Important: Chrome will be removing support for Chrome Apps on Windows, Mac, and Linux. Chrome OS will continue to support Chrome Apps. Additionally, Chrome and the Web Store will continue to support extensions on all platforms. Read the announcement and learn more about migrating your app.
Transitioning from Chrome apps on Windows, Mac, and Linux
Chrome packaged and hosted apps will be discontinued on Windows, Mac, and Linux over the course of now and early 2018. For more information, refer to the August 2016 Chromium blog post. This transition does not apply to Chrome OS, where Chrome packaged and hosted apps will remain supported and maintained for the foreseeable future.
To transition away from a Chrome packaged or hosted app on Windows, Mac, and Linux, the following options are available.
Chrome packaged apps
Recommended migration options for packaged apps are listed in order from simplest to most complicated.
Build a web app
Building a web app on top of the web platform is the ideal way to reach users across platforms.
We are investigating ways to improve the migration path to the web for developers that depend on exclusive Chrome App APIs - in particular the Sockets, HID, fileSystem and Serial APIs. If there are web platform features you need that are only available in select browsers, you can use feature detection to gracefully degrade or include explanatory text when your app is run in a browser that doesn’t support a particular API. If there are gaps in the web platform for your application, please let us know.
In the future, Progressive Web Apps (PWAs) will become installable on desktop platforms. Users will be able to install web apps to the desktop and launch them via icons and shortcuts, similar to the way that Chrome Apps can be installed today. This functionality is expected to launch sometime during 2018.
Build an extension-enhanced web page
If there is a capability that your Chrome App has that the regular web platform can’t provide, it might be available as an extension API. In this case, you can continue using a regular web app together with an externally connectable extension your web app can send messages to. Building a web app is typically preferable to this, because this approach forces users to install an extension that is only useful on your site and can cause increased friction.
As Chrome extensions can’t be run on other browsers, you should detect when required functionality isn’t available and provide explanatory text for users on these other browsers.
Build an extension
Depending on the user experience you want to provide, it might make sense to convert your app over to an extension. For example you could provide a browser action button which shows a small popup window for your user interface, or navigates to a page your extension provides. For some apps, this might be a good enough user experience to be a viable solution.
Note there are significant costs to this approach. Users that do not use Chrome will not be supported, and there will be significant friction for your users to install your extension.
Q: My Chrome App has many users. What’s the best way to migrate them to my new web app and deprecate my Chrome App?
You will still be able to publish updates to your Chrome App until the runtime is removed in early 2018. We recommend updating your app to include a message indicating that your app has been discontinued, and that users should visit your website going forward (see example below). You can also include an “uninstall” button that calls the uninstallSelf function.
Q: I want to continue to have a UI affordance that enables users to launch my app from the taskbar/dock/app shelf. What should I do?
This is still possible with web apps, but it is up to the user. On all desktop platforms except Mac, users can add ‘bookmark apps’ to their shelf / desktop, which optionally run your web app in its own window.
In the future, Progressive Web Apps (PWAs) will become installable on desktop platforms. Users will be able to install web apps to the desktop and launch them via icons and shortcuts, similar to the way that Chrome Apps can be installed today. This functionality is expected to launch sometime during 2018.Q: I want my app to run in its own window, not in a tab. Is this something I can do with a web app?
Yes, but it is up to the user. On desktop platforms other than OS X, users can add ‘bookmark apps’ to their shelf / desktop, which optionally run your web app in its own window.
Q: My app supports keyboard shortcuts. Can the web support this also?
Web apps can support keyboard shortcuts by listening to the various key events (e.g. keydown), but the shortcuts you can support are limited. For example, you cannot intercept Ctrl+N, Ctrl+T etc. as these are used by the browser. Or, these might be shortcuts that are handled by the operating system (e.g. Alt+Tab on Windows).
Work has begun to allow developers to intercept these additional shortcuts on the web, but this is not yet ready for use.
The Extension Commands API supports richer keyboard shortcuts which operate across the entire browser, which may enable new capabilities if transitioning to an extension is possible.
Q: My app needs to run when the user is offline. Can I do that with a web app?
Yes, you can use service workers to make your site run offline.
Q: My app uses the chrome.gcm API to receive push messages from the cloud. Can the open web do that?
Yes, the Push API and Notifications API provide the ability to show your user push notifications even when the page is closed.
Q: My app needs to be able to show notifications to the user even when it has no windows open. Can I do that with the open web?
Yes, the Push API and Notifications API provide this functionality.
Q: My app uses chrome.bluetooth/chrome.usb to talk to a Bluetooth/USB device. Can this be done on the open web?
Q: My app uses the chrome.fileSystem API to read and write user-specified files and / or directories. Can this be done on the open web?
In general, no. The open web can read single files that the user opens, but cannot retain access to those files, write to those files, or have any access to directories.
If it is critical for your app to read and write directories (e.g. it is a text editor with a folder view), you will need to either have a native helper app and extension combo, or create a native app.
Q: My app uses the chrome sockets API to do networking. Can this be done with the open web?
You might be able to do what you need with WebSockets. However, to use this, you will likely need to change the remote end of your connection. If that isn’t possible (e.g. you’re connecting over an established protocol like IRC or SSH), you’ll need some sort of native component.
Chrome hosted apps
Q: My hosted app uses the notifications permission. How do I do that on the web?
Q: My hosted app uses the unlimitedStorage permission. How do I do that on the web?
The unlimitedStorage permission ensured that data you store was ‘Persistent’, which means it can only be cleared manually by the user.
The recommended alternative is to use Persistent Storage in your web app. As of Chrome 52, Persistent Storage is available as an Origin Trial. To use this API in Chrome Stable, you will need to request a token and insert it in your web application.
Q: My hosted app uses the geolocation permission. How do I do that on the web?
The HTML5 geolocation API can be used in web apps to locate a user’s position.
Note: As of Chrome 50, the Geolocation API will only work on secure contexts such as HTTPS. If your site is hosted on an non-secure origin (such as HTTP), the requests to get the user’s location will no longer function.
Q: My hosted app uses the background permission. How do I do that on the web?
To ensure minimal power consumption we have been careful about introducing a generic method for sites to run in the background on user’s devices.
As of Chrome 49, sites can ensure actions taken by the user are synced to the server using the Background Sync API. This API allows sites to run some limited code in the background when the device re-connects to the internet, even if the tab has since been closed. Note that this API doesn’t allow for the service worker to be woken up periodically.
In the future, we hope to introduce a quota-limited ability for sites to wake up their service worker by sending a push message but without being required to show the user a notification. You can follow the progress of this work by starring this issue in the Chromium Issues Tracker.
Q: My hosted app uses the clipboardWrite permission. How do I do that on the web?
As of Chrome 43, websites are now able to programmatically trigger cut and copy.
Q: My hosted app uses the clipboardRead permission. How do I do that on the web?
We are currently working on bringing the ability to read the clipboard to the web and you can star this issue to be kept up to date. Until then you may consider building an extension with the clipboardRead permission to provide the functionality to your web app.