Develop Reference

What is different about xmlhttprequest in Firefox

My code works in Chrome and Safari, but it hangs in FF.
I removed the parts of the code that aren't necessary.
I used console commands to show how far the first loop gets, and it will do the second log fine right before the xhr open and send commands.
If the open/send commands are present the loop only happens once, if I remove the open/send commands the loop completes successfully.
Currently using FF 62nightly, but this issue has plagued me since Quantum has come out and I'm now trying to figure out why it doesn't work right.
for (i = 0; i < length; i++) {
(function(i) {
// new XMLHttpRequest
xhr[i] = new XMLHttpRequest();
// gets machine url from href tag
url = rows[i].getElementsByTagName("td")[0].getElementsByTagName('a')[0].getAttribute('href');
// Insert the desired values at the end of each row;
// will try to make this customizable later as well
insertVNC[i] = rows[i].insertCell(-1);
insertSerial[i] = rows[i].insertCell(-1);
insertVersion[i] = rows[i].insertCell(-1);
insertFreeDiskSpace[i] = rows[i].insertCell(-1);
// the fun part: this function takes each url, loads it in the background,
// retrieves the values needed, and then discards the page once the function is complete;
// In theory you could add whatever you want without taking significantly longer
// as long as it's on this page
console.log(i);
xhr[i].onreadystatechange = function() {
if (xhr[i].readyState == 4 && xhr[i].status == 200) {
}
};
//"Get" the "Url"... true means asyncrhonous
console.log(url);
xhr[i].open("GET", url, true);
xhr[i].send(null);
})(i); //end for loop
}

I cannot tell you why it gives issues in Firefox. I would not trust sending off arbitrarily many requests from any browser
I would personally try this instead since it will not fire off the next one until one is finished
const urls = [...document.querySelectorAll("tr>td:nth-child(0) a")].map(x => x.href);
let cnt=0;
function getUrl() {
console.log(urls[cnt]);
xhr[i].open("GET", urls[cnt], true);
xhr[i].send(null);
}
let xhr = new XMLHttpRequest();
xhr.onreadystatechange = function() {
if (xhr[i].readyState == 4 && xhr[i].status == 200) {
if (cnt>urls.length) getUrl();
cnt++;
}
}
getUrl();

Combine two AJAX without jQuery

This is what I have and it works about 80% of the time but every once in a while it will crash. Can any combine these better? or at least check that both have returned a valid result before displaying. I tried setting as two separate functions and running them in tandem and that didn't work either for some reason. I would like to avoid jQuery if possible. Thanks
function doit(str){
var color = document.getElementById('button').value;
if (str == "") {
document.getElementById("Div1").innerHTML = "";
document.getElementById("Div2").innerHTML = "";
return;
} else {
if (window.XMLHttpRequest) {
ajaxRequest = new XMLHttpRequest();
ajaxRequesttwo = new XMLHttpRequest();
} else {
ajaxRequest = new ActiveXObject("Microsoft.XMLHTTP");
ajaxRequesttwo = new ActiveXObject("Microsoft.XMLHTTP");
}
ajaxRequest.onreadystatechange = function() {
if (ajaxRequest.readyState == 4 && ajaxRequest.status == 200) {
document.getElementById("Div1").innerHTML = ajaxRequesttwo.responseText;
document.getElementById("Div2").innerHTML = ajaxRequest.responseText;
}
}
ajaxRequesttwo.open("GET","ajax/page1.php?r="+str+"&c="+color,true);
ajaxRequest.open("GET","ajax/page2.php?q="+str,true);
ajaxRequesttwo.send();
ajaxRequest.send();
}
}

If I understand correctly, you want to execute two AJAX requests, and set the output to the two divs after both are complete.
Option 1: Only let the latest AJAX request handler (can be either request) do the output display.
function doitnow(str) {
// DOM elements / values
var div1 = document.getElementById("Div1");
var div2 = document.getElementById("Div2");
var color = document.getElementById('button').value;
// reset results
if (str == "") {
div1.innerHTML = "";
div2.innerHTML = "";
return;
}
// Prepare AJAX requests & handlers
var ajaxRequest1 = new XMLHttpRequest();
var ajaxRequest2 = new XMLHttpRequest();
var result1, result2; // to store results of requests
ajaxRequest1.onreadystatechange = function() {
if (ajaxRequest1.readyState == 4 && ajaxRequest1.status == 200) {
result1 = ajaxRequest1.responseText;
attemptDisplay();
}
};
ajaxRequest2.onreadystatechange = function() {
if (ajaxRequest2.readyState == 4 && ajaxRequest2.status == 200) {
result2 = ajaxRequest2.responseText;
attemptDisplay();
}
};
function attemptDisplay() {
// display only if both results are set
if (typeof result1 !== 'undefined' && typeof result2 !== 'undefined') {
div1.innerHTML = result1;
div2.innerHTML = result2;
}
}
// Fire AJAX
ajaxRequest2.open("GET","ajax/page1.php?r="+str+"&c="+color,true);
ajaxRequest1.open("GET","ajax/page2.php?q="+str,true);
ajaxRequest2.send();
ajaxRequest1.send();
} //doitnow()
This should work. I didn't do the disgusting ActiveXObject thing because really, developers really shouldn't have to do this type of cross-browser "polyfilling" in the meat of the application. Use a proper XHR polyfill and henceforth, you may pretend that the legacy MS IE's AJAX construct doesn't exist. I recommend: https://github.com/Financial-Times/polyfill-service/blob/master/polyfills/XMLHttpRequest/polyfill.js
Option 2: ES-6 Promises
I know you're not keen on jQuery (and probably other libraries), but if you don't mind polyfilling your environment with ES-6 Promises, that can be an elegant solution as well, particularly through the use of Promise.all. It's essentially the same idea as Option 1, but abstracted away by the cleaner promises interface.

You're only checking the ready state of one of the two Ajax requests. What happens if the other isn't ready when the first one is? Perhaps try adding another handler separately. Also, unless you're supporting very old versions of IE (I think below 8 even) you don't need to create an ActiveXObject as a fallback.
ajaxRequesttwo.onreadystatechange = function(){...}

AJAX div not updating without alert call with error: b.data is undefined

First of all, any suggestions on rewriting my title?
Issue:
I have an AJAX updatable div, when my error checking alert calls are in, everything works perfect. But when I remove a specific one, the DIV never gets updated. I know it sounds confusing, but code is as follows, and then I will explain further.
AJAX SCRIPT
var xmlhttp
/*#cc_on #*/
/*#if (#_jscript_version >= 5)
try {
xmlhttp=new ActiveXObject("Msxml2.XMLHTTP")
} catch (e) {
try {
xmlhttp=new ActiveXObject("Microsoft.XMLHTTP")
} catch (E) {
xmlhttp=false
}
}
#else
xmlhttp=false
#end #*/
if (!xmlhttp && typeof XMLHttpRequest!='undefined') {
try {
xmlhttp = new XMLHttpRequest();
} catch (e) {
xmlhttp=false
}
}
function myXMLHttpRequest() {
var xmlhttplocal;
try {
xmlhttplocal= new ActiveXObject("Msxml2.XMLHTTP")
} catch (e) {
try {
xmlhttplocal= new ActiveXObject("Microsoft.XMLHTTP")
} catch (E) {
xmlhttplocal=false;
}
}
if (!xmlhttplocal && typeof XMLHttpRequest!='undefined') {
try {
var xmlhttplocal = new XMLHttpRequest();
} catch (e) {
var xmlhttplocal=false;
alert('couldn\'t create xmlhttp object');
}
}
return(xmlhttplocal);
}
function sndReq(page,key,includesDir,changeDiv,parameterString) {
var divToChange = document.getElementById(changeDiv); // the Div that the data will be put into
// Place loading image in container DIV
divToChange.innerHTML = '<div class="loading">Loading</div>';
if (includesDir == 1){
//Find Current Working Directory. Use to find path to call other files from
var myloc = window.location.href;
var locarray = myloc.split("/");
delete locarray[(locarray.length-1)];
var arraytext = locarray.join("/");
xmlhttp.open("POST","AJAXCaller.php",true);
xmlhttp.setRequestHeader("Content-type","application/x-www-form-urlencoded");
xmlhttp.send(parameterString);
} else {
}
xmlhttp.onreadystatechange = handleResponse(changeDiv);
xmlhttp.send(null);
}
function handleResponse(changeDiv) {
/* ======== If I remove the following line script halts ================*/
/* ======== If line stays here, script executes perfectly ==============*/
alert('to changetext 1\n'+xmlhttp.responseText);
/* =========End of line removal issue =================================*/
if(xmlhttp.readyState == 4){
if (xmlhttp.status == 200){
var response = xmlhttp.responseText;
var update = new Array();
if(response.indexOf('|') != -1) {
update = response.split('|');
changeText(update[0], update[1]);
} else {
changeText(changeDiv, response);
}
} //End IF xmlhttp.status == 200
}
}
function changeText( div2show, text ) {
// Detect Browser
var IE = (document.all) ? 1 : 0;
var DOM = 0;
if (parseInt(navigator.appVersion) >=5) {DOM=1};
// Grab the content from the requested "div" and show it in the "container"
if (DOM) {
var viewer = document.getElementById(div2show);
viewer.innerHTML = text;
} else if(IE) {
document.all[div2show].innerHTML = text;
}
}
When I check my Firefox error console, this error ONLY appears when I remove that alert as defined in the code:
Timestamp: 5/30/2012 5:07:55 PM
Error: b.data is undefined
Source File: http://cdn.sstatic.net/js/wmd.js?v=cfd2b283af83
Line: 92
I am an advanced PHP/mySQL developer, but have been trying hard to grasp AJAX/JavaScript. Doing tutorials like mad. So please be descriptive in comments/answers so I can use them as a reference for learning...
Why would displaying an alert box alter code execution (for the better!) in any way?
NEW ERRORS - Google Chrome and Firefox Console (sigh...)
Uncaught Error: INVALID_STATE_ERR: DOM Exception 11
sndReqAJAX.js:89
element.onclick
Line 89 is the following (verified by Google Chrome Console)
xmlhttp.send(null);
Everything I find on the web refers to extremely complex issue regarding DOM objects not existing... This wouldn't apply here, would it?

First, the problem. This is the line:
xmlhttp.onreadystatechange = handleResponse(changeDiv);
The reason this is wrong is that xmlhttp.onreadystatechange should be a function - you need to assign a function to the property, what you are doing is calling the function and assigning the return value to the property. This is not in itself a problem, as long as your function returns a function. Which it doesn't.
If you're used to working with PHP (especially if your used to working with PHP <5.3) you may not be used to this concept. Javascript has support for closures in a way that anyone who doesn't use them much will find confusing.
What the line needs to look like is this:
xmlhttp.onreadystatechange = handleResponse;
By appending (changeDiv) you are calling the function, whereas what you need to do is simply pass it, like you would any other value.
Now, where it gets complicated is that you want to pass an argument that is local to the scope of the calling function. There are a number of ways to handle this, but a cursory look at your code tells me that handleResponse() is not used anywhere else, so it would be better to define this as a closure and not litter the global scope with a named event handler. This also overcomes the variable scoping problem:
xmlhttp.onreadystatechange = function() {
if (xmlhttp.readyState == 4) {
if (xmlhttp.status == 200) {
var response = xmlhttp.responseText;
var update = []; // "new Array()" is baaaad
if (response.indexOf('|') != -1) {
update = response.split('|');
changeText(update[0], update[1]);
} else {
changeText(changeDiv, response);
}
} //End IF xmlhttp.status == 200
}
};
Replace the aforementioned offending line with that block of code, and remove the handleResponse() function definition, and that should solve the immediate problem. Now, as to why the alert() "fixes" your original code - this is a little hard to explain, but I shall have a go... give me a minute to inspect the code properly
Attempt at a full and comprehensible explanation abandoned. If anyone wants one, post a comment and I'll have another go at it when I've had some sleep...

The entity name must immediately follow the '&' (...)

Seems like a common error, but none of the solutions I've found searching have proved successfull (replacing & with & is one). I have a simple Javascript (AJAX) that includes a couple of double if statements. The script works fine in both Chrome and Firefox, but not in IE9 which reports an error on the same line as Netbeans (the entity name must immediately follow the '&' (...)).
Hoping that someone here can spot the error or provide clues to where I shoud look.
function getNames(str) {
var xmlhttp;
if (str.length == 0) {
document.getElementById("txtHint").innerHTML = "";
return;
}
if (window.XMLHttpRequest) { // code for IE7+, Firefox, Chrome, Opera, Safari
xmlhttp = new XMLHttpRequest();
} else { // code for IE6, IE5
xmlhttp = new ActiveXObject("Microsoft.XMLHTTP");
}
xmlhttp.onreadystatechange = function () {
// The line below is what produces the error.
if (xmlhttp.status == 200 && xmlhttp.readyState == 4) {
var arr = xmlhttp.responseText.split(",");
var ul = document.getElementById("names");
var li = document.createElement("li");
if (ul.hasChildNodes()) {
while (ul.childNodes.length >= 1) {
ul.removeChild(ul.firstChild);
}
}
for (var i = 0; i < arr.length; i++) {
if (arr[i] != ":#") {
var li = document.createElement("li");
li.innerHTML = newListItem = arr[i];
ul.insertBefore(li, ul.getElementsByTagName("li")[0]);
}
}
}
}
xmlhttp.open("GET", "./ext/get_names.php?q=" + str, true);
xmlhttp.send();
}

You can enclose the script within a CDATA section:
<script type="text/javascript">
//<![CDATA[
document.write("&&");
//]]>
</script>

Unless you have a particular reason, the cleanest solution is usually moving your javascript code to an external file.
Besides avoiding most odd problems with browsers interpreting included javascript differently, it also makes things more cacheable.

Jukka K. Korpela explained an IE9 quirks mode behavior in a comment to this answer
This is bug in IE 9 (in Quirks Mode), since by the spefications, &#x2022 is valid here. But it has always been good and recommended practice to end all entity references (like ) and character references (like é) with a semicolon, partly because omitting it has triggered various browser bugs.
I disagree on what the specification says, but IE9 quirks mode may be the culprit.

AJAX responseXML errors

I've been having some weird issues when it comes to make an AJAX request and handling the response.
I am making an ajax call for an xml file. however when i get the response the xhr.responseText property works fine in firefox but not in IE.
Another thing is that I am trying to access the xhr.responseXML as XMLDocument, but it tells me in firefox it tells me that xhr.responseXML is undefined in ie it doesnt even show me the undefined error or displays the output.
This is the code I am using to make the request:
var ajaxReq = function(url, callback) {
//initialize the xhr object and settings
var xhr = window.ActiveXObject ?
new ActiveXObject("Microsoft.XMLHTTP") : new XMLHttpRequest(),
//set the successful connection function
httpSuccess = function(xhr) {
try {
// IE error sometimes returns 1223 when it should be 204
// so treat it as success, see XMLHTTPRequest #1450
// this code is taken from the jQuery library with some modification.
return !xhr.status && xhr.status == 0 ||
(xhr.status >= 200 && xhr.status < 300) ||
xhr.status == 304 || xhr.status == 1223;
} catch (e) { }
return false;
};
//making sure the request is created
if (!xhr) {
return 404; // Not Found
}
//setting the function that is going to be called after the request is made
xhr.onreadystatechange = function() {
if (!httpSuccess(xhr)) {
return 503; //Service Unavailable
}
if (xhr.responseXML != null && xhr.responseText != null &&
xhr.responseXML != undefined && xhr.responseText != undefined) {
callback(xhr);
}
};
//open request call
xhr.open('GET', url, true);
//setup the headers
try {
xhr.setRequestHeader("Content-Type", "application/x-www-form-urlencoded");
xhr.setRequestHeader("Accept", "text/xml, application/xml, text/plain");
} catch ( ex ) {
window.alert('error' + ex.toString());
}
//send the request
try {
xhr.send('');
} catch (e) {
return 400; //bad request
}
return xhr;
};
and this is how i am calling the function to test for results:
window.onload = function() {
ajaxReq('ConferenceRoomSchedules.xml', function(xhr) {
//in firefox this line works fine,
//but in ie it doesnt not even showing an error
window.document.getElementById('schedule').innerHTML = xhr.responseText;
//firefox says ''xhr.responseXML is undefined'.
//and ie doesn't even show error or even alerts it.
window.alert(xhr.reponseXML.documentElement.nodeName);
});
}
This is also my first attempt to work with AJAX so there might be something that I am not looking at right.
I've been searching crazy for any indications of why or how to fix it, but no luck there.
any ideas would be great.
EDIT:
I know this would be better with a framework, but the boss doesn't want to add a framework for just an ajax functionality ('just' is not a fair word for ajax :P). So I am doing it with pure javascript.
The XML file is well-formed, I see it well in the web browser, but for completion this is the testing file I am using:
<?xml version="1.0" encoding="utf-8"?>
<rooms>
<room id="Blue_Room">
<administrator>somebody#department</administrator>
<schedule>
<event>
<requester>
<name>Johnny Bravo</name>
<email>jbravo#department</email>
</requester>
<date>2009/09/03</date>
<start_time>11:00:00 GMT-0600</start_time>
<end_time>12:00:00 GMT-0600</end_time>
</event>
</schedule>
</room>
<room id="Red_Room">
<administrator>somebody#department</administrator>
<schedule>
</schedule>
</room>
<room id="Yellow_Room">
<administrator>somebody#department</administrator>
<schedule>
</schedule>
</room>
</rooms>
EDIT 2:
Well the good news is that I convinced my boss to use jQuery, the bad news is that AJAX still perplexes me. I'll read more about it just for curiousity. Thanks for the tips and I gave the answer credit to Heat Miser because he was the closest working tip.

I was having the same problem a few years ago, then I gave up on responseXML and began always using responseText. This parsing function has always worked for me:
function parseXml(xmlText){
try{
var text = xmlText;
//text = replaceAll(text,"<","<");
//text = replaceAll(text,">",">");
//text = replaceAll(text,""","\"");
//alert(text);
//var myWin = window.open('','win','resize=yes,scrollbars=yes');
//myWin.document.getElementsByTagName('body')[0].innerHTML = text;
if (typeof DOMParser != "undefined") {
// Mozilla, Firefox, and related browsers
var parser=new DOMParser();
var doc=parser.parseFromString(text,"text/xml");
//alert(text);
return doc;
}else if (typeof ActiveXObject != "undefined") {
// Internet Explorer.
var doc = new ActiveXObject("Microsoft.XMLDOM"); // Create an empty document
doc.loadXML(text); // Parse text into it
return doc; // Return it
}else{
// As a last resort, try loading the document from a data: URL
// This is supposed to work in Safari. Thanks to Manos Batsis and
// his Sarissa library (sarissa.sourceforge.net) for this technique.
var url = "data:text/xml;charset=utf-8," + encodeURIComponent(text);
var request = new XMLHttpRequest();
request.open("GET", url, false);
request.send(null);
return request.responseXML;
}
}catch(err){
alert("There was a problem parsing the xml:\n" + err.message);
}
}
With this XMLHttpRequest Object:
// The XMLHttpRequest class object
debug = false;
function Request (url,oFunction,type) {
this.funct = "";
// this.req = "";
this.url = url;
this.oFunction = oFunction;
this.type = type;
this.doXmlhttp = doXmlhttp;
this.loadXMLDoc = loadXMLDoc;
}
function doXmlhttp() {
//var funct = "";
if (this.type == 'text') {
this.funct = this.oFunction + '(req.responseText)';
} else {
this.funct = this.oFunction + '(req.responseXML)';
}
this.loadXMLDoc();
return false;
}
function loadXMLDoc() {
//alert(url);
var functionA = this.funct;
var req;
req = false;
function processReqChange() {
// alert('reqChange is being called');
// only if req shows "loaded"
if (req.readyState == 4) {
// only if "OK"
if (req.status == 200) {
// ...processing statements go here...
eval(functionA);
if(debug){
var debugWin = window.open('','aWindow','width=600,height=600,scrollbars=yes');
debugWin.document.body.innerHTML = req.responseText;
}
} else {
alert("There was a problem retrieving the data:\n" +
req.statusText + '\nstatus: ' + req.status);
if(debug){
var debugWin = window.open('','aWindow','width=600,height=600,scrollbars=yes');
debugWin.document.body.innerHTML = req.responseText;
}
}
}
}
// branch for native XMLHttpRequest object
if(window.XMLHttpRequest) {
try {
req = new XMLHttpRequest();
} catch(e) {
req = false;
}
// branch for IE/Windows ActiveX version
} else if(window.ActiveXObject) {
try {
req = new ActiveXObject("Msxml2.XMLHTTP");
} catch(e) {
try {
req = new ActiveXObject("Microsoft.XMLHTTP");
} catch(e) {
req = false;
}
}
}
if(req) {
req.onreadystatechange = processReqChange;
if(this.url.length > 2000){
var urlSpl = this.url.split('?');
req.open("POST",urlSpl[0],true);
req.setRequestHeader('Content-Type','application/x-www-form-urlencoded');
req.send(urlSpl[1]);
} else {
req.open("GET", this.url, true);
req.send("");
}
}
}
function browserSniffer(){
if(navigator.userAgent.toLowerCase().indexOf("msie") != -1){
if(navigator.userAgent.toLowerCase().indexOf("6")){
return 8;
}else{
return 1;
}
}
if(navigator.userAgent.toLowerCase().indexOf("firefox") != -1){
return 2;
}
if(navigator.userAgent.toLowerCase().indexOf("opera") != -1){
return 3;
}
if(navigator.userAgent.toLowerCase().indexOf("safari") != -1){
return 4;
}
return 5;
}
Granted, this is very old code, but it is still working for me on a site I built a few years ago. I agree with everyone else though I typically use a framework nowadays so I don't have to use this code or anything like it anymore.
You can ignore some of the particulars with the split, etc... in the Request onreadystate function. It was supposed to convert the request to a post if it was longer than a certain length, but I just decided it was always better to do a post.

This problem occurs mostly when content type is mis-detected by the browser or it's not sent correctly.
Its easier to just override it:
var request = new XMLHttpRequest();
request.open("GET", url, false);
request.overrideMimeType("text/xml");
request.send(null);
return request.responseXML;
Not sure why... This problem occurs only with Safari and Chrome (WebKit browsers, the server sends the headers correctly).

Are you calling the URL relative to the current document? Since IE would be using the ActiveXObject, it might need an absolute path, for example:
http://some.url/ConferenceRoomSchedules.xml
As for the XML, are you sure it's well-formed? Does it load in an XML editor, for instance?

What I can suggest you is to take a look at frameworks that hide and manage these cross-browser issues for you (in a reliable way). A good point here is jQuery. Doing these things yourself can become quite difficult and complex.
This may be what you need.
//Edit:
This is how the w3school shows it:
function ajaxFunction()
{
var xmlhttp;
if (window.XMLHttpRequest)
{
// code for IE7+, Firefox, Chrome, Opera, Safari
xmlhttp=new XMLHttpRequest();
}
else if (window.ActiveXObject)
{
// code for IE6, IE5
xmlhttp=new ActiveXObject("Microsoft.XMLHTTP");
}
else
{
alert("Your browser does not support XMLHTTP!");
}
}

To avoid your cross browser problems (and save yourself coding a lot of items that a strong community has already developed, tested, and reviewed), you should select a javascript library. JQuery and Dojo are great choices.

I believe that your web server need to serve correct response headers with 'ConferenceRoomSchedules.xml' e.g. Content-Type: text/xml or any other xml type.

The answer provided by Aron in https://stackoverflow.com/a/2081466/657416 is from my point of view the simplest (and the best) one. Here is my working code:
ajax = ajaxRequest();
ajax.overrideMimeType("text/xml");
ajax.open("GET", myurl;