khairie [dot] net

Creating Symbolic Links in Linux

2012-01-18T22:18:00.001+08:00

This one is a pretty easy one, but I always forgot which one comes first - the symlink name or the target file/folder.

So, here's how you create a symlink in Linux:-

ln -s [target directory or file] [symlink shortcut]

For example:-

ln -s /usr/local/lib/example /example

Fix the Netbeans Master Password prompt on Ubuntu 64 bit

2012-01-05T09:17:00.000+08:00

Refer: http://wiki.netbeans.org/FaqMasterPasswordDialog

I've been having this prompt for quite a while everytime I run Netbeans, and it's REALLY annoying. Googled a bit, and found out about the wiki above. So, here's what I did to solve it.

Note: for Netbeans 7, you can just disable the master password by using the options below in your netbeans.conf (this should be in your <netbeans installation path>/etc/netbeans.conf):-

-J-Dnetbeans.keyring.no.master=true

For Netbeans 6 though, you have to troubleshoot what's causing the problem. First, add this option to netbeans.conf:-

-J-Dorg.netbeans.modules.keyring.level=FINE

Then start Netbeans, and you will STILL get the master password prompt. Check the IDE log (View -> IDE Log). Look for what's causing the error. In my case, I found this in my IDE Log:

java.lang.UnsatisfiedLinkError: Unable to load library 'gnome-keyring': libgnome-keyring.so: cannot open shared object file: No such file or directory
        at com.sun.jna.NativeLibrary.loadLibrary(NativeLibrary.java:145)
        at com.sun.jna.NativeLibrary.getInstance(NativeLibrary.java:188)
        at com.sun.jna.Library$Handler.(Library.java:123)
...

I tried to search fo that libgnome-keyring.so file in my Ubuntu installation:-

khairi@ubuntu:~$ locate libgnome-keyring.so
/usr/lib/libgnome-keyring.so.0
/usr/lib/libgnome-keyring.so.0.1.1
/usr/lib32/libgnome-keyring.so
/usr/lib32/libgnome-keyring.so.0
/usr/lib32/libgnome-keyring.so.0.1.1

Ahah! I'm missing that libgnome-keyring.so file in my 64 bit library. That's why Netbeans is prompting me for the master password, coz it can't find my Gnome Keyring.

This is an easy fix. Just create a soft symlink to the actual file. Here's how:-

khairi@ubuntu:~$ cd /usr/lib
khairi@ubuntu:/usr/lib$ sudo ln -s libgnome-keyring.so.0 libgnome-keyring.so
khairi@ubuntu:/usr/lib$ ls -l libgnome-keyring*
lrwxrwxrwx 1 root root     21 2012-01-05 09:06 libgnome-keyring.so -> libgnome-keyring.so.0
lrwxrwxrwx 1 root root     25 2011-10-19 09:05 libgnome-keyring.so.0 -> libgnome-keyring.so.0.1.1
-rw-r--r-- 1 root root 139144 2011-09-26 21:45 libgnome-keyring.so.0.1.1
khairi@ubuntu:/usr/lib$ ls -l ../lib32/libgnome-keyring*
lrwxrwxrwx 1 root root     21 2011-10-19 09:06 ../lib32/libgnome-keyring.so -> libgnome-keyring.so.0
lrwxrwxrwx 1 root root     25 2011-10-19 09:06 ../lib32/libgnome-keyring.so.0 -> libgnome-keyring.so.0.1.1
-rw-r--r-- 1 root root 133412 2011-09-26 21:44 ../lib32/libgnome-keyring.so.0.1.1

That should fix it.

Cheers!

SVN revert all other properties changes after an SVN merge

2011-10-19T08:15:00.000+08:00

Usually after doing an SVN merge, there will be a lot of SVN properties changes to our local copy. If we committed these changes to our repo, we will have headache later on to resolve the issues (at least that's what my colleague told me).

Thus, in order to revert all other properties changes (except for the things that we really change) after doing an SVN merge, type the following in shell prompt:-

svn status | grep "^ M" | awk '{print $2}' | xargs svn revert

PHP Pretty Backtrace

2011-10-09T13:19:00.000+08:00

Sometimes during development, you want to look at the execution stack to see what's being passed around, and from where it's being called. That's when PHP's debug_backtrace function becomes useful. At other times, you probably only want a few of those information for debugging. Here's how I filtered those info to get what I really want:-

$log = debug_backtrace();
for ($i=0; $i < sizeof($log); $i++) {
  $dump[$i]['file'] = $log[$i]['file'];
  $dump[$i]['function'] = $log[$i]['function'];
  $dump[$i]['line'] = $log[$i]['line'];
}
error_log(__METHOD__.":".__LINE__.print_r($dump,true));

I've also found this blog post, which does similar, but only takes out the args argument.

$trace = debug_backtrace();
foreach ($trace as &$t) {
  unset($t['args']);
}
echo "<pre>".print_r ($trace,true)."</pre>";
exit();

Happy debugging!

How to change workspace in Netbeans

2011-09-09T16:29:00.002+08:00

By default, Netbeans' workspace is located at

/home/<username>/.netbeans/<version>

or (for Windows users):-

C:\Users\<username>\.netbeans\<version>

If you want to work with different workspace, then you'll need to use the --userdir switch when executing Netbeans' shortcut.

In Linux, right click on your Netbeans' shortcut, and edit the command to be something like this:-

/bin/sh "/path/to/NetBeans x.x/bin/netbeans" --userdir /path/to/your/new/workspace

For Windows users, right click on your Netbeans' shortcut, and select Properties.
In the Target textbox, add the --userdir parameter to the very end. It should look something like this:-

"C:\Program Files\NetBeans x.x\bin\netbeans.exe" --userdir C:\path\to\new_workspace

Tips: You can create multiple shortcuts for different workspaces ;-)

References:-
http://stackoverflow.com/questions/973557/is-there-a-netbeans-equivalent-to-eclipses-workspace
http://mangstacular.blogspot.com/2009/03/netbeans-and-workspaces.html

Syntax Highlighting in Blogger/Blogspot

2011-08-25T22:40:00.010+08:00

One of the challenges I faced when moving to Blogger's platform is to get syntax highlighting work correctly for all my previous posts. After digging around Google for a while, I found out about Alex Gorbatchev's SyntaxHighlighter. Here's how I did it for my blog here (remember to replace <version> with the appropriate version):

1) Go to Blogger's Design>Edit Html page.
2) In the <head> section, include the style sheets and related scripts.

<link href='http://alexgorbatchev.com/pub/sh/<version>/styles/shCore.css' rel='stylesheet' type='text/css'/> 
<link href='http://alexgorbatchev.com/pub/sh/<version>/styles/shThemeDefault.css' rel='stylesheet' type='text/css'/> 
<script src='http://alexgorbatchev.com/pub/sh/<version>/scripts/shCore.js' type='text/javascript'></script> 
<script src='http://alexgorbatchev.com/pub/sh/<version>/scripts/shAutoloader.js' type='text/javascript'></script>

3) Right before </body>, add this snippet so that autoloader works.

<script language='javascript' type='text/javascript'>
function path() {
  var args = arguments, result = [];

  for(var i = 0; i &lt; args.length; i++)
      result.push(args[i].replace('@', 'http://alexgorbatchev.com/pub/sh/<version>/scripts/'));
  return result
};

SyntaxHighlighter.autoloader.apply(null, path(
  'applescript            @shBrushAppleScript.js',
  'actionscript3 as3      @shBrushAS3.js',
  'bash shell             @shBrushBash.js',
  'coldfusion cf          @shBrushColdFusion.js',
  'cpp c                  @shBrushCpp.js',
  'c# c-sharp csharp      @shBrushCSharp.js',
  'css                    @shBrushCss.js',
  'delphi pascal          @shBrushDelphi.js',
  'diff patch pas         @shBrushDiff.js',
  'erl erlang             @shBrushErlang.js',
  'groovy                 @shBrushGroovy.js',
  'java                   @shBrushJava.js',
  'jfx javafx             @shBrushJavaFX.js',
  'js jscript javascript  @shBrushJScript.js',
  'perl pl                @shBrushPerl.js',
  'php                    @shBrushPhp.js',
  'text plain             @shBrushPlain.js',
  'ps powershell          @shBrushPowerShell.js',
  'py python              @shBrushPython.js',
  'ruby rails ror rb      @shBrushRuby.js',
  'sass scss              @shBrushSass.js',
  'scala                  @shBrushScala.js',
  'sql                    @shBrushSql.js',
  'vb vbnet               @shBrushVb.js',
  'xml xhtml xslt html    @shBrushXml.js'
));

  SyntaxHighlighter.config.bloggerMode = true;
  SyntaxHighlighter.all();
</script>

4) Start adding code snippets with <pre> tags. Example:-

<pre class="brush: js">
alert('This is an example!');
</pre>

5) The result of the above code would be:-

alert('This is an example!');

That's it! You should now have syntax highlighting in your blog ;-)

p/s: You can also view the source of this blog to see what I have included here.

Fix ugly characters in emacs shell buffer

2011-07-29T15:50:00.003+08:00

After installing Emacs on a fresh installation of Ubuntu (maybe even other linux distros), I encountered this ugly-looking characters filling up my emacs shell buffer:-

khairie@khairi-ubuntu:~$ ls
^[[01;34mDesktop^[[0m   ^[[01;34mDownloads^[[0m  examples.desktop  ^[[01;34mPictures^[[0m  ^[[01;34mVideos^[[0m
^[[01;34mDocuments^[[0m  ^[[01;34mMusic^[[0m  ^[[01;34mPublic^[[0m    ^[[01;34mUbuntu One^[[0m     
khairie@khairi-ubuntu:~$

This is caused by the coloring being enabled in Ubuntu shell. Referring to this post, you can actually turn on the coloring inside emacs as well. Type the following command in emacs.

M-x ansi-color-for-comint-mode-on

Or even better, edit your .emacs file, and add in the following line:-

(add-hook 'shell-mode-hook 'ansi-color-for-comint-mode-on)

That should fix it. Cheers ;-)

Finished migrating to blogger...

2011-07-28T09:39:00.000+08:00

Yup, finally finished migrating all my posts to blogger.

Why switch? Just because I don't have the time to maintain my own blogging platform. Drupal is good. Excellent in fact. But I just don't have the time to keep updating and upgrading them, especially when there are security releases (unless if I can get paid to do it :-p).

Anyhow, hope I can continue blogging after taking a whole year 'timeout'. Hehe..

Bash : Compound condition using && (AND) and || (OR) operators

2010-06-21T17:47:00.001+08:00

To have compound conditions in bash using && or ||:-

if [ $VAR1 -eq 1 ] && [ "$VAR2" == "true" ]
then
  # do something
fi

if [ "$VAR1" == "String 1" ] || [ "$VAR2" == "String 2" ]
then
  # do something
fi

Reference: http://www.museum.state.il.us/ismdepts/library/linuxguides/abs-guide/ops.html#ANDOR

Bash : Compare Strings With Spaces

2010-06-15T13:01:00.009+08:00

To compare two strings that contain spaces in bash, enclose them with a double quote (e.g. "$VAR" ). Example below:-

if [ "$VAR" == "this is a string with spaces" ]
then
  # do something
fi

Reference: http://hacktux.com/bash/string

Exclude files/directories when using linux tar command

2010-03-02T01:06:00.003+08:00

To exclude files/folders when archiving using tar in linux, use the following command:-

tar -czvf filename.tar.gz --exclude 'path1/*' --exclude 'path2/*' --exclude 'path3/*' folder_or_files_to_be_archived

References:-
http://www.wallpaperama.com/forums/how-to-exclude-a-directory-in-a-tar-linux-command-t6486.html
http://www.cyberciti.biz/faq/exclude-certain-files-when-creating-a-tarball-using-tar-command/
http://www.linuxtopia.org/online_books/linux_tool_guides/tar_user_guide/problems-with-exclude.html

Configure multiple log4j instances in different contexts

2010-02-25T23:15:00.020+08:00

The problem:
We needed to use different log4j.properties file for each context.

We have one library (directory.jar) which resides in tomcat's shared lib that uses log4j. We also have three contexts (core, admin & user contexts) that will be using log4j for logging too.

Initially, we put the log4j library in tomcat's shared lib, and specify log4j.properties for each context's web.xml. But we found out that only one context is able to initialize the log4j properly. The other two contexts doesn't seem to be able to use log4j properly.

After googling around for quite some time, I found out this post that suggests a fix for my problem. It uses log4j-contrib so that log4j can properly select the log4j.properties file based on it's own context.

So, here's the quick run-down on how to do it:-

1) Place log4j.jar in tomcat's shared lib.
2) Download log4j-contrib and place it in tomcat's shared lib as well.
3) For each context that needed to use log4j, edit their web.xml and add the following entries as first entry of their appropriate sections. First, add this context-param entry:-

<context-param>
   <param-name>log4jConfigLocations</param-name>
   <param-value>log4j.properties</param-value>
</context-param>

Then, add the listener (make it the first listener):-

<listener>
   <listener-class>com.mathieucarbou.log4j.web.ContextLoggingListener</listener-class>
</listener>

4) Edit the log4j.properties for each context as you wish. They should be under /WEB-INF/classes/log4j.properties for each context.

Walla! Now log4j is working happily like it should =)

Drupal Blog Page Not Found

2009-09-29T16:29:00.010+08:00

If you have the problem of not being able to see the Blog Module page after enabling it, and you are using Clean URL for your site, then it's *probably* because you have accidentally created a subdomain with the name 'blog' (e.g. http://blog.example.com), or you have a subdirectory named 'blog' on your website's root folder (www, htdocs or public_html). You should be able to see Drupal's Blog Module after you remove the subdomain / subdirectory with that same name.

The reason you cannot view the Blog Module at your website's URL (e.g. http://www.example.com/blog) when you have a subdomain / subdirectory with the same name is because when using Clean URL, Drupal rewrites the Blog Module address (http://www.example.com/?q=blog to http://www.example.com/blog) using .htaccess file. This causes confusion on the webserver, because when we try to access http://www.example.com/blog, it doesn't know which URL it is suppose to display. It could be the URL that Drupal has rewritten, or it could also be the subdirectory that you have in your web root folder.

The same concept also apply with all the URL Aliases that you have created with your Drupal. Make sure you don't have a folder with the same name as your aliases, or else you will get a Page Not Found error.

Hope this helps

XAMPP Virtual Hosts in Ubuntu

2009-08-15T16:16:00.017+08:00

Quick HOWTO to enable Virtual Hosts with XAMPP on Ubuntu (or any *nix distribution).

sudo gedit /opt/lampp/etc/httpd.conf

Uncomment the line below (it should be towards the end of the file):

Include etc/extra/httpd-vhosts.conf

Save & Close the httpd.conf file.

sudo gedit /opt/lampp/etc/extra/httpd-vhosts.conf

Make sure the line below is uncommented:

NameVirtualHost *:80

Add/Edit the following piece of code:

<VirtualHost *:80>
    ServerAdmin admin@localhost
    DocumentRoot /path/to/vhosts/folder/site.local
    ServerName site.local
    ErrorLog /path/to/vhosts/logs/site.local-error_log
    CustomLog /path/to/vhosts/logs/site.local-access_log common
</VirtualHost>

Save & Close the httpd-vhosts.conf file.

sudo gedit /etc/hosts

Add the following line to the hosts file:

127.0.0.1 site.local

Save & Close the hosts file.

sudo /opt/lampp/lampp restart

Open up web browser, and go to http://site.local to see if it's working.

Cheers.

References:
http://www.dimitritaghon.be/2008/07/xampp-and-virtual-hosts/
http://mintarticles.com/read/articles_about_software/multiple_virtual_hosts_in_xampplampp_in_ubuntu,977/

Format USB Thumb drive with NTFS filesystem

2009-06-03T09:07:00.006+08:00

In Windows, we all know that we can't format our thumb drive to NTFS by default. There are no options for NTFS when we're trying to format our thumb drive. But that doesn't mean that we can't have NTFS filesystem on our thumb drive. All we have to do is use the Convert utility to change the thumb drive's filesystem to NTFS. Here's how:

1) Open your Command Prompt
2) Assuming the drive for your thumb drive is "X", type the following line in command prompt:-

CONVERT X: /fs:ntfs

That's all to it. You don't even have to format your thumb drive to have an NTFS filesystem on it. Your data will still remains even after converting to NTFS. Note that this is a one way process, you will NOT be able to convert back to FAT32 once you have convert your thumb drive to NTFS. If you want to go back to FAT32, you will have to format your thumb drive (make sure you back up all your data first).

Have fun!

Note: I will NOT be responsible if something goes wrong with your thumb drive, or your Windows for that matter. This post is for reference only.

ASP.NET Client-side Validation & Confirmation Popup

2009-05-22T15:38:00.018+08:00

The other day I was working on a project for our client. They wanted some client-side validation to be done on the page to avoid postbacks, since postbacks could take REALLY long on a slow connection. They also wanted a javascript confirmation popup before processing the form.

In the page, I had a few RequiredFieldValidators, as well as some RegularExpressionValidators. Initially, I modified the submit button to include the confirmation message, and check whether the page is valid or not. Below is a snippet of the ASP.NET button that I had in my page.

<asp:button text="Submit" onclientclick="return (Page_IsValid && confirm('Are you sure?'));" onclick="Button1_Click" runat="server" id="Button1"></asp:button>

Note: Page_IsValid is the client-side version of Page.IsValid that we are used to in codebehind.

Now, here's the problem I was facing the other day. After the page had loaded successfully, I immediately clicked on the submit button to test the validation on the page. Supposedly, error messages should be displayed because the form was not completed, and I have RequiredFieldValidators and RegularExpressionValidators all over the page. But instead, the javascript confirmation popped up. After I clicked OK, then only the error messages came out. Subsequent clicks on the Submit button did not pop up the confirmation message anymore.

From my observation, it seems that Page_IsValid is returning true the first time the page loads, and holds true until we click the submit button the first time. This means that the client-side validation is only being done after our javascript snippet that we have in the OnClientClick event.

A few Google searches showed me that I have to manually call a javascript function Page_ClientValidate() to validate the form. This function is included with the .NET Framework. By calling this function, it will force to update the Page_IsValid variable. The working code in my aspx now is:

<asp:button text="Submit" onclientclick="Page_ClientValidate(); return (Page_IsValid && confirm('Are you sure?'));" onclick="Button1_Click" runat="server" id="Button1"></asp:button>

Hope that helps someone out there.

VB.NET vs C# : The Differences

2009-04-10T13:01:00.000+08:00

For those who are switching from VB.Net to C#, or just switching between the two of them back and forth (that's what I do at work LOL. Had to know both of them 'coz some of our projects are in VB.Net, but we're in the transition to move to C# with new projects), here's a basic summary of the differences between the two languages that you need to know. Note that this list is far from complete, but they will definitely useful for basic operations at the very least. This list was given to me a long time ago by one of my lecturer.

Note: You can scroll the table below left-right to be able to fully view them.

Features	VB.NET	C#
Case Sensitivity	Not case-sensitive. Example: response.write("Hello") ' OK	Case-sensitive. Example: response.write("Hello"); // Error Response.Write("Hello"); // OK
Functional Blocks	Use beginning and ending statements to declare functional blocks of code. Example: Sub Show(ByVal str As String) Response.Write(str) End Sub	Use curly braces to declare functional blocks of code. Example: void Show(string str) { Response.Write(str); }
Type Conversion	Implicit type conversions are permitted by default. Example: Dim x As Integer x = 3.15 ' OK	Implicit type conversions are limited to operation that are guaranteed not to lose information, such as converting from int to float. Example: int x = 0; float y = x; // OK Other type conversions are performed explicitly by casting. Example: float y = 3.14; int x = (int)y; // Cast OK or by using type conversion methods. Example: string z = y.ToString();
Comments	Always start with an apostrophe ('). Example: ' This is a comment	Three types: 1) Block comment. Example: /* This is a comment. */ 2) Inline comment. Example: // This is a comment. 3) Documentation. Example: /// /// This is a comment. ///
Arrays	Array elements are specified using parentheses. Example: arrFruit(0) = "Apple"	Array elements are specified using square brackets. Example: arrFruit[0] = "Apple";
Methods	You can omit parenthesis after method names if arguments are omitted. Example: str = objX.ToString	You must include parenthesis after all methods. Example: str = objX.ToString();
Statement Termination	Terminated by using carriage return. Example: Response.Write("Hello")	Terminated by a semicolon(;). Example: Response.Write("Hello");
Statement Continuation	Continued using underscore (_). Example: int x = System.Math.PI * _ intRadius	Continued until semicolon (;), and can span multiple lines if needed. Example: int x = System.Math.PI * intRadius;
String Operator	Use the ampersand (&) or plus (+) to join strings. Example: strFruit = "Apples " & _ "Oranges"	Use plus (+) sign to join strings. Example: strFruit = "Apples " + "Oranges";
Comparison Operators	Use =, >, <, >=, <=, <> to compare values. Example: If intX >= 5 Then	Use ==, >, <, >=, <=, != to compare values. Example: if (intX >= 5)
Negation	Use the Not keyword to express logical negation. Example: If Not IsPostBack Then	Use the ! operator. Example: if (!IsPostBack)
Object Comparison	Use the Is keyword to compare object variables. Example: If objX Is objY Then	Use == to compare object variables. Example: if (objX == objY)
Object Existence	Use the Nothing keyword or the IsNothing() function to check whether an object exists. Example: If IsNothing(objX) Then	Use the null keyword to check whether an object exists. Example: if (objX == null)

How to remove/delete XmlNode from an XmlDocument in C#

2009-03-17T12:27:00.000+08:00

Suppose we have an XML like below:-

<parent>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
    <childnode3>some text 3</childnode3>
  </children>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
    <childnode3>some text 3</childnode3>
  </children>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
    <childnode3>some text 3</childnode3>
  </children>
</parent>

Now, for some reason (whatever the reason is), we want our application to remove the node childnode3 from all children nodes. Here’s one method to accomplish this task:-

XmlNode nodeList = xmldoc.SelectSingleNode("//parent");

foreach (XmlNode node in nodeList) {
  XmlNode tempNode = node.SelectSingleNode(”//children/childnode3″);

  if (tempNode != null)
    tempNode.ParentNode.RemoveChild(tempNode);
}

The resulting XML would be like this:

<parent>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
  </children>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
  </children>
  <children>
    <childnode1>some text 1</childnode1>
    <childnode2>some text 2</childnode2>
  </children>
</parent>

Reference: http://channel9.msdn.com/forums/TechOff/77233-Deleting-an-XML-node-in-C/?CommentID=77236

How to find controls in Repeater header/footer template in C#

2009-03-16T12:52:00.000+08:00

Here's one method to find controls in Repeater header/footer template in ASP.NET. The codes below are in C#.

To find a control in Repeater header template:-

Literal litControl = (Literal)repeater1.Controls[0].Controls[0].FindControl("litControl");

To find a control in Repeater footer template:-

Literal litControl = (Literal)repeater1.Controls[repeater1.Controls.Count - 1].Controls[0].FindControl("litControl");

Reference: http://forums.asp.net/t/1193229.aspx#2494827

vbproj / csproj requires Policy Management

2009-01-13T12:45:00.000+08:00

Last Friday, my colleague was having this problem opening our VB Project at client’s site. We brought the source code from our development server in the office. The error message says:

The project 'http://localhost/myproj.vbproj‘ requires Policy Management. Open this project in Visual Studio .NET Enterprise Developer or Visual Studio.NET Enterprise Architect.

Googling gave us so many method of trying to solve this issue. Some even say that we can’t open a project that was created using Visual Studio.NET Enterprise Edition from a Visual Studio.NET Standard Edition. But I think that's not entirely true. Here’s what I did to be able to re-open our project:-

1) Make a backup of your myproj.vbproj (or myproj.csproj if it's a C# Project).
2) Open myproj.vbproj in Notepad (or any of your favourite text editor).
3) Remove the line with "OwnerKey".
4) Save that myproj.vbproj, and try to open it in Visual Studio.NET again.

Hope that helps.

Request Filtering module is configured to deny a request that exceeds request content length

2008-12-28T14:09:00.035+08:00

It's been a while since I last wrote on this blog. Been busy with work + real life. Anyway, here's what I encountered a couple of days ago.

We were working on a module that would allow users to upload files to the server (the file is stored in SQL Server as VARBINARY). The maximum allowed file size to be uploaded was made configurable through web.config file. We set those to be 50mb. When we tried to upload a 36mb file, IIS7 gave us this error:

Most google search results says that we have to add the following entry to our web.config file (the value for maxAllowedContentLength could be different from below):-

<system.webServer>
  <security>
    <requestFiltering>
      <requestLimits maxAllowedContentLength="2000000000"/>
    </requestFiltering>
  </security>
</system.webServer>

The problem is, this setting will NOT work with .NET 1.1 (oh yeah, you read that right! I was working on .NET 1.1... **sigh**) on IIS7 because it does not know the <system.webServer> property. Further searching yield us with the solution. Here's the solution:-

1) Open notepad (as Administrator)
2) Open the file:

%windir%\system32\inetsrv\config\applicationHost.config

3) Find:

</requestFiltering>

4) Add before that </requestFiltering> line.

<requestLimits maxAllowedContentLength="2000000000" />

5) Final applicationHost.config file should be (only related part shown):

      <requestLimits maxAllowedContentLength="2000000000" />
    </requestFiltering>
  </security>

An error occurred while parsing EntityName. Line X, position Y.

2008-12-11T23:17:00.007+08:00

So today I was at client's site providing Technical Support and finalizing Enhancement Project that was done for the past few months. One of the task was for me to develop a Windows Service, which will grab the user's updated profile from a table in SQL Server 2000 and submit them to a Web Service developed by another vendor. I am using SOAP in my Windows Service to communicate with the Web Service on the other side. Since SOAP is using an XML format in their request and response, I had to grab the data from the SQL Server and build a SOAP Request XML before sending it to the Web Service. Previously it ran just fine, until it hit an error that says:-

An error occurred while parsing EntityName. Line X, position Y.

No matter how many retries, it keeps on hitting the same error and bail out of my Windows Service. After googling around for a bit, it turns out that one of the content in the user's profile column (address field) contains an ampersand (&). The content of the column looks something like this:-

Mailbox 3 & 4

Referring to this website, I found out that I have to replace the ampersand (&) sign into an equivalent HTML entity (&). The reason? I'm not so sure, but I think it has something to do on how .NET interprets the XML string that was built.

I was so lazy to go and recompile my Windows Service just to fix that tiny bug. So, what I did was to change my Stored Procedure in SQL Server 2000 so that it returns the data with '&' replaced with '&'. This is how I did it in my Stored Procedure (this is just an example):-

SELECT REPLACE(column_name, '&', '&amp;') AS column_name FROM table_name WHERE some_conditions

That fixed my problem without having to recompile my Windows Service again. If I were to recompile again, then my code (in C#) would be something like this (refer to this website):-

public string EscapeXml (string input) {
  string output = string.Empty;
 
  output = input.Replace("&", "&amp;");
  output = output.Replace("'", "&apos;");
  output = output.Replace("\"", "&quot;");
  output = output.Replace(">", "&gt;");
  output = output.Replace("<", "&lt;");
 
  return output;
}

Remove Java Quick Starter from Firefox

2008-11-27T10:25:00.012+08:00

This problem has actually happen quite a while for me. Everytime I plug in FirefoxPortable in my office PC, it will automatically install Java Quick Starter 1.0 extension on the first run. What annoy me the most is, this extension CANNOT be uninstalled from Firefox Extension menu, and it is NOT located in the extensions folder in Firefox Profile.

After googling around, I found out this post at MozillaZine forum. It seems that this situation happens after I updated my JRE to Version 6 Update 10 a while back. By default, after installing JRE6u10, Java will install the Java Quick Starter extension to your Firefox through Windows Registry. So, if you want to get rid of this extension, follow this steps:

WARNING!!! Do NOT perform these steps, unless you REALLY know what you are doing!!

1) Open up Windows Registry.
2) Navigate to "HKEY_LOCAL_MACHINE\SOFTWARE\Mozilla\Firefox\extensions".
3) Remove the entry named "jqs" (or something similar).
4) Restart your Firefox.

Beware of your old hard drives…

2006-04-04T16:04:00.000+08:00

I was so bored when I encountered this page at PC World. It started with a quiz to find out whether something you know about computer is a fact of a fiction. I read until the last question:-

True or false? I’m ready to give away my old computer after I’ve deleted all the files.

False. Deleting files isn't good enough. You'd be amazed what people can pull from the hard drives of secondhand or donated computers if the hard drive hasn't been completely reformatted and Windows reinstalled. (And if you pass along the PC to a family member, you're probably not going to do this.)

PC World did a story where we found Social Security numbers, income tax returns, and other personal and important information on old, discarded hard drives. Once you pass along your old PC, you have no idea where it will travel down the line. To erase your info, try a free utility program, such as Eraser, to zap documents into oblivion. There's also Webroot's Window Washer for cleaning up the many little nuggets that lurk elsewhere.

After reading that part of the quiz, it leads me to this page. I realized something crucial regarding security and personal data. Even after formatting the hard drives, people can still access the data of previous owner of the hard drive if it is not proplerly destroyed. Go ahead and read the page, you’ll probably end up surprised like I did.

Here’s the page — PC World - Hard Drives Exposed.