A Flexible AJAX Framework

AJAX = Asyncronous (but not always), JavaScript, XML (Text or JSON). These technologies have been around almost as long as the Internet itself, but have only recently been used together to change the way that web applications are built.

XMLHTTPRequest is an API used mainly by browser-based Javascript applications to send and retrieve data from servers. It was developed originally by Microsoft for Exchange Server’s Outlook Web Access, but it has since been widely supported in browsers and is the heart of AJAX dynamic Web applications.

I’ve messed with many of the frameworks available online, and while they are very good, I’ve routinely found issues or features that they lack. Over time I’ve cobbled together my own framework, I fully expect this to change over time, but publish it here for your consideration!

The XSS methods (and “otherHost” variable) are provided as MSIE6 allows for Cross-site Scripting here. MSIE7 and Mozilla will indicate errors if connections are attempted to a ‘non-originating host’. I’ve exploited this vulnerability in a few applications, as such I provide the test for you too!

NOTE: The example response program is written in PHP, but should be easily ported to any other language.

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html>

<head>

<title>AJAX Example</title>

<script type=”text/javascript”>

var xbusy = true;

var otherHost = ‘http://www.giantgeek.com’; // DO NOT USE the ending slash!

var xhr = null;

/**

* Check for existance on DOM browsers (Mozilla, etc.)

* @return xhr

*/

function ajaxCheckDOM(){

var myxhr = null;

if(window.XMLHttpRequest) {

try {

myxhr = new XMLHttpRequest();

}

catch(e) {}

}

return myxhr;

}

/**

* Check for existance on Windows/MSIE (prior to MSIE7 which is now DOM)

* Evaluate using – new ActiveXObject(“Microsoft.XMLDOM”);

* @return xhr

*/

function ajaxCheckActiveX(){

var myxhr = null;

//if(window.ActiveXObject){

try {

myxhr = new ActiveXObject(“Msxml2.XMLHTTP”);

}

catch(e) {

try {

myxhr = new ActiveXObject(“Microsoft.XMLHTTP”);

}

catch(e) {}

}

//}

return myxhr;

}

/**

* This is a default response hook for AJAX, you SHOULD create your own as it’s only for DEMO

* @param obj – the clicked item

* @param customObj – user defined

*/

function ajaxResponseHook(obj,customObj){

var txt = xhr.responseText;// NOTE: xhr.responseXML is also valid

popStatus(‘AJAX Response value [‘ + txt + ‘|’ + customObj +’]’,”);

}

/**

* This is a common Response handler AJAX, you SHOULD NOT need to modify, use a ‘custom’ ajaxhook function to process the responses.

* @param obj – the clicked item

* @param ajaxhook Function (default will be used if undefined)

* @param customObj (optional – left for developer implementation, passed to the ajaxhook)

*/

function ajaxResponse(obj,ajaxhook,customObj){

var status=”;

if(xhr!=undefined){

var state=xhr.readyState;

if(state!=undefined){

if(state==4){

var code = xhr.status;

status = xhr.statusText;

if(code==200){

popStatus(‘AJAX Response Received.’,”);

if(ajaxhook==undefined){

ajaxhook = function(){ ajaxResponseHook(obj,customObj); }

}

ajaxhook(obj,customObj);

ajaxBusy(obj,false);

} else {

popStatus(‘AJAX Error ‘ + code + ‘ ‘ + status + ‘.’,”);

}

ajaxBusy(obj,false);

xhr=null; /* MSIE6 leak fix */

}

}

}

}

/**

* NOTE: MSIE6-7 supports XSS url’s (be careful!)

* @param obj – the clicked item

* @param url – the GET url params (FQDN)

* @param async (true or false) – false will LOCK browser until response – use cautiously!

* @param callback Function – allows for customized response handling

*/

function ajaxObj(obj,url,async,callback){

if(xbusy == true){

popStatus(‘AJAX BUSY, Please Retry.’,”);

} else {

if(callback==undefined){

callback = function(){ ajaxResponse(obj,hook,”); }

}

ajaxInit(obj,url,async,callback);

}

}

/**

* This is a Non-Caching implementation of ajaxObj() to show how you can avoid the MSIE caching issue.

* NOTE: You can use similar approaches to cache per page or session.

* @param obj – the clicked item

* @param url – the GET url params (FQDN)

* @param async (true or false) – false will LOCK browser until response – use cautiously!

* @param callback Function – allows for customized response handling

*/

function ajaxObjNoCache(obj,url,async,callback){

var cacheBuster=uniqueUrl(url);// damn MSIE!

ajaxObj(obj,cacheBuster,async,callback)

}

/**

* Initializes the AJAX operation

* @param obj – item clicked

* @param url – FQDN

* @param async (true/false) – for locking

* @param callback Function – provided for customization.

*/

function ajaxInit(obj,url,async,callback){

xhr=ajaxCheckDOM();

if(!xhr){

xhr=ajaxCheckActiveX();

}

if(xhr){

ajaxBusy(obj,true);

popStatus(‘AJAX Start.’,”);

xhr.onreadystatechange = callback; // method call

try {

xhr.open(‘GET’,url,async); /* POST */

//xhr.setRequestHeader(‘Content-Type’,’application/x-www-form-urlencoded’);

xhr.send(”); /* null */

}

catch(e)

{

if(xhr){

var code = xhr.status;

var status = xhr.statusText;

popStatus(‘AJAX Client SECURITY ‘ + navigator.appName +’ ‘+ navigator.appVersion + ‘ ‘ + code + ‘ ‘ + status + ‘.’,”);

}

}

ajaxBusy(obj,false);

} else {

popStatus(‘AJAX Client ERROR.’,”);

}

}

function ajaxBusy(obj,yn){

if(yn){

swapStyleObj(obj,’idle’,’busy’);

} else {

swapStyleObj(obj,’busy’,’idle’);

}

xbusyInd(yn,”);

}

function xmillis(){

return new Date().getTime();

}

function swapStyleObj(obj,oldCSS,newCSS){

if(obj!=undefined) {

var current=obj.className;

if(current!=undefined){

var txtOld = current.replace(newCSS,’ ‘);//no doubles

var txtMid = txtOld.replace(oldCSS, ‘ ‘);

var txtNew = (txtMid + ‘ ‘ + newCSS);

obj.className = txtNew;

} else {

obj.className = newCSS;

}

}

}

function xbusyInd(valnew, msg){

xbusy = valnew;

if(xbusy==true){

document.body.style.cursor=’wait’;

if(msg != ”){

top.window.defaultStatus=msg;

popStatus(msg,”);

}

showDiv(‘throbber’);

//blockScreen();

} else {

document.body.style.cursor=’default’;

hideDiv(‘throbber’);

//unBlockScreen();

}

}

function showDiv(id) { //show a div

var obj=xgetHelper(id);

if(obj!=null){ obj.style.display=”block”;}

//xrestart();

}

function hideDiv(id) { //hide a div

var obj=xgetHelper(id);

if(obj!=null){ obj.style.display=”none”;}

//xrestart();

}

function popStatus(txt,title){

popit(‘statusdyn’,’statusarrow’,’statusul’,’statusdiv’,txt,false,title,”);

}

function popit(dynId,arrowId,ulId,divId,txt,expand,title,cssCls){

popText(ulId,txt,title,cssCls);

showDiv(divId);

if(expand == true){

var arrowObj=xgetHelper(arrowId); if(arrowObj!=null){ arrowObj.style.backgroundPosition=’2px -106px’; }

showDiv(dynId);

}

}

function popText(id,txt,title,cssCls){

var obj = xgetHelper(id);

if(obj != null){

var oldHTML = obj.innerHTML;

var cls=”; if(cssCls!=”){ cls=’ class=”‘ + cls +'”‘; }

var htm = ‘<‘+’li’+ cls +’>’ + txt + ‘<‘+’/’+’li’+’>’ + oldHTML;

obj.innerHTML = htm;

}

}

function xgetHelper(id){

var obj = null;

try {

obj = document.getElementById(id);

} catch(z) {

var dummy=alert(“Error:” + z);

}

return obj;

}

function arrowTog(objectID,arrow) {

var obj = xgetHelper(objectID);

if(obj!=null){

if (obj.style.display ==’block’) {

arrow.style.backgroundPosition = ‘2px -21px’;}

else {arrow.style.backgroundPosition = ‘2px -106px’;}

objTog(objectID);

}

return false;

}

function objTog(objectID) {

var obj = xgetHelper(objectID);

if(obj!=null){

if (obj.style.display ==’block’) obj.style.display=’none’;

else {obj.style.display=’block’;}

}

return false;

}

function headTog(objectID,arrow) {

var obj = xgetHelper(objectID);

if(obj!=null){

if (obj.style.display ==’block’) {

arrow.style.borderBottomWidth = ‘1px’;}

else {arrow.style.borderBottomWidth = ‘0px’;}

arrowTog(objectID,arrow);

}

return false;

}

/*

* adds timestamp to URLs to make them unique

* @param URL String

*/

function uniqueUrl(x){

return urlAppender(x,’.cache’,xmillis());

}

/*

* helps to add parms to the url

* @param URL String

* @param aname String

* @param avalue String

*/

function urlAppender(x,aname,avalue){

var delim = “?”;

if(x.indexOf(“?”) >=0) { delim = “&”; }

return x + delim + aname + ‘=’ + avalue;

}

function xload(){

hideDiv(‘throbber’);

xbusy=false;

}

function testAjax(obj){

var callback=function(){ ajaxResponse(obj,null,’testAjax’); }

var x = ajaxObjNoCache(obj,’/ajax.php’,true,callback);

}

function testAjaxParms(obj){

var callback=function(){ ajaxResponse(obj,null,’testAjaxParms’); }

var x = ajaxObjNoCache(obj,’/ajax.php?testing=Y’,true,callback);

}

function testAjaxXSS(obj){

var callback=function(){ ajaxResponse(obj,null,’testAjaxXSS’); }

var x = ajaxObjNoCache(obj,otherHost+’/ajax.php’,true,callback,otherHost);

}

function testAjaxHook(obj){

var hook=function(){ customAjaxHook(obj,’ajaxhookTestMessageObect’); }

var callback=function(){ ajaxResponse(obj,hook,’testAjaxHook’); }

var x = ajaxObjNoCache(obj,’/ajax.php’,true,callback);

}

function testAjaxHookXSS(obj){

var hook=function(){ customAjaxHook(obj,’ajaxhookXSSMessageObect’); }

var callback=function(){ ajaxResponse(obj,hook,’testAjaxHookXSS’); }

var x = ajaxObjNoCache(obj,otherHost+’/ajax.php’,true,callback);

}

/**

* This is a custom implemenation, the customObj COULD be used for anything (perhaps delay measurement!)

* @param obj – the clicked item

* @param customObj – user defined

*/

function customAjaxHook(obj,customObj){

var xml = xhr.responseXML;// NOTE: xhr.responseText is also valid

var tmp = ‘DEMO customAjaxHook [‘ + customObj + ‘]\n’ + xml;

alert(tmp);

}

</script>

<style type=”text/css”>

.busy { color:red; }

.idle { color:black; }

</style>

</head>

<body onload=”xload();”><!– onbeforeunload=”alert(‘before’);” onunload=”alert(‘after’);” –>

<div id=”throbber”>WORKING!</div>

<div id=”statusdiv” class=”dyn” style=”display:none;”>

<h3><a id=”statusarrow” onclick=”headTog(‘statusdyn’,this);” href=”javascript:void(0);”>Status</a></h3>

<fieldset id=”statusdyn” style=”display:block;background-color:#ffffcc;”>

<div id=”statusBox” class=”box”>

<ul id=”statusul” class=”error”>

<li id=”ajaxStatus” style=”list-style:none;display:none;”>FILLER</li>

</ul>

</div>

</fieldset>

</div>

<h3><a href=”javascript:void(0);” onclick=”testAjax(this);”>TEST</a></h3>

<h3><a href=”javascript:void(0);” onclick=”testAjaxParms(this);”>TESTPARMS</a></h3>

<h3><a href=”javascript:void(0);” onclick=”testAjaxXSS(this);”>TESTXSS</a></h3>

<h3><a href=”javascript:void(0);” onclick=”testAjaxHook(this);”>TESTHOOK</a></h3>

<h3><a href=”javascript:void(0);” onclick=”testAjaxHookXSS(this);”>TESTHOOKXSS</a></h3>

<p><a href=”index.php”>RELOAD</a></p>

</body>

</html>

Response program (ajax.php)

<?php

header(“Cache-Control: no-store”);

header(“Charset: UTF-8”);

header(“Content-Type: text/xml”);

echo(“<?xml version=\”1.0\” encoding=\”UTF-8\”?>\n”);

?>

<test><?php echo( gmdate(“D, d M Y H:i:s”) ) ?> </test>

What are the MSIE Bugs you might ask… these are all in the ActiveX implementation of XmlHttpRequest, they are:

  1. Caching – for some reason, unless you define a ‘unique’ URL for each request, MSIE6 will respond from the cache, even if the server is unreachable!
  2. XSS – I discussed this above, MSIE6 will allow you to connect to ANY HOST with AJAX!
  3. Memory leak – if you don’t nullify the XHR object (for MSIE6), the application will leak memory.

Cheers!

Href Links as HTTP POST (complex DOM solution)

The earlier post, while easy to implement, has some well known security issues. Now lets get around them. First we’ll remove the FORM from the HTML itself, and instead build it dynamically and insert it into the BODY via the DOM and then submit it with JavaScript.

Again, if you’ve already implemented my prior solutions, this is just a small code refactor.

<html>
<head>
<title>Example FORM Post - DOM</title>
<script type="text/javascript">
/*
* Uses "location.replace()" vs. "location.href()" for all valid links.
* 'replace' has side-effect of 'restricting' back-button, or 'location'
* @param obj Object clicked
* @param x URL
*/
function xlinkObj(obj,url){
//Consider replacing w/ "return xlinkFrm(obj,url); " if you want POST behavior.
window.location.replace(uniqueUrl(url));
return false;
}
/*
* uses a FORM for the requested URL, could be POSTed!
* This is a simple solution, more complex solution COULD build the FORM and then parse attributes into INPUT's
* NOTE: you probably SHOULD NOT use this for external links, unless you intend for them to receive your params!
* @param obj Object clicked (NOT USED in this Example)
* @param x URL
*/
function xlinkFrm(obj,x){
var url=uniqueUrl(x);
var frmObj=xlinkFrmHelp(url);
if(frmObj!=null){
frmObj.submit();
}else{
alert('ERROR!');
}
return false;
}
/*
* Expects URL with queryString as param href
* @param x URL
* @return obj Object of the generated FORM
*/
function xlinkFrmHelp(x){
var rc=null;
try{
var ar = x.split("?");
var act = ar[0];
var str = ar[1];
var id="frm" + xmillis();
var oFORM=document.createElement("form");
oFORM.setAttribute("id",id);
oFORM.setAttribute("method","post");
oFORM.setAttribute("action",act);
if(str!=null){
var parms=str.split('&');
for(i=0; i < parms.length; i++){
var parm=parms[i];
var pair=parm.split('=');
var oINPUT=document.createElement("input");
oINPUT.setAttribute("type","hidden");
oINPUT.setAttribute("name",pair[0]);
oINPUT.setAttribute("value",pair[1]);
oFORM.appendChild(oINPUT);
}
}
var oBODY=document.getElementsByTagName("body")[0];
oBODY.appendChild(oFORM);
rc=oFORM;
}catch(e){
alert("Error"+e);
}
return rc;
}
/*
* generates a timestamp as a number
*/
function xmillis(){
return new Date().getTime();
}
/*
* adds timestamp to URLs to make them unique
* @param URL String
*/
function uniqueUrl(x){
return urlAppender(x,'time',xmillis());
}
/*
* helps to add parms to the url
* @param URL String
* @param aname String
* @param avalue String
*/
function urlAppender(x,aname,avalue){
var delim = "?";
if(x.indexOf("?") >=0) { delim = "&"; }
return x + delim + aname + '=' + avalue;
}
/*
* Abstracts "document.getElementById()" with appropriate error handling.
* @param id String
* @returns Object (NULL when not found!)
*/
function xgetHelper(id){
var obj = null;
try {
obj = document.getElementById(id);
} catch(z) {
var dummy=alert("Error:" + z);
}
return obj;
}
</script>
<a href="javascript:void(0);" onclick="return xlinkObj(this,'index.php');">REFRESH</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'index.php');">TEST-POST</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'index.php?a=b');">TEST-POST2</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'http://www.skotfred.com/hello');">TEST-XSS</a>
</body>
</html>

Cheers!

Href Links as HTTP POST (simple solution)

Another interesting challenge, The standard <a href=”…”></a> style link performs an HTTP GET, however you might want to perform a POST in some cases. HTML does not natively support this behavior, but it can be accomplished in JavaScript. If you have already implemented some of my security ‘hacks’ from previous posts this is only a small change. As usual I’ve included the minimum code required for this task in the example, but you should be able to easily merge the different features back into this!

Later on (in a different post), I’ll expand this to make it even more secure as this solution simply puts all of the URL into the FORM ‘action’ attribute and it would be better to pass them in the FORM body itself to hide them from the URL shown in the browser.

<html>
<head>
<title>Link to FORM POST simple example</title>
<script type="text/javascript">
/*
* Uses "location.replace()" vs. "location.href()" for all valid links.
* 'replace' has side-effect of 'restricting' back-button, or 'location'
* @param obj Object clicked (NOT used in this example)
* @param x URL
*/
function xlinkObj(obj,url){
//Consider replacing w/ "return xlinkFrm(obj,url); " if you want POST behavior in all cases!
window.location.replace(uniqueUrl(url));
return false;
}
/*
* uses a FORM for the requested URL, could be POST'ed!
* This is a simple solution, using an existing empty FORM on the page.
* A more complex and secure solution COULD build the FORM dynamically and then parse attributes into INPUT's
* NOTE: you probably SHOULD NOT use this for external links, unless you intend for them to receive your params!
* @param obj Object clicked (NOT used in this example)
* @param x URL
*/
function xlinkFrm(obj,x){
var frmObj=xgetHelper('frmXlink');
if(frmObj!=null){
frmObj.action=uniqueUrl(x);
frmObj.submit();
}else{
alert('ERROR!');
}
return false;
}
/*
* generates a timestamp as a number
*/
function xmillis(){
return new Date().getTime();
}
/*
* adds timestamp to URLs to make them unique
* @param URL String
*/
function uniqueUrl(x){
return urlAppender(x,'time',xmillis());
}
/*
* helps to add parms to the url
* @param URL String
* @param aname String
* @param avalue String

*/
function urlAppender(x,aname,avalue){
var delim = "?";
if(x.indexOf("?") >=0) { delim = "&"; }
return x + delim + aname + '=' + avalue;
}
/*
* Abstracts "document.getElementById()" with appropriate error handling.
* @param id String
* @returns Object (NULL when not found!)
*/
function xgetHelper(id){
var obj = null;
try {
obj = document.getElementById(id);
} catch(z) {
var dummy=alert("Error:" + z);
}
return obj;
}
</script>
</head>
<body>
<form id="frmXlink" action="#" method="post"></form>
<a href="javascript:void(0);" onclick="return xlinkObj(this,'index.html');">REFRESH</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'index.html');">TEST-POST</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'index.html?a=b');">TEST-POST PARMS</a>
<a href="javascript:void(0);" onclick="return xlinkFrm(this,'http://www.giantgeek.com/hello');">TEST-XSS</a>
</body>
</html>

This just uses an “empty” FORM in the page and uses the new ‘xlinkFrm()’ method to copy the URL to the FORM ‘action’.

Like i said, this is a simple solution as the params are still on the URL making them less secure. I’ll be refactoring it to parse the params to dynamically build the FORM (that will no longer be hardcoded).

Cheers!

What is a SLD (Second Level Domain)?

I just wrote about the TLD, so this naturally follows:

A SLD includes the TLD name and further identifies the owning organization of a URL.

Second-level domains can be divided into further levels. These subdomains sometimes represent different computers within an organization, but are many times the same machine with different aliases.

Examples:

www.giantgeek.com
mail.giantgeek.com

TLD =  .com
SLD = giantgeek.com
Subdomains = www.gianteek.com & mail.giantgeek.com

NOTE: Refer to HTTP/1.1 for details on how IP addresses, routing and webservers are impacted by this.

Interesting enough, some one character SLD’s do exist,  x.com for example.

References:

Cheers!

What is a TLD (Top Level Domain)?

I occasionally get this question, as many technical people don’t fully understand it.

A TLD is ‘actually” the last section of a URL (owned by the domain registrars themselves).

Examples:

  • .com
  • .net
  • .org
  • .us (and other 2 digit country codes)

You can find a full list of TLD’s here:

Various other TLD’s have been proposed through the years, here are a few common ones (that are in various states of approval or implementation):

  • .museum
  • .info
  • .xxx
  • .biz
  • .mobi

References:

Cheers!

Open source planetarium software

One of the first applications I ever saw on a (original) IBM-PC was planetarium.  Additionally, in my “Navy Days” I became quite interested in astronomy, as the skies are never quite as clear as they are in the middle of the ocean!

Stellarium is a free open source and cross-platform planetarium for your computer. It shows a photo-realistic sky full of stars, constellations, planets, nebulas in realtime 3D, just like what you see with the naked eye, binoculars or a telescope. Other items like ground, landscape, atmosphere, and constellation illustrations can also be added.  You can even see the Sun and Moon in their actual locations.

Notable is that you can also update the existing star charts with improved information so that even more stars can be seen in the application.

References:

Another good competitor that you may prefer more for various reasons is Celestia, it can be found at:

Happy sky-gazing, even if it’s only virtual!

MSIE’s flawed SSL implementation

This has been quite frustrating. It seems that Microsoft has again ventured from complying with the industry web standards in this space too!

The comments from the Apache HTTP 2.x ‘http-ssl.conf’ files say it all:

#   SSL Protocol Adjustments:
#   The safe and default but still SSL/TLS standard compliant shutdown
#   approach is that mod_ssl sends the close notify alert but doesn't wait for
#   the close notify alert from client. When you need a different shutdown
#   approach you can use one of the following variables:
#   o ssl-unclean-shutdown:
#     This forces an unclean shutdown when the connection is closed, i.e. no
#     SSL close notify alert is send or allowed to received.  This violates
#     the SSL/TLS standard but is needed for some brain-dead browsers. Use
#     this when you receive I/O errors because of the standard approach where
#     mod_ssl sends the close notify alert.
#   o ssl-accurate-shutdown:
#     This forces an accurate shutdown when the connection is closed, i.e. a
#     SSL close notify alert is send and mod_ssl waits for the close notify
#     alert of the client. This is 100% SSL/TLS standard compliant, but in
#     practice often causes hanging connections with brain-dead browsers. Use
#     this only for browsers where you know that their SSL implementation
#     works correctly.
#   Notice: Most problems of broken clients are also related to the HTTP
#   keep-alive facility, so you usually additionally want to disable
#   keep-alive for those clients, too. Use variable "nokeepalive" for this.
#   Similarly, one has to force some clients to use HTTP/1.0 to workaround
#   their broken HTTP/1.1 implementation. Use variables "downgrade-1.0" and
#   "force-response-1.0" for this.

SetEnvIf User-Agent ".*MSIE.*" \
nokeepalive ssl-unclean-shutdown \
downgrade-1.0 force-response-1.0

A little further research indicates that MSIE6 has (probably) partially fixed this (the HTTP/1.0 & KeepAlive issues), so the updated config should use a Regular Expression to look like…

SetEnvIf User-Agent ".*MSIE [1-5].*"
 nokeepalive ssl-unclean-shutdown
 downgrade-1.0 force-response-1.0
SetEnvIf User-Agent ".*MSIE [6-9].*"
 ssl-unclean-shutdown

Related Information:

Cheers!

Pretty Good Privacy (PGP)

I’ve used PGP (Pretty Good Privacy) since I was in college. It provides for both digital signatures and strong encryption and content without the user having to go make extraordinary effort. The process uses what is known as Public Key Encryption and uses a Web Of Trust to certify individual users.

For years I used the original PGP 2.6.2, 5.x and 6.x products that were available as freeware. After PGP was acquired by a much larger commercial entity, most development has shifted to the open-source community that makes it available as GnuPG aka GPG.

There are several plugins available for common Email Clients such as Thunderbird and Outlook to natively integrate the functions into those applications. Additionally plugins are available for Firefox to enable encryption and signing of WebMail services such as GMail (Google Mail).

My public keys are available online at http://www.giantgeek.com/pgpkeys.asc, http://www.skotfred.com/pgpkeys.asc, or through most of the keyservers.

References:

I look forward to your signed/encrypted emails,
Cheers.

MSIE6 CSS issue ‘dotted’ behaves like ‘dashed’

Another fix in MSIE7 (broken before), ‘dotted’ is now implemented, in MSIE6 dotted had the same visual representation as ‘dashed’.

This explains why you might expect to see a line of “……” that appear to be “——“, even when you’re absolutely positive that you’re CSS is correct!
CSS:

border:1px dotted #fff;

HTML Example:

<html>
<head>
<title>dotted-dashed Example</title>
<style type=”text/css”>
fieldset {background-color:#fcfcfc;
width:95%;
padding:15px 10px 0 10px;margin:0 0 20px 0;
border:1px solid #999;
border-top-width:2px;
overflow:hidden;}
fieldset div.buttons {clear:both; padding-top:10px;padding-bottom:10px;margin:3px 0 0 0;border-top:1px dotted #b5b5b5;text-align:left;}
</style>
</head>
<body>
<fieldset>
Some form fields go here…
<div class=”buttons”>
Some buttons go here…
</div>
</fieldset>
</body>
</html>

Cheers!

Clientside Session Timeout’s

There comes a time in web application development that you need to ‘timeout’ idle users. This comes in a variety of ways, here’s a few common reasons that you may desire this activity.

  • Security – you don’t want to leave sensitive data on a users screen when they’ve gone to lunch or left for the day.
  • Server Resources – persisting/keeping an active ‘session’ available on the server takes resources (the exact type varies, but this is usually database, memory or file resources)
  • Server ‘enforced’ session timeout’s and the potential errors and lost data experienced by the users in that circumstance.

My personal approach to this has evolved over time, here’s a brief synopsis:

  1. Use standard server-side session timeout, often leading to a bad user experience when they loose data on a form submit.
  2. Use META REFRESH…where timeout is in seconds, in this example it’s 60 seconds (1 minute).
    <meta http-equiv="refresh" content="60;url=http://www.giantgeek.com/" />
  3. Use javascript 'timeout' (problem is that this is not 'measureable')
    
    <script type="text/javascript">
    setTimeout("javascript:myTimeout();",minutes*60000); // code minutes
    </script>
  4. Use javascript countdown timer and custom code event.

<html>
<head>
<title>Timeout example</title>
<script type=”text/javascript”>
var build=’testing’;
var timerID = 0;
var loadTime = null;
var stopTime = null;
function xload(){
loadTime=grvMillis();
grvWindowStatus(build);
grvSetTimeout();
}
function xclose(){
grvWindowStatus(”);
}
function grvMillis(){
return new Date().getTime();
}
// Start timer
function grvTimerUpdate(){
timerID = grvTimerClear(timerID);
if(loadTime == null){
loadTime=grvMillis();// Start Time
}
// Calculate Current Time in seconds
var timeNow = grvMillis();

var think = calcMinSec( calcTimeDiff(timeNow,loadTime) );
var remain = calcMinSec( calcTimeDiff(stopTime,timeNow) );
grvWindowStatus(build + ” ” + think + ” ” + remain );
timerID = setTimeout(“grvTimerUpdate()”,1000);
}
function calcMinSec(diff){
var mm = removeDecimal(diff/60);
var ss = zeroPad(removeDecimal(diff-(mm*60)),2);
return (mm + “:” + ss);
}
function calcTimeDiff(tmpStart,tmpStop){
var diff = (tmpStart – tmpStop)/1000;
return diff;
}
function removeDecimal(val){
var rc=””;
val = val + “”;
if(val!=””){
var pos = val.indexOf(“.”);
if(pos > -1){
rc=val.substr(0,val.indexOf(“.”));
} else {
rc=val;
}
}
return rc;
}
function zeroPad(x,sz){
x = x + “”;
while(x.length < sz){
x = “0” + x;
}
return x;
}
function grvTimerClear(x){ // this clears a timer from the queue
if(x){
clearTimeout(x);
x = 0;
}
return x;
}
function grvSetTimeout(){
var min=45; xID=grvTimeout(“javascript:grvTimeoutUSER()”,min); // EXAMPLE: this could be conditional!
stopTime = grvCalculateTimeout(min);
grvTimerUpdate();
}
function grvCalculateTimeout(mins){
var timeNow = grvMillis();
var exp = timeNow + (mins*60*1000);
var timeExp = new Date(exp).getTime();
return timeExp;
}
function grvTimeout(x,minutes){ // this sets a timer(request) in a queue
return setTimeout(x,minutes*60000);
}
function grvTimeoutUSER(){
alert(‘Session Inactivity Timeout [USER]’);
// DO WHAT YOU NEED TO HERE!}
function grvWindowStatus(txt){
window.defaultStatus=txt;
}
</script>
</head>
<body onunload=”xclose();” onload=”xload();”>
</body>
</html>

Another benefit of this last solution is that you also have access to the user “Think Time” and can therefore measure how long the user spends on a given page.

Cheers!