I want to add data to the current webpage with jQuery before I print the webpage. The reason is that I have a digital CV on my website and don't want to have my personal information on the website available to crawlers etc. I created an XHR request to get my personal information before printing and want to inject the data into HTML.
I found this script to find out with jQuery when a user is printing. This works fine in general but the problem is that the script works "too slow", meaning that the data will not be appended before the print dialogue pops up. This is my HTML code:
<div>
<span id="cvnameplaceholder">My Name</span><br />
<span id="cvstreetplaceholder"></span><br />
<span id="cvcityplaceholder"></span><br /><br />
<span id="cvemailplaceholder"></span>
</div>
And this is the jQuery code which returns an array like {street: 'ABC', city: 'DEF', email: 'mail#example.com}
$(document).ready(function() {
var beforePrint = function() {
$.post("./api/RequestHandler.php", {request : "getCVInformation"}, function (response) {
response = JSON.parse(response);
$('#cvstreetplaceholder').text(response.street);
$('#cvcityplaceholder').text(response.city);
$('#cvemailplaceholder').text(response.email);
})
};
if (window.matchMedia) {
var mediaQueryList = window.matchMedia('print');
mediaQueryList.addListener(function(mql) {
if (mql.matches) {
beforePrint();
}
});
}
window.onbeforeprint = beforePrint;
}());
When I hit the print button a second time, the information is there. So is there any possibility to add some text before showing the actual print dialogue?
A good number of crawlers now also either parse the js, or execute it and parse the result, so this may not keep your info private.
That said, it looks as though your problem is that you're not stopping the event, you're just hooking into it. The event listener is calling beforePrint, but it's not stopping anything else from also triggering on that event - in this case, the actual print dialog.
Additionally, you're loading the data via an ajax request, which is asynch. If you were somehow storing the data in a local javascript variable and populating the page, you might get lucky and have the text loaded in before the print dialog actually opened. But you're running into the same kind of problem that's mentioned on the page linked from that other answer, https://www.tjvantoll.com/2012/06/15/detecting-print-requests-with-javascript/ , where someone had a mix of low and high-res images actually printing. That is, the ajax request is sent before the print event bubbles up and triggers the print dialog, but the response is coming back after the browser's already rendered the page for printing.
You need to use stop the event from propagating (https://api.jquery.com/event.stoppropagation/ ), then re-trigger the print event within the post's callback, after you actually have the info. That's why you're seeing the info in there that second time; it's actually the info you loaded the first time that you're seeing.
Here's an example of catching and blocking the event, then re-triggering it after your data's arrived and been loaded:
$(document).ready(function() {
var beforePrint = function(e) {
e.stopPropagation();
$.post("./api/RequestHandler.php", {request : "getCVInformation"}, function (response) {
response = JSON.parse(response);
$('#cvstreetplaceholder').text(response.street);
$('#cvcityplaceholder').text(response.city);
$('#cvemailplaceholder').text(response.email);
window.print();
});
};
if (window.matchMedia) {
window.matchMedia('print').addListener(function(mql) {
if (mql.matches) {
beforePrint();
}
});
}
window.onbeforeprint = beforePrint;
}());
It's worth noting, though, that as mentioned on that linked page, the onbeforeprint event is IE5+ and Firefox 6+, window.matchMedia is Chrome 9+ and Safari 5.1, while Opera doesn't support either, and both seem to have various unsolved issues. So you very well might still end up with a printed CV that's lacking your contact info.
I know this does not directly answer the question, but you could iframe in your CV content and use nofollow/noindex to prevent indexing. This solution deals with appending content, but it does not use jQuery.
index.html
<html>
<head>
<style>
iframe {
border: none;
}
.cv {
display: none;
}
#media print {
.cv {
display: block;
}
}
</style>
</head>
<div class="cv">
<iframe src="cv.html"></iframe>
</div>
</html>
cv.html
<html>
<head>
<meta name="robots" content="noindex,nofollow">
</head>
<body>
My CV Details
</body>
</html>
Related
Some websites have lots of images, so lazyloading seems appropiate to reduce load times and data consumption. But what if you also need to support printing for that website?
I mean, you can try to detect the print event and then load the images, with something like this:
HTML
<img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7">
Note: this is a one by one pixels gif dummy image.
JavaScript
window.addEventListener('DOMContentLoaded', () => {
img = document.querySelector('img');
var isPrinting = window.matchMedia('print');
isPrinting.addListener((media) => {
if (media.matches) {
img.src = 'http://unsplash.it/500/300/?image=705';
}
})
});
Note: if you try this in a code playground, remove the DOMContentLoaded event (or simply fork these: JSFiddle | Codepen).
Note: I didn't event bother with the onbeforeprint and onafterprint for obvious reasons.
This will work fine if the image is cached, but then again that's precisely not the point. The image/s should all load and then appear in the printing screen.
Do you have any ideas? Has anyone successfully implemented a print-ready lazyloading plugin?
Update
I've tried redirecting the user after the print dialog is detected, to a flagged version of the website a.k.a website.com?print=true where lazyloading is deactivated and all images load normally.
This method is improved by applying the window.print() method in this flagged print-ready version of the page, opening a new print dialog once all images are finished loading, and showing a "wait for it" message in the meantime at the top of the page.
Important note: this method was tested in Chrome, it does not work in Firefox nor Edge (hence why this is not an answer, but a testimony).
It works in Chrome beacuse the print dialog closes when you redirect to another website (in this case same url but flagged). In Edge and Firefox the print dialog is an actual window and it does not close it, making it pretty unusable.
Based on your desired functionality, I'm not quite sure what you want to do is feasible. As a developer we don't really have control over a users browser. Here are my thoughts as to why this isn't fully possible.
Hooking the event to go and load your missing images won't let you guarantee images will make it from the server into your page. More specifically, the PDF generated for your print preview is going to get generated before your image(s) is done loading, the img.src = "..." is asynchronous. You'd run into similar issues with onbeforeprint as well, unfortunately. Sometimes it works, sometimes it does not (example, your fiddle worked when testing in safari, but did not in Chrome)
You cannot stall or stop the print call -- you can't force the browser to wait for your image to finish loading in the lazy loading context. (I read something about using alerts to achieve this once, but it seemed really hacky to me, was more of a deterrent to printing than stalling)
You cannot force img.src to get that data synchronously in a lazy-load context. There are some methods of doing this, but they are clever hacks -- referenced as pure evil and may not work in always. I found another link with a similar approach
So we have a problem, if the images are not loaded by the time print event is fired, we cannot force the browser to wait until they are done. Sure, we can hook and go get those images on print, but as above points out, we cannot wait for those resources to load before the print preview pops up.
Potential solution (inspired by links in point three as well as this link)
You could almost get away with doing a synchronous XMLHttpRequest. Syncrhonous XMLHTTPRequests will not let you change the responseType, they are always strings. However, you could convert the string value to an arrayBuffer encode it to a base-64 encoded string, and set the src to a dataURL (see the link that referenced clever hacks) -- however, when I tried this I got an error in the jsfiddle -- so it would be possible, if things were configured correctly, in theory. I'm hesitant to say yes you can, since I wasn't able to get the fiddle working with the following (but it's a route you could explore!).
var xhr = new XMLHttpRequest();
xhr.open("GET","http://unsplash.it/500/300/?image=705",false);
xhr.send(null);
if (request.status === 200) {
//we cannot change the resposne type in synchronous XMLHTTPRequests
//we can convert the string into a dataURL though
var arr = new Uint8Array(this.response);
// Convert the int array to a binary string
// We have to use apply() as we are converting an *array*
// and String.fromCharCode() takes one or more single values, not
// an array.
var raw = String.fromCharCode.apply(null,arr);
// This is supported in modern browsers
var b64=btoa(raw);
var dataURL="data:image/jpeg;base64,"+b64;
img.src = dataURL;
}
Work around to enhance the user experience
Something you could do is have some text that only displays in the print version of your page (via #print css media) that says "images are still loading, cancel your print request and try again" and when the images are finished loading, remove that "still waiting on resources try again message" from the DOM. Farther, you could wrap your main content inside an element that inverses the display to none when content is not loaded, so all you see is that message in the print preview dialog.
Going off of the code you posted this could look something like the following (see updated jsfiddle):
CSS
.printing-not-ready-message{
display:none;
}
#media print{
.printing-not-ready-message{
display:block;
}
.do-not-print-content{
display:none;
}
}
HTML
<div class="printing-not-ready-message">
Images are still loading please cancel your preview and try again shortly.
</div>
<div class="do-not-print-content">
<h1>Welcome to my Lazy Page</h1>
<img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7">
<p>Insert some comment about picture</p>
</div>
JavaScript
window.addEventListener('DOMContentLoaded', () => {
img = document.querySelector('img');
var isPrinting = window.matchMedia('print');
isPrinting.addListener((media) => {
if (media.matches) {
img.src = 'http://unsplash.it/500/300/?image=705';
//depending on how the lazy loading is done, the following might
//exist in some other call, should happen after all images are loaded.
//There is only 1 image in this example so this code can be called here.
img.onload = ()=>{
document.querySelector(".printing-not-ready-message").remove();
document.querySelector(".do-not-print-content").className=""
}
}
})
});
I'm the author of the vanilla-lazyload script and I've recently developed a feature that makes print of all images possible!
Tested cross browser using this repo code which is live here.
Take a look and let me know what you think!
I'm open to pull requests on GitHub of course.
I wrote a lazy loading jquery plugin that supports showing images on print using the window.onbeforeprint events and mediaQueryListeners.
https://github.com/msigley/Unveil-EX/
//declare custom onbeforeprint method
const customOnBeforePrint = () => {
const smoothScroll = (h) => {
let i = h || 0;
if (i < 200) {
setTimeout(() => {
window.scrollTo(window.scrollY, window.scrollY + i);
smoothScroll(i + 10);
}, 10);
}
};
let height = document.body.scrollHeight;
let newHeight;
while (true) {
smoothScroll(100);
if (newHeight === height) break;
height = newHeight;
}
};
//override the onbeforeprint method
window.onbeforeprint = customOnBeforePrint;
Copy&Paste that block into devtool's console and then try to click print button. That workaround is working for me.
For whoever is in the same boat as I was: when using the browser native loading="lazy", you can simply remove that attribute when printing is going to happen. Below is my jQuery implementation.
window.onbeforeprint = function () {
$('img').each(function () {
$(this).removeAttr('loading')
});
}
Chrome will then just load all images and they will show up when printing.
I'm currently writing a little program that generates an html file and opens it with the default browser to start multiple downloads.
I don't want to open a tab/window for every download, so creating hidden iframes for the downloads seemed like a good solution.
I'm using onload on the iframes to find out if the download prompts for each download have shown up yet. This approach seems to be very unreliable in the Internet Explorer though.
So I'm wondering if there is there a way to fix this or maybe a better approach?
(Without libraries please.)
Here is my html/js code:
<!DOCTYPE html>
<!-- saved from url=(0016)http://localhost -->
<html><head>
<meta content="text/html;charset=utf-8" http-equiv="Content-Type">
<meta content="utf-8" http-equiv="encoding">
<title>Downloads</title>
<script>
"use strict";
var downloadsInfo = {
"http://download-installer.cdn.mozilla.net/pub/firefox/releases/26.0/win32/en-US/Firefox%20Setup%2026.0.exe":"Status: Connecting",
"http://download.piriform.com/ccsetup410.exe":"Status: Connecting"
};
var i = 0;
var iv = setInterval(function() {
i = ++i % 4;
var j = 0;
var finished = true;
for (var key in downloadsInfo) {
var value = downloadsInfo[key];
if (value != "Status: Download Started!") {
value = value+Array(i+1).join(".");
finished = false;
}
document.getElementsByTagName("div")[j].innerHTML = key+"<br/>"+value;
j = j+1;
}
if (finished) {
alert('Done! You can close this window/tab now.');
clearInterval(iv);
}
}, 800);
</script>
</head><body>
<h3>Please wait for your downloads to start and do not reload this site.</h3>
<div></div> <br/><br/>
<div></div> <br/><br/>
<iframe src="http://download-installer.cdn.mozilla.net/pub/firefox/releases/26.0/win32/en-US/Firefox%20Setup%2026.0.exe" onload="downloadsInfo['http://download-installer.cdn.mozilla.net/pub/firefox/releases/26.0/win32/en-US/Firefox%20Setup%2026.0.exe'] = 'Status: Download Started!';" style="display:none"></iframe>
<iframe src="http://download.piriform.com/ccsetup410.exe" onload="downloadsInfo['http://download.piriform.com/ccsetup410.exe'] = 'Status: Download Started!';" style="display:none"></iframe>
</body></html>
Quite simply you can't know whether a native browser download started. Every browser has different ways this is handled, the user may set up his browser to prompt the location or he might just let it auto download to the Downloads folder (the default in most browsers nowadays). If he's prompting for a location he might cancel by mistake, yet your setup would still claim the download started. So, no, there is no way whatsoever to reliably inform the user that they can close a tab once all downloads are started/finished... provided that you use the native browser download mechanism.
The way to achieve this effect would be possibly by first downloading the file using Javascript (requiring you to have access to those files, hotlinking to third party files is of course not an option then). To see this in action try downloading a file from mega.nz. I was planning on writing up how to do this by hand, but there is already a nice (quite outdated) answer outlining this.
If the intention is only to ensure that the download has started you could implement a trigger on the back end to note when the file has been accessed. In it's simplest form this would look like:
Page download.html requests file.php?location=[...]&randomHash=1234
Once file.php is actually loaded it will set a flag in memory or the database that randomHash id 1234 has started.
file.php redirects the page with a 302 header to the actual file location.
download.html checks periodically using Ajax whether flag randomHash=1234 has been raised. If so it knows the download has started.
Indeed IE is reported to not always behave nicely with the onload event handler of iframes. There is an active bug tracker record opened.
The problem is discussed in a number of places around the web, and what seems to be the most reliable solution is to have an indirect download with nested iframes: the iframe loads a HTML file with an iframe that loads the file to download. The reason for that is that IE does not seem to like iframes that point to something else than HTML. So if you have the possibility to do that in your program:
For each file to download, generate a HTML file with a body that looks like this:
<iframe src="http://filetodownload.exe" style="display:none"></iframe>
Store this file in a temporary folder, e.g. C:\tmp\filetodownload.html
In your "master" generated HTML file, replace the iframe source with this intermediate file:
<iframe src="C:\tmp\filetodownload.html"
onload="downloadsInfo['http://filetodownload.exe']='Status: Download Started!';"
style="display:none"></iframe>
That may do the trick. But following IE's tradition, this could or could not work depending on the case...
If it does not work, some solutions that have proved useful include:
Put the onload handler in a function, and write in the definition of the iframe: onload="return theonloadfunction()" (even if the function does not return anything)
Instead of using the onload attribute, attach the event handler in javascript, like so:
iframe = document.getElementById("theiframeid")
iframe.attachEvent("onload", theonloadfunction);`
Note that attachEvent is for IE only. If you want to support other browsers you will have to detect it and use addEventListener for the non-IE cases.
Finally, you may try combinations of two or more of these solutions :)
<html>
<head>
<meta content = 'text/html;charset=utf-8' http-equiv = 'Content-Type'>
<meta content = 'utf-8' http-equiv = 'encoding'>
<script>
/*
First, I removed the setInterval(). Since you rely on the onload property we can aswell just check it on each onload.
Second, I changed your downloadsInfo to an object array.
Also be aware while testing, that some browsers cache your cancel/block choice and do not reask again for the same url.
Additionally firefox does not fire on frame downloads.
Furthermore the alert in your test might not show for overlapping or setting reasons.
*/
var downloadsInfo = [
{url: "http://download-installer.cdn.mozilla.net/pub/firefox/releases/26.0/win32/en-US/Firefox%20Setup%2026.0.exe", Status: "Connecting"},
{url: "http://download.piriform.com/ccsetup410.exe", Status: "Connecting"}
];
//IE has a problem in sometimes merely firing the onload propery once, which we bypass by dynamically creating them
//It is also less limited.
function iframeConnect(){
for(var i=0, j=downloadsInfo.length; i<j; i++){
var tF = document.createElement('iframe');
tF.arrayIndex = i; //For convenience
tF.style.display = 'none';
//Normal load event, working in ie8-11, chrome, safari
tF.onload = function(){
iframeExecuted(this.arrayIndex);
};
//Workaround for firefox, opera and some ie9
tF.addEventListener('DOMSubtreeModified', function(){
iframeExecuted(this.arrayIndex);
}, false);
document.body.appendChild(tF);
tF.src = downloadsInfo[i].url;
}
}
function iframeExecuted(i){
downloadsInfo[i].Status = 'Executed';
var tStatus = iframeFinished();
var tE = document.querySelector('h3');
if (tStatus.Done) tE.innerHTML = 'Finished'
else tE.innerHTML = 'Processed ' + tStatus.Processed + ' of ' + tStatus.Started;
}
function iframeFinished(){
for(var i=0, j=downloadsInfo.length; i<j; i++){
if (downloadsInfo[i].Status != 'Executed') break;
}
//Note that the Processed value is not accurate, yet it solves is testing purpose.
return {Done: (i == j), Processed: i, Started: j}
}
</script>
</head>
<body onload = 'iframeConnect()'>
<h3>Please wait for your downloads to start and do not reload this site.</h3>
</body>
</html>
Is there a working example how you can use the zxing Barcode Scanner from a web page?
Referring to this documentation:
https://github.com/zxing/zxing/wiki/Scanning-From-Web-Pages
shouldn't the following test code work?
function Test1()
{
$.ajax(
{
url: "zxing://scan/?ret=http%3A%2F%2Ffoo.com%2Fproducts%2F%7BCODE%7D%2Fdescription&SCAN_FORMATS=UPC_A,EAN_13",
success:function()
{
alert("success");
},
error:function()
{
alert("error");
}
});
}
function Test2()
{
$.ajax(
{
url: "http://zxing.appspot.com/scan?ret=http%3A%2F%2Ffoo.com%2Fproducts%2F%7BCODE%7D%2Fdescription&SCAN_FORMATS=UPC_A,EAN_13",
success:function()
{
alert("success");
},
error:function()
{
alert("error");
}
});
}
<script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.1/jquery.min.js"></script>
<button id="button1" onClick="Test1();">Test 1</button>
<br>
<br>
<button id="button2" onClick="Test2();">Test 2</button>
I keep getting "error" on my Android 4.4.2 Samsung Galaxy TabPro and Samsung Galaxy S4.
I've tried the stock browser, Chrome, Firefox and Dolphin Browser.
Even http://zxing.appspot.com/scan doesn't work as it always asks me to install the (already installed) app.
Any help would be much appreciated.
ZXing isn't designed to work with AJAX. Instead, it works by opening a parsed URL in the default browser. The behavior of the browser is mainly what's responsible for the user experience from that point forward.
There are several methods posted regarding this; unfortunately, there is no one method that will work for every browser.
Some browsers, when you open them from the command line, will check to see if the URL is already opened in another tab, and if so, will use that tab instead of a new one. This will cause a "onhashchange" event if the zxing link contains "zxing://scan/?ret=mytab.html#{CODE}".
Other browsers don't perform such a check, so we wind up with multiple tabs, all having the same URL (with the exception of the hash), and none of them raising the "hashchanged" event. For those browsers, we need to re-use the page from cache if possible (to prevent network traffic on every scan), and change the localStorage value to what the hash is. If the browser is capable of listening for the "storage" event, we can use that to trigger the code.
The code below works with Chrome, the intrinsic Android browser, and Firefox. It might work with others, but I haven't tried. One Firefox caveat, though, is that the scanner window will only close if the about:config setting "dom.allow_scripts_to_close_windows" is set to "true".
** This was edited to work better with multiple pages that allow scans, and now you can use have different hashes without interfering with the code. **
NEW VERSION 12/19/16
<!DOCTYPE html>
<HTML>
<HEAD>
<script type="text/javascript">
if(window.location.hash.substr(1,2) == "zx"){
var bc = window.location.hash.substr(3);
localStorage["barcode"] = decodeURI(window.location.hash.substr(3))
window.close();
self.close();
window.location.href = "about:blank";//In case self.close isn't allowed
}
</script>
<SCRIPT type="text/javascript" >
var changingHash = false;
function onbarcode(event){
switch(event.type){
case "hashchange":{
if(changingHash == true){
return;
}
var hash = window.location.hash;
if(hash.substr(0,3) == "#zx"){
hash = window.location.hash.substr(3);
changingHash = true;
window.location.hash = event.oldURL.split("\#")[1] || ""
changingHash = false;
processBarcode(hash);
}
break;
}
case "storage":{
window.focus();
if(event.key == "barcode"){
window.removeEventListener("storage", onbarcode, false);
processBarcode(event.newValue);
}
break;
}
default:{
console.log(event)
break;
}
}
}
window.addEventListener("hashchange", onbarcode, false);
function getScan(){
var href = window.location.href;
var ptr = href.lastIndexOf("#");
if(ptr>0){
href = href.substr(0,ptr);
}
window.addEventListener("storage", onbarcode, false);
setTimeout('window.removeEventListener("storage", onbarcode, false)', 15000);
localStorage.removeItem("barcode");
//window.open (href + "#zx" + new Date().toString());
if(navigator.userAgent.match(/Firefox/i)){
//Used for Firefox. If Chrome uses this, it raises the "hashchanged" event only.
window.location.href = ("zxing://scan/?ret=" + encodeURIComponent(href + "#zx{CODE}"));
}else{
//Used for Chrome. If Firefox uses this, it leaves the scan window open.
window.open ("zxing://scan/?ret=" + encodeURIComponent(href + "#zx{CODE}"));
}
}
function processBarcode(bc){
document.getElementById("scans").innerHTML += "<div>" + bc + "</div>";
//put your code in place of the line above.
}
</SCRIPT>
<META name="viewport" content="width=device-width, initial-scale=1" />
</HEAD>
<BODY>
<INPUT id=barcode type=text >
<INPUT style="width:100px;height:100px" type=button value="Scan" onclick="getScan();">
<div id="scans"></div>
</BODY>
</HTML>
You can make a JS include file for the top block of script, and include it on all the pages where you need scanning capabilities.
Then in the body of your document, you can set an event somewhere to call getZxing(), which will call processBarcode(barcode) that you write into your page. Included is a simple one for example's sake.
Side Note: The first time you run zxing from your page, you'll be asked to choose a default app. Make sure you chose the same browser that you're running the page from. Additionally, if you previously selected a default broswer for zxing and want to change which browser you use for zxing, you'll need to clear defaults from your other browsers.
Many thanks to #sean-owen for his hard work and fantastic product.
UPDATE 12/19/16
Ok, I did a slightly more robust version that works well with Firefox and Chrome. A couple of things I discovered:
Chrome will use the Storage event if the scanner is not set to open Chrome automatically, and will use the Hash event after it becomes default.
Firefox will never use the Hash event, but opens an extra window unless you call the scanner with window.location.href (Thanks, #Roland)
There are a couple of other anomalies, but no deal breakers.
I left the "zx" prefix in the hash, so that the code could delineate between scanner hashes and regular hashes. If you leave it in there, you'll not notice it in the processBarcode function, and non-zx hashes will operate as expected.
I've the same problem depicted in iOS 5 pauses JavaScript when tab is not active thread.
My question is if I can be noticed when come back to the paused tab.
onfocus and onblur events don't work on to the to be paused tab.
The code:
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<script src="http://code.jquery.com/jquery-1.7.1.min.js" ></script>
<script type="text/javascript">
window.onblur = function () {
console.log("blur");
$("#whatevent").append("blur<br/>");
}
window.onfocus = function () {
console.log("focus");
$("#whatevent").append("focus<br/>");
}
window.onunload = function () {
console.log("unload");
$("#whatevent").append("unload<br/>");
}
window.onload = function () {
console.log("load");
$("#whatevent").append("load<br/>");
}
</script>
</head>
<body>
<div id="whatevent"></div>
</body>
</html>
none but onload (but only the first time I load the page) events works on iPad when I switch tab.
Someone posed this very question over two years ago with this query. Unfortunately, it was met with only a couple of answers, one of which seems to be the only method to achieve this effect. Until Apple can implement the full Page Visibility API in mobile safari, I'm left with using this custom object I created that will use the API and fall back to a heart beat ticker if it's unavailable. However, as far as I can tell, there is no great way to check an imminent tab switch.
Here's a basic fiddle of the object demonstrating its only real method. It essentially just accepts a handler function for a focus event that gets fired whenever the browser returns the source tab. The fallback is hacky at best and will fire not just on a page re-entry but whenever scripting stops for longer than the timer threshold; which could be whenever the keyboard is visible, on scroll, or if a running script prevents the requestAnimationFrame from firing. Since scrolling is the most common behavior, I've added a handler that resets the last saved time so that the focus event refrains from firing.
This is the main portion of the script that includes the "hacky" method as described above:
_that.onFocus = function(handler, params) {
var hiddenProp = getHiddenProp();
console.log("Hidden prop: " + hiddenProp);
if (hiddenProp) {
var evtName = hiddenProp.replace(/[H|h]idden/, "") + "visibilitychange";
document.addEventListener(evtName, function(e) {
if (isHidden()) {
handler(e, params);
}
}, false);
}else {
var handlerObj = {"handler": handler};
if (params !== undefined) {handlerObj.params = params}
_handlers.push(handlerObj);
startLoop();
}
};
The rest may be read in the fiddle. In order to see the fallback you'll have to use a tablet (why else would you be needing this function without one?).
Note that the .onFocus method may accept an array of params for its second param that it will then pass to your event handler. This means that your event handler will always have an event object for it's first param (or null if the API is not supported) and your array of params as its second param.
Also not that this code has been tested for all of a couple hours so it may be prone to glitchiness. I would appreciate any constructive criticism to make it production worthy until Mobile Safari gets its butt in gear.
First, the background:
I'm working in Tapestry 4, so the HTML for any given page is stitched together from various bits and pieces of HTML scattered throughout the application. For the component I'm working on I don't have the <body> tag so I can't give it an onload attribute.
The component has an input element that needs focus when the page loads. Does anyone know a way to set the focus to a file input (or any other text-type input) on page load without access to the body tag?
I've tried inserting script into the body like
document.body.setAttribute('onload', 'setFocus()')
(where setFocus is a function setting the focus to the file input element), but that didn't work. I can't say I was surprised by that though.
EDIT:
As has been stated, I do indeed need to do this with a page component. I ended up adding file-type inputs to the script we use for giving focus to the first editable and visible input on a page. In researching this problem I haven't found any security issues with doing this.
<script>
window.onload = function() {
document.getElementById('search_query').select();
//document.getElementById('search_query').value = '';
// where 'search_query' will be the id of the input element
};
</script>
must be useful i think !!!
This has worked well for me:
<script>
function getLastFormElem(){
var fID = document.forms.length -1;
var f = document.forms[fID];
var eID = f.elements.length -1;
return f.elements[eID];
}
</script>
<input name="whatever" id="maybesetmaybenot" type="text"/>
<!-- any other code except more form tags -->
<script>getLastFormElem().focus();</script>
you can give the window an onload handler
window.onload = setFocus;
I think you have a fundamental problem with your encapsulation. Although in most cases you could attach an event handler to the onload event - see http://ejohn.org/projects/flexible-javascript-events/ by John Resig for how to do this, setFocus needs to be managed by a page component since you can't have two components on your page requiring that they get the focus when the page loads.
Try play with tabstop attribute
First of all, the input file is no the same as the other inputs, you need to keep this in mind.... thats for security reasons. When the input file get focus it should be read only or the browser should popup a dialog to choose some file.
Now, for the other inputs you could try some onload event on some of your elements...(not only the body have the onload event) or you could use inline javascript in the middle of the html. If you put javascript code without telling that is a function it gets executes while the browser reads it. Something like:
<script type="text/javascript">
function yourFunction()
{
...;
};
alert('hello world!");
yourFunction();
</script>
The function will be executed after the alert just when the browser reads it.
If you can, you should use jQuery to do your javascript. It will make your live soooo much easy.... :)
With jQuery could be done like this:
$(function() {
$("input:file").eq(0).focus()
})
With plain javascript could be done like this:
var oldWindowOnload = window.onload; // be nice with other uses of onload
window.onload = function() {
var form = document.forms[0];
for(i=0; i < form.length; i++) {
if (form[i].type == "file") {
form[i].focus();
}
}
oldWindowOnload();
}
For more elaborate solution with plain javascript see Set Focus to First Input on Web Page on CodeProject.
Scunliffe's solution has a usability advantage.
When page scripts are loading slowly, calling focus() from "onLoad" event makes a very nasty page "jump" if user scrolls away the page. So this is a more user friendly approach:
<input id="..."></input>
... really small piece of HTML ...
<script>getTheDesiredInput().focus();</script>