I'm trying to pass the active dom element when the contextmenu is clicked from my background script to a script that is being called through chrome.tabs.executeScript. I can pass booleans and strings just fine, but i always get an error when i pass dom elements. I'm starting to think it's not possible.
//doScripts function called from browser action
chrome.browserAction.onClicked.addListener(function(tab) {
doScripts(true, null);
});
//doScripts function called from context menu click
function getClickHandler(info, tab) {
var currTarg = document.activeElement;
console.log("currTarg = " + currTarg);
doScripts(false, currTarg);
}
//i reference doingBrowserAction and contextTarg in myscript.js
function doScripts(context, targ){
chrome.tabs.executeScript(null, {code: "var doingBrowserAction = "+context+"; var contextTarg = "+targ+";"}, function(){
chrome.tabs.executeScript(null, {file: "js/myscript.js"}, function(){
//all injected
});
});
}
//setup context menu
chrome.contextMenus.create({
"title" : "DESTROY!",
"type" : "normal",
"contexts" : ["page","selection","link","editable","image","video","audio"],
"onclick" : getClickHandler
});
i reference doingBrowserAction and contextTarg in myscript.js. I know what i'm trying to do is possible because the adblock extension does it, but having a hard time figuring out how. thanks in advance.
You cannot get a direct reference to a content script's DOM element from the background page, because the background page runs in the extension's process, and the content script runs in the tab's process. See also https://code.google.com/p/chromium/issues/detail?id=39507.
The document.activeElement property in the background page refers to the active element in the background page's document. As you can imagine, this value is quite useless.
If you query the state of the currently right-clicked element, bind an event in the content script. In the next example, I've chosen the contextmenu event, because context menus can also be opened through the keyboard.
This example adds a context menu option that removes the last active element from the document.
// content script
var lastElementContext;
document.addEventListener('contextmenu', function(event) {
lastElementContext = event.target;
}, true);
chrome.runtime.onMessage.addListener(function(message, sender, sendResponse) {
if (lastElementContext && lastElementContext.parentNode) {
lastElementContext.parentNode.removeChild(lastElementContext);
lastElementContext = null;
}
});
Background script:
chrome.contextMenus.create({
title: 'DESTROY!',
contexts: ['page', 'link', 'editable', 'image', 'video', 'audio'],
onclick: function(info, tab) {
chrome.tabs.sendMessage(tab.id, 'doDestroy');
}
});
Related
I am developing a chrome extension. I am trying to get the selected/highlighted text from the active tab and do something with it. For now, say all the extension does is writing what the selected text is on a popup. I can't seem to do it. I tried a lot of methods. content script, background scripts. Nothing works. in my manifest.jsonI have permissions for activeTab, contextMenus. I tried multiple functions that take the selected text but nothing works. Example of some functions
const text = (window.getSelection) ?
window.getSelection().toString() :
document.selection.createRange().text;
console.log(text)
chrome.contextMenus.create({
id: 'selectionGetter',
title: 'send selected text',
contexts: ['selection'],
});
chrome.browserAction.onClicked.addListener(function(tab) {
chrome.tabs.sendRequest(tab.id, {method: "getSelection"},
function(response){
const url=response.url;
const subject=response.subject;
const body = response.body;
console.log(body)
});
});
chrome.browserAction.onClicked.addListener(function(tab) {
chrome.tabs.sendRequest(tab.id, {method: "getSelection"},
function(response){
sendServiceRequest(response.data);
});
});
Would love your help
See How to access the webpage DOM rather than the extension page DOM? for full explanation, here's a quick illustration:
chrome.tabs.executeScript({code: 'getSelection().toString()'}, ([sel] = []) => {
if (!chrome.runtime.lastError) {
document.body.textContent = 'Selection: ' + sel;
}
});
If you have default_popup declared in manifest.json then put this code in popup.js, and put <script src=popup.js></script> in popup.html. It will run each time the popup is shown.
Otherwise put it inside chrome.browserAction.onClicked listener in the background script.
My Google Chrome Extension currently retrieves the ImageURL which was right-clicked from the menu, but that's not enough. I also need to do this:
Get the DIV Element inside which this specific Image was clicked.
Get some text strings stored inside this DIV, or in sub-DIVs, next
to the clicked image.
So far I have this, which is working:
background.js
// Add context menu for Images
chrome.contextMenus.create({
"title": "Get this Image",
"contexts": ["image"],
"onclick" : getInfo
});
function getInfo(e)
{
var imgURL = e.srcUrl;
// Set up an event which Content will receive
chrome.tabs.query({active: true, currentWindow: true}, function(tabs) {
chrome.tabs.sendMessage(tabs[0].id, {url: imgURL}, function(response) {
alert('Response in Background');
});
});
}
contentscript.js
// Listener for Background Event Sends
chrome.runtime.onMessage.addListener(
function(request, sender, sendResponse) {
// print Image Src URL in a custom Content-created place
divHTML = request.url;
document.getElementById('infoDiv').innerHTML = divHTML;
});
Where do I do the extra work to analyze the DOM tree and retrieve extra surrounding elements? In the Background or in the Content?
So I'm basically developing an automated click-paste-and-upload system for mutiple texts and files inside a google page.
This method helped me get the instances of objects that I'm looking for: buttons, textboxes, richtextboxes, etc.
Now I want to work with them.
So for example I know the id of a button , and the function subscribed to its click event. How do I trigger the click event from the extension ? I've tried injecting a script with the click event handler (discovered with DOM inspector) at "document_startup" but I don't get an error or anything else.
Here's the content script! The loggerhead function should have inserted the script but I don't think it did. What might be the reason for the blow code not giving anything?
// Runs a function for every added DOM element that matches a filter
// filter -- either function(DOM_node){/*...*/}, returns true or false
// OR a jQuery selector
// callback -- function(DOM_node){/*...*/}
function watchNodes(filter, callback){
observer = new MutationObserver( function (mutations) {
mutations.forEach( function (mutation){
if(typeof filter === "function"){
$(mutation.addedNodes).filter(
function(i){ return filter(this); }
).each(
function(i){ callback(this); }
);
} else {
$(mutation.addedNodes).filter(filter).each(
function(i){ callback(this); }
);
}
});
});
// For every added element, a mutation will be processed
// with mutation.taget == parent
// and mutation.addedNodes containing the added element
observer.observe(document, { subtree: true, childList: true });
}
function loggerhead(node) {
console.log("passhead");
//also inject jquery
var jqueryEl = document.createElement('script');
jqueryEl.setAttribute('src', chrome.extension.getURL('jquery-1.11.1.min.js'));
jqueryEl.setAttribute('type', 'text/javascript');
var scriptEl = document.createElement('script');
scriptEl.setAttribute('src', chrome.extension.getURL('script.js'));
scriptEl.setAttribute('type', 'text/javascript');
node.appendChild(jqueryEl);
node.appendChild(scriptEl);
}
watchNodes("head", loggerhead);
// method not working
//var gmailHead = jQuery("head", document).get(0);
script.js contains the function of subscribed to the click event of the button that I've managed to find through the DOM inspector:
function Cdb(b){return function(){if(Vbb()){return Ddb(b,this,arguments)}else{var a=Ddb(b,this,arguments);a!=null&&(a=a.val);return a}}}
You should try to call the existing click handler like
buttonElement.click()
First of all, I'm not sure if my title describes the problem correctly... I did search but didn't find anything that helped me out...
The project I'm working on has an #orderList. All orders have a delete option. After an order gets deleted the list is updated.
Sounds simple... I ran into a problem though.
/**
* Data returned at the end of selecting some options
*/
$.post(myUrl, $('#myForm').serialize(), function(data) {
// I build the orderlist
// The data returned is a JSON object holding session data (including orders)
buildOrderList(data);
...
// Do some other work
});
/*
* function to build the html list
*/
function buildOrderList(data) {
// Empty list
$('#orderList').empty();
// The click handler for the delete button is in here because it needs the data object
$(document).on('click', '[id^=delete_]', function() {
// Get the orderId from the delete button
var orderId = $(this).attr('id').split('_');
orderId = orderId['1'];
// I call the delete function
deleteOrder(orderId, data);
});
var html = '';
// Loop the data object
$.each(data, function(key,val){
...
// Put html code needed in var html
...
});
$('#orderList').append(html);
}
/*
* function to delete an order
*/
function deleteOrder(orderId, data) {
// Because of it depends on other 'products' in the list if the user can
// simply delete it, I use a jQuery dialog to give him some options.
// These options I send to a php script so it knows what should be deleted.
// This fires when a user clicks on the 'delete' button from a dialog.
// The dialog uses data to show options but does not change the value of data.
switch(data.type) {
case 'A':
delMsg += '<p>Some message for case A</p>';
delMsg += '<select>with some options for case A</select>';
$('#wizard_dialog').append(delMsg);
$('#wizard_dialog').dialog('option', 'buttons', [
{ text: "Delete", click: function() {
$.post(myUrl, $('#myDeleteOptions').serialize(), function(newData) {
// Now the returned data is the updated session data
// So I build the orderList again...
buildOrderList(newData);
...
// Do some other work
});
$( this ).dialog( "close" );
$(this).html(''); }},
{ text: "Cancel", click: function() { $( this ).dialog("close"); $(this).html(''); }}
] );
break;
case 'B':
// Do the same thing but different text and <select> elements
break;
}
}
The orderList updates correctly, however if I try to delete another order, the jQuery dialog gives me the option for the current (correct product) AND the option for the product that previously owned the id of the current. (Hope I didn't loose anyone in my attempt to explain the problem)
The main question is how to 'refresh' the data send to buildOrderList.
Since I call the function in a new $.post with fresh data object returned it should work, shouldn't it?
/**
* Enable the JQuery dialog
* (#wizard_dialog)
* this is the init (note that I only open the dialog in deleteOrder() and set text and buttons according to the data send to deleteOrder() )
*/
$('#wizard_dialog').dialog({
autoOpen: false,
resizable: false,
modal: true,
dialogClass: "no-close",
open: function() {
$('.ui-dialog-buttonpane').find('button:contains("Annuleren")').addClass('cancelButtonClass');
$('.ui-dialog-buttonpane').find('button:contains("Verwijderen")').addClass('deleteButtonClass');
$('.ui-dialog :button').blur(); // Because it is dangerous to put focus on 'OK' button
$('.ui-widget-overlay').css('position', 'fixed'); // Fixing overlay (else in wrong position?)
if ($(document).height() > $(window).height()) {
var scrollTop = ($('html').scrollTop()) ? $('html').scrollTop() : $('body').scrollTop(); // Works for Chrome, Firefox, IE...
$('html').addClass('noscroll').css('top',-scrollTop); // Prevent scroll without hiding the bar (thus preventing page to shift)
}
},
close: function() {
$('.ui-widget-overlay').css('position', 'absolute'); // Brake overlay again
var scrollTop = parseInt($('html').css('top'));
$('html').removeClass('noscroll'); // Allow scrolling again
$('html,body').scrollTop(-scrollTop);
$('#wizard_dialog').html('');
}
});
EDIT:
Because the problem could be in the dialog I added some code.
In the first code block I changed deleteOrder();
ANSWER
The solution was rather simple. I forgot to turn the click handler off before I added the new one. This returned the previous event and the new event.
$(document).off('click', '[id^=delete_]').on('click', '[id^=delete_]', function() {
// Get the orderId from the delete button
var orderId = $(this).attr('id').split('_');
orderId = orderId['1'];
// I call the delete function
deleteOrder(orderId, data);
});
I want to be able to change the icon of my extension according to what site I am currently browsing. How can I listen for changes in tab focus?
I think I've figured this one out. You need two listeners. One to sense when the tab has been changed, one to sense when it's been updated. And then they both can trigger the same function too run. Here's what would be in the background file...
function changeIcon() {
//query the information on the active tab
chrome.tabs.query({active: true}, function(tab){
//pull the url from that information
var url=tab[0].url;
//do whatever you need to do with the URL
//alert(url);
//change the icon
chrome.browserAction.setIcon({path: 'pathToIcon'});
});
}
//listen for new tab to be activated
chrome.tabs.onActivated.addListener(function(activeInfo) {
changeIcon();
});
//listen for current tab to be changed
chrome.tabs.onUpdated.addListener(function(tabId, changeInfo, tab) {
changeIcon();
});
Simply register for tab update notifications in your background page:
chrome.tabs.onUpdated.addListener(function(tabId, changeInfo, tab)
{
if (changeInfo.status == "loading")
{
var url = tab.url;
var iconPath = ???
chrome.pageAction.setIcon({tabId: tabId, path: iconPath});
}
});
This handler will be called whenever a tab changes location. You don't need to care which tab is currently selected because you will have defined a different icon for each tab. Still, if you want to do it - http://code.google.com/chrome/extensions/tabs.html#event-onSelectionChanged is the way to go.