Skip to content

lawlietmester/webextension

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Cross-browser Firefox/Chrome webextension

Hello! This project created to make united syntax for chrome/browser objects, promise-based with some enchantments (callback behaviour in most cases won't work). So in most cases you will write one predictable code. Deprecated features in most cases are not supported. This script does not modify original browser/chrome objects.

Keep in mind that some features like Proxy API have absolutely different realizations in FF and Chrome - so you can not write one code for them both, you need to create browser split in your code.

In comparison with similar FF library, this library supports Chrome-only features like desktopCapture.

How to use

Install in npm using

npm i crossbrowser-webextension

If you need standalone file version - use Browser.js. It defines window.Browser object.

Code requirements

Your environment must support JS features listed below:

  • Object.assign
  • Promise

How to make browser split

if( typeof browser === 'undefined' ){
  // Chrome
}
else{
  // FF
}

For more complicated cases (like embedded webextestions for FF + Chrome):

let vChrome = () => {
  // Return some object
};
let vFF = () => {
  // Return some object
};
return typeof browser === 'undefined' ? vChrome() : vFF();

Syntax features

Zero arguments support

In some cases you are using empty object as first arguments. For such cases you can use zero arguments instead of empty object.

Example: Old code:

chrome.tabs.query({}, tabs => {})

New code:

Browser.tabs.query().then( tabs => {})

List of methods with zero arguments support:

  • browser.browserAction.getBadgeText
  • browser.browserAction.getTitle
  • browser.browserAction.getPopup
  • browser.browserAction.getBadgeBackgroundColor
  • browser.sidebarAction.getTitle
  • browser.sidebarAction.getPanel
  • browser.tabs.query

You can use .get and .clear without arguments. All 3 methods are promise-based. As for 54th FF, there is no onChange object in it.

browserAction

.getBadgeText, .getTitle, .getPopup, .getBadgeBackgroundColor could be ue used with tabId as first argument. Like:

Browser.browserAction.getPopup( 5 ).then( url => {
  // Use url
});

browserAction.setBadgeText

You can use it with text as argument. Like:

Browser.browserAction.setBadgeText( 'Icon text' );

browserAction.removeBadgeText (no arguments)

This is alias of browser.browserAction.setBadgeText({ 'text': '' })

contextualIdentities.query

You can use it with name as argument. Like:

Browser.contextualIdentities.query( 'name' ).then( identities => {
  // Some code
});

desktopCapture.chooseDesktopMedia

This method returns Promise with desktopMediaRequestId property. Promised resolved with options object, containing streamId property. (Due to promise cannot be resolved with 2 arguments)

history

.addUrl, .getVisits, .deleteUrl could be ue used with url as first argument. Like:

Browser.history.addUrl( 'http://mysite.com' ).then( url => {
  // Addition successfull
});

identity.removeCachedAuthToken

You can use it with token as argument. Like:

Browser.identity.removeCachedAuthToken( 'f8k48fk48fk' ).then( () => {
  // Removal complete
});

privacy.network

If .webRTCIPHandlingPolicy exist, deprecated features like .webRTCNonProxiedUdpEnabled and .webRTCMultipleRoutesEnabled are not provided.

pageAction

.getTitle, .getPopup could be ue used with tabId as first argument. Like:

Browser.pageAction.getTitle( 5 ).then( title => {
  // Use title
});

permissions.request

In Chrome browser.permissions.request automatically will add all new available APIs to Browser object.

proxy

proxy.onError works as proxy.onProxyError when proxy.onError undefined

sidebarAction

.getPanel, .getTitle could be ue used with tabId as first argument. Like:

Browser.sidebarAction.getTitle( 5 ).then( title => {
  // Use title
});

tabs.create

You can use it with URL as first argument. Like:

Browser.tabs.create( 'http://myurl.com/' ).then( tabInfo => {
  // Use tabInfo
});

tabs.reload

You can use reloadProperties argument as boolean. You can use tabs array (several tabs). Like:

Browser.tabs.reload( [ 7, 12, 50 ], true ).then( () => {
  // Use title
});

But keep in mind if at least one of tabs does not exist - promise will be rejected.

webRequest.onAuthRequired.addListener

You can use string in extraInfoSpec parameter. Like:

Browser.webRequest.onAuthRequired.addListener(
  () => {},
  { 'urls': [ '<all_urls>' ] },
  'blocking'
);

For synchronous request pass "blocking" in the extraInfoSpec parameter. For asynchronous request pass "asyncBlocking" in the extraInfoSpec parameter and return Promise in listener. Not documented part of all APIs: if you want to pass not desired asynchronous request use resolve() without any arguments.

Answers on some questions

  • Why this script does not based on Proxy object? By work I need to support Chrome 38+ which does not support Proxy object. In time it will be Proxy object based.

  • What about support of Edge? Edge is now equal to Chrome ;)

TODO

  • Write about difference in usage between background and popup for onmessage