This is a simplified example:
I use two online dictionaries, but each one has only a set of features I need. However it's a pain to click through both and type in the words in two places.
I would like to write my own javascript webpage, that loads elements from the two sites, i.e. the input field for the word, and then displays the results from the two sites on the same page.
My script would need to read elements from one page, and use that data to
manipulate the other page.
When I tried researching this I encountered the same origin policy, but apparently chrome there's a workaround. My question is is this possible via browsers and javascript or would I have to look at a completely different technology like Python and webscraping.
If it's just to consult the two websites for the same term at the same time, you could have your webpage with two iframes and set their src attributes like the following:
$("#search").click(function () {
var term = $("#term").val();
$("#dictionary").attr("src", "http://www.dictionary.com/browse/" + term);
$("#free").attr("src", "https://www.thefreedictionary.com/" + term);
});
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.min.js"></script>
<input id="term" type="text" placeholder="search dictionaries">
<input id="search" type="button" value="Search"><br>
<iframe id="dictionary" src="about:blank" style="width:45vw;height:500px;"></iframe>
<iframe id="free" src="about:blank" style="width:45vw;height:500px"></iframe>
The URL of the website is their API. Just look at the address bar after you've done a search on their site to find out how to construct it yourself.
Bonus
I myself use different websites like this, but via custom commands to automatically open multiple tabs in the browser (which you can also do via Javascript).
I use commands on linux like the following (declared in .bashrc):
function d {
firefox "http://www.dictionary.com/browse/$1+$2+$3+$4+$5"&exit
}
which means, without firefox even being open, I can open a terminal (with a keyboard shortcut) and type d test to launch firefox and search a dictionary for test. If firefox is already running it'll be brought to the foreground and the website will open in a new tab.
In windows you can add batch files (*.bat) or command files (*.cmd) to do the same thing. In c:\windows you can create a file called d.cmd with:
start "" "http://www.dictionary.com/browse/%1+%2+%3+%4+%5"
then use the keyboard shortcut Win+R to open the run dialog and type d test.
The run dialog used to be in the Start Menu, they've probably hidden it on your version of windows.
Yes, you can do that with JS if the dictionaries have endpoints you can interact with. (Or at least that would be the easiest). And most of them do. You can often enough simply google dictionary name + API.
If you have access to that you can fire off 2 requests to fetch data which you can then display on your page once it is back.
Related
I would like to write simple scripts which after I have already opened site ( I dont wanna script to open it) press two buttons and insert data in comment section after pressing f.ex. 'g' button. I am completly new in that kind of programming so any help will be nice( also link to good tutorials).
webBrowser1.Document.GetElementById("User").SetAttribute("value", textBox1.Text);
webBrowser1.Document.GetElementById("but").InvokeMember("click");
I am aware of those 2 functions i will use but how to instantiate them on already opened page by pressing a button? (If thats important deafult used browser is opera).
You should use something like Selenium (http://www.seleniumhq.org/) which is a browser automation framework.
Selenium scripts can be written in many languages (including c#) and the scripts can be run on a variety of browsers. There is even browser plugins for creating scripts my recording a macro - no code required!
This is much more robust that using a browser control embedded in an app as that is only a cut down version of internet explorer I believe.
This is a rough sample of selenium in c#
using OpenQA.Selenium;
using OpenQA.Selenium.IE;
using OpenQA.Selenium.Support.UI;
var options = new InternetExplorerOptions();
options.IntroduceInstabilityByIgnoringProtectedModeSettings = true;
Driver = new InternetExplorerDriver(options);
Driver.Navigate().GoToUrl("yourURL");
Driver.FindElement(By.Id("User")).SendKeys("<your text>");
Driver.FindElement(By.Id("but")).Click();
I need to create a simple static web part - the only dynamic part is the current user login name (needs to be included in a URL parameter in a link).
I'd rather just use a content editor web part instead of a custom web part that I'll need to deploy. In fact, I already did that using JavaScript and ActiveX (WScript.Shell - full code below)
Is there a better, more native way to do it? What I don't like about the ActiveX approach is that it requires more liberal configuration of IE, and even when I enable everything ActiveX-related in the security settings there is still a prompt that needs to be clicked.
Cross-browser support is not a major issue, it's intranet - but will be a nice extra, too.
One way I can think of is to scrape the username from the top-right hand corner with jQuery (or just JavaScript), but is there something even cleaner?
Or maybe someone has already done the scraping and can share the code to save me some time ;-)
Here's my current solution:
<script language="javascript">
function GetUserName()
{
var wshell = new ActiveXObject("WScript.Shell");
var username = wshell.ExpandEnvironmentStrings("%USERNAME%");
var link = document.getElementById('linkId');
link.href = link.href + username.toUpperCase();
}
</script>
<P align=center>
<a id="linkId" onclick="GetUserName();" href="<my_target_URL>?UserID=">open username-based URL</a>
</P>
Include this Web part token in your content editor web part:
_LogonUser_
It outputs the same value as Request.ServerVariables("LOGON_USER")
I've done this before including in a hidden span and plain old document.getElementById to get the innertext.
My conclusion is that there's no better way to do that (which is also sufficiently easy).
I checked the source of the page and the username is not anywhere there, so can't be scraped by JavaScript/jQuery. The browser doesn't seem to be sending it with the request (as it does with your local IP and other client-related information), so can't be obtained from headers either.
The only other approach I can conceive is calling a web service from the client-side that would be an overkill for my scenario.
This question already has answers here:
How to disable (View Source) and (Ctrl + C ) from my site
(10 answers)
Closed 9 years ago.
I'm looking for disable keyboard script to protect hidden content.
It is not possible. A user will always be able to view your source since he needs to download it in order to render the page.
There are more ways to view source than what you are trying to prevent:
Using firebug
Using wget
Right clicking on content choosing 'view source'
Using the menu option
Via man in the middle
probably more...
This is not very complicated but totally unreliable. Its same with all other Javascript protections.
First the trick (IE incompatible):
function denyKey(event) {
var code = event.keyCode;
if(event.ctrlKey) {
if(code==85)
return false;
}
}
window.addEventListener("keydown", denyKey);
My code is just scratch, it is not cross-browser. This is where to get keyCodes. I did not put much effort in the code since I want to discourage you from using it.
Once you send data to user, he can read the data unless you encrypt them without giving him the key. This means any:
Javascript authentication
Secret loading pages
Javascript "Wait before download..."
Blocked mouse buttons
..can and will be bypassed by the user.
I have a bookmarklet to unblock mouse buttons for example.
That is not possible, even if it was, it would have been a horrible protection. Even I could write a simple script that fetches the source of an arbitrary page. Everything that the client sees, is 'view source'-able (somebody edit that). Only server-side code is safe. Even if it was only possible to view your page through a real browser (but you can't make it so) you'll probably overlook a accelerator key, or other shortcut. If you don't want the client to see some code, don't give it to him! Keep it server-side (and not in a .txt file, that's accessible too) or don't keep it.
I manage an ASP.NET site where they want to open multiple reports at the click of a button (preferably each in its own tab). The reports are saved as URLs for now, so I just need to open multiple windows with those urls.
I am attempting to do so using javascript (namely window.open). Here is an example of what I am doing (though I removed the actual urls):
<html>
<body>
<button onclick="openLinks();return false;">Open both links</button>
<script type="text/javascript">
function openLinks(){
var FirstWindow = window.open('');
FirstWindow.location = 'https://myssrsReportURL/rs:Format=EXCEL';
window.open('https://myssrsReportURL/rs:Format=EXCEL');
}
</script>
</body>
</html>
The first 2 lines of the js method above has the advantage of forcing the url to open in a new tab versus a new window. However, in the example above, when I set the location after opening a blank window, I receive the file download error IE "blocked this site from downloading":
Where as the second url opens (in a new window) without issue (I get a file download prompt).
Why does the second method work (I receive a file download prompt) while the first doesn't (I receive an IE security warning) if they are opening the same url? Is there some way around this using javascript (or jQuery)?
The customer is requiring that a single button click open all these reports. I can not zip them all up in one request because the reports are generated by SSRS upon a get request to a specific url (the ones I am attempting to open in new tabs).
Note, the question is not, "how do I open multiple tabs" it's why does the behavior discrepancy exist between window.open('') versus window.open('url')
There is no JavaScript magic trick to trick browsers into opening multiple tabs. And if there is...it will be stopped as it is a security vulnerability (download/tabs spam denial of service).
There is a better solution, which works on every browser. You can even choose between JavaScript and ASP.NET.
Create multiple iframe elements and show one at a time by using on-page tabs/links/buttons (simple onclick action to reveal one iframe and hide the rest). For downloads, this method has the advantage of not losing the current page if a server side script fails (so instead of 500 Internal Error or blank page the user remains on the current page).
For multiple downloads, why not zip the files on the server to have a single download prompt? It makes sense to not annoy users with many many download prompts.
Below is the HTML
<a id="LnkEmail" onclick="doMailto('d#s.com');" href="javascript:void(0);">
<span id="LblEmail">ABC</span></a>
Javascript
<script type="text/javascript">
function doMailto(EmailAddress) {
document.location.href = window.open('mailto:' + EmailAddress, 'new window');
}
</script>
In FireFox, it opens the image on clicking the span like below.
Query - In IE 8 - Nothing happens on clicking it. Any Idea ?
The popup selection feature is native to Firefox and is NOT an available feature in Internet Explorer, as Internet Explorer handles association directly from Windows, your operating system. So, whichever program is meant to handle mailto: links on your computer is what will open (most typically, Outlook Express). There is no consistent way to avoid this as you cannot control what a user decides to open that protocol with. MY suggestion is to write a POST asp.net contact form. I'm not an ASP.NET developer myself, but I found this tutorial for you: http://www.jimcobooks.com/tutorials/emailform1/default.aspx
To test this theory: try finding a computer without any mail client (no outlook, outlook express, etc.) Internet Explorer will then prompt for a program to open the protocol.
Another test (the way I tested) I set up Google Chrome to handle all mailto:requests and forward them into my Gmail Webmail interface. When I tested your link, and modified your windowname in jsfiddle ( http://jsfiddle.net/sHYW8/2/ ), Windows asked me if IE could open Google Chrome to Handle the Protocol.
Short answer: what you ask is technically impossible unless you force all your users to install a third party addon for IE. This is the result of Internet Explorer being a part of the Windows Operating System, and Mozilla Firefox is a third party browser that is forced to handle protocols in its own way.
UPDATE
I found a jQuery plugin that uses the API for Gmail, Yahoo! and MSN. It's not a popup, but more of a rollover. I think this is going to be your closest bet.
http://kevin-cantwell.github.com/webmailto/
Good thing for you is that implementation seems easy enough. I would look at the bottom example, it looks pretty slick.
try this:
function doMailto(EmailAddress) {
document.location.href = 'mailto:' + EmailAddress;
}
I think your IE is preventing pop up windows created by javascript.
Just to be clear...
Adriano's suggestion of just using a normal html tag would also work.
Like this:
<a id="LnkEmail" href="mailto:d#s.com">
And as Vishal and Kyle Macey tried to explain:
That "Launch Application" window that pops up in Firefox... that is not a window you can create from a web page. That is Firefox's own window that it shows when a mailto: link is clicked. IE does not offer the same type of window. It usually just opens your default mail client (in your case it would probably be Outlook).
and finally...
Javascript is not the same as JQuery.
JQuery is written in Javascript but JQuery is NOT Javascript.
For IE 7 and 8 only you can't use any space in the window name. Try to change your code to:
window.open('mailto:' + EmailAddress, 'Mail');
If you really want such a list there is a way with pure javascript although it might not be the exact same experience as you currently have in firefox. What you could do is create a modal dialog with javascript showing a number of popular webmail clients and an option "default system client" instead of "Microsoft Office Outlook". The "Choose an Application" would be impossible to include as well. Next, if the user selects the native client you would simply trigger a mailto link as you currently do and in case the user selects for example gmail you trigger a window open of a link along the lines of
https://mail.google.com/mail/?view=cm&fs=1&tf=1&source=mailto&to=info#example.com&body=the+body+of+your+message
with your own variables of course from your mailto link. You would have to figure out the relevant links for different webmail services yourself, but as far as I know, most have these kind of links and gmail and yahoo have for sure.
Below is the working code as you mention
window.open('mailto:' + EmailAddress, 'newwindow');
its working but like FF IE not provide you option to choose mail engine.
If you want to run your code you have to set the default program for mail using set default program.
And you can set only Outlook as the default program. In out look you can bind any thing like yahoo or gamil that way you can use your mailto code for IE.
I think You have to doing coding for that because IE not provide any add on like FF.
For That first you have to chekc if default client mai lis there or not by following code
RegistryKey hkey = Registry.ClassesRoot.OpenSubKey(
"mailto\shell\open\command", false);
if this key is null then no default client is there. so you have to show the mail provides list on popup. and selected provider you have to set as default client mail.
http://msdn.microsoft.com/en-us/library/microsoft.win32.registry.classesroot(v=vs.90).aspx
using above link you can find list of mailto registered on machine for list display.
http://www.pcreview.co.uk/forums/re-add-dword-value-registry-t1401434.html
this link show how to set value in registry.
then execute your mailto code.
The mailing list is an utility features provided by Firefox only. You may or may not not find one software's feature on another similar one. If you don't, you should settle for a work around.
Try to remember that in firefox once the user selects a default mail client, you will not get the popup anymore. So there is no use of attempting to create a solution, that is not going to be permanent.
To trim down your requirement, you are trying to select the mail client of the user. But a website cannot changed the system settings of the user, its simply not allowed. Why? Because it opens many vulnerabilities to the user, if this was somehow allowed.