Develop Reference

Scraping from specific part of html based on right click

I am building a chrome extension whereby I want to be able to right click on a certain part of a page and then scrape some info from it. So using chrome.contextMenus I'd like to be able to only scrape from the element(one of it's attributes) where I've right clicked. Sort of like the behaviour in chrome when you right click somewhere on a page and select inspect it will open the element view on the page element you right clicked. The reason I want to do this is because there will be a number of similar type elements with different ids(attribute) so I want to be able get only the id of the particular element I'm interested in.
Is this even possible?
I was looking though the chrome.contextMenus documentation and I'm wondering if I know the element type(article)could I set the context menu on that and get the id that is stored in it that way?

I'd say your extension has to remember what was the last element under the right click. And content-script suits well for it.
background.js
chrome.contextMenus.create({
title: "Get ID",
id: "menu1",
onclick: function (info, tab) {
// send message about context manu action
chrome.tabs.sendMessage(tab.id, {
msg: 'get_id'
}, {
frameId: info.frameId
});
}
});
content-script.js
let lastClickedEl = null;
// remember last clicked element
document.body.addEventListener('mousedown', function (e) {
if (e.button === 2) { // right click to an element
lastClickedEl = e.target;
}
});
// receive message about context menu action
chrome.runtime.onMessage.addListener(function (request) {
if(request.msg === 'get_id') {
console.log(lastClickedEl.id); // your code here
}
});

jQuery calling div element that has been loaded by ajax, or inserted via html() method

It's my first post on stackoverflow. I've searched similiar questions here and found a couple of answers but couldn't really find a solution to this particular problem that would help me.
I have a webpage which loads main contents by ajax. Simply like this:
function loadContent(content) {
if(localStorage.content != content) {
$("#content #content_loading").css("display", "block");
}
var userID = Cookies.get("UserID");
$.ajax({
url: '../game/data/load_content.php',
type: 'post',
data : { ID : userID, Content : content },
success: function(response) {
$("#content #content_loading").css("display", "none");
$("#content #import").html(response);
localStorage.content = content;
$("#header").html("<div class='header_text'>"+content+"</div>");
}
}); }
It loads other ajax functions, html and css. Since I have thousands and thousands lines of code simple things get trickier. Now I simply want to create an universal 'close' button for popup windows. All of the popup windows are in a box, and the close button is inside the box header. Now I want to close all popup windows with a single function:
$('.close').click(function() {
$(this).parent().parent().fadeOut();
});
This simply selects the parent of the close element, which is header and then parent of that parent which is the whole box. One of the popup functions looks like this:
function showPopup(header, content) {
$("#popup_header").html(header+"<div class='close'></div>");
$("#popup_content").html(content);
$("#popup").fadeIn(300);
}
This function is included in the main document (<script src="script"></script>).
Then the other popup is directly loaded upon loadContent(content) function, so it's loaded with ajax call. And it's simply HTML that looks like this:
<div id="nearby_players">
<div class="header">PLAYERS NEARBY <div class="close"></div></div>
<ul> </ul>
</div>
Now if I insert the 'close' function upon click in the document that ajax is loading it will work. And if I change the loadPopup() function to this:
function showPopup(header, content) {
$("#popup_header").html(header+"<div class='close'></div>");
$("#popup_content").html(content);
$("#popup").fadeIn(300);
$(".close").click(function() {
$(this).parent().parent().fadeOut();
});
}
It works too. But what I want to do is to create a single click function attached to the main document that will close all the possible popups that are being loaded on the webpage or already are on the webpage. I thought it was a problem since 'close' element was an ID not a class. And since it should be unique I've changed it to class. So my question is. How do I refer to all of the elements with class 'close' whether they are being loaded with ajax, and then within that ajax they get loaded again with another ajax and so on. And also the popups that are already inserted into document when the webpage gets loaded?
How do I add these elements to the DOM so jQuery actually finds it?
Regards,
Hazes!

You create elements dynamically, which means that events are not attached to them.
Please read how to attach events to dynamically created elements: http://api.jquery.com/live/

How can I detect page navigation without reloading?

I have script, what work one time when window loaded, it work on this page, but site use some navigation links what not fully reload page (see this answer for example: Modify the URL without reloading the page). How can I detect that and run my script again?
I have one idea: storing URL (without anchor) in variable and check it periodically with current url, but I think this is bad solution. May be you know better one?
JavaScript or JQuery is possible to use.

Use window.onpopstate or window.onpushstate if u are using pushState or replaceState ( from ur given example).
ex:-
To Navigate Without reload ( you already did this )
// Your code to fetch new URL body and title
// update the DOM then use following code to update URL also (don't use location.href, otherwise the page reloads)
// sorry u already know this because u provided the example ;)
let data = { title : "Current Title",
body : document.body.innerHTML" } // to store current page data
window.history.pushState(data, 0, "newURL");
To detect navigation ( i.e., you wanna do )
window.onpushstate: when above code runs for navigation to a new url and load new content without reload ...
window.onpushstate(function() {
// detects ur navigation
})
window.onpopstate: when you press back button
window.onpopstate(function (e) {
let { data } = e.state;
// data object that u gave in pushState method when u called ur last / latest pushState method...
// use this data to retrieve previous data content and title
let { title, body } = data;
document.title = title;
document.body.innerHTML = body
})
for more detail mdn docs

That's because the new pages are either
1 ) Already at the ready and simply being brought in-sight by jQuery
2 ) Ajax called in.
If you scout for your navigation (the links you click on to go to the other page), you should find click me or so.
If you look for wherever this is is bound (i.e.: $('#navigation a').on("click", function(){});, you can simply wrap your script within a function, and trigger this function together with loading the new page every time. (after it, obviously).
I wish I could be more clear, but you did not provide any code yourself, so I have absolutely no idea of what kind of example I should be giving here.
-- the point: Those page changes are triggered by something in your javascript. Find the trigger that makes the page-change happen, and simply insert myCustomFunction();.
If you want to make your bindings update with a new DOM, you could use this:
function setBindings(){
//removing the old bindings prevents the "click" from triggering twice.
$('a').off("click");
$('a').on("click", function(){
//load page and such here
//Apply script you want to run here
setbindings(); //rerun the function to set the bindings.
});
}

I think you are looking for hashchanges you can listen to this event onhashchange
window.onhashchange = function(e) {
var sublink = window.location.hash.substring(1);
//do your thing here
}
You can also check what updated the url after the hashchange
var sublink = window.location.hash.substring(1);

I think the URL of script is cached，do you used Ajax get method？if it is，please
like this write url "wwww.baidu.com?"+Math.random();if not is ,in the firefox ,you can used pageshow event.

Creating and structuring the index page

I have a test website www.lemonbrush.com is has the two menu items "Home" and "About".
When we click on the "About" menu, the About page is loaded dynamical, which is great, but when I type the About page's url directly http://www.lemonbrush.com/about.html into the address bar, my whole plan goes for six.
I need some guidance in how shall I structure and load the pages so that the whole header and navigation are included even when we use the direct URL in the address bar.
My skills are HTML and Javascript.
Please see the following screen shots.

When you're clicking the menu item you are updating the page with the data, but when you are going to the link directly you are just getting the data. one way around this is to have common elements for the page, ie. navigation menus, In a javascript file and then use a script tag with a link where you want it on the page.

so, since i thought it would be nice for my project, to have a usable browser history and bookmarkable subpages, i yesterday tried the approach from my comments of the OP.
step 1: change the anchors in your navigation bar to something like that:
home
about
needles to say, that they don't need to be inside the same parent element. those links could be anywhere.
step 2: remove the on click handler from your script, that reacts on the navigation bar clicks!
step 3: create a function, to listen to the onhashchange event. (this has to be done once, on the main html page)
//create a function for the onhashchange event
window.onhashchange = function()
{
//check if a location hash is used
if(window.location.hash != null)
{
//store current location hash in a variable
requestedPage = location.hash;
//remove the '#' from the location hash value
requestedPage = requestedPage.replace('#','');
//load content from subpage into your content frame
$('#contentFrame').load('./' + requestedPage + '.html', function( response, status, xhr )
{
//catch 'file not found' errors and load start page
if ( status == "error" ) {
$('#mainPanel').load('./start.html');
}
});
};
for your page, you have of course to adapt, use the correct selectors and the correct file names.
you then would call your page via www.mydomain.com, and every subdomain via www.mydomain.com/#subPage

Google Analytics does not track link click events

I am currently using analytics.js (the newer version of GA) and I am trying to track all types of events from my website, including when a user clicks on an anchor tag pointing to an external URL.
I am currently using this setup:
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-KEY-HERE', { 'alwaysSendReferrer': true, 'allowAnchor': true });
and I send the events when hitting a link like this:
$(document).on("click", ".anchor-class", function (event) {
label = //get data from anchor element here ...
ga('send', 'event', 'Link Clicked', 'Click Details', label);;
return;
}
});
and this does not send anything to GA (even though the event handler is calling the ga(send...) method).
How ever, if I use this exact technique for but with event.preventDefault(); at the beginning of the function, the event is sent and it appears in GA dashboard.
Is there some setting that I missed in order to make this work properly?

Use the hitCallback function:
$(document).on('click','a', function(event){
event.preventDefault();
var label = $(this).attr('href');
ga('send', 'event', 'Link Clicked', 'Click Details', label, {
'hitCallback': function(){
window.location.href = label;
}
});
});

As pointed out by Blexy, the correct way to do this is to use a hit callback. However, you also need to take into account that users may block Google Analytics using some privacy protection tool such as Ghostery, in which case the hit callback will never be executed. The following article explains how to implement this correctly:
http://veithen.github.io/2015/01/24/outbound-link-tracking.html

We currently had this issue and ported our analytics code off of the website and into GTM. Another issue is that we have hundreds of sites that cannot have the new code released to them to deprecate the on-page analytics, but we already had GTM on them.
We were able to find the jQuery events that were bound to the click event and write code in GTM that removes the jQuery event on those clicked buttons via the exact event handler. Then we were able to apply standard GTM click triggers and tags so we didn't get double eventing.
Assuming you can easily remove the code from the page the following should work great using GTM.
This will fire an event off to analytics when a user clicks a given element and it will make any navigation wait until the corresponding tags have finished firing first before letting the page navigate away.
GTM Implementation
This is the quick and standard way now with the newer GTM.
Pre-Setup
Variables
You will need access to the Click Element Built-In variable. You can enable this under Variables --> Configure --> Click Element.
Also enable the Page Path or Page URL Built-In variable. This is later used to help you determine which pages to run the trigger on.
It looks like you're trying to get some text off of the clicked element for the event. If so you should create a Custom JavaScript variable. This just gets the inner text of the element, but you could also get an attribute or whatever other data you're looking for at this step.
Name: Click Element - Inner Text
Variable Type: Custom JavaScript
Custom JavaScript: function() { return {{Click Element}}.innerText; }
Triggers
Create a new trigger
Trigger Type: Click - Just Links
Wait for Tags: Enable this, this is the magic you're looking for.
Max wait time: set this to what you feel is appropriate and if you can live with the possibility that you may loose some analytics for a better user experience. Imagine what the wait time may be for a 3G user on a mobile device. Is 5000ms enough, not enough, maybe 10000ms. The user probably understands their connection is bad.
Enable this trigger when all of these conditions are true:
NOTE: you should only run this on pages that you need to run it on. Google isn't very clear if there is a performance loss and their "Learn More" says nothing about it.
If you need this to run on all pages though, configure it like so:
Page Path matches RegEx .*
Otherwise you should write something like:
Page Path matches RegEx ^/path/my-page$ specific page
Page Path matches RegEx ^/path/my-page/.*
NOTE: I'm not sure if those regex are correct, I'm not sure if you'll get a path with a proceeding or appended forward slash / or if you need anchors - normally it's best to be explicit on anchors so you don't get any funny business.
This trigger fires on: Some Link Clicks
Fire this trigger when an Event occurs and all of these conditions are true. Choose one below that fits your needs and change it how you need it.
Click Element matches CSS selector a
or maybe something more specific?
Click Element matches CSS selector .container .calls-to-action a
maybe only on external links? Assuming all internal links are relevant pathing.
Click Element matches CSS selector a[href^="http"]
Tags
Create a new tag
Tag Type: Event
Category: Link Clicked
Action: Click Details
Label: {{Click Element - Inner Text}}
Google Analytics Settings: best to use these for reusability and consistency rather than manually setting them up.
Firing Triggers: The Link Clicking Trigger you created above.