record<string,string>;meta:unknown;}functionmatch(method:string,pathname:string)=>Route|null;

This synchronous function takes an HTTP method and path name and returns an object representing the best match, ornull if no match is found.

• params - a{ key: value } collection of any path parameters for the route
• meta - metadata for the route

asyncfunctioninvoke(route:Route,request:Request,platform:any)=>Promise<Response|void>;

This asynchronous function takes a route object returned byRun.match the request and platform data and returns a response in the same way theRun.fetch does.

Express example:

importexpressfrom"express";import*asRunfrom"@marko/run/router";express().use((req,res)=>{constmatchedRoute=Run.match(req.method,req.path);if(matchedRoute){req.match=matchedRoute;}})// ...other middleware.use(async(req,res,next)=>{// Check if a route was previously matchedif(!req.match){next();return;}constrequest=// ...code to create a WHATWG Request from `req`constresponse=awaitRun.invoke(req.match,request,{      req,      res});if(response){// ...code to apply a response to `res`}else{next();}}).listen(3000);

marko/run provides a global namespaceMarkoRun with the following types:

MarkoRun.Handler - Type that represents a handler function to be exported by a +handler or +middleware file

MarkoRun.Route - Type of the route's params and metadata

MarkoRun.Context - Type of the request context object in a handler and$global in your Marko files. This type can be extended using TypeScript's module and interface merging by declaring aContext interface on the@marko/run module within your applcation code

declare module"@marko/run"{interfaceContext{customPropery:MyCustomThing// will be globally defined on MarkoRun.Context}}

MarkoRun.Platform - Type of the platform object provided by the adapter in use. This interface can be extended in that same way asContext (see above) by declaring aPlatform interface:

declare module"@marko/run"{interfacePlatform{customPropery:MyCustomThing// will be globally defined on MarkoRun.Platform}}

If aTSConfig file is discovered in the project root, the Vite plugin will automatically generate a .d.ts file which provides more specific types for each of your middleware, handlers, layouts, and pages. This file will be generated at.marko-run/routes.d.ts whenever the project is built - including dev.

Note TypeScript will not include this file by default. You should use theMarko VSCode plugin andadd it in your tsconfig.

These types are replaced with more specific versions per routable file:

MarkoRun.Handler

• Overrides context with specific MarkoRun.Context

MarkoRun.Route

• Adds specific parameters and meta types
• In middleware and layouts which are used in many routes, this type will be a union of all possible routes that the file will see

MarkoRun.Context

• In middleware and layouts which are used in many routes, this type will be a union of all possible routes that the file will see.
• When an adapter is used, it can provide types for the platform

• Error handling
• Error component
• Redirect component

Once stable @marko/run will replace @marko/serve and improves upon that project in several critical ways.

1. Special "route files" (e.g.+page.marko) improve the developer ergonomics quite substantially. While they may cause a double take initially, making these explicit allows colocating additional components, tests, stories, config, utilities, and whatever else you need alongside the page components. With@marko/serve it was far too easy to "accidentally" serve some of your test fixtures :see-no-evil:.
2. @marko/serve was built around Webpack. Since Webpack doesn't have great support for SSR, a lot of this work was up to us. This was not only a maintenance burden but also lead to some rough edges such as no HMR support (just full page reloading in dev). By switching to Vite with its first-class SSR support, things all come together much more smoothly.
3. @marko/serve was primarily designed with a node target in mind. @marko/run instead supports an "adapter" model, where you can author in web standard APIs and build your application to run in Node, Deno, Netlify, Cloudflare, and even a static site.
4. The programmatic API of@marko/serve left some to be desired which made it difficult to integrate into existing development servers and projects. Because of this, public-facing applications at eBay (the largest consumer of Marko) were not able to bring in@marko/serve. With@marko/run we've worked from the ground up to ensure a flexible enough programmatic API to allow embedding in existing complex applications. Because of this, we're confident that@marko/run will see much more use than@marko/serve and more investment from us!
5. Built-in layout management. Strictly speaking, Marko does not need the concept of+layout.marko "route file". If you've used Marko before you know it's very easy to treat layouts as normal components. But by bringing these layouts into the router we're able to reduce the amount of JavaScript naively sent to the browser, reduce the amount of boilerplate, and prime ourselves for some plans we have for after Marko 6 is out 👀.

There's more of course, but we're committed to making@marko/run thebest way to build a Marko application.