Wednesday, January 19, 2011

Visual studio resolve usings.

I had the absolute joy of hanging out at the Tampa .NET Developer Group Meeting last night. Here is a tip I mentioned briefly at the meeting around resolving and optmizing namespaces.

I know a lot of us are hardcore ReSharper users and can do all kinds of jedi-like moves in the Visual Studio 2008 IDE. However, get us without ReSharper at times and we just stare at the keyboard in bewilderment wondering why namespaces aren't resolving, usings are being optimized, and refactorings are not taking place at blinding speeds, etc.

It turns out, Visual Studio 2008 actually has good support for resolving namespaces and optmizing using statements that can get you the functionality if you are not using ReSharper.



Resolving Namespaces

When you are writing code and Visual Studio places a small red notification rectangle at the end of the class,







Pressing Ctrl + . will bring up a context-sensitive menu that allows you to add a using statement or optionally fully qualify the path to the class.







Clicking the Enter Key will automatically add using System.Collections.Generic; with the other using statements, requiring no touch of the mouse.

This also works with attributes as well. Adding an attribute that needs a using statement to qualify its namespace will cause the same red notification rectangle to appear:







and pressing Ctrl + . will allow you to add the using statement, etc.







Optimizing, Removing, and Sorting Unused Using Statements

The other nice thing that ReSharper does is remove unused using statements using Ctrl+Alt+O.

We can get that using Visual Studio 2008, because you may have noticed the cool context-sensitive Organize Usings Option:







Very, very cool, but this needs to have a shortcut because we will be using it often. I want to use the familar Ctrl+Alt+O shortcut that I get from ReSharper ( you can choose your own ) so I need to map the shortcut to the Edit.RemoveAndSort Command in the keyboard options:







Now when I type Ctrl+Alt+O in the code editor it will remove all unused using statements in the current file as well as sort those using statements that are being used.

Very cool! Hope this helps.

Monday, January 3, 2011

A simple distributed lock with memcached

When you have a cluster of web application servers, you often need to coordinate the activity of your servers to avoid the same expensive work being done at the same time when a condition triggers it.

Most people use memcached as a simple key/value store but it can also be used as a simple distributed lock manager: along with the put(key, value) operation, it also has an add(key, value) operation that succeeds only if the cache wasn't already holding a value for the key.

Locking then becomes easy:

if (cache.add("lock:xyz", "1", System.currentTimeMillis() + 60000)) 
{   
 try 
 {
      doSomeExpensiveStuff();  
 } 
 finally 
 {  
    cache.delete("lock:xyz");   
 } 
} else {
   // someone else is doing the expensive stuff 
} 

The code above tries to get the lock by adding a dumb value for our lock's idenfitier, with an expiration of one minute. This is the lock lease time, and should be more than the estimated maximum time for the lengthy operation. This avoids the lock being held forever if ever things go really bad such as your server crashing.

Once the operation is completed, we delete the lock, et voilĂ .

If you want the system to be rock-solid, you should check that you still own the lock before deleting it (in case the lease time expired), but in most cases this simple approach works nicely.

And if the expensive operation resets in the database the condition that triggered it, the lock should be released once the transaction has been committed to prevent a race condition in the time interval between the end of the expensive operation and the actual commit that would allow other servers to restart the same work. Spring's transaction synchronization helps doing that.

Registering an Application to a URL protocol

Registering an Application to a URL Protocol

The About Asynchronous Pluggable Protocols article describes how to develop handlers for URL protocols. In some cases, it may be desirable to invoke another application to handle a custom protocol. To do so, register the existing application as a URL Protocol handler. Once the application has successfully launched, it can use command-line parameters to retrieve the URL that launched it. These settings apply to protocol handlers launched from within Windows Internet Explorer and from Windows Explorer using the Run... command (Windows logo key+R).

security note Security Alert Applications that handle URL protocols must consider how to respond to malicious data. Because handler applications can receive data from untrusted sources, the URL and other parameter values passed to the application may contain malicious data that attempts to exploit the handling application.

This topic contains the following sections:

Registering the Application Handling the Custom Protocol

To register an application to handle a particular URL protocol, add a new key, along with the appropriate subkeys and values, to HKEY_CLASSES_ROOT. The root key must match the protocol scheme that is being added. For instance, to add an "alert:" protocol, add an alert key to HKEY_CLASSES_ROOT, as follows:

HKEY_CLASSES_ROOT
alert
URL Protocol = ""

Under this new key, the URL Protocol string value indicates that this key declares a custom protocol handler. Without this key, the handler application will not launch. The value should be an empty string.

Keys should also be added for DefaultIcon and shell. The Default string value of the DefaultIcon key must be the file name to use as an icon for this new URL protocol. The string takes the form "path, iconindex" with a maximum length of MAX_PATH. The name of the first key under the shell key should be an action verb, such as open. Under this key, a command key or aDDEEXEC key indicate how the handler should be invoked. The values under the command and DDEEXEC keys describe how to launch the application handling the new protocol.

Finally, the Default string value should contain the display name of the new protocol. The following example shows how to register an application, alert.exe in this case, to handle the alertprotocol.

HKEY_CLASSES_ROOT
alert
(Default) = "URL:Alert Protocol"
URL Protocol = ""
DefaultIcon
(Default) = "alert.exe,1"
shell
open
command
(Default) = "C:\Program Files\Alert\alert.exe" "%1"

When a user clicks a link registered to your custom URL protocol, Internet Explorer launches the registered URL protocol handler. If the specified open command specified in the registry contains a %1 parameter, Internet Explorer passes the URL to the registered protocol handler application.

Launching the Handler

By adding the above settings to the registry, navigating to URLs such as alert:Hello%20World would cause an attempt to launch alert.exe with the complete URL on the command line. Internet Explorer decodes the URL, but the Windows Run... command does not. If a URL contains spaces, it may be split across more than one argument on the command line.

For example, if the link above is followed through Internet Explorer, the command line would be:

"C:\Program Files\Alert\alert.exe" "alert:Hello World" 

If this link is followed through Windows Explorer, the Windows Run command, or some other application, the command line would be:

"C:\Program Files\Alert\alert.exe" "alert:Hello%20World" 

Because Internet Explorer will decode all percent-encoded octets in the URL before passing the URL to ShellExecute, URLs such as alert:%3F? will be given to the alert application protocol handler as alert:??. The handler won't know that the first question mark was percent-encoded. To avoid this issue, application protocol handlers and their associated URL scheme must not rely on encoding. If encoding is necessary, protocol handlers should use another type of encoding that is compatible with URL syntax, such as Base64 encoding. Double percent-encoding is not a perfect solution either; if the application protocol URL isn't processed by Internet Explorer, it will not be decoded.

When ShellExecute executes the application protocol handler with the URL on the command line, any non-encoded spaces, quotes, and slashes in the URL will be interpreted as part of the command line. This means that if you use C/C++'s argc and argv to determine the arguments passed to your application, the URL may be broken across multiple parameters. To mitigate this issue:

  • Avoid spaces, quotes, or backslashes in your URL
  • Quote the %1 in the registration ("%1" as written in the 'alert' example registration)

However, avoidance doesn't completely solve the problem of quotes in the URL or a backslash at the end of the URL.

Security Issues

As noted above, the URL that is passed to an application protocol handler might be broken across multiple parameters. Malicious parties could use additional quote or backslash characters to pass additional command line parameters. For this reason, application protocol handlers should assume that any parameters on the command line could come from malicious parties, and carefully validate them. Applications that could initiate dangerous actions based on external data must first confirm those actions with the user. In addition, handling applications should be tested with URLs that are overly long or contain unexpected (or undesirable) character sequences.

For more information, please see Writing Secure Code.

Example Protocol Handler

The following sample code contains a simple C# console application demonstrating one way to implement a protocol handler for the alert protocol.

using System; using System.Collections.Generic; using System.Text;  namespace Alert {   class Program   {     static string ProcessInput(string s)     {        // TODO Verify and validate the input         // string as appropriate for your application.        return s;     }      static void Main(string[] args)     {       Console.WriteLine("Alert.exe invoked with the following parameters.\r\n");       Console.WriteLine("Raw command-line: \n\t" + Environment.CommandLine);        Console.WriteLine("\n\nArguments:\n");       foreach (string s in args)       {         Console.WriteLine("\t" + ProcessInput(s));       }       Console.WriteLine("\nPress any key to continue...");       Console.ReadKey();     }   } } 

When invoked with the URL alert:"Hello%20World" (note extra quotes) from Internet Explorer, the program responds with:

Alert.exe invoked with the following parameters.  Raw command-line:         "C:\Program Files\Alert\alert.exe" "alert:"Hello World""   Arguments:          alert:Hello         World  Press any key to continue...