Generate Dummy Placeholder Images in the WordPress Export File Instantly

It’s a common problem for the theme developers who sell their themes in the marketplaces and ships a data file with it, that contains exported data from the demo WordPress site. The problem is, this content file contains links to the images used in the demo. And those images, which the author uses to decorate his demo sites, may not be freely distributable.

So, theme developers have to create another site with placeholder images and then export data from this site. It’s a double work. It’s a boring work as well.

To save from this pain, I’ve created a site called DimgX which is available at that can deliver placeholder images on the fly. Using this service is pretty straight forward. If your image url is all you have to do is add a at the end of the image domain, like

So if the url of the original image is which displays like this Artwork By Greg Christman

It’s placeholder image’s url will be and it will show like this – Tada! You don’t have to worry about the original image’s dimension or anything. Everything will be handled by DimgX on the fly.

Using this feature, DimgX allows you to upload the exported content file from your WordPress site, which has an xml extension, and then download a modified content file having all the attachment urls replaced using It’s pretty handy

DimgX replaces all the <guid> and <wp:attachment_url> in the export file, which contains links to the image files, and adds at the end of those domain names in those image urls.

While creating a placeholder image, DimgX needs to calculate the size of the original image, to properly draw a placeholder with the same dimension. And in PHP, you know the only way to do it using getImageSize() function. People usually download whole image and then calculate these dimensions but that’s is a pretty expensive process when you have to download lots of bigger images. Fortunately, all these popular image formats contains these dimension in the first few KB in their header. So DimX doesn’t download the whole image but only a chunk of it. And then it uses that chunk of the header data to calculate this size. It makes the whole process fast, and less resource extensive. And while doing this, it also saves a good amount of bandwidth. To feed your appetite, here’s a part of that routine

switch ($extension) {
    case "jpg" :
    case "jpeg" :
        $range = "0-50000";
    case "png":
        $range = "0-10000";
    case "gif":
        $range = "0-10";
        $range = "0-15000";

$ch = curl_init ($url);
curl_setopt($ch, CURLOPT_HEADER, 0);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_BINARYTRANSFER,1);
curl_setopt($ch, CURLOPT_RANGE, $range);

And here is the regular expression pattern that gets this complex URL replacement job beautifully done

$pattern1 = "/<guid([^>]+)>http://([^/]+)/([S]+)(.jpg|png|gif)</guid>/";
$pattern2 = "/<wp:attachment_url>http://([^/]+)/([S]+)(.jpg|png|gif)</wp:attachment_url>/";
$new_data = preg_replace($pattern1,'<guid${1}>http://${2}${3}${4}</guid>',$data);
$new_data = preg_replace($pattern2,'<wp:attachment_url>http://${1}${2}${3}</wp:attachment_url>',$new_data);

Anyway, feel free to use DimgX at and let me know how you feel about this service. Right now, it can convert image urls that is served using http://, because wildcard SSLs are not cheap. But it’s in our roadmap. Stay tuned :)

Installing gearmand, libgearman and pecl gearman extension for php from source in Debian 6 & 7

I had a pretty rough evening today. No, not because the waiter forgot to add sugar in my tea, but because it was so boring to go through the trial and errors of installing gearman daemon and pecl gearman extension for php.

First I tried with the pretty-obvious-awe-fucking-some way.

apt-get update
apt-get install gearman 
pecl install gearman

In above steps, “apt-get install gearman” ran smoothly. But then “pecl install gearman” failed saying that it doesn’t have libgearman. Eh? so I installed libgearman4 (in Debian Squeezy. It was libgearman6 in Wheezy) and libgearman-dev. But pecl failed again to compile the extension saying that libgearman is missing. Strange! So I googled about this issue and found that it could be a link error where a symbolic link might be required of “/usr/include/libgearman” as “/usr/include/libgearman-1.0”. Still no luck.

So I decided to compile gearmand from the source. And libgearman is also included with that, so it will fix two errors at once. I downloaded the latest gearman package from launchpad and tried to compile it

tar -zxvf gearmand-1.1.11.tar.gz
cd gearmand-1.1.11

Failure 1:
At this step, configure stopped showing the following error

configure: error: cannot find Boost headers version >= 1.39.0

To fix this, I had to install “libboost-all-dev” by following command

apt-get install libboost-all-dev

And I tried to compile gearman and it failed again

Failure 2:
At this step, configure stopped showing that it cannot find gperf. That’s fine – I have installed gperf and tried to configure gearman again

apt-get install gperf

Failure 3:
Now it failed again, showing that libevent is missing. Hmm! Had to fix it anyway

apt-get install libevent-dev

Failure 4:
Heck! Another failure. Now it’s showing that it can’t find libuuid. This part was a little tricky to solve, but finally fixed with the following package

apt-get install uuid-dev

Let’s configure again. And sweet that the configure script ran smoothly. Let’s compile using make

Failure 5:
Grrr! At this point the make script failed with a bunch of text, where the following lines were at the top

libgearman/ In function 'void custom_backtrace()':
libgearman/ sorry, unimplemented: Graphite loop optimizations can only be used if the libcloog-ppl0 package is installed

So it cannot find a library named libcloog-ppl. Let’s fix this problem by

apt-get install libcloog-ppl-dev

Now I’ve tried to run the make script, and it was good. So i also ran make install to complete the installation.

make install

Now gearmand and libgearman both are installed. So I tried to install pecl-gearman with the following extension and voila! it worked. No more missing libgearman anymore.

pecl install gearman

Now all I had to do is add the line “” in my php.ini .

The process was tedious and boring and took me more time than writing this article. If you have seen “Despicable Me 2” when Lucy and Gru went to ElMacho’s restaurant and were attacked by that crazy chicken and finally Lucy exclaimed “What’s wrong with that chicken!”

I really wanted to say “What’s wrong with this chicken” after gearman was installed at last.


connecting to flickr using PHP and PECL oAuth extension

flickr, one of the world dominant in photo sharing service, has added support to oAuth recently. Though their old authentication system still works (marked as deprecated but not discontinued) I find it’s wise to use their oAuth system for the better future proof application development.

PHP has a nice extension to perform oAuth dance in it’s pecl repository – pecl oAuth. Here is the code to connect to flickr using this extension. The only catch I found and took me more than 30 minutes to figure out a failed attempt, is you will have to append the permission flag in it’s oAuth authorization url. Pass either one of these permission flags “read”,”write” or “delete” as “&perms=” (flickr.php, line # 20) and then the redirection will be successful. Register your Flickr application from here, you will find your consumer key and secret key in next page in the App Garden.

While creating new application in flickr app garden, point the callback url to your flickr.php – thats it.

source code of config.php


And here is the source code of flickr.php

 * flickr authentication script based on
 * pecl oauth extension
$oauthc = new OAuth($oauth['flickr']['consumerkey'],
if(empty($_SESSION['frequest_token_secret'])) {
    //get the request token and store it
    $request_token_info = $oauthc->getRequestToken($oauth['flickr']['requesttokenurl']); //get request token
    $_SESSION['frequest_token_secret'] = $request_token_info['oauth_token_secret'];
    header("Location: {$oauth['flickr']['authurl']}?oauth_token=".$request_token_info['oauth_token']."&perms=read");//forward user to authorize url with appropriate permission flag
else if(empty($_SESSION['faccess_oauth_token'])) {
    //get the access token - dont forget to save it 
    $request_token_secret = $_SESSION['frequest_token_secret'];
    $oauthc->setToken($_REQUEST['oauth_token'],$request_token_secret);//user allowed the app, so u
    $access_token_info = $oauthc->getAccessToken($oauth['flickr']['accesstokenurl']);
    $_SESSION['faccess_oauth_token']= $access_token_info['oauth_token'];
    $_SESSION['faccess_oauth_token_secret']= $access_token_info['oauth_token_secret'];
if(isset($_SESSION['faccess_oauth_token'])) {
    //now fetch current users profile
    $access_token = $_SESSION['faccess_oauth_token'];
    $access_token_secret =$_SESSION['faccess_oauth_token_secret'];
    $data = $oauthc->fetch('');
    $response_info = $oauthc->getLastResponse();
    echo "<pre>";
    echo "</pre>";

You can download the complete package from here (Thanks

Note: I have written a similar post to demonstrate connecting to twitter and linkedin via their oAuth protocol which you can find here :) widget in MiproApps – why did it require special care…

In MiproApps, our Visual Facebook Fanpage Desiger from Leevio, everything is built on top of a scalable plugin based architecture. Every plugin manages it’s data using a central plugin manager. Most of these data are isolated from each other, stored and served by the plugin manager without any special coding required from plugin developers. That makes everything simple. As storage and serving is fully managed by Plugin Manager, it helps us to cache, validate and sanitize user data properly from a single place.

But when we decided to add support for, there comes a challenge. We have asked for username and password for account from our users to pull out the data from their shared folders and files. “PASSWORD” – and that is the thing everyone thinks twice before providing to a third party. Everyone cares about their personal data security.

In MiproApps every data collected from user are submitted to storage service via Ajax request. And we simply cant send plain password collected by users in an Ajax request. What we did in this case is we had signed user’s sensitive information using a 128 bit public key (RSA) in client side. The private key is stored securely in our server and that encrypted information is decrypted only in server side. So client application has just the public key. widget in Facebook Fanpages powered by MiproApps widget in Facebook Fanpages powered by MiproApps

There were other challenges as well, while we went to implement this encryption in client side by Javascript and Decryption by PHP. Unfortunately PHPs Mcrypt doesnt support RSA, and Zend Framework doesnt provide any component for that. And there was a trick when you encrypt your data in JS. You must add a null byte at the end of your data, otherwise PHP cant decrypt it.

We have used RSA library (a combination of RSA, BigInt and Barett Library) from Ohdave and used the Crypt_RSA library from PEAR. And it also required us to install bigint PECL extension.

You can see some example code at here and you can use the RSA public/private key pair either by openssl shell command or the RSA key generator from ohdave.

The service layer is working smooth. Plugin developers doesnt need to bother about encryption and decryption. Everything was managed transparently under the hood. And so far we are only developing our plugins, data is secured and safe. Users can add their publicly shared files directly in their facebook fanpage. For a sample output, you can check out my page. You will find the component at the bottom right corner.

Prelude to foundation: Its time to go for a better PHP Framework

I remember those old days when I had to write everything by myself. I wrote huge libraries to work with MySQL. Then I learned PostgreSQL and SQLite but didn’t rewrote my old library to work with those, I was running short of time. So I forsake the opportunity to write a db library which works with them. What I did was plain code relevant to database specific portions. Oh ya, that was a long time ago.

Soon after that I came to know adoDB which made my dream come true. I was so much happy getting my all db specific works done in a much more smarter way. I get rid of database portability issues. I was very happy that time.

I learned smarty soon after I realize that my codes are getting ugly with all the inline HTMLs and PHPs. Nothing could be smarter than separating the presentation logic from the business layer. I am a big smarty fan since that time. It saves my sleep for many nights.

But again I am recurrently suffering from maintainability issues. I was not surprised to find that my code is becoming huge unmanageable giant and it takes huge time for refactoring the application. I was very sad those days. Oh what a disaster that was.

When working with my team members located remote places, I fall into a deep shit. How can we manage and track the changes done by us? Even I was getting strange code in my routine which I bet was not written by me!! It was a terrific job Continue reading Prelude to foundation: Its time to go for a better PHP Framework

Just PHP will give you nothing… unless you upgrade yourself

I have been working with PHP for more than 3yrs (I believe still I am beginner in this category) – I was present in several interview board. Which things disappointed me most is the “lack of eagerness” to learn what comes new. Sometime developers thinks that learning Only PHP will help them to get lucrative jobs!! OMG

Specially in BD most of the time PHP developers plays multiple roles in the companies, they are developer, they are template designers, they are HTML coders, they are DBA, they are PMs….what the heck. Only few companies have different people for these roles.

How far you can go just learning PHP (RAW code, in ancient style, that means PHP+HTML together, yak!!)? You have to have knowledge in CSS, JS, Frameworks, Multiple DBS

When it comes the question of CSS, you should maintain a list of websites where from you can get updates. Dont learn the CSS from book, but goto websites and see what is happening… If you are still using Tables to design Continue reading Just PHP will give you nothing… unless you upgrade yourself

Caching – Save visitors time and Balance the load

Caching is a major process of speeding up your application. By caching the output of a potentially large scale operation and displaying that output later from that cache saves a lot of time. First if you have cached output of a large scale operation, just check whether the cache is out-of-date or not. If it succeeds all the parameters, display the cache which saves that time consuming procedures. These large scale operation may include fetching content from an remote source using CURL or any stream wrapper, or a database operation, or even a time consuming graph/chart generation.

There are several types of caching, like opcode caching, output caching etc. Compiler level caching are out of scope in this article. We are simply talking about the output caching. In this step PEAR::Cache or PEAR::Cache_Lite may save really a great amount of time. Lets take a look how.

In the following example we are using PEAR::Cache. You can also use Cache_Lite which functions great by sacrificing a bit performance.


$cache = new Cache_Output("file");
if (!$cache->start($cacheid, "samplegroup"))
	//perform large scale operation
        echo "Hello";
	echo "Printing from Cache";

//display the cached content
echo $cache->get($cacheid);

First time, the output is generated as usually. But from next time the output is displayed from cached bypassing the large scale operation.

You can flush content of any cached group by accessing it’s flush() method which looks like below

$cache->flush( “samplegroup”);

Thats it.