Code Monkey home page Code Monkey logo

svelte-adapter-firebase's Introduction

⚠️ WARNING: this project is considered to be in ALPHA until SvelteKit is available for general use and the Adapter API is stable!

SvelteKit adapter Firebase social preview

svelte-adapter-firebase

GitHub Release npm Test CodeQL

Firebase adapter for SvelteKit. Support for:

✔️ SSR on Cloud Run
✔️ SSR on Cloud Functions
✔️ Integrates with existing JavaScript or TypeScript Cloud Functions!
✔️ Local production testing with Firebase Emulator
✔️ Multiple Hosting Sites

Utilise the Firebase Hosting CDN with dynamic content served by SvelteKit on Cloud Functions or Cloud Run!

Contents

Quickstart

In your standard SvelteKit project:

  • npm i -D svelte-adapter-firebase

  • add adapter to svelte.config.cjs:

    const firebaseAdapter = require("svelte-adapter-firebase");
    
    /** @type {import('@sveltejs/kit').Config} */
    module.exports = {
      kit: {
        adapter: firebaseAdapter(),
        target: "#svelte",
      },
    };
  • npm run build

  • Follow further instructions output by the adapter to prepare for deployment.

This adapter reads your firebase.json to determine if the Firebase Hosting site is using Cloud Functions or Cloud Run and outputs the server pieces accordingly. Static assets are output to the public directory for the Hosting site config.

Please read the docs carefully!

Configuration Overview

Adapter options:

  • hostingSite
    • required when firebase.json.hosting is an array (contains many site configurations)
    • default: no default value
  • sourceRewriteMatch
    • used to lookup the rewrite rule used for SSR
    • default: **
  • firebaseJson
    • path to your firebase.json, relative from where svelte build is called
    • default: ./firebase.json
  • cloudRunBuildDir
    • output dir of Cloud Run service, relative from your svelte app
    • default: ./.${run.serviceId} where run.serviceId is pulled from the firebase.json rewrite rule

Details

Quickstart outlines the steps most commonly used with a single SvelteKit app. Here we go into the details of each configuration and how it interacts with the firebase.json config.

The 3 step process is:

  1. select Hosting config from firebase.json. If more than one site match against hostingSite
  2. output static assets to the directory in the public field
  3. identify the rewrite rule for SSR to determine Cloud Function or Cloud Run output. The rewrite rule is determined by a lookup of the rewrites.source against sourceRewriteMatch

firebase.json Configurations

Due to the relaxed rules of firebase.json we can have many valid configs. At a minimum, one or more Hosting sites is required with an associated Functions config if a Cloud Function rewrite is used. These are the combintations:

single Hosting site with Cloud Function rewrite
{
  "hosting": {
    "public": "<someDir>",
    "rewrites": [
      {
        "source": "**",
        "function": "<functionName>"
      }
    ]
  },
  "functions": {
    "source": "<anotherDir>"
  }
}
multiple Hosting site with Cloud Function rewrite
{
  "hosting": [
    {
      "site": "blog",
      "public": "<someDir>",
      "rewrites": [
        {
          "source": "**",
          "function": "<functionName>"
        }
      ]
    },
    {
      // another site config
    }
  ],
  "functions": {
    "source": "<anotherDir>"
  }
}

To correctly lookup the blog site, hostingSite will need to be set in svelte.config.cjs:

const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ hostingSite: "blog" }),
    target: "#svelte",
  },
};
single Hosting site with Cloud Run rewrite
{
  "hosting": {
    "public": "<someDir>",
    "rewrites": [
      {
        "source": "**",
        "run": {
          "serviceId": "<cloudRunServiceId>"
        }
      }
    ]
  }
}
multiple Hosting site with Cloud Run rewrite
{
  "hosting": [
    {
      "site": "blog",
      "public": "<someDir>",
      "rewrites": [
        {
          "source": "**",
          "run": {
            "serviceId": "<cloudRunServiceId>"
          }
        }
      ]
    },
    {
      // another site config
    }
  ]
}

To correctly lookup the blog site, hostingSite will need to be set in svelte.config.cjs:

const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ hostingSite: "blog" }),
    target: "#svelte",
  },
};

Adapter Configurations

Detailed examples of the adapter configuration options.

hostingSite

If the firebase.json.hosting is an array of sites, then you must provide a site with hostingSite to correctly match against. For example:

// firebase.json
{
  "hosting": [
    {
      "site": "blog",
      "public": "<someDir>",
      "rewrites": [
        {
          "source": "**",
          "run": {
            "serviceId": "<cloudRunServiceId>"
          }
        }
      ]
    },
    {
      "site": "adminPanel",
      "public": "<anotherDir>"
    }
  ]
}
const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ hostingSite: "blog" }),
    target: "#svelte",
  },
};
sourceRewriteMatch

If the rewrite source pattern is not **, then svelte.config.cjs sourceRewriteMatch will need to be set to match your desired rewrite rule. For example:

// firebase.json
{
  "hosting": {
    "public": "<someDir>",
    "rewrites": [
      {
        "source": "/blog/**",
        "run": {
          "serviceId": "<cloudRunServiceId>"
        }
      }
    ]
  }
}
const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ sourceRewriteMatch: "/blog/**" }),
    target: "#svelte",
  },
};
firebaseJson

If the firebase.json file is not in the directory you run svelte build, then you can set a relative path in svelte.config.cjs:

.gitignore
firebase.json
app/                    <-- svelte build run in this dir
	package.json
	svelte.config.cjs
	src/
anotherApp/
	index.html
	index.css
functions/
	package.json
	index.js
const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ firebaseJson: "../firebase.json" }),
    target: "#svelte",
  },
};
cloudRunBuildDir

By default, a Node.js Cloud Run service is output to the directory named after the run.serviceId prefixed with a . relative to the dir in which svelte build was executed. IE: ./.${run.serviceId}. So with this config:

// firebase.json
{
  "hosting": {
    "public": "public",
    "rewrites": [
      {
        "source": "**",
        "run": {
          "serviceId": "mySiteSSR"
        }
      }
    ]
  }
}

will result in this output:

.mySiteSSR/         <--- Cloud Run service code
public/             <--- Hosting static assets
firebase.json
package.json
svelte.config.cjs
src/
	app.html
	routes/
		index.svelte
functions/
	package.json
	index.js

If you wish to customise this output dir, then you can specify it in the adapter config:

const firebaseAdapter = require("svelte-adapter-firebase");

/** @type {import('@sveltejs/kit').Config} */
module.exports = {
  kit: {
    adapter: firebaseAdapter({ cloudRunBuildDir: ".special/ssr/output/dir" }),
    target: "#svelte",
  },
};

Cloud Function

With this firebase.json and functions/ dir in a standard SvelteKit app structure and default svelte-adapter-firebase config:

// firebase.json
{
  "hosting": {
    "public": "myApp",
    "rewrites": [
      {
        "source": "**",
        "function": "ssrServer"
      }
    ],
    "predeploy": ["npm run build"]
  },
  "functions": {
    "source": "functions"
  }
}
firebase.json ("public": "myApp")
package.json
svelte.config.cjs
src/
	app.html
	routes/
		index.svelte
functions/
	package.json ("main": "index.js")
	index.js
	sveltekit/		<-- Server assets
myApp/				<-- Static assets to go to Firebase Hosting CDN

The firebase.json functions.source dir is used to find functions/package.json whose main field is used to find the Cloud Function build dir. This is used as the server asset output dir.

TypeScript Cloud Functions

Because we use the above method to determine the output dir the server assets are output to the correct place when using TypeScript.

firebase.json ("public": "myApp")
package.json
svelte.config.cjs
src/
	app.html
	routes/
		index.svelte
functions/
	package.json ("main": "lib/index.js")
	index.ts
	lib/
		index.js
		sveltekit/	<-- Server assets output to functions/lib
myApp/				<-- Static assets to go to Firebase Hosting CDN
Output with Multiple Sites

In a multi-site setup, the site name from hosting config in firebase.json is used as the server output dir:

firebase.json ("site": "myCoolSite","public": "myApp")
package.json
svelte.config.cjs
src/
	app.html
	routes/
		index.svelte
functions/
	package.json
	index.js
	myCoolSite/		<-- Server assets
myApp/				<-- Static assets to go to Firebase Hosting CDN

The final piece is to write the actual Cloud Function source code to reference the output server assets. The code is printed during svelte build and should be placed in your index.js or index.ts manually.

This is a flexible solution that allows integrating with other Cloud Functions in your project. You can edit the provided code as you see fit. The import/require of the generated code will not change unless you change the firebase.json.hosting.site or package.json.main fields, so you shouldn't need to update this code after adding it.

Cloud Function Firebase Emulator local Testing

Test your production build locally before pushing to git or deploying!

  • build your app: svelte-kit build
  • install Function dependencies: pnpm install --prefix functions
  • start the emulator: firebase emulators:start

Cloud Function Deployment

firebase deploy, that is all.

Cloud Function Caveats

As package.json gains dependencies for your Svelte app you may need to copy some of these to functions/package.json depending on the server-side functionality and how SvelteKit bundles your dependencies. ℹ️ This should be re-evaluated again in the future as it changes depending on SvelteKit and may not exist as a caveat in 1.0.0 release.

Cloud Run

With this firebase.json a standard SvelteKit app structure and default svelte-adapter-firebase config:

// firebase.json
{
  "hosting": {
    "public": "<someDir>",
    "rewrites": [
      {
        "source": "**",
        "run": {
          "serviceId": "mySiteSSR"
        }
      }
    ]
  }
}

will result in this output:

.mySiteSSR/         <--- This contains the Cloud Run service code
firebase.json
package.json
svelte.config.cjs
src/
	app.html
	routes/
		index.svelte
functions/
	package.json
	index.js

See the official Hosting/Cloud Run docs here for more setup information (enabling required APIs etc).

For those interested, we support Cloud Run with the same JS code as Cloud Functions, via the NodeJS Functions Framework and reliance on the Node.js 14 Buildpacks, which is what essentially powers Cloud Functions.

Cloud Run Local Testing

Cloud Run cannot be tested locally with the Firebase Emulator. However, you can still build and run it locally with gcloud cli:

gcloud beta code dev --builder

This will build the container using the Google Node 14 buildpack image, run the image locally, and rebuild the image on code changes. For more details, see - https://cloud.google.com/run/docs/testing/local#cloud-sdk_

When you route to the hosted image you should be able to navigate your Cloud Run app but your CDN hosted resources (css, images, etc) will not load properly. This can be used as a sanity check

Cloud Run Deployment

gcloud CLI is required to build & deploy Cloud Run services. The recommended build & deploy command for Cloud Run will be output when the adapter is run.

gcloud beta run deploy ${serviceId} --platform managed --region ${region} --source ${serverOutputDir} --allow-unauthenticated

Notably, this command builds and deploys your container, which is traditionally a two step process for container runtimes. You can orchestrate your deployment however you wish. Feel free to modify the command with any other Cloud Run features you may want to use, like increasing the concurrency or setting min_instances

This deploy command uses Cloud Build and the aforementioned Buildpacks and Functions Framework.

⚠️ Each build of your app will require firebase deploy --only hosting alongside your Cloud Run deployment as your CDN content will need to be updated as the filename hashes of static resources is rewritten on each svelte-kit build

Cloud Run Caveats

  • testing of a Cloud Run service with Firebase Hosting CDN and other backend features (PubSub/Cloud Functions) will require a full deployment to a Firebase project. The suggestion is a dev project per engineer and manual deployments to each env to test. No environment per PR here.

Function vs Run

Choice is a good thing, hopefully this comparison table helps you decide which compute environment is best for your application:

Feature Functions Run
Firebase Emulator Integration ✔️
Hosting CDN content deployed with Compute resources ✔️
Cold start mitigations ✔️

Cloud Functions seems do be a better default.

Non-goals

Write Cloud Function code directly into .js file instead of printing in console.

Firebase Cloud Functions have a long history of people configuring their index files completely differently, some even generating them from directories. Accommodating these would be a headache. Instead, all we look for is a match against this string, ${name} =, where name is your Cloud Functions name. We may make this configurable to a specific file in future.

Additionally, this allows for users to customise their Firebase Cloud Function API like runWith() options for memory/CPU and VPC/Ingress/Egress configuration settings, without complex support for options in the adapter. This keeps the Function config where it should, close to the executing code.

Handle the deployment of the app to Firebase.

Firebase apps consist of many different services with the CLI providing optional deployments. We do not want to dictate full deployments with your frontend nor perform partial deployments if it does not fit your app. The only option then is to leave it to you 🎉

Custom Docker images

We support Cloud Run with the same JS code as Cloud Functions, via the NodeJS Functions Framework and reliance on the Node14 Buildpacks, which is what essentially powers Cloud Functions. Diverging from this would make supporting Cloud Run more difficult. The idea behind the support is to allow usage of Cloud Run features, for example, min_instances and concurrent requests which both reduce cold starts (if that matters to you, use the CDN!)

We are open to discussion on this topic, but it was not a goal when setting out to build this adapter (consider the cost/benefit).

FAQ

Why is the Cloud Function code output to the terminal for me to add manually instead of being written to functions/index.js?

See non-goals Write Cloud Function code directly into .js file instead of printing in console.

Firebase libs in SvelteKit routes

This adapter does not try to solve the issue of using Firebase libraries in SvelteKit routes. These routes are compiled by the SvelteKit pipeline and there are many issues as ESM support in Firebase libs is not released and won't be for a long time.

Using Firebase libs in SvelteKit endpoints/routes may have been resolved with sveltejs/kit#490, however this does not mean the output is compatible with the Cloud Function runtime (investigation pending) and therefore compatible with svelte-adapter-firebase. Our recommendation is to prefer using Firebase Cloud Functions for API routes over SvelteKit routes.

Cold Starts

Depends on your application load. Typically, Cloud Functions will require more instances to handle the same number of requests as each Cloud Run as each CR instance can handle up to 250 (default maximum at the time of writing). Though in my experience, bumping the memory/cpu configuration dramatically reduces the response times.

Since the purpose of using this adapter is to leverage the Firebase Hosting CDN, you should consider improving the user experience with targetted caching/TTLs.

If cold starts are still an issue for your application, Cloud Run has support for min_instances which will keep x number of instances warm. This incurs additional costs. See the Cloud Run documentation for more.

Caveats

  • Firebase Hosting Preview Channels currently lacks first-party support for SSR applications. This adapter doesn't attempt to remedy this issue and doesn't produce a different SSR Function/Run for preview channel deployments.
  • ⚠️ while you can specify the region for both, Cloud Run in firebase.json and Cloud Functions in runWith({}) config, us-central1 is the only valid region for Firebase Hosting rewrites, other regions will error. The official warning about this can be found in these docs.
  • 1.0.0 will not be published until the SvelteKit Adapter API is declared stable and SvelteKit is released for general use.

Contributing

Contributions of any kind welcome, just follow the guidelines!

Short version:

git clone https://github.com/jthegedus/svelte-adapter-firebase.git
asdf install
pnpm i

external contributions

  • Cloud Function validation code linked in utils.js is from two different sources which indicates that it is being validated by firebase-tools in two separate places. PR a fix there.

svelte-adapter-firebase's People

Contributors

jthegedus avatar semantic-release-bot avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.