Setting up PlantUML for coding C4 architecture diagrams - Video Tutorial

At last I published a video tutorial. As always its very difficult to get the first job done, if its a different type of job.

The tutorial is about setting Visual Studio Code with PlantUML to code C4 architecture diagrams  as explained in one of my previous post.

Since I don't have the right recording equipment and not a native English speaker, I used a text to speech service. It is AWS->Polly service. I didn't feel somehow the voices in Azure are good. The voice is US English Male voice named Matthew.

You tube link is given below and also embedded.
https://www.youtube.com/watch?v=Zt3Bj1HMJ8g

Javascript window.external.notify to communicate to WebBrowser control

How to communicate from web page loaded inside browser control to the hosting application

There are so many scenarios, where we may need to show web page inside our native applications such as Windows forms, WPF, Windows Phone, Android, iPhone etc... One of the scenario is to integrate with external authentication providers.

If we look at any third party authentication providers such as Google, Facebook, Azure ACS they don't let us take the user authentication details via our forms and get it authenticated using their web services. Instead the are strict in accepting the user name and password only through their web pages. The reason is security. If each and every application starts accepting user's Facebook or Google user name and password, there are chances that some of those apps will store the credentials locally and it might be miss used. That is the issue which is addressed by only allowing web page login even if we are in native applications.

The solution is, if we are developing even WPF or native mobile application and want to get external authentication we need to show their web page in our application. And once the authentication happens inside the browser, the native application can receive the security token and use it for subsequent operations.

For showing identity provider login page, we need to rely on the browser control available in the corresponding platform. If its WPF the browser control name is WebBrowser, in Android its named as WebView and in iOS its UIWebView. It is simple. First part is over, now our native application need to know when the authentication is completed, so that it can remove the browser control and obtain the token. This is the main topic of this post. In other words, the authentication happened inside the browser and browser need to to notify that event to the host application via some techniques.

The same technique, we can reuse the same technique if we want to notify any event happens inside the web page loaded inside the browser control.

Communication from HTML5 web page to WPF

WPF Browser control class has a property called ObjectForScripting. Assign any object to this property and it will be available as the window.external object in Javascript. So whatever methods are available in the class of the object assigned, those can be invoked from Javascript. For example, we created a class called HTMLInteropClass and that class has a method named MyMethod(). If we assign an object of that class to the WebBrowser.ObjectForScirpting property, we can invoke the MyMethod() written in C# from javascript using window.external.MyMethod() statement. Obviously the javascript should be downloaded from server or injected during run time to be running inside the browser control. Steps are as follows.

1. Develop the interop class. Have the method inside it, which we are intended to call from javascript.
2. Create object of that class and assign to the ObjectForScripting property of the WPF WebBrowser control.
3. Point web browser control to the web page which is having invocation code. ie window.external.<methodname>()
4. Develop the web page which is having the invocation code

More details here.

Communication from HTML5 web page to Android

In Android the web browser control name changes. But the handling is kind of same as of WPF browser control. Steps given below

1. Develop a class in Java for Android application with method which will be called from javascript.
2. In the onCreate() method of activity capture the webView reference and relate the object of above class with javascript using addJavascrtiptInterface() method.
3. Make sure the Javascript is changed to call the method, present in the class created in step 1.
4. Point the WebView to the page which contains the code to invoke our Android method.

More details here.

Communication from HTML5 web page to Windows phone

Windows phone uses a different way than WPF. It uses event based method invocation. ie We need to listen to the ScriptNotify event of WebBrowser. Steps below

1. Subscribe to the ScriptNotify event.
2. In the event handler e.Value gives the value sent from javascript
3. The javascript code can invoke this method using window.external.notify("value");

This can be used in Silverlight as is. More details here.

Communication from HTML5 web page to iPhone

Here again the way is different. We cannot handle the window.external object directly. Need to follow a workaround. Whenever we want to communicate to the iOS UIWebView, we need to navigate to a dummy url and catch the navigation event in objective C code. We can use the webViewDidFinishLoad handler of UIWebView to capture the event. Steps below.

1. Write code in JS to redirect the window.external invocations to a dummy url with data in its query string.
2. Handle the webViewDidFinishLoad of UIWebView control and write the code in it.

More details here.

Android - Making Facebook login works in HTML5 app hosted inside WebView

Context - Develop one app for all machines / platforms.

The software stake holders always want to target a large audience regardless of what machine / plat form their clients uses. That is one of the reason why software are often written as web application. If we write as web application, at least in theory it can run in any machine whether it is laptop,tablet or mobile also regardless of the OS Windows / Linux / Mac used in those machines. Another advantage going with web is the less development cost as we are developing one application. Its the trend among the developers / organizations.

There are so many arguments going on whether we should go with this HTML5 web apps to address this problem in case we are targeting mobile platforms only. Some people argue that we should go with Xamarin kind of converging platforms. But that won't help to run the app in laptops.

Another requirement from the stakeholder will be to have native web app even if we have working web application. That is to make sure the application is present in the app stores in the way standard mobile applications are present and users don't need to worry about the web application url. They just need to install the mobile app from the store.

So only one way to meet the above requirement of running in desktops, laptops and as native mobile app is to develop application using HTML5 and create native wrapper applications which uses WebBrowerControl / WebView to show the HTML5 web site inside the native mobile app.

But will that make our the development life easier? If the application is very simple, it is easy. But when the application gets more features, problems will starting popping up. Below is one of the problem we faced in such a scenario.

Problem

The application needs to have Facebook login. It is easy to setup that using the Facebook javascript sdk. It worked well when we try this from laptop and the mobile browser. But when the same is tried from native Android wrapper we got an issue. After Facebook login, it ends up in a blank white page. 

The analysis leads to Android WebView limitations. The Facebook login SDK normally creates browser window child window and after login it will close that window and control will be passed to parent window. But when we show the same page using Android WebView control. That javascript popup window closing is not working.

Solution

The solution is as follows 

1. Show the FB login popup in a different WebView control
2. After the login success, close that window. 

How do we achieve that? Below are the tasks to be done.

Task 1 - Show login popup in different WebView

There is already a WebView in the application. We need to catch the javascript popup show event happened inside the web page and show that popup in a different WebView control. To do that we need to follow steps below

1. Subclass WebChromeClient and override it's onCreateWindow method. Name it as UriWebChromeClient
2. From the onCreateWindow method create object of new WebView control and add to the view.
1. When we create the second WebView control, associate that to our custom WebViewClient which we are going to create in Task 2. That is required to close this second WebView.
3. The new UriWebChromeClient class object needs to be connected to the existing WebView by calling webViewObj.setWebChromeClient() Method

Task 2 - How to know FB authentication is success to hide WebView

Now we need to think how the Android web view knows whether the FB login is successful. It can only be determined by looking at a particular URL(https://m.facebook.com/v2.0/dialog/oauth) where the FB will be redirecting after login success. How do the Android web view knows that the web redirection happens? It can be done by sub classing WebViewClient class. Steps below

1. Subclass the WebViewClient class.Name it as UriWebViewClient
2. Override the onPageFinished() method of WebViewClient in the new inherited UriWebViewClient class and check for the url. If the URL is "https://m.facebook.com/v2.0/dialog/oauth", we can decide that the FB authentication is completed and its time to hide the second WebView.
3. Override the shouldOverrideUrlLoading() method too and do the similar.

The code is mentioned in this SO link .Only difference from this link is the onPageFinished method overriding.

Reference links

• What is WebViewClient and WebChromeClient subclassing? These are the extensibility points for Android WebView control. More details http://developer.android.com/reference/android/webkit/WebView.html

Why my HDInsight Hadoop job failed

This post assumes that the reader had read my previous post about Hadoop HDInsight Installation issues. My aim was to run a simple Hadoop Map Reduce task in Win 7 - 64 bit machine using HDInsight

When I run my first Hadoop application from Visual Studio, I got the below result in console windows with a status of 'Job failed'.

---

Output folder exists.. deleting.
File dependencies to include with job:
[Auto-detected] <drive>:\Joy\Code\DotNet\HDInsight\HadoopTest\bin\Debug\HadoopTest.vsh
ost.exe
[Auto-detected] <drive>:\Joy\Code\DotNet\HDInsight\HadoopTest\bin\Debug\HadoopTest.exe

[Auto-detected] <drive>:\Joy\Code\DotNet\HDInsight\HadoopTest\bin\Debug\Microsoft.Hado
op.MapReduce.dll
[Auto-detected] <drive>:\Joy\Code\DotNet\HDInsight\HadoopTest\bin\Debug\Microsoft.Hado
op.WebClient.dll
[Auto-detected] <drive>:\Joy\Code\DotNet\HDInsight\HadoopTest\bin\Debug\Newtonsoft.Jso
n.dll
packageJobJar: [] [/<drive>:/Hadoop/hadoop-1.1.0-SNAPSHOT/lib/hadoop-streaming.jar] <drive>:
\Users\<username>\AppData\Local\Temp\streamjob4051103269437643633.jar tmpDir=null
13/10/21 22:19:07 WARN security.ShellBasedUnixGroupsMapping: got exception tryin
g to get groups for user <logged in user's name without domain\>
org.apache.hadoop.util.Shell$ExitCodeException: GetLocalGroupsForUser error (222
1): The user name could not be found.

        at org.apache.hadoop.util.Shell.runCommand(Shell.java:454)
        at org.apache.hadoop.util.Shell.run(Shell.java:369)
        at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:
573)
        at org.apache.hadoop.util.Shell.execCommand(Shell.java:659)
        at org.apache.hadoop.util.Shell.execCommand(Shell.java:642)
        at org.apache.hadoop.security.ShellBasedUnixGroupsMapping.getUserGroups(
ShellBasedUnixGroupsMapping.java:65)
        at org.apache.hadoop.security.ShellBasedUnixGroupsMapping.getGroups(Shel
lBasedUnixGroupsMapping.java:41)
        at org.apache.hadoop.security.Groups.getGroups(Groups.java:79)
        at org.apache.hadoop.security.UserGroupInformation.getGroupNames(UserGro
upInformation.java:1041)
        at org.apache.hadoop.mapreduce.JobSubmissionFiles.getStagingDir(JobSubmi
ssionFiles.java:107)
        at org.apache.hadoop.mapred.JobClient$2.run(JobClient.java:872)
        at org.apache.hadoop.mapred.JobClient$2.run(JobClient.java:866)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInforma
tion.java:1136)
        at org.apache.hadoop.mapred.JobClient.submitJobInternal(JobClient.java:8
66)
        at org.apache.hadoop.mapred.JobClient.submitJob(JobClient.java:840)
        at org.apache.hadoop.streaming.StreamJob.submitAndMonitorJob(StreamJob.j
ava:917)
        at org.apache.hadoop.streaming.StreamJob.run(StreamJob.java:122)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:65)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:79)
        at org.apache.hadoop.streaming.HadoopStreaming.main(HadoopStreaming.java
:50)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.
java:39)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces
sorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at org.apache.hadoop.util.RunJar.main(RunJar.java:156)
13/10/21 22:19:07 WARN security.UserGroupInformation: No groups available for us
er joyg
java.io.IOException: No groups found for user <user name>
        at org.apache.hadoop.security.Groups.getGroups(Groups.java:81)
        at org.apache.hadoop.security.UserGroupInformation.getGroupNames(UserGro
upInformation.java:1041)
        at org.apache.hadoop.mapreduce.JobSubmissionFiles.getStagingDir(JobSubmi
ssionFiles.java:107)
        at org.apache.hadoop.mapred.JobClient$2.run(JobClient.java:872)
        at org.apache.hadoop.mapred.JobClient$2.run(JobClient.java:866)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInforma
tion.java:1136)
        at org.apache.hadoop.mapred.JobClient.submitJobInternal(JobClient.java:8
66)
        at org.apache.hadoop.mapred.JobClient.submitJob(JobClient.java:840)
        at org.apache.hadoop.streaming.StreamJob.submitAndMonitorJob(StreamJob.j
ava:917)
        at org.apache.hadoop.streaming.StreamJob.run(StreamJob.java:122)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:65)
        at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:79)
        at org.apache.hadoop.streaming.HadoopStreaming.main(HadoopStreaming.java
:50)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.
java:39)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces
sorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at org.apache.hadoop.util.RunJar.main(RunJar.java:156)
13/10/21 22:19:08 INFO util.NativeCodeLoader: Loaded the native-hadoop library
13/10/21 22:19:08 WARN snappy.LoadSnappy: Snappy native library not loaded
13/10/21 22:19:08 INFO mapred.FileInputFormat: Total input paths to process : 4
13/10/21 22:19:08 INFO streaming.StreamJob: getLocalDirs(): [c:\hadoop\HDFS\mapr
ed\local]
13/10/21 22:19:08 INFO streaming.StreamJob: Running job: job_201310212213_0001
13/10/21 22:19:08 INFO streaming.StreamJob: To kill this job, run:
13/10/21 22:19:08 INFO streaming.StreamJob: C:\Hadoop\hadoop-1.1.0-SNAPSHOT/bin/
hadoop job  -Dmapred.job.tracker=localhost:50300 -kill job_201310212213_0001
13/10/21 22:19:08 INFO streaming.StreamJob: Tracking URL: http://127.0.0.1:50030
/jobdetails.jsp?jobid=job_201310212213_0001
13/10/21 22:19:10 INFO streaming.StreamJob:  map 0%  reduce 0%
13/10/21 22:19:42 INFO streaming.StreamJob:  map 100%  reduce 100%
13/10/21 22:19:42 INFO streaming.StreamJob: To kill this job, run:
13/10/21 22:19:42 INFO streaming.StreamJob: C:\Hadoop\hadoop-1.1.0-SNAPSHOT/bin/
hadoop job  -Dmapred.job.tracker=localhost:50300 -kill job_201310212213_0001
13/10/21 22:19:42 INFO streaming.StreamJob: Tracking URL: http://127.0.0.1:50030
/jobdetails.jsp?jobid=job_201310212213_0001
13/10/21 22:19:42 ERROR streaming.StreamJob: Job not successful. Error: # of fai
led Map Tasks exceeded allowed limit. FailedCount: 1. LastFailedTask: task_20131
0212213_0001_m_000000
13/10/21 22:19:42 INFO streaming.StreamJob: killJob...
Streaming Command Failed!

---

Did you get any thing from this log? There were basically 2 problems. 

1. WARN - Not able to get groups 
2. ERROR Job not successful

Error :Job not successful

Lets first focus on the error as always.The starting point of this investigation should be the log and where we can find the logs? Here comes the job tracker. Notice the tracking URL mentioned in the output http://127.0.0.1:50030/jobdetails.jsp?jobid=job_201310212213_0001

Browse to this page

We can easily find a table like the below one in that page

Kind	% Complete	Num Tasks	Pending	Running	Complete	Killed	Failed/Killed Task Attempts
map	100.00%	4	0	0	0	4	2 / 0
reduce	100.00%	1	0	0	0	1	0 / 0

This table gives the details about map reduce tasks executed. In the last column (Failed/Killed Task Attempts) we can see a link. Its highlighted in blue above for reference. Navigate to that link to get more information.

Now we are entering more detailed log page. There will be a table which describes about attempts, tasks error etc..Our target is the last column titled as 'Logs'. Select the 'Last 4KB' link to navigate to there.

This takes us to a page with URL of the below kind
http://<FQDN>:50060/tasklog?attemptid=attempt_201310212213_0001_m_000000_0&start=-4097

Welcome to .Net!!! Here we can see logs which tells more about our .net application. Earlier we were seeing Java related call stack. an example log is pasted below

---

Unhandled Exception: Microsoft.Hadoop.MapReduce.StreamingException: The user type could not be loaded. DLL=HadoopTest.exe, Type=HadoopTest.SentenceMapper ---> 

System.BadImageFormatException: Could not load file or assembly 'file:///c:\hadoop\HDFS\mapred\local\taskTracker\joylocal\jobcache\job_201310290904_0001\attempt_201310290904_0001_m_000000_0\work\HadoopTest.exe' or one of its dependencies. An attempt was made to load a program with an incorrect format.
   

at System.Reflection.RuntimeAssembly._nLoad(AssemblyName fileName, String codeBase, Evidence assemblySecurity, RuntimeAssembly locationHint, StackCrawlMark& stackMark, IntPtr pPrivHostBinder, Boolean throwOnFileNotFound, Boolean forIntrospection, Boolean suppressSecurityChecks)
   

at System.Reflection.RuntimeAssembly.InternalLoadAssemblyName(AssemblyName assemblyRef, Evidence assemblySecurity, RuntimeAssembly reqAssembly, StackCrawlMark& stackMark, IntPtr pPrivHostBinder, Boolean throwOnFileNotFound, Boolean forIntrospection, Boolean suppressSecurityChecks)
   

at System.Reflection.RuntimeAssembly.InternalLoadAssemblyName(AssemblyName assemblyRef, Evidence assemblySecurity, RuntimeAssembly reqAssembly, StackCrawlMark& stackMark, Boolean throwOnFileNotFound, Boolean forIntrospection, Boolean suppressSecurityChecks)
   

at System.Reflection.RuntimeAssembly.InternalLoadFrom(String assemblyFile, Evidence securityEvidence, Byte[] hashValue, AssemblyHashAlgorithm hashAlgorithm, Boolean forIntrospection, Boolean suppressSecurityChecks, StackCrawlMark& stackMark)
   

at System.Reflection.Assembly.LoadFrom(String assemblyFile)   

at Microsoft.Hadoop.MapReduce.MapperMain.Main()
   --- End of inner exception stack trace ---
at Microsoft.Hadoop.MapReduce.MapperMain.Main()
at Microsoft.HdInsight.MapDriver.Program.Main(String[] args)

java.lang.RuntimeException: PipeMapRed.waitOutputThreads(): subprocess failed with code -532462766
 at org.apache.hadoop.streaming.PipeMapRed.waitOutputThreads(PipeMapRed.java:362)

---

This tells that, when the Apache Hadoop framework which is written in Java tried to invoke our MapReduce tasks which is written in C#.Net, there was an exception in our .Net code related to the executable's image format.

The easiest method to solve this is to google the error message. As usual, the first google result was point to stackoverflow.com and it solved my problem. Below are the causes of that problem

1. The program is not running as admin;
2. The executables are missing (paths in log files)
3. Whether .net framework is working for the corresponding version?
4. The building target is x86 instead of x64

Lesson 1

The .Net assembly we are creating for Hadoop MapReduce tasks should be targeting to 64bit environment.

WARN - got exception trying to get groups for user ****

This took so much time. As the type says WARN, it was not causing any issues to the execution of Hadoop MapR jobs.But still I felt this is something interesting to solve. The first aim was to understand the warning. Basically it says, 'it is not able to retrieve windows groups of user who is running the program'. As I have logged in to this system with my company login (company domain\username), it shows my user name without the domain name prefix.

Attempt 1 - Editing dfs.web.ui

I tried googling the error message. The links said that, I need to edit the hdfs-site.xml which is present in the location <install drive>:\Hadoop\hadoop-1.1.0-SNAPSHOT\conf\hdfs-site.xml

I modified the dfs.web.ui section in that file but no luck.Actually that section itself was not there.So I added it. This file is something to be studied later.

Attempt 2 - Understanding the Hadoop source code

Now we don't have any chance. We need to go back to the past days where internet was not there to google. Really I had programmed 2 of my academic projects without googling as internet was not common at that time. This will be really hard to digest for the people who starts coding with google.

If we are true programmers and confident in our algorithmic skills, understanding source code will not be a problem even if we are seeing that language for the first time. This case is more simple. Hadoop is open source written in Java and anybody can see the source code.

As we can see in the stack trace, there is a Shell class which is executing the command which is retrieving the groups of user. What is the command and where from its coming? Only way is to browse the source code of Util.Shell class available in the below link

http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/Shell.java?view=markup

Here is a small tricky part or the point where people who don't know the programming principles needs to spend more time. The method name is runCommand() which doesn't tell anything about users or groups. Yes the source code is written by great programmers who follows SOLID principles where each class and method will be doing single responsibility. The command which is used to get the user groups will be coming from somewhere outside of this method. More precisely, in one of the previous methods in the call stack.

When we manually inspect the call stack, we can see the logic of retrieving the groups of the user may be more likely present in the class ShellBasedUnixGroupsMapping and method getGroups().If we further digg down the code path, we can see that, it is calling Shell.getGroupsForUserCommandShell.getGroupsForUserCommand() method to get the command. Lets see the code of getGroupsForUserCommand method

---

public static String[] getGroupsForUserCommand(final String user) {
   //'groups username' command return is non-consistent across different unixes
   return (WINDOWS)? new String[] { WINUTILS, "groups", "-F", "\"" + user + "\""}
                   : new String [] {"bash", "-c", "id -Gn " + user};
}

---

WINUTIL.exe command

We can assume that WINDOWS token will be true as we are running it in windows machine. When its called, it will return the command as follows.

winutils groups -F <user name as it reached via param>

In our case the chance for user name format as short name is 100%, because the log says it. Can we execute the command ourself in the same way how system is doing? To do that we need to get the path to WINUTILS.exe file

What is the path to winutil command file? Simple search the Hadoop folder. We can find at <install drive>:\Hadoop\hadoop-1.1.0-SNAPSHOT\bin

Now we can try to executing the command
<drive>:\Hadoop\hadoop-1.1.0-SNAPSHOT\bin>winutils groups -F <short username> (This failed)

<drive>:\Hadoop\hadoop-1.1.0-SNAPSHOT\bin>winutils groups -F <DOMAIN>\username (Success)

We can reach to a conclusion that this command will not work and nobody should be able to execute MapReduce tasks. But people claims that they are able to execute Hadoop tasks using HDInsight. Are there any more tests remaining?

Lets try executing the command with hadoop username. .\Hadoop is a local user which is created when we install Hadoop.

<drive>:\Hadoop\hadoop-1.1.0-SNAPSHOT\bin>winutils groups -F hadoop (Success)

Solution - Execute the HDInsight Hadoop task in the context of local user

It gives the light. When we try to run Hadoop application in the context of local user it may work. Immediately I created a local user in the machine and run the application. It worked !!!

So where its trimming the user name? Or which method says the user groups needs to be retrieved based on short user name? Some more source code analysis revealed it. Its the UserGroupInformation.getGroupNames() method

---

public synchronized String[] getGroupNames() {
   ensureInitialized();
   try {
     List<String> result = groups.getGroups(getShortUserName());
     return result.toArray(new String[result.size()]);
   } catch (IOException ie) {
     LOG.warn("No groups available for user " + getShortUserName());
     return new String[0];
   }
}

---

Now things are clear. I am finding a way to confirm this with Hadoop developers.

"The literal Octal 08 (digit 8) of type int is out of range" Java compiler error

This error came when my wife was doing simple programming in Java using Eclipse IDE. She was trying to do 'number to words conversion' where hard coded int input = 8; can be converted to "Eight" but not int input = 08;

It clearly says the compiler is treating this as octal literal .But this is really new to me as I am more working with C# & VB.Net which don't have this feature. After a google we were able to find the list of supported literals in Java.

http://en.wikibooks.org/wiki/Java_Programming/Literals

Leading 0 - Ocatal eg: 010 -> 8
Leaxing 0x - Hex eg: 0xA -> 10
Leading 0B - Binary eg: 0B11 -> 3 (Supported in latest Java versions >=7)

Where is vjslib.dll located?

Recently I had to prepare for a session related to "Programming for non programmers". I thought of starting with what is computer and programming using some day to day examples. But there is no meaning in talking 1 hour about programming to such an audience who comes from non programming background. To obey the famous "A picture is worth a thousand words" saying, I thought of giving a demo of programming to the audience. But it was tougher than showing a demo in a session about Node.JS or DSL because in normal technical sessions we can show demo using the corresponding language and the related tool. Here which language I should choose? In what tool I can select? Will the surroundings of Visual Studio take the attention of these people from the code window? 

Karel the Robot learns programming

These questions lead me to the famous Karel programming language. The first time, I came to know about Karel the Robot is in the programming paper CS106a of Stanford university syllabus. Don't think that, I studied in Stanford but attended their free online programming course via iTunesU iPad application. But most of the Karel runtime environments are available in Java and Java is not setup in our machines. Googled for online simulators but failed. So decided to go with one .net implementation found in the below site.

http://www.acthompson.net/DotNet/Karel.htm

Location of vjslib.dll

I downloaded the project but it failed during build. The reasons was simple. Its based on Visual J# which was included in earlier days of .Net and slowly removed from Visual Studio suite. It needs a dll named vjslib.dll. I searched for all the folders in my machine related to .Net 2.0 but cannot find. 

Finally google gave me the location of the Microsoft Visual J# Version 2.0 Redistributable Package which contains the vjslib.dll file. Installed it and things started working.Below the location where we can find the vjslib.dll file

C:\Windows\Microsoft.NET\Framework\v2.0.50727\vjslib.dll

Location of Microsoft Visual J# Version 2.0 Redistributable Package 
http://www.microsoft.com/en-us/download/details.aspx?id=4712

I don't think this post will be useful if you are developing a new project. But useful, if you had to maintain or work on legacy code base or ended up in a sample which is created using VJ# assemblies.