Monday, March 25, 2013

Ruminating on Server Side Push Techniques

The ability to push data from the server to the web browser has always been a pipe-dream for many web architects. Jotting down the various techniques that I used in the past and the new technologies on the horizon that would enable server side push.
  • Long Polling (Comet): For the last few years, this technique has been most popular and is used behind the scenes by multiple ajax frameworks, such as DoJo, WebSphere Ajax toolkit, etc. The fundamental concept behind this technique is for the server to hold on to the request and not respond till there is some data. Once the data is ready, push the data to the browser as the HTTP Response. After getting the response, the client would again make a new poll request and wait for the response. Hence the term - "long polling". 
  • Persistent Connections / Incomplete Response: Another technique in which the server never ends the response stream, but always keeps it open. There is a  special MIME type called multipart/x-mixed-replace, that is supported by most browsers expect IE :) This MIME type enables the server to keep the response stream open and send data in deltas to the browser.
  • HTML 5 WebSockets: The new HTML 5 specification brings to us the power of WebSockets that enable full-duplex bidirectional data flow between browsers and servers. Very soon, we would have all browsers/servers supporting this. 

Monday, March 18, 2013

What is the actual service running behind svchost.exe

I knew that a lot of Windows Services (available as DLLs) run under the host process svchost.exe during start-up. But is there any way to find out what is the actual mapping service behind each svchost.exe? Sometimes, the svchost.exe process occupies a lot of CPU/memory resources and we need to know the actual service behind it.

The answer is very simple on Windows 7. Press "Ctr+Shift+Esc" to open the Task Manager.
Click on "Show processes from all users". Just right click on any svchost.exe process and in the context menu, select "Go To Service". You would be redirected to the Services Tab, wherein the appropriate service would be highlighted.

Another nifty way is to use the following command on the cmd prompt:
tasklist /svc /fi "imagename eq svchost.exe"

Tuesday, March 12, 2013

Behind the scenes..using OAuth

Found the following cool article on the web that explains how OAuth works behind the scenes..

OAuth 2.0 is essentially an authentication & authorization framework that enables a third-party application to obtain limited access to any HTTP service (web application or web service). It essentially is a protocol specification (a token-passing mechanism) that allows users to control which applications have access to their data without revealing their passwords or other credentials.Thus it can also be used for delegated authentication as mentioned here.

OAuth is also very useful when you are exposing APIs that third party applications may use. For e.g. all Google APIs can now be accessed using OAuth 2.0 protocol specification. In fact, for web-sites and mobile apps running on Android/iOS, Google has released a solution called as Google+ Sign-In for delegating authentication to Google. More information is available here:

The basic steps for any application to use OAuth is to first register/create a Client ID (client key) on the OAuth Authorization Server (e.g. Google, Facebook) along with a secret. (This is the crux of the solution, which I had missed in my earlier understanding :) Since the application is registered with the Service Provider, it can make requests now for access to services.) Then create a request token that would be authorized. Finally create a new pair of access tokens that would be used to access the services.
To understand these concepts, Google has also made a cool web app called OAuth PlayGround, where developers can play around with OAuth requests.

A good illustration for OAuth is provided on the Magento website here

Thursday, March 07, 2013

Jigsaw puzzles in PowerPoint and Visio

Found this cool tutorial on the web that can be used to make jigsaw puzzles in PowerPoint or Visio. One of my friends actually used this technique to create a good visualization on technology building blocks.

Monday, March 04, 2013

Long file names on Windows

Just spend the last 30 mins in total frustration on the way Windows 7 handles long file names. I was essentially trying to copy "LifeRay Social Office" portal folder structure from one location to the other.

On my Windows 7 desktop, the copy command from Explorer won't just work ! No error message, no warning, just that the window disappears. I did a remote desktop to the server and tried to copy from there. On the Windows Server 2000 box, I atleast got an error message - "Cannot copy file". But that's it, no information on why copy did not work.

I debugged further and tried to copy each individual file and only then did I get a meaningful error message - "The file Name(s) would be too long for the destination folder." So essentially the total path (string-length) of the files were long enough for Windows to go awry.

A quick google search showed that this is a core windows problem. Windows Explorer (File Explorer in Windows 8 and Windows Server 2012) uses ANSI_API calls which is limited to 260 characters in the paths. There are some hot fixes available as a patch on windows, but did not try them yet.

So what are the options then? MS has released a tool called as RoboCopy that can handle this problem. Another popular tool is LongPathTool. In my case, fortunately I had JDK installed on my box. I used the jar command to inflate/deflate the folder structure between copying and it worked like a charm :) Strangely WinZip on Windows 7 did not work as it threw some weird error long file names.

There is another headache due to long file names. You also cannot delete such directories form the windows explorer !. I tried using the rmdir command from the command prompt and thankfully that worked !!!