Wednesday, 12 July 2017

Using ES6 Generators to implement Where Skip Take

This article will present the use of ES6 Generators to implement in Javascript some handy functions for collections (arrays) that can filter and limit the size of that result. The technique will use the new ES6 generator functions.

ES6 Generators are functions of Ecmascript 6 that let's the programmer take control of iterators using the yield keyword and use custom logic.

We will use Traceur to support ES6 syntax. Traceur is a Javascript compiler in the form of a library. In production you should instead use a transpiler, that rewrites your ES6 syntax into ES2015, which is more supported by current web browsers. One transpiler is Babel.

First off we include Traceur into the Html code:

<!DOCTYPE html>
<html lang="en">
    <head>
        <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
        <script src="https://google.github.io/traceur-compiler/bin/traceur.js"></script>
        <script src="https://google.github.io/traceur-compiler/bin/BrowserSystem.js"></script>
        <script src="https://google.github.io/traceur-compiler/src/bootstrap.js"></script>
        <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.2.1/jquery.min.js"></script>
        
    </head>
    <body bgcolor="teal" style="font-family:Verdana, Courier,Helvetica, Arial">
        
        <h1 style="color:yellow;font-family:Arial">ES6 module demo</h1>



We now have the compiler loaded into our HTML page and can make use of ES6 Generator functions and other ES6 syntax. First lets look at the WHERE function in Javascript. Think of it as WHERE in Linq, we will also use ES6 Arrow functions here, which is an important functional programming concept and syntax that ES6 introduces.


   <script type="text/javascript">

         var where = function* (predicate, collection){
                for (let item of collection){                  
                    if (predicate(item)){                   
                        yield item;      
                    }
                }
            }




We assign where here to the function defined at the right hand side. Note the asterisk (star) here. This is what a generator function looks like in ES6. It must not be mixed with function pointers as in C++, the star here is just to separate generator functions with ordinary functions. On a operational level, generator functions are just like iterators in C# that hey hand off control from the function to the program that calls the functions. Generator functions are usually used in iteration loops, which will be seen later on in this article.

The where function passes in a predicate (an arrow function that returns a boolean which is true or false of course) and if that predicate on the current item of the collection is true, yield will return that item. The sample also uses the of operator of the collection passed in.

Next off, let us look at the take function. This function will take a desired number of the result set and end iteration early if a result set is found. On a very large collection, this is very efficient. Imagine searching a huge collection for a desired item defined by an arrow function and exit early if for example one item is found. Then take(1, mycoll) will return control to the calling program when that item is found.


            var take = function* (size, collection){

            if (size < 1) 
                return;

            let count = 0;

            for (let item of collection){
                yield item;
                count = count + 1;
                if (count >= size){
                    return;
                }
            }
        }



Make note that we end the looping here by doing a return when the amount of items return is the desired size specified. Next off, let us look at the Skip function. We could do a reserve on the passed in collection, and use the take function. Instead the Skip method will use tailored logic to iterate on the passed in collection from start to finish, skipping the specified number of items first.

  var skip = function* (size, collection){
         
            let skippedItems = 0;

            for (let item of collection){
                skippedItems = skippedItems + 1;
                //console.log(item);
                //console.log(skippedItems);
                
                if (skippedItems <= size){
                    continue;
                }
                else {
                    yield item;
                }
            }

        }
The sample HTML code below tests out the JS code above using ES6 arrow functions and ES 6 generator functions on a simple array and looping through with our new powerful functional programming constructs that let us search even huge collections with filtering and skipping or taking the desired result.

         
            var coll = [ "Tammy", "Tom", "Betty", "Marge", "Joey", "Tim", "Jane", "Tommy" ];

            //console.log(coll);
            
            for (let i of take(2, where(n => n.startsWith("T"),coll))){
                   $("body").append("<li>" + i + "</li>");
            }

Next off, we will combing all the uses of WHERE, SKIP and TAKE to both filter and provide paging functionality. Consider this code:


            var coll = [ "Tammy", "Tom", "Betty", "Marge", "Joey", "Tim", "Jane", "Tommy" ];

            console.log(coll);

            let pageSize = 1;
            let pageIndex = 1;
            
            for (let i of take(pageSize, skip(pageIndex*pageSize, where(n => n.startsWith("T"),coll)))) {
                   $("body").append("<li>" + i + "</li>");
            }
                  
Here we use the following to get a PAGE of our result by using the formula:

PAGE_INDEX = TAKE(PAGE_SIZE) of (SKIP(PAGE_INDEX * PAGESIZE) of COLLECTION WHERE Predicate(x)). It is impressive that we can implement Linq like functionality by using ES6 Generator functions in ES6 sticking to ES6 syntax. This promises that Javascript in the future will be a very versatile language when it comes to functional programming. We use Traceur in the meantime to support ES6 syntax in different web browsers. I have included a Plunk in the link below so that you can test this out yourself.

ES6 Generator functional programming sample

Sunday, 9 July 2017

Ecmascript 6 Modules in Javascript

Javascript is a programming (or scripting) language known for its wide use. It is mature by the fact that it is used for many years since its creation in the mid 90s. But at the same time Javascript or just JS is immature by the fact that it lacks a lot of features built into the language. This is an observed by the large number of different libraries to add features to JS that programmers using other programming languages take for granted. For example, modules and classes are not something easy to create with JS.
The good thing is that JS is finally evolving in large steps now with Ecmascript 6 or ES6 with a common standard that many browsers vendors can agree upon. There are different ways to run JS code with ES6 features. One can use a transpiler such as Babel that will rewrite the JS code with ES6 scripts into to compatible ES2015 syntax which more browsers support. Or one can use a javascript library that contains polyfills and fills to support ES6 code in browsers. The sample code in this article has been tested with Internet Explorer 11 and is available as a Plunk. The following url contains the running demo:
Plunk with ES6 modules

First in this sample a class called Employee is created. This uses ES6 new class feature.

        export class Employee {

 
 
 constructor (name) {
  //console.log("Constructor of Employee");
  this._name = name; 
 }

 get name() {
  return this._name;
 }

 doWork(){
  return `${this.name} is working`;
 }
}

ES6 classes can export a class and then be imported in other classes and build up modules. The employee class also uses ES6 template string feature. Note that the backtick quote is used and ${..} is used to refer to variables. The next class then imports the Employee class. Note that you must change url here to match your Plunk in the running editor to adjust the session url that Plunkr uses to give unique urls. ES6 can support static urls of course.

import {Employee} from "http://run.plnkr.co/gIljKiTbjZ0IOX8b/Employee.js"
export class Company {


 hire(...names) {
  this.employees = names.map(n => new Employee(n));
 }

 doWork(){
  console.log(this.employees);
  $("body").append("<ul>");
  for (let e of this.employees){
   console.log(e);
   $("body").append(`<li style='list-style-type:circle'>${e.doWork()}` + "</li>");
  }
  $("body").append(">/ul<");
  return 1;
 }
}

Again the class keyword is used to define a class and then exported to be used in another class or module with the export keyword. To import the Employee class the import keyword. The Company class above uses the rest operator (...) to allow passing in an arbitrary number of elements. Then the map operator built into Arrays in ES6 is used to return a mapped array. The use of the let operator is used and also the of operator of iterable collections.
Finally the sample HTML code below is used to define the use of an ES6 module. Here the script tag with the tag module is used.


<!DOCTYPE html>
<html lang="en">
    <head>
        <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
        <script src="https://google.github.io/traceur-compiler/bin/traceur.js"></script>
        <script src="https://google.github.io/traceur-compiler/bin/BrowserSystem.js"></script>
        <script src="https://google.github.io/traceur-compiler/src/bootstrap.js"></script>
        <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.2.1/jquery.min.js"></script>
        
    </head>
    <body bgcolor="teal" style="font-family:Verdana, Courier,Helvetica, Arial">
        
        <h1 style="color:yellow;font-family:Arial">ES6 module demo</h1>
        <script type="module">
        import {Company} from "http://localhost/babeldemo/src/Company.js"
        var c = new Company();
        c.hire("Tim", "Tom", "Betty", "Maggie");
        c.doWork();
        </script>
        
    </body>
</html>

Note that Traceur is a good option to test out ES6 features. In production, using a transpiler such as Babel to create a ES2015 compatible JS code is probably much better since there is a performance cost with Traceur. The reason Traceur is used in this sample is to show how an old browser such as Internet Explorer 11 can run ES6 code with a Javascript library such as Traceur.

To see a compatibility matrix of which browsers supports which ES6, see the Kangax table at the following url:
https://kangax.github.io/compat-table/es6/
With the help of Traceur more browsers can run more feature of ES6 and you can test out and use ES6 in your development or production environments and structure your code with classes and modules for example. Make note that the current state of Javascript module loaders of browsers will be improved in the future. Until then, additional frameworks such as AMD or Require.Js can be used to support module loading in more complex scenarios.

Sunday, 14 May 2017

Redis Image Cache Provider

Caching images in web sites and web applications is possible to achieve using many methods. This article presents an image cache provider to use in MVC applications using Redis as a caching server.

First off, create a new MVC web application and add the Nuget package RedisImageCacher (created by me):
RedisImageCacher

Install-Package RedisImageCacher

This Nuget package will also add in the dependencies for libraries ServiceStack.Redis and Microsoft.Web.RedisOutputCacheProvider.


Next off, one will need a Redis server. It is possible to install a local Redis server in Windows using the binaries here:

Redis 2.6 Windows (32/64 bits) on GitHub from MSOpenTech.


Tips on how to use Redis client and server and code in this article is written in another article of mine:
ASP.NET and Redis

Download Redis 2.6 and start it up on your local developer box. Start the server from a command-line for example :

C:\redis26\redis-server

Start a Redis client also from the commandline e.g. :
C:\redis26\redis-cli



  <caching>
      <outputCache defaultProvider="MyRedisOutputCache">
        <providers>
          <!-- Either use 'connectionString' and provide all parameters as string OR use 'host','port','accessKey','ssl','connectionTimeoutInMilliseconds' and 'operationTimeoutInMilliseconds'. -->
          <!-- 'databaseId' and 'applicationName' can be used with both options. -->
          <!--
          <add name="MyRedisOutputCache" 
            host = "127.0.0.1" [String]
            port = "" [number]
            accessKey = "" [String]
            ssl = "false" [true|false]
            databaseId = "0" [number]
            applicationName = "" [String]
            connectionTimeoutInMilliseconds = "5000" [number]
            operationTimeoutInMilliseconds = "1000" [number]
            connectionString = "<Valid StackExchange.Redis connection string>" [String]
            loggingClassName = "<Assembly qualified class name that contains logging method specified below>" [String]
            loggingMethodName = "<Logging method should be defined in loggingClass. It should be public, static, does not take any parameters and should have a return type of System.IO.TextWriter.>" [String]
          />
          -->
          <!-- For more details check https://github.com/Azure/aspnet-redis-providers/wiki -->
          <!-- Either use 'connectionString' OR 'settingsClassName' and 'settingsMethodName' OR use 'host','port','accessKey','ssl','connectionTimeoutInMilliseconds' and 'operationTimeoutInMilliseconds'. -->
          <!-- 'databaseId' and 'applicationName' can be used with both options. -->
          <!--
          <add name="MyRedisOutputCache" 
            host = "127.0.0.1" [String]
            port = "" [number]
            accessKey = "" [String]
            ssl = "false" [true|false]
            databaseId = "0" [number]
            applicationName = "" [String]
            connectionTimeoutInMilliseconds = "5000" [number]
            operationTimeoutInMilliseconds = "1000" [number]
            connectionString = "<Valid StackExchange.Redis connection string>" [String]
            settingsClassName = "<Assembly qualified class name that contains settings method specified below. Which basically return 'connectionString' value>" [String]
            settingsMethodName = "<Settings method should be defined in settingsClass. It should be public, static, does not take any parameters and should have a return type of 'String', which is basically 'connectionString' value.>" [String]
            loggingClassName = "<Assembly qualified class name that contains logging method specified below>" [String]
            loggingMethodName = "<Logging method should be defined in loggingClass. It should be public, static, does not take any parameters and should have a return type of System.IO.TextWriter.>" [String]
            redisSerializerType = "<Assembly qualified class name that implements Microsoft.Web.Redis.ISerializer>" [String]
          /> -->
          <add name="MyRedisOutputCache" type="Microsoft.Web.Redis.RedisOutputCacheProvider" host="localhost" accessKey="" applicationName="RedisCache" port="6379" ssl="false" />
        </providers>
      </outputCache>
    </caching>
    



It is also possible to control how long time Image Data is to be kept in the cache in seconds. In web.config:
  <add key="RedisImageCacheTimeout" value="10" />



This sets a timeout of caching image data to 10 seconds for example. It is possible to change this of course, using 600 will cache image data for 10 minutes for example.


To cache an image, the following razor code show an example of its use:
 <img src="@Url.Action("ShowImage", "Images", new { id = Url.Encode("Croatia6-jpg") })" width="500" />
Using the syntax "filename"-"extension" like "myfile1-jpg" and using the embedded controller ImageController and action method ShowImage will also cache the image to Redis. A parameter of id is provided to the file to cache and load or retrieve from cache directly. This will be taken care of by the embedded images controller in the library. It is possible to use different image file formats, such as JPG, GIF, PNG and BMP.


Inside redis-cli Redis Client we can see that loading the image in the example will add a cache item in Redis:
redis 127.0.0.1:6379> keys * 1) "RedisCache/IMAGEBANK/Content/Images/Croatia6.jpg"

Note that images must be put into the Content/Images folder of your MVC solution! A configuration of changing this can be added later to the library. We can also ask Redis how long the cache item will exist until it expires and is removed by using TTL (Time-To-Live):

redis 127.0.0.1:6379> ttl "RedisCache/IMAGEBANK/Content/Images/Croatia6.jpg" (integer) 2


Note that TTL will show -1 if the cache item is expired. Redis client got many other handful commands to control the Redis memory cache, see Redis documentation here: Redis documentation The way the RedisImageCacher is saving data is actually done very generic and can be extended to support other data and cache this. For implementation details, see: ASP.NET and Redis


To download the source code for the RedisImageCacher and Demo library a download Mercurial repository is available on Bitbucket here:

RedisImageCacher Bitbucket repo


Or if you have Mercurial (hg) installed, from a command-line issue:
hg clone https://bitbucket.org/toreaurstad/redisimagecacher

Friday, 14 April 2017

AsyncStateMachine in .NET

This article will present a custom AsyncStateMachine that shows how it is possible to create as async awaitable state machine manually. Let's first consider an easy async-await example and implement it using a custom async state machine instead of the default one in .NET. The purpose is to glance more into the inner workings of async. The following code is what is going to be implemented.

   class Program
    {
        static void Main(string[] args)
        {
            try
            {
                CallFooAsync();
            }
            catch (AggregateException ae)
            {
                Console.WriteLine(string.Join(",",
                    ae.InnerExceptions.ToList().Select(e => e.Message).ToArray()));
            }

            Console.WriteLine("Hit any key to continue..");
            Console.ReadKey();
        }

        public static async void CallFooAsync()
        {
            int foo = await FooAsync();
            Console.WriteLine(foo);
        }

        static async Task<int> FooAsync()
        {
            await Task.Delay(5000);
            return 42;
        }


The Main method of the console application is not allowed to have async modifier, so it was necessary to introduce an extra method here. Next off, here is an example of an AsyncStateMachine at its simplest form. It is implemented as a struct.

        struct FooAsyncStateMachine : IAsyncStateMachine
        {

            private int state; 

            public AsyncTaskMethodBuilder<int> methodBuilder;
            private TaskAwaiter awaiter;

            public void MoveNext()
            {
                //State machine
                try
                {
                    if (state == 0)
                    {
                        awaiter = Task.Delay(millisecondsDelay: 5000).GetAwaiter();
                        if (awaiter.IsCompleted)
                        {
                            state = 1;
                            goto state1;
                        }
                        else
                        {
                            state = 1;
                            methodBuilder.AwaitUnsafeOnCompleted(ref awaiter, ref this);
                        }
                        return;
                    }

                    state1:
                    if (state == 1)
                    {                     
                        methodBuilder.SetResult(42);
                        return;
                    }
                }
                catch (Exception ex)
                {
                    methodBuilder.SetException(ex);
                    return;
                }
            }

            public void SetStateMachine(IAsyncStateMachine stateMachine)
            {
                methodBuilder.SetStateMachine(stateMachine);
            }

        }

The IAsyncStateMachine interface has a method called void MoveNext(). This method is called when the async method starts up. This method can be called multiple times. It contains a state variable to control which step is next. The AsyncMethodBuilder is used to control the result returned from the async method. The method SetResult sets the result returned and the method SetException inside a catch sets an exception, if encountered. Thet method SetStateMachine sets the statemachine struct as the state machine struct itself and is also part of the IAsyncStateMachine. Now that the AsyncStateMachine is defined, it is time to kick it off.

 static Task FooAsync2()
        {
            var stateMachine = new FooAsyncStateMachine();
            stateMachine.methodBuilder = AsyncTaskMethodBuilder<int>.Create();
            stateMachine.methodBuilder.Start(ref stateMachine);
            return stateMachine.methodBuilder.Task;
        }

The AsyncMethodBuilder inside the IAsyncStateMachine is using the Create method and Start method and returns a Task. In this article the async state machine is hard coded and not made general. Of course in .NET the AsyncStateMachine used is far more complex. But creating an async state machine yourself is a nice introduction to taking a deeper look into what async really is. Async-await is rewriting your code and could have been implemented in earlier versions of .NET. Async-await not changes the functionality of your code, but the way it is executed. It provides a way to halt a method and free up the thread, for example the UI thread and run code in the background, then continue later on again. It also allow parallel code.

Resources:

Saturday, 3 December 2016

Parallell execution with threads in C# - Old Stars finder

This article will present parallell execution of threads in C# to find old stars in a star formation known as W5 in the constellation of Cassiopeia with the Spitzer Space telescope. The code is from the book "C# Multithreaded and Parallell programming" by Packt Publishing by author Rodney Ringler et. al.

using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;
using System.Threading;
using System.Drawing.Imaging;


namespace OldStarsFinder
{
    public partial class Form1 : Form
    {

        //The number of processors or cores available in the computer for this application
        private int priProcessorCount = Environment.ProcessorCount;
        //The bitmaps list
        private List<Bitmap> prloBitmapList;
        //The long list with the old stars count 
        private List<long> prliOldStarsCount;
        //The threads list
        private List<Thread> prloThreadList;
        //The original huge infrared bitmap portrait
        Bitmap proOriginalBitmap; 

        public Form1()
        {
            InitializeComponent();
        }

        public bool IsOldStar(Color poPixelColor)
        {
            //Hue between 150 and 258
            //Saturation more than 0.10 
            //Brightness more than 0.90
            return ((poPixelColor.GetHue() >= 150 && (poPixelColor.GetHue() <= 258)) &&
                (poPixelColor.GetSaturation() >= 0.10) &&
                (poPixelColor.GetBrightness() >= 0.90)); 
        }

        private Bitmap CropBitmap(Bitmap proBitmap, Rectangle proRectangle)
        {
            //Create a new bitmap copying the portion of the original defined by proRectangle and keeping the PixelFormat 
            var loCroppedBitmap = proBitmap.Clone(proRectangle, proBitmap.PixelFormat);
            //Return the cropped bitmap 
            return loCroppedBitmap;
        }

        private void ThreadOldStarsFinder(object poThreadParameter)
        {
            //Retrieve the thread number reeived in object poThreadParameter 
            int liThreadNumber = (int) poThreadParameter;
            //The pixel matrix (bitmap) row number (Y)
            int liRow;
            //The pixel matrix (bitmap col number (X)
            int liCol;
            //The pixel color 
            Color loPixelColor;
            //Get my bitmap part from the bitmap list 
            Bitmap loBitmap = prloBitmapList[liThreadNumber];

            //Reset my old stars counter 
            prliOldStarsCount[liThreadNumber] = 0;
            //Iterate through each pixel matrix (bitmap) row 
            for (liRow = 0; liRow < loBitmap.Height; liRow++)
            {
                //Iterate through each pixel matrix (bitmap) cols 
                for(liCol = 0; liCol < loBitmap.Width; liCol++)
                {
                    //Get the pixel color for liCol and liRow 
                    loPixelColor = loBitmap.GetPixel(liCol, liRow);
                    //Get the pixel color for liCol and liRow 
                    if (IsOldStar(loPixelColor))
                    {
                        //The color range correspons to an old star
                        //Change its color to a pure blue 
                        loBitmap.SetPixel(liCol, liRow, Color.Blue);
                        //Increase the old stars counter 
                        prliOldStarsCount[liThreadNumber]++;
                    }
                    else
                    {
                        loBitmap.SetPixel(liCol, liRow, Color.FromArgb(128, loPixelColor));
                    }
                }
            }
            //Simulate heavy processing
            Random rnd = new Random();
            Thread.Sleep(rnd.Next(2000, 2500)); 
        }

        private void WaitForThreadsToDie()
        {
            //A bool flag 
            bool lbContinue = true;
            int liDeadThreads = 0;
            int liThreadNumber;
            while (lbContinue)
            {
                for(liThreadNumber = 0; liThreadNumber < priProcessorCount; liThreadNumber++)
                {
                    if (prloThreadList[liThreadNumber].IsAlive)
                    {
                        //One of the threads is still alive
                        //exit the for loop and sleep 100 milliseconds 
                        break;
                    }
                    else
                    {
                        //Increase the dead threads count 
                        liDeadThreads++;

                        progressBar1.Value = (int) ((liDeadThreads * 1.0 / priProcessorCount * 1.0) * 100.0);
                    }
                }

                if (liDeadThreads == priProcessorCount)
                {
                    //All the threads are dead, exit the while loop 
                    break; 
                }
                Thread.Sleep(100);
                liDeadThreads = 0; 
            }
        }

        private void ShowBitmapWithOldStars()
        {
            int liThreadNumber;
            //Each bitmap portion 
            Bitmap loBitmap;
            //The starting row in each iteration 
            int liStartRow = 0;

            //Calculate each bitmap's height 
            int liEachBitmapHeight = ((int) (proOriginalBitmap.Height / priProcessorCount)) + 1;

            //Create a new bitmap with the whole width and height 
            loBitmap = new Bitmap(proOriginalBitmap.Width, proOriginalBitmap.Height);
            Graphics g = Graphics.FromImage((Image) loBitmap);
            g.InterpolationMode = System.Drawing.Drawing2D.InterpolationMode.HighQualityBicubic;

            for (liThreadNumber = 0; liThreadNumber < priProcessorCount; liThreadNumber++)
            {
                //Draw each portion in its corresponding absolute starting row 
                g.DrawImage(prloBitmapList[liThreadNumber], 0, liStartRow);
                //Increase the starting row 
                liStartRow += liEachBitmapHeight;        
            }

            //Show the bitmap in the PictureBox picStarsBitmap 
            picStarsBitmap.Image = loBitmap;

            g.Dispose(); 
        }

        private void butFindOldStars_Click(object sender, EventArgs e)
        {
            progressBar1.Visible = true;

            proOriginalBitmap = new Bitmap(pictureBox1.Image);

            //Thread number 
            int liThreadNumber;
            //Create the thread list, the long list and the bitmap list 
            prloThreadList = new List<Thread>(priProcessorCount);
            prliOldStarsCount = new List<long>(priProcessorCount);
            prloBitmapList = new List<Bitmap>(priProcessorCount);

            int liStartRow = 0;

            int liEachBitmapHeight = ((int) (proOriginalBitmap.Height / priProcessorCount)) + 1;

            int liHeightToAdd = proOriginalBitmap.Height;
            Bitmap loBitmap; 

            //Initialize the threads 

            for (liThreadNumber = 0; liThreadNumber < priProcessorCount; liThreadNumber++)
            {
                //Just to occupy the number 
                prliOldStarsCount.Add(0); 

                if (liEachBitmapHeight > liHeightToAdd)
                {
                    //The last bitmap height perhaps is less than the other bitmap height
                    liEachBitmapHeight = liHeightToAdd; 
                }

                loBitmap = CropBitmap(proOriginalBitmap, new Rectangle(0, liStartRow, proOriginalBitmap.Width, liEachBitmapHeight));
                liHeightToAdd -= liEachBitmapHeight;
                liStartRow += liEachBitmapHeight;
                prloBitmapList.Add(loBitmap);

  

                //Add the new thread, with a parameterized start (to allow parameters)
                prloThreadList.Add(new Thread(new ParameterizedThreadStart(ThreadOldStarsFinder))); 
            }

            //Now, start the threads
            for (liThreadNumber = 0; liThreadNumber < priProcessorCount; liThreadNumber++)
            {
                prloThreadList[liThreadNumber].Start(liThreadNumber); 
            }

            WaitForThreadsToDie();

            ShowBitmapWithOldStars();

            progressBar1.Visible = false;
        }

    }
}

The code is available as a Windows Forms application in a Visual 2015 Solution available for download (zip) below:
Old Stars Finder (.zip) W5 image (NASA website): W5 image

Sunday, 27 November 2016

Making a simple accordion in Bootstrap 3


Creating an accordion for a web site is a breeze with Bootstrap 3. Just including the Bootstrap CSS and Javascript files and jQuery, we can start building an accordion. The accordion is a menu that shows one menu item a time. These menu items are panels and an accordion resembles a tab control with tabs, but is vertically stacked default. The following HTML page renders a simple accordion with Boostrap, CSS, Javascript and HTML.

<!DOCTYPE html>
<html>
<head>
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/css/bootstrap.min.css">
  <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.1.1/jquery.min.js"></script>
  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js"></script>
</head>
<body>

<div class="container">

  <div id="accordion" class="panel-group">
  
   <div class="panel panel-success">
    <div class="panel-heading">
     <h4 class="panel-title"><a href="#collapse1" data-toggle="collapse" data-parent="#accordion">Collapsible panel 1</a></h4>
    </div>
    <div id="collapse1" class="panel-collapse collapse">
     <div class="panel-body"><div class="well">This is a nice collapsible panel.</div><p>This is a test.</p></div>
     <div class="panel-footer">Panel footer</div>
    </div>   
   </div>
   
    <div class="panel panel-warning">
    <div class="panel-heading">
     <h4 class="panel-title"><a href="#collapse2" data-toggle="collapse" data-parent="#accordion">Collapsible panel 2</a></h4>
    </div>
    <div id="collapse2" class="panel-collapse collapse">
     <div class="panel-body">This is another nice panel</div>
     <div class="panel-footer">Hey a panel footer too!</div>
    </div>   
   </div>
   
    <div class="panel panel-default">
    <div class="panel-heading">
     <h4 class="panel-title"><a href="#collapse3" data-toggle="collapse" data-parent="#accordion">Collapsible panel 3</a></h4>
    </div>
    <div id="collapse3" class="panel-collapse collapse">
     <div class="panel-body">Omg a third panel!</div>
     <div class="panel-footer">Let's have another Panel footer too!</div>
    </div>   
   </div>
  
  </div>  
  
</div>

</body>
</html>

Make note that we here use the data-parent HTML5 extension to point to the parent element to get the accordion effect of only showing one panel at a time. Each panel consists of a panel with a panel heading, having a panel title and then a panel body and finally a panel footer. We use the CSS framework of Bootstrap to achieve this.

Sunday, 6 November 2016

Twitter 3 Bootstrap Autocomplete control for MVC 5 - Typeahead.js



This article will present a reusable user control for MVC applications using Twitter Bootstrap autocomplete feature. This feature is known as the Twitter Bootstrap typeahead. There is a lot of articles covering this topic on the Internet, my version will present a simple reusable control using MVC html helper that generates input fields of type text and hidden, i.e. a textbox and a hidden field to save the value. The MVC model binder will therefore be able to save the selected value in the list showing up in the autocomplete-enabled textbox into the target property specified. Of course, your needs for autocomplete feature will vary. The control described in this article will suit many developer's needs as they call up a controller action to get the desired data and then filter the drop down and also use the Bloodhound engine of Twitter to mark up the matches quite nicely. The end result is a very useful and nicely styled autocomplete textbox! It also supports keyboard navigation with arrow keys and Enter! Read on! Let's first look at the MVC Html helper itself first:

using System;
using System.Linq.Expressions;
using System.Web.Mvc;

namespace TwitterBootstrapAutoCompleteControl.HtmlHelpers
{

    public static class CustomMvcHelpers
    {

        public static MvcHtmlString AutoCompleteFor<TModel, TResult>(this HtmlHelper<TModel> htmlHelper,
            Expression<Func<TModel, TResult>> propertyToSet, string fetchUrl)
        {
            var metaData = ModelMetadata.FromLambdaExpression(propertyToSet, htmlHelper.ViewData);
            string propertyName = metaData.PropertyName;
            string jsComponent = string.Format(
            @"
              <input type ='hidden' id='{0}' />   
              <input type='text' id='{1}' class='typeahead form-control' placeholder='Search some values' />             
              <script type='text/javascript'> 
              <!-- AutoCompleteFor -->
              $(function() {{

                var suggestionEngine = new Bloodhound({{
                limit: 300,
                datumTokenizer: function(datum) {{
                    Bloodhound.tokenizers.obj.whitespace('value')
                }},
                queryTokenizer: Bloodhound.tokenizers.whitespace,
                remote:
                {{
                    url: '{2}',
                    filter: function(response) {{
                    var matches = [];
                    $.map(response, function(item) {{
                            var query = $('#{1}').val().toLowerCase();
                            var itemKey = item.Text.toLowerCase();                          
                            if (itemKey.indexOf(query) >= 0)
                            {{
                                matches.push(item);
                                //console.log(item);
                            }}
                        }});

                        return matches;
                    }}
                }}
            }});

            suggestionEngine.initialize();

            $('#{1}').typeahead({{
                hint: true,
                highlight: true,
                minLength: 1,           
        }}   , {{
                limit: 30,
                displayKey: 'Text',
                source: suggestionEngine.ttAdapter(),
                filter: function(data) {{
                    console.log(data);
                    return data;
                }},            
                templates:
                {{
                    suggestion: function(data) {{
                        return '<p>' + data.Text + '</p>';
                    }},
                empty: [
                '<div>',
                'No results matching',
                '</div>'
                ].join('\n'),
            }}
            }});

            $('#{1}').bind('typeahead:select', function(ev, suggestion) {{
             //console.log('Selection: ' + suggestion.Text);
         
            $('#{0}').val(suggestion.Id); 

           }});

          }});

        </script>

     ", propertyName, propertyName + "TextBox", fetchUrl);


            return MvcHtmlString.Create(jsComponent);
        }

    }

}

The MVC html helper will generate the HTML and the javascript that is required to generate a textbox and a hidden field with the autocomplete feature. Your MVC solution needs to include both jQuery and Twitter Bootstrap, plus the Twitter typeahead.js Nuget packages. In addition, you need to include the Bloodhound javascript file. Let's look at a controller action return Json data to our Html helper, which will use javascript to call that method:

        public ActionResult SomeData()
        {
            var countries = new List
            {
            new IdTextItem {Id = "US", Text = "United States"},
            new IdTextItem {Id = "CA", Text = "Canada"},
            new IdTextItem {Id = "AF", Text = "Afghanistan"},
            new IdTextItem {Id = "AL", Text = "Albania"},
            new IdTextItem {Id = "DZ", Text = "Algeria"},
            new IdTextItem {Id = "DS", Text = "American Samoa"},
            new IdTextItem {Id = "AD", Text = "Andorra"},
            new IdTextItem {Id = "AO", Text = "Angola"},
            new IdTextItem {Id = "AI", Text = "Anguilla"},
            new IdTextItem {Id = "AQ", Text = "Antarctica"},
            new IdTextItem {Id = "AG", Text = "Antigua and/or Barbuda"}
            };

            return Json(countries, JsonRequestBehavior.AllowGet);

        }

The Json method returns a JsonResult that is called once by this html helper. We filter the data on the client as can be seen in the Html Helper code. Let's look at the script bundle added in BundleConfig

   bundles.Add(new ScriptBundle("~/bundles/typeahead").Include(
                "~/Scripts/bloodhound.js",
                "~/Scripts/typeahead.bundle.js"
                ));

This Html helper will fetch data on load, while it is the text that the user is typing that filters the autocomplete list. If you need to pass in a text and use that in your fetchUrl, the html helper's filtering most likely can be adjusted, also look into the prefetch property of the typeahead. We also move up jQuery bundle to the top as the default MVC template has this bundle in the footer and not in the header, as will be required by the typeahead feature. Place in the section of _Layout.cshtml: @Scripts.Render("~/bundles/jquery") And in the bottom of the element of that same file: @Scripts.Render("~/bundles/typeahead") The html helper expects that your class contains the properties Id and Text. You can of course use an anonymous type to avoid creating a new type in your C#-code. The type could be a string or an integer for the Id.

 public class IdTextItem
    {

        public string Id { get; set; }

        public string Text { get; set; }

    }

We also do some adjustments to the style, since the default setting of the typeahead is borishingly looking. Add the following css styling to your view - or better - in a standalone css to be included in _Layout.cshtml. Save the following content into typeahead.css file which you place in the Content folder:
 

.tt-query {
  -webkit-box-shadow: inset 0 1px 1px rgba(0, 0, 0, 0.075);
     -moz-box-shadow: inset 0 1px 1px rgba(0, 0, 0, 0.075);
          box-shadow: inset 0 1px 1px rgba(0, 0, 0, 0.075);
}

.tt-hint {
  color: #999
}

.tt-menu {    /* used to be tt-dropdown-menu in older versions */
  width: 422px;
  margin-top: 4px;
  cursor: pointer;
  padding: 4px 0;
  background-color: #fff;
  border: 1px solid #ccc;
  border: 1px solid rgba(0, 0, 0, 0.2);
  -webkit-border-radius: 4px;
     -moz-border-radius: 4px;
          border-radius: 4px;
  -webkit-box-shadow: 0 5px 10px rgba(0,0,0,.2);
     -moz-box-shadow: 0 5px 10px rgba(0,0,0,.2);
          box-shadow: 0 5px 10px rgba(0,0,0,.2);
}

.tt-suggestion {
  padding: 3px 20px;
  line-height: 24px;
}

.tt-suggestion.tt-cursor,.tt-suggestion:hover {
  color: #fff;
  background-color: #0097cf;

}

.tt-suggestion p {
  margin: 0;
}

Then adjust the StyleBundle in BundleConfig to include this css file:
      bundles.Add(new StyleBundle("~/Content/css").Include(
                      "~/Content/bootstrap.css",
                      "~/Content/site.css",
                      "~/Content/typeahead.css"));
Finally, here is an example of how to use this Html Helper:

<div class="row">
    <div class="col-md-4">    
        <h2>Autocomplete control html helper:</h2>   
        @Html.AutoCompleteFor(m => m.SomeProperty, Url.Action("SomeData", "Home"))
    </div>
</div>

The call to the html helper provides as the first argument the property of the Model of the MVC View and the second argument is an url to the action to fetch the data. This HTML helper should match a lot of developer's needs, but can of course be adjusted. The benefit of using a MVC Html helper is that you get reusability. You avoid having to fiddle with Javascript for each field you want to add to your MVC view where you want some autocomplete feature. Maybe you want to adjust the HTML helper to fit your needs. I have provided a link to a zip file of this Html Helper in a Visual Studio 2015 below, let me know if there are some tips or improvement you have in case you evaluate and test out this Html helper and find improvements. Note that the chosen value in the autocomplete list is set to a hidden field. The textbox will be named "propertyname"TextBox and the hidden field will be named "propertyname"

Download the source code for the autocomplete control (VS 2015 solution):

Download zip file [.zip | 31,7 MB] Reading material:

Monday, 3 October 2016

Disposing objects instantiated by MEF

Experienced developers that has worked with the official extensibility framework in .NET, the Managed Extensibility Framework (MEF) allows the composition of different parts into more composite parts through composition. MEF has got similarities to other IoC framework, where you register components and then make use of them in other components. However, with MEF there is a caveat and an important one too!
MEF beautifully abstracts away the IoC container and lets you specify parts that can be epxorted and imported. But if you inspect your application or system with a memory profiler such as JetBrains DotMemory or Red Gate Memory Profiler, you soon find out that much of the memory used by your applications is not properly disposed, i.e freed up after use. This is the case for nonshared (non-singleton) objects that are exported and then instantiated (imported). This means that your application will through continued use hold more and more memory. It will leak memory. By inspecting the memory dependency chain, one can see that MEF is the reason why the nonshared objects instantiated by MEF is not released, even after the objects are issued to be disposed.

I use in this code the ServiceLocator found with the Enterprise Library. Make note that my code will break up the dependency chain that hinders the object, but it does not mean that necessarily objects will be disposed right away. After all, .NET is managed and decides itself when objects are really to be disposed. But if you strive with releasing objects that are tied to memory even after use and also use MEF, read on.

I use the Factory pattern here to instantiate objects. I also use the new feature in .NET 4.5 that is called the ExportLifeTimeContext. I also use the ExportFactory in MEF inside a class called ExportFactoryInstantiator that does actual instantiation of the objects and keeping a track of these ExportLifeTimeContext objects. As noted, you need at least .NET 4.5 to make this work. For .NET 4.0 users, sorry - you are out of luck as far as I know. Upgrade your application to .NET 4.5 if possible and get the newer version of MEF.

The code below shows how you can accomplish control over memory resources using MEF:

MefFactory.cs

using System;
using System.Collections.Concurrent;
using System.Collections.Generic;
using System.ComponentModel.Composition;
using System.Linq;
using Microsoft.Practices.ServiceLocation;

namespace SomeAcme.Client.Infrastructure.IoC
{

    /// <summary>
    /// Factory for MEF Parts that is able to actually dispose MEF instantiated objects and get around bug in MEF 
    /// where objects never gets properly GC-ed when they should dispose
    /// </summary>
    /// <typeparam name="T"></typeparam>
    [Export]
    [PartCreationPolicy(CreationPolicy.Shared)]
    public class MefFactory<T> : IPartImportsSatisfiedNotification 
    {

        /// <summary>
        /// Backlog that keeps track of mef parts that are instantiated via this factory 
        /// </summary>
        private static readonly ConcurrentBag<ExportLifetimeContext<T>> MefParts = new ConcurrentBag<ExportLifetimeContext<T>>();

        /// <summary>
        /// Disposes parts added to the mef factory backlog of type T
        /// </summary>
        public static void DisposeMefParts()
        {
            ExportLifetimeContext<T> item;
            while (MefParts.TryTake(out item))
            {
                if (item != null)
                    item.Dispose();
            }
        }

        /// <summary>
        /// Disposes parts added to the mef factory backlog of type T by a given predicate condition
        /// </summary>
        public static void DisposeMefParts(Predicate<T> condition)
        {
            ExportLifetimeContext<T> item;
            List<ExportLifetimeContext<T>> lifeTimeProlonged = new List<ExportLifetimeContext<T>>();
            while (MefParts.TryTake(out item))
            {
                if (item != null && condition(item.Value))
                    item.Dispose();
                else 
                    lifeTimeProlonged.Add(item);
            }
            if (lifeTimeProlonged.Any())
            {
                //Add back again the parts not matching condition to the Concurrent bag
                foreach (var part in lifeTimeProlonged)
                {
                    MefParts.Add(part);
                }
            }
        }

        public void OnImportsSatisfied()
        {
            //marker interface
        }
   
        /// <summary>
        /// Resolves the mef part
        /// </summary>
        /// <returns></returns>
        public static T Resolve()
        {
            var factoryInstantiator = ServiceLocator.Current.GetInstance<ExportFactoryInstantiator<T>>();
            MefParts.Add(factoryInstantiator.Lifetime);
            return factoryInstantiator.Instance;
        }

    }
}



using System.ComponentModel.Composition;

namespace SomeAcme.Client.Infrastructure.IoC
{

    [Export]
    [PartCreationPolicy(CreationPolicy.NonShared)]
    public class ExportFactoryInstantiator<T> : IPartImportsSatisfiedNotification
    {

        [Import]
        public ExportFactory<T> Factory { get; set; }

        public T Instance { get; private set; }

        private ExportLifetimeContext<T> _lifeTime;

        public ExportLifetimeContext<T> Lifetime
        {
            get { return _lifeTime; }
        } 

        public void OnImportsSatisfied()
        {
            _lifeTime = Factory.CreateExport();
            Instance = _lifeTime.Value;
        }

        public bool DisposeOnDemand()
        {
            if (_lifeTime == null)
                return false;
            _lifeTime.Dispose();
            return Instance == null;
        }

    }

}

To instantiate an object, you do:

 var somepart = MefFactory.Resolve();

When you are done using the object you can dispose it with:

 MefFactory.DisposeMefParts(); 

Please note, you can use a Predicate here to filter out which object you want to keep and which ones to dispose.

And once more, the immediate disposal of the object is not guaranteed, since GC will still control the true lifetime of objects. You can use GC.Collect(); to force releasing disposed objects, but that will usually degrade application performance.

But the techniques shown here will over time really improve your application by gaining control on the memory footprint your application uses.

Resources

[1] Enterprise Library: https://msdn.microsoft.com/library/cc467894.aspx
[2] ServiceLocator class: https://msdn.microsoft.com/en-us/library/microsoft.practices.servicelocation.servicelocator(v=pandp.51).aspx
[3] ServiceLocator pattern: https://msdn.microsoft.com/en-us/library/ff648968.aspx
[4] Managed Extensibility Framework: https://msdn.microsoft.com/en-us/library/dd460648(v=vs.110).aspx