ClickOffline.com

At last I got my Google wave account activated yesterday, but it didn’t carry any additional invites that can be passed on to others. Its a bit tricky if it doesn’t carry additional invites, since you wont be able to share and evaluate this technology with others. Please take a look at some of the Screenshots of Google Wave.

Google Wave Homepage

A Public Wave

The Wave API will allow plug-in developers to develop extensible plug-ins. Sudoko plug-in was quite interesting. So was the Google Maps plug-in, the live connectivity was just too good.

Playing Sudoko in GW

GW Error Messages

Google Wave Extensions

On a overview, Google wave seems to show the light of how the next generation of communications will look like. This technology has the prospect to replace email altogether (am i hyping it too much ??). But on an overview i am not able to define Google wave properly, yes it is a collaboration platform, but its a mix of email, instant messaging, blogs and other things. I will not term Google wave as a social networking platform/tool, because this is more sort of a communications platform, more to do with interactions and collaboration, and real time information exchange (more reason being, it could attract an access ban by employers, if termed so.. he he ). I will take a look at API’s now, and will try to post if i am able to build something out of it.

Often when working with large datasets, you will come across issues related to intense event overload for objects like table cells. This could translate to poor client JavaScript performance. The root cause for the performance lag is the large number of events that are associated and handled by web page at the client. Lets consider a scenario of highlighting cells in a table as follows

HTML

<table id="resultSetTable">
<tr id="row1">
<td class="nameCell"> UPS </td>
<td class="priceCell"> 2500 </td>
</tr>
<tr id="row2">
<td class="nameCell"> Mouse</td>
<td class="priceCell"> 300 </td>
</tr>
.
.
.
<tr id="row859">
<td class="nameCell"> Keyboard </td>
<td class="priceCell"> 250 </td>
</tr>
</table>

Javascript to highlight the selected cell

<script type="text/javascript">
$(function(){
$('.nameCell').click(function(){
if ($(this).css('background-color') != 'rgb(255, 0, 0)')
$(this).css('background-color', 'red');
else
$(this).css('background-color', 'green');
});

});
</script>

Although the code above works properly, the inefficiency of the above list code can be identifeid if the code is put through large datasets, since the javascript has to assign the event handler to all cells of the class “nameCell”, the Javascript engine gets overloaded with events, in a scenario of a few thousand rows, there are equal number of events being declared and handled simultaneously. This in some cases might cause the browser to crash. I got the following error when I tested this code with a large dataset on Firefox.

A script on this page may be busy, or it may have stopped responding. You can stop the script now, open the script in the debugger, or let the script continue.
Script: jquery-1.3.2.min.js:18

An alternate way to handle this problem will be to assign the event handler to a larger parent object, and to execute the event handler by validating the properties of the event. In this case the larger parent object is the Table. We can drill down to the class of the clicked cell of the table, see if it matches the ‘nameCell’ class type and then run the instructions, based on the validation. This mechanism is known as event delegation.

<script type="text/javascript">
$(function(){
$('table').click(function(event) {
var $thisCell, $tgt = $(event.target);
if ($tgt.is('td')) {
$thisCell = $tgt;
} else if ($tgt.parents('td').length) {
$thisCell = $tgt.parents('td:first');
}
// now do something with $thisCell
if ($($thisCell).attr("class")=="nameCell")
{
if ($($thisCell).css('background-color') != 'rgb(255, 0, 0)')
$($thisCell).css('background-color', 'red');
else
$($thisCell).css('background-color', 'green');
}
});
});
</script>

The web application saves lots of client computing power by avoiding the replication of thousands of events by delegating the event to a parent object. Proper use of event delegation in jQuery can help in excellent performance optimization of the web application.

References

• Working with Events by Karl Swedberg -
  http://www.learningjquery.com/2008/03/working-with-events-part-1

A lot of you may very well know what happens behind the scenes when a webpage is requested or being served, this article will explain the details and internals of what goes behind when you type http://www.clickoffline.com/index.php

The Internet Architecture

What is meant by internet is a collection of networks that interconnect a large number of content providing servers with the end users. Internet typically is a network of much smaller networks which are interconnected. Any network/machine that is connected to the internet can provide and retrieve content from the internet.

Since internet is a network, just like any other network it supports a wide range of protocols, one of the most frequently used protocols is the HTTP protocol. The HTTP (Hyper Text Transfer Protocol) is the main medium for content retrieval and delivery over the internet. Some of the other protocols that are frequently used are as follows

• FTP – File Transfer Protocol
• SMTP – Simple Mail Transfer Protocol
• POP – Post Office Protocol
• IMAP – Internet Message Access Protocol
• HTTPS – Secure HTTP Protocol

All these protocols are dependent on the internet as the backbone for providing the infrastructure needed to communicate. However these protocols may be used in a wide range of networks, not just the internet, for example, you may run an ftp server in an intranet (local LAN), which is not exactly internet.

DNS Service

Internet also has services that are used by other applications. DNS or Domain Name System is the one of the most important one. DNS works like a telephone directory. DNS typically is a collection of servers, which have a list of domain names matched with their internet IP addresses. Since every machine in the internet can be assigned with a unique IP address, and remembering the IP address is not so easy. DNS helps in referencing these Domains with IP addresses

A sample DNS table is shown below.

So when you request a page from Google.com, the browser sends a request to the DNS server to find a match for ‘www.google.com’, the DNS servers will in turn, return the IP address of the respective Domain Name by doing a looking in the DNS table. When this operation is performed in reverse, i.e identifying the default domain name from a given IP address, it is referred to as Reverse DNS Lookup or rDNS.

Steps and processes involved in obtaining an html webpage

Request a web page

Once you type the URL in your browser, which may be IE or Firefox; the steps as described above happen sequentially before we get to see the page on our browsers. Let’s look at each of these steps in detail.

Components of the URL

The first phase is to understand the URL or identifying the components of the URL. A typical URL consists of the following components

•    Protocol
•    Domain Name
•    Resource Name

However there are other components that can be plugged into a URL which include, usernames, passwords and port numbers; we can ignore them as they are not used prominently.

The figure above explains the components of the URL. Now were ready to send the HTTP Request.

Construction of the HTTP Request

The browser composes a request called the HTTP Request that should be sent to the server to retrieve information from the server.

Sample

GET /index.php HTTP/1.1
Host: www.clickoffline.com

Since HTTP is a connection-less protocol, all information related to the session are sent along the request every time, this includes, resource name, protocol version, browser name, version and OS, supported content types, cookies if any, etc.

Identifying the server

Before sending out a HTTP request, we will need to identify the remote server; this is done by identifying the IP address using DNS for the domain specified in the URL. The HTTP Request is sent to eth remote server by establishing a route between the local machine and the remote server.

Wait For Response

The client browser waits for the response for the server.

Analysis/ Rendering of HTTP Response

The HTTP Response consists of a HTTP Response Header and the HTTP Response Data.
The Response Headers provide information related to the HTTP Response, the most important field is the Response code. Different response codes mean different statuses

The browser analyzes the HTTP response and renders the response one the screen, any dependencies like Images, CSS, java scripts that arise from the source html page follow the same processes to load int eh client, except that the URLs of the  resources is loaded automatically by the client browser instead of a user typing it.

HTTP/1.1 200 OK
Date: Wed, 18 Feb 2009 16:19:57 GMT
Server: Apache/2.2.11 (Unix)
X-Powered-By: PHP/5.2.8
X-Pingback: http://www.clickoffline.com/xmlrpc.php
Keep-Alive: timeout=15, max=100
Connection: Keep-Alive
Transfer-Encoding: chunked
Content-Type: text/html; charset=UTF-8

Advanced Resources
•    Specfications for URL (RFC 1738): http://www.w3.org/Addressing/URL/url-spec.txt
•    Specifications for DNS (RFC 1035): http://www.ietf.org/rfc/rfc1035.txt
•    HTTP Protocol Details: http://en.wikipedia.org/wiki/HTTP

Disclaimer: The views expressed on this webpage are my personal views and do not necessarily reflect views of my employer.

I had this weird problem, where i had to secure a web server, to which access form the local network is granted by default, but if anyone who is not in the local network tires to access a web page, we had to validate him for username and password.

After some googling, i found this configuration that allowed me to secure the apache directory, this is the configuration

<Directory /var/www/html/123/>

AuthType Digest
AuthName "Intranet"
AuthDigestDomain /
BrowserMatch "MSIE" AuthDigestEnableQueryStringHack=On

AuthDigestProvider file
AuthUserFile /physical/path/to/.digest_pw
Require valid-user
Order Allow,Deny
Allow from 192.168.0.0/255.255.255.0

Satisfy any

</Directory>

There are typically two kinds of authorizations here,
1) Allow from Directive
2) AuthUserFile Directive

The Allow from directive allows traffic from the specified ip range, and the AuthUserFile validates allows the validation of user credentials, if he/she has given them. The Satisfy any directive allows apache to validate any one of teh above conditions to grant access to the specified resources.

There can be many instances where you would like to preload images in a HTML primarily for caching in order to serve the page better. This can be achieved by the use of Image objects in JavaScript. Image Object allow dynamic creation and manipulation of images in JavaScript, it is support by IE 4+ and Firefox.

Preloading Image

Preloading an image can be done by invoking a small JavaScript, that creates an image object and loads an image from the URL specified in the src parameter. See below

<head>
<script type="text/javascript">
<!--
image01= new Image();
image01.src="http://www.google.co.in/intl/en_com/images/logo_plain.png";
image02= new Image();
image02.src="http://labs.google.com/images/labs_logo2.gif";
//-->
</script>
</head>

The most optimal place to trigger the preloaded is on the body onLoad(), so that the images will start loading once the page has completed loading.

Changing images in JavaScript

The most immediate function after loading/caching and image is to use it somewhere in the html page, this can be achieved easing using JavaScript
See Full Example below

<html>
<head>
<title>Page title</title>
<script language="JavaScript">
objImage1 = new Image();
objImage2 = new Image();
objImage3 = new Image();
cnt = 0;
function preLoadImages(){
// preload the image file
objImage1.src='http://www.google.com/logos/mlk09.gif';
objImage2.src='http://www.google.com/logos/newyear09.gif';
objImage3.src='http://www.google.com/logos/olympics08_opening.gif';
}

function changeImage() {
cnt++;
if (cnt==1) {
document.images['im'].src = objImage1.src;
}
else if (cnt==2) {
document.images['im'].src = objImage2.src;
}
else if (cnt==3) {
document.images['im'].src = objImage3.src;
cnt=0;
}
}
</script>
</head>
<body onLoad="preLoadImages()">
<form name="myWebForm">
<img name="im" src="http://www.google.com/logos/olympics06_opening.gif"><br>
<input type="button" name="Prev" value="Switch Image " onClick="changeImage()">
</form>
</body>
</html>