chrome.devtools.inspectedWindow.eval - frameURL - google-chrome-extension

I am trying to get the selected element to the sidebar pane in my chrome extension.
It's working fine if the page has no frames when the element is in the frame, it's not working.
As per the document I have to pass the frameURL, but how do I get the frame or Iframe URL?
Thank you.
Note: This issue is duplicate that was opened in 3 years ago, but still no solution there, so re-opening it again.
In devtools.js
chrome.devtools.panels.elements.createSidebarPane(name, (panel) => {
// listen for the elements changes
function updatePanel() {
chrome.devtools.inspectedWindow.eval("parseDOM($0)", {
frameURL: // how to pass dynamic
useContentScriptContext: true
}, (result, exceptipon) => {
if (result) {
console.log(result)
}
if (exceptipon) {
console.log(exceptipon)
}
});
}
chrome.devtools.panels.elements.onSelectionChanged.addListener(updatePanel);
});

I ran into this as well. I ended up needing to add a content_script on each page/iframe and a background page to help pass messages between devtools and content scripts.
The key bit is that in the devtools page, we should ask the content_scripts to send back what their current url is. For every content script that was registered, we can then call chrome.devtools.inspectedWindow.eval("setSelectedElement($0)", { useContentScriptContext: true, frameURL: msg.iframe } );
Or in full:
chrome.devtools.panels.elements.createSidebarPane( "example", function( sidebar ) {
const port = chrome.extension.connect({ name: "example-name" });
// announce to content scripts that they should message back with their frame urls
port.postMessage( 'SIDEBAR_INIT' );
port.onMessage.addListener(function ( msg) {
if ( msg.iframe ) {
// register with the correct frame url
chrome.devtools.panels.elements.onSelectionChanged.addListener(
() => {
chrome.devtools.inspectedWindow.eval("setSelectedElement($0)", { useContentScriptContext: true, frameURL: msg.iframe } );
}
);
} else {
// otherwise assume other messages from content scripts should update the sidebar
sidebar.setObject( msg );
}
} );
}
);
Then in the content_script, we should only process the event if we notice that the last selected element ($0) is different, since each frame on the page will also handle this.
let lastElement;
function setSelectedElement( element ) {
// if the selected element is the same, let handlers in other iframe contexts handle it instead.
if ( element !== lastElement ) {
lastElement = element;
// Pass back the object we'd like to set on the sidebar
chrome.extension.sendMessage( nextSidebarObject( element ) );
}
}
There's a bit of setup, including manifest changes, so see this PR for a full example:
https://github.com/gwwar/z-context/pull/21

You can found url of the frame this way:
document.querySelectorAll('iframe')[0].src
Assuming there is at lease one iframe.
Please note, you cannot use useContentScriptContext: true, as it will make the script execute as a context page (per documentation) and it will be in a separate sandboxed environment.
I had a slightly different problem, but it might be helpful for your case too, I was dynamically inserting an iframe to a page, and then tried to eval a script in it. Here the code that worked:
let win = chrome.devtools.inspectedWindow
let code = `
(function () {
let doc = window.document
let insertFrm = doc.createElement('IFRAME')
insertFrm.src = 'about:runner'
body.appendChild(insertFrm)
})()`
win.eval(code, function (result, error) {
if (error) {
console.log('Eror in insertFrame(), result:', result)
console.error(error)
} else {
let code = `
(function () {
let doc = window.document
let sc = doc.createElement('script')
sc.src = '${chrome.runtime.getURL('views/index.js')}'
doc.head.appendChild(sc)
})()`
win.eval(code, { frameURL: 'about:bela-runner' }, function (result, error) {
if (error) {
console.log('Eror in insertFrame(), result:', result)
console.error(error)
}
})
}
})

Related

Chrome Extension API Calls order and DOM Information

I'm working on an extension that is supposed to extract information from the DOM based specific classes/tags,etc, then allow the user to save the information as a CSV file.
I'm getting stuck on a couple of places and haven't been able to find answers to questions similar enough.
Where I am tripped up at is:
1) Making sure that the page has completely loaded so the chrome.tabs.query doesn't return null a couple of times before the promise actually succeeds and allows the blocksF to successfully inject. I have tried placing it within a settimeout function but the chrome api doesn't seem to work within such the function.
2) Saving the extracted information so when the user moves onto a new page, the information is still there. I'm not sure if I should use the chrome.storage api call or simply save the information as an array and keep passing it through. It's just text, so I don't believe that it should take up too much space.
Then main function of the background.js is below.
let mainfunc = chrome.tabs.onUpdated.addListener(
async(id, tab) => {
if (buttonOn == true) {
let actTab = await chrome.tabs.query({
active: true,
currentWindow: true,
status: "complete"
}).catch(console.log(console.error()));
if (!actTab) {
console.log("Could not get URL. Turn extension off and on again.");
} else {
console.log("Tab information recieved.")
};
console.log(actTab);
let blocksF = chrome.scripting.executeScript({
target: { tabId: actTab[0]['id'] },
func: createBlocks
})
.catch(console.error)
if (!blocksF) {
console.log("Something went wrong.")
} else {
console.log("Buttons have been created.")
};
/*
Adds listeners and should return value of the works array if the user chose to get the information
*/
let listenersF = chrome.scripting.executeScript({
target: { tabId: actTab[0]['id'] },
func: loadListeners
})
.catch(console.error)
if (!listenersF) {
console.log("Listeners failed to load.")
} else {
console.log("Listeners loaded successfully.")
};
console.log(listenersF)
};
});
Information from the DOM is extracted through an event listener on a div/button that is added. The event listener is added within the loadListeners function.
let workArr = document.getElementById("getInfo").addEventListener("click", () => {
let domAr = Array.from(
document.querySelectorAll(<class 1>, <class 2>),
el => {
return el.textContent
}
);
let newAr = []
for (let i = 0; i < domAr.length; i++) {
if (i % 2 == 0) {
newAr.push([domAr[i], domAr[i + 1]])
}
}
newAr.forEach((work, i) => {
let table = document.getElementById('extTable');
let row = document.createElement("tr");
row.appendChild(document.createElement("td")).textContent = work[0];
row.appendChild(document.createElement("td")).textContent = work[1];
table.appendChild(row);
});
return newAr
I've been stuck on this for a couple of weeks now. Any help would be appreciated. Thank you!
There are several issues.
chrome methods return a Promise in MV3 so you need to await it or chain on it via then.
tabs.onUpdated listener's parameters are different. The second one is a change info which you can check for status instead of polling the active tab, moreover the update may happen while the tab is inactive.
catch(console.log(console.error())) doesn't do anything useful because it immediately calls these two functions so it's equivalent to catch(undefined)
Using return newArr inside a DOM event listener doesn't do anything useful because the caller of this listener is the internal DOM event dispatcher which doesn't use the returned value. Instead, your injected func should return a Promise and call resolve inside the listener when done. This requires Chrome 98 which added support for resolving Promise returned by the injected function.
chrome.tabs.onUpdated.addListener(onTabUpdated);
async function onTabUpdated(tabId, info, tab) {
if (info.status === 'complete' &&
/^https?:\/\/(www\.)?example\.com\//.test(tab.url) &&
await exec(tabId, createBlocks)) {
const [{result}] = await exec(tabId, loadListeners);
console.log(result);
// here you can save it in chrome.storage if necessary
}
}
function exec(tabId, func) {
// console.error returns `undefined` so we don't need try/catch,
// because executeScript is always an array of objects on success
return chrome.scripting.executeScript({target: {tabId}, func})
.catch(console.error);
}
function loadListeners() {
return new Promise(resolve => {
document.getElementById('getInfo').addEventListener('click', () => {
const result = [];
// ...add items to result
resolve(result);
});
});
}

Ask for a variable defined in renderer from preload

With:
contextIsolation = true
nodeIntegration = false
What is the best way to ask for the value of a variable defined in the renderer context from preload context?
I'm developing an application for managing photo collections.
In my main.js there is a menuItem to import photos, the click function for this menu entry is:
click() {
dialog.showOpenDialog(mainWindow, {
title: 'Import',
properties: ['openFile', 'multiSelections']
}).then(result => {
if (result.filePaths.length) {
mainWindow.webContents.send('import', result.filePaths);
}
});
}
in preload.js:
ipcRenderer.on('import', (ev, origins) => {
origins.forEach(origin => {
files.importPhoto(origin, dir);
});
});
the 'dir' attribute in preload function is the destination folder to copy the photos and is the value I need to get from the app.js in the renderer.
If you want to be able to modify dir in your preload.js file, you should expose your ipcRenderer.on code as a function through the contextBridge. Ie.
preload.js
const { contextBridge, ipcRenderer } = require("electron")
contextBridge.exposeInMainWorld("api",
{
importFunc: function(dir) {
// clear out any existing listeners if this method
// is called multiple times from the app.js/renderer code
ipcRenderer.removeAllListeners("import");
ipcRenderer.on("import", (ev, origins) => {
origins.forEach(origin => {
files.importPhoto(origin, dir);
});
});
}
}
)
In your app.js/renderer code, you could call this function like so when you have the right value for dir. (Once this method is run, your MenuItem click should pull in the right value).
window.api.importFunc("C:\\myfolder");

Chrome extension background script sometimes does not run after install or update

I have had recent reports of a chrome extension that I develop that stops working after an update or a fresh install. The background script seems to not start at all.
There is no response to messages sent to it from the content scripts.
There is no process for it in the task manager.
Opening background page from chrome://extensions does not show any activity in the console, or show any source files.
Profiling, memory snapshot buttons are disabled.
Once this issue appears, it persists for the chrome profile even after reloading or uninstalling/reinstalling the extension.
Restarting chrome resolves the problem.
The issue has been seen on chrome v79. But I cannot say for sure that it is exclusive to this version, as the issue is difficult to reproduce and seemingly random.
Has anyone seen such an issue, or has any ideas what to look for? I am happy to update my question with any new info I have or with any info you need.
Edit:
Here is my webNavigation listener, which is used to inject content scripts. This handler is wired up in the 'root' context of the background script (not asynchronously inside an event handler)
chrome.webNavigation.onCompleted.addListener((details) ⇒ {
if(details.frameId === 0) {
injectScript(
'js/contentScript.js',
details.tabId,
details.frameId,
details.url
).catch((e) ⇒ {});
}
}
The injectScript function is as follows
export const injectScript = ƒ (scriptPath,tab,frame,tabUrl) {
return new Promise((res,rej) ⇒ {
let options = {
file : scriptPath,
allFrames : false,
frameId : frame,
matchAboutBlank: false,
runAt : 'document_idle',
};
const cb = ƒ () {
if (chrome.runtime.lastError) {
let err = new Error('Could not inject script');
capture(err,{
...options,
tabUrl,
lastError : chrome.runtime.lastError.message,
});
rej(err);
}else{
res();
}
};
if (tabUrl.indexOf('.salesforce.com') !== -1) {
window.setTimeout(() => {
chrome.tabs.executeScript(tab,options,cb);
},500);
}else{
chrome.tabs.executeScript(tab,options,cb);
}
});
};
Note above, the capture function reports the error to a backend and I cannot see it being reported there as well. Cannot add a breakpoint in code because no source appears in the background page, as noted above.
A background service worker is loaded when it is needed, and unloaded when it goes idle.
https://developer.chrome.com/docs/extensions/mv3/service_workers/
You can use the following methods:
// Keep heartbeat
let heartTimer;
const keepAlive = () => {
heartTimer && clearTimeout(heartTimer);
heartTimer = setTimeout(() => {
chrome.tabs.query({ active: true, currentWindow: true }, function (tabs) {
console.info('[heartbeat]')
tabs.length && chrome.tabs.sendMessage(
tabs[0].id,
{ action: "heartbeat" }
);
});
keepAlive();
}, 10000);
};
keepAlive();

Click event does nothing when triggered

When I trigger a .click() event in a non-headless mode in puppeteer, nothing happens, not even an error.. "non-headless mode so i could visually monitor what is being clicked"
const scraper = {
test: async () => {
let browser, page;
try {
browser = await puppeteer.launch({
headless: false,
args: ["--no-sandbox", "--disable-setuid-sandbox"]
});
page = await browser.newPage();
} catch (err) {
console.log(err);
}
try {
await page.goto("https://www.betking.com/sports/s/eventOdds/1-840-841-0-0,1-1107-1108-0-0,1-835-3775-0-0,", {
waitUntil: "domcontentloaded"
});
console.log("scraping, wait...");
} catch (err) {
console.log(err);
}
console.log("waiting....");
try {
await page.waitFor('.eventsWrapper');
} catch (err) {
console.log(err, err.response);
}
try {
let oddsListData = await page.evaluate(async () => {
let regionAreaContainer = document.querySelectorAll('.areaContainer.region .regionGroup > .regionAreas > div:first-child > .area:nth-child(5)');
regionAreaContainer = Array.prototype.slice.call(regionAreaContainer);
let t = []; //Used to monitor the element being clicked
regionAreaContainer.forEach(async (region) => {
let dat = await region.querySelector('div');
dat.innerHTML === "GG/NG" ? t.push(dat.innerHTML) : false; //Used to confirm that the right element is being clicked
dat.innerHTML === "GG/NG" ? dat.click() : false;
})
return t;
})
console.log(oddsListData);
} catch (err) {
console.log(err);
}
}
}
I expect it to click the specified button and load in some dynamic data on the page.
In Chrome's console, I get the error
Transition Rejection($id: 1 type: 2, message: The transition has been superseded by a different transition, detail: Transition#3( 'sportsMultipleEvents'{"eventMarketIds":"1-840-841-0-0,1-1107-1108-0-0,1-835-3775-0-0,"} -> 'sportsMultipleEvents'{"eventMarketIds":"1-840-841-0-0,1-1107-1108-0-0,1-835-3775-535-14,"} ))
Problem
Behaving non-human-like by executing code like element.click() (inside the page context) or element.value = '..' (see this answer for a similar problem) seems to be problematic for Angular applications. You want to try to behave more human-like by using puppeteer functions like page.click() as they simulate a "real" mouse click instead of just triggering the element's click event.
In addition the page seems to rebuild parts of the page whenever one of the items is clicked. Therefore, you need to execute the selector again after each click.
Code sample
To behave more human-like and requery the elements after each click you can change the latter part of your code to something like this:
let list = await page.$x("//div[div/text() = 'GG/NG']");
for (let i = 0; i < list.length; i++) {
await list[i].click();
// give the page some time and then query the selectors again
await page.waitFor(500);
list = await page.$x("//div[div/text() = 'GG/NG']");
}
This code uses an XPath expression to query the div elements which contain another div element with the given text. After that, a click is simulated on the element and then the contents of the page are queried another time to respect the change of the DOM elements.
Here might be a less confusing way to click those:
for(var div of document.querySelectorAll('div')){
if(div.innerHTML === 'GG/NG') div.click()
}

How to tell if a script is run as content script or background script?

In a Chrome extension, a script may be included as a content script or background script.
Most stuff it does is the same, but there are some would vary according to different context.
The question is, how could a script tell which context it is being run at?
Thank you.
I think this is a fairly robust version that worked in my initial tests and does not require a slower try catch, and it identifies at least the three primary contexts of a chrome extension, and should let you know if you are on the base page as well.
av = {};
av.Env = {
isChromeExt: function(){
return !!(window['chrome'] && window['chrome']['extension'])
},
getContext: function(){
var loc = window.location.href;
if(!!(window['chrome'] && window['chrome']['extension'])){
if(/^chrome/.test(loc)){
if(window == chrome.extension.getBackgroundPage()){
return 'background';
}else{
return 'extension';
}
}else if( /^https?/.test(loc) ){
return 'content';
}
}else{
return window.location.protocol.replace(':','');
}
}
};
Well I managed to work out this:
var scriptContext = function() {
try {
if (chrome.bookmarks) {
return "background";
}
else {
return "content";
}
}
catch (e) {
return "content";
}
}
It's because an exception would be thrown if the content script tries to access the chrome.* parts except chrome.extension.
Reference: http://code.google.com/chrome/extensions/content_scripts.html
The best solution I've found to this problem comes from over here.
const isBackground = () => location.protocol === 'chrome-extension:'
The background service worker at Manifest v3 does not contain a window.
I use this as part of my extension error handling which reloads the content scripts, when i receive an Extension context invalidated error:
...
if (!self.window) {
console.warn('Background error: \n', error);
} else {
location.reload();
}
...

Resources