Tips for Microsoft Developers

AI skills needed for today's jobs

2024-05-17T15:25:00.000+09:30

I searched the seek.com.au site today for all jobs advertised in Australia with the keyword "AI". It showed total of 836 jobs. The interesting part was that they were not just "Information & Communication Technology" jobs but all other sorts of job. The screenshot below shows the top of the list:

Thinking about a new accounting job? You need to know AI!

Of course the sector with highest number was "Information & Communication Technology" with total of 359 jobs.

The list below shows where these 359 jobs are mostly located:

Sydney: 154

Melbourne: 89

Brisbane: 39

Perth: 27

Adelaide: 15

For my home town Adelaide, out of 15 jobs advertised these were subcategories:

Developer/Programmer: 5

Engineering -Software: 3

Management: 2

Consultant: 2

Architect: 1

Business/System Analysis: 1

Program & project Management: 1

For the top category of "Developer/Programmer" these were places that advertised for work:

NovaWorks - $95,000-$129,999

Expert Ease AI - $95 per hour

Neo Analytics $180,000- $200,000

MyOSH - (not declared)

Comunet (not declared)

The job offered by "Expert Ease AI" which was titled: "Full-Stack Developer/ AI Prompt Programmer for Conversational AI Platform" had these requirements:

We Are Seeking:

Experienced Full-Stack Developer with a strong background in Python, Django, React, Prompt Programming, Large Language Models (LLMs), storage and retrieval of vector embeddings, Machine Learning, Few-shot prompting, In-context learning, and AI automations.

Internationalization of .NET applications

2020-11-18T09:33:00.002+10:30

NGettext Nuget package is used to add internationalization (i18n) support to your .NET application, i.e. translating labels and strings used in your application to other languages. You can use Poedit application to mange translations to different languages. For each language you would have a different po file which Poedit can save as mo file which is used by your application.

The code example below uses NGettext package to provide translation service:

using NGettext;

using System.Globalization;

using System.IO;

namespace YourApp

{

public static class Translate

{

static Catalog catalog;

static CultureInfo currentCulture;

public static void SetTranslationLocale(string locale = "")

{

string fileName = "en_GB.mo";

switch (locale.ToLower())

{

case "de":

fileName = "de_DE.mo";

break;

case "en":

fileName = "en_GB.mo";

break;

case "es":

fileName = "es_ES.mo";

break;

case "fr":

fileName = "fr_FR.mo";

break;

case "pt":

fileName = "pt_PT.mo";

break;

case "it":

fileName = "it_IT.mo";

break;

}

currentCulture = new CultureInfo(locale);

FileStream moFileStream = File.OpenRead(Path.Combine(@"..\..\Resources\Translations", fileName));

catalog = new Catalog(moFileStream, currentCulture);

}

public static string TranslateToCurrentLocale(string stringToTranslate)

{

return catalog.GetString(stringToTranslate);

}

ML.NET Model Builder

2020-09-04T15:25:00.003+09:30

Today I came across ML.NET Model Builder in Visual Studio. This enables programmers using .NET to build, train and use a model in their programs for a variety of AI related projects such as classification. For example imagine you have a large number of pictures of 4 different types of flowers say roses, sun flowers, tulips and daisies. You want to build a Machine Learning model that goes through these images and after training builds a model which you can feed an image of a flower to and it tells you which type of flower it is. That is exactly what you can do by using ML.NET model builder in your application. The good news for this particular example is that the list of free pictures of flowers is already available that you can download and train your mode with.

This link shows you step by step how you can set up your Machine Leaning experiment using ML.NET

https://devblogs.microsoft.com/dotnet/ml-net-model-builder-is-now-a-part-of-visual-studio/

SQL Refresher

2020-08-14T15:12:00.000+09:30

Code developers rarely need to set up a SQL database and generate queries. This activity is normally done once in the early stages of a project and not performed much later on when the project enters maintenance or feature addition/enhancement phase which is like 80 to 90 percent of the time. As a result of this and because SQL query generation is not an every day activity the developers tend to forget some of the tricky parts of SQL queries, for example those involving GROUP and CASE statements. Recently I came across sqlzoo.net site which provided a nice refresher for me for SQL queries. It comes with a number of questions that you can write your answer online in SQL and see if the query you write passes the test.

Using Docker in Windows along Virtual Box

2020-08-08T16:22:00.001+09:30

If you have tried to install "Docket Desktop for Windows" to run Docker containers on your Windows machine you might have noticed that since it enables HyperV it prevents Virtual Box from running your virtual machines. If you have some virtual machines managed by Virtual Box then you need to install "Docker Toolbox" instead. "Docker Toolbox" does not use HyperV. The other benefit of using "Docket Toolbox" instead of "Docker Desktop" is that since it does not need HyperV it can be run on Windows 10 Home edition and you do not need to have Window 10 Pro..

JavaScript promises do not fulfil in the order they were called

2020-02-05T14:14:00.004+10:30

I learned a lesson today, or more accurately I had to remind myself of a fact on how JavaScript promises work. My front-end application was not working as expected. The tables were not populated with the data. The data was provided by a service as a promise. On the then() part of that promises I was populating my table. The promises were called in a loop, the index of the loop was passed as a parameter to the callback function. Since the loop started with zero and went up to a certain size, I assumed that the handlers are also called one by one in the same order and I was relying on the index number passed to reach the set max value. While debugging I realised that callbacks are not executed in the same order that promises are called, After all they are asynchronous operations with no guarantee of timing of their execution. The solution was to remove the check for the index and instead rely on the number of times the callback is called in total and take actions based on that.

Using an AngularJS service in Angular

2020-01-15T12:26:00.003+10:30

We had an AngularJS application that we decided to gradually add Angular components to it. This can be done using Angular Upgrade. If an existing AngularJS service is needed in an Angular component the following steps need to be performed. Note that in this example I am going to use an AngularJS service called EventService in my Angular framework:

1. Create a file called event-service-provider.ts in yourpath with this content:

export function EventServiceFactory(i: any) {
return i.get('EventService');
}

export const EventServiceProvider = {
provide: 'EventService',
useFactory: EventServiceFactory,
deps: ['$injector']
};

2. Update app.module.ts by adding sections highlighted below:

import { EventServiceProvider } from 'yourpath/event-service.provider';

.....
providers: [ ...
EventServiceProvider,
...

]

3. In a component that you want to use this service do the following:

import { Inject } from "@angular/core";

export class YourComponent {
private EventService: any;
constructor(@Inject('EventService') EventService: any){
this.EventService = EventService;
EventService.subscribe(null, 'region:updated', (region) =>{ this.translateLanguage(region.id);}, false);
}

In this particular example subscribe is a function of EventService. we are subscribing to region:update event. When that event happens we call a method called translateLanguage() in this class passing it the value of region.id we got from EventService.

A template ASP.NET Core project with MVC, Web API and SignalR

2019-08-15T11:20:00.000+09:30

Today I created a Visual Studio ASP.NET Core 2.2 project that includes all these items:
1. MVC
2. Web API
3. Entity Framework Core
4. SignalR
It is stored at
https://github.com/mbahrami/ASP.NET-Core-with-MVC-Web-API-and-SignalR
The SignalR component would broadcast a message to all clients when one client sends a message similar to a chat application. I had a Python application running on a Raspberry PI implementing a web socket client which would listen to the messages sent by web clients and would turn an LED on or off upon receiving of those messages.

Raspberry PI "No wireless interface found"

2019-08-04T14:45:00.005+09:30

I was working on a "Remote Control" project in which a user would set a switch on a web site to on or off and then an application running on a Raspberry Pi constantly checking the status of that switch (from a web API)would turn an LED on or off to reflect the status of that switch.

Everything was going OK until at the last minute my Raspberry Pi could not connect to the internet. It was showing “No wireless interface found”. To fix that I needed to write this in /etc/network/interfaces file:

# Include files from /etc/network/interfaces.d:

source-directory /etc/network/interfaces.d

auto lo

iface lo inet loopback

auto eth0

iface eth0 inet manual

allow-hotplug wlan0

iface wlan0 inet manual

and reboot the Raspberry PI.

Programming Languages Frameworks

2019-07-01T11:34:00.001+09:30

In the last two previous posts I listed the current most popular or used programming languages based on the IEEE investigations and the GitHub statistics. For someone interested in becoming a software developer it might seem that if they learn and master a programming language such as JavaScript or C++ then they can immediately work on any project using one of these languages. That used to be the case may be 10 years ago but these days on top of each programming language we also have a programming language framework.

Frameworks extend a programming language and provide a particular way of performing different tasks. Each framework has its own structure and convention of programming which takes some time for a developer to learn and become familiar with.

For JavaScript we currently have Angular.js, Vue.js, React.js, Ember.js, Backbone.js, and Aurelia.js just to name a few.

In C++ we have Qt and Boost as probably the most widely used frameworks.

To be a competent software developer these days one has to not only be proficient in a programming language but also the main frameworks used by that programming language.

The top 10 programming languages (IEEE Spectrum rankings)

2019-06-19T12:38:00.001+09:30

In a previous post I listed the top 10 programming languages based on GitHub usage. The IEEE Spectrum has a different ranking based on search, social media, job listing trends,
GitHub repositories, and mentions in journal articles. Details are at
IEEE 2018 Top Programming Languages

The IEEE ranking is shown below:

Javascript is the top language in the GitHub listing but is ranked 8th in IEEE listing. Probably a Javascript piece of code is more likely to be deposited to the GitHub than a Python or C++ piece of code.

The IEEE listing also indicates the domain a language is used in (being web, mobile, desktop or embedded).

The top 10 most used programming languages

2019-06-16T13:27:00.000+09:30

The top 10 most used programming languages based on Github statistics is shown below:

As the above graph shows Javascript, Java, Python and PHP have been the top 4 languages used in the last 4 years.

I imagine in the future we would see PHP declining in usage as I don't see any good reason these days for using it in a new application. Client side frameworks such as Angular and React are the architecture of choice these days.

Typescript would continue to increase in usage, partly because of Angular now being Typescript based.

Objective C has lost its top 10 position in last 2 years. New mobile application are more likely to be Cordova (Javascript) or Xamarin(C#) based. Even for iOS specific systems Swift has replaced Objective C but Swift is still not a top 10 programming language.

Shell and C++ have received more attention lately which I think is due to wider usage of Linux based systems in recent years.

Increasing the size of DatePicker icon in WPF

2018-12-04T12:50:00.000+10:30

The default DatePicker control in WPF looks like this:

In a desktop application you can click on the icon (which has number 15 on it) to get the calendar to pick a date. That works OK on a desktop application with a mouse but on a touch screen device the default icon is too small for easy touch which would annoy the user.

When you add a DatePicker to your WPF application the code that you get is the following:

<Grid>

</Grid>

You cannot change the size of the icon in this code. You need the extended code behind for the template used by DatePicker. To get that you need to right click on the DatePicker control in Visual Studio and select Edit Templates -> Edit a Copy ... as shown below:

This will first bring up a dialog as shown below

It will then add the template code to your file. In the extended code the line highlighted below represents the button containing the icon.

You can extend the width in this line from default 20 pixel to say 30. Also find the Grid that contains this button and change its width and Height values.

The default values are Height =”18” and Width=19” that you can increase (to say 30) for better usability on touch screen devices.

Electron Framework

2018-10-05T09:14:00.005+09:30

You might have heard of Cordova framework where languages and tools used for developing web sites (i.e. HTML, Javascript and CSS) are used to make mobile phone native applications. This concept can be extended further to desktop application using Electron framework.

Electron framework allows you to develop native Window, Linux or Mac application using the same tools that are used in developing web applications. By using this approach, the development team only need to learn and master one way of developing UI. The other benefit is that applications developed for desktop can easily be adjusted to be displayed on web instead of whole new rewrite. As you would expect an Electron application feels like a web site although being a desktop application.

You can think of Electron framework as a web server and a browser combined in one application. The web server side is a slightly modified Node.js and the browser is a slightly modified Chromium. Chromium is the open source version of Chrome that the developer community can use. The two browsers share the majority of code and features. Node.js is also open source so the whole Electron framework is based on open source.

To run Electron, make sure you have installed Git and Node.js. Then follow these to get a skeleton Electron based application working on your computer.

# Clone this repository

git clone https://github.com/electron/electron-quick-start

# Go into the repository

cd electron-quick-start

# Install dependencies

npm install

# Run the app

npm start

Electron is developed and maintained by Github and is used in many well-known desktop applications such as Slack, Spotify and Visual Studio Code to name a few.

The initial release was 2013 and the latest version is 3.0.0. For more info go to https://github.com/electron/electron and https://electronjs.org/

Enabling Remote Access to SQL Server

2018-05-31T11:24:00.003+09:30

When you install SQL Server on a server, remote access to that is disabled by default as a security measure. To enable remote access to a SQL Server you need to perform the following steps:

1. When installing SQL Server on a machine, tick the box to install “Client Tools Connectivity” too . With that you get “SQL Server Configuration Manager” that you need later.

2. In “SQL Server Configuration Manager” under connections select “TCP/IP enable”. Then restart SQL Server service.

3. In “SQL Server Configuration Manager” start “SQL Server Browser” service.

4. In Windows fire walls settings open Inbound and Outbound channels for (SQL Server – port 14343 TCP/IP) and (SQL Server Browser – port 1434 UDP).

New Angular Template Project in Visual Studio 2017

2017-12-19T16:46:00.000+10:30

The latest update of Visual Studio 2017 (version 15.5.2) has new template projects for ASP.NET Core applications.

In Visual Studio 2017 when you select File > New > Project, you now get this

If you select the subcategory Web Site then you get the following:

In the original screen if we select "ASP.NET Core Web Application" as shown below:

We will be given the following choices:

As you see in above screenshot you can now choose between a number of options including Web API, React.js and Angular. If you choose Angular you will get a project template that can run without any modifications and will bring this web page:

As the template indicates, this application uses ASP.NET Core for the back end and Angular for the front end. It also uses Webpack for the front end building and bundling and Bootstrap for layout and styling.

The solution in the Solution Explorer looks like this:

Software Technology Jobs in Australian Capital Cities

2017-12-13T13:40:00.002+10:30

seek.com.au is the most popular job hunting site in Australia. I did a search on the number of jobs that are advertised for Australia capital cities on 12/12/2017 on this site using different technologies as key words. The result is shown below:

Two things to note from this chart:

Firstly, availability of software jobs is very much dependent on the city. For example there are a lot more jobs available in Sydney than any other city in Australia. That is expected as Sydney has the highest population but even when you take into account the population the advertised software engineering positions in Sydney are a lot more than a city like Adelaide. Sydney has a population of 5 million compared with Adelaide which is only 1.2 million so you expect Sydney to have 5 times more jobs than Adelaide but as you can see from the above graph the ratio is 14 times for Java and 15 times for HTML.

The other point to notice is the the technology mix in demand and whether it is worth your time to learn and master a technology which compared with other technologies has a lower level of demand in the job market at the moment.

Web API with Angular – Part 6 – Angular Service Connection to Back End

2017-11-29T16:26:00.002+10:30

In the previous posts we created an Angular service with a method called getPerson(). That method was returning a fixed text. At this post we connect that angular service to the back end so getPerson()gets its data from the back end.

Our Web API and the angular project are running on different port numbers (50135 for Web API in this example vs 4200 for angular). This constitute a Cross Origin Resource Sharing (CORS) situation. For this to work we need to modify our Web API to enable CORS. First we need to install CORS package by typing:

Install-Package Microsoft.AspNet.WebApi.Cors

Then in file App_Start/WebApiConfig.cs add the following code to the WebApiConfig.Register method.

config.EnableCors();

And add the following attribute to the PeopleController class:

[EnableCors(origins: "http://localhost:4200", headers: "*", methods: "*")]

This will allow any ajax request from localhost:4200 to go through.

A service method like getPerson() in angular cannot be a synchronous one. It takes time for a request to get to the back end and the result be returned back. getPerson() can take a callback, or it can turn a Promise or an Observable.

Callbacks generally result in messy code with limited error handling. Promises and Observables provides an alternative to callbacks resulting in a cleaner code. They both create an abstractions that help us deal with the asynchronous nature of our applications. AngularJS is based on Promises but Angular 2 uses Observables because they can define both the setup and tear-down steps and they can be cancelled or re-tried.

To use Observable in PersonService we first need to define a Person class (in person.ts). The structure of this class would be similar to Person.c model in our back end:

Now we need to import that and Observable and of symbols from RxJS to person.service.ts :

import { Person } from './person';

import { Observable } from 'rxjs/Observable';

import { of } from 'rxjs/observable/of';

Also replace the getPerson method with this one.

getPerson(): Observable<Person[]> {

return of(Persons);

}

We need to reflect this change in person.component.ts.

From this:

getPerson(): void {

this.person = this.personService.getPerson();

}

To this:

getPerson(): void {

this.personService.getPerson()

.subscribe(person => this.person = person);

}

Angular communicates with the back end server through HttpClient. To make HttpClient available we need to import it into AppModule and and add it to the @NgModule.imports array.The result app.module.ts is shown below:

Then import httpClient to person.service.ts, inject it into constructor and modify the getPerson() method. The result is shown below:

With these modifications in place the angular service can use the Web API as its data source as shown below, note the name Fred Green comes form Web API:

The latest version of files are shown below:

person.ts:

person.service.ts:

person.component.html:

person.component.ts:

app.module.ts:

app.component.html:

Web API with Angular – Part 5 – OData Query

2017-11-28T10:21:00.000+10:30

In this section of series of related posts we perform OData query on our Web API. Fiddler is the tool we will use for this exercise.

We have added two persons to our database by updating our Seed() function as shown below:

After updating this function we need to run update-database command in package manager console so the database gets updated.

When we perform

GET /odata/People

The result would be as shown in the screenshot below:

Now say we want to get only the top 1 of the list, then according to OData convention that would be

GET /odata/People?$top=1

The result is shown below:

To order the list by LastName the message would be

GET /odata/People?$orderby=LastName

And result would be as shown below:

Please refer to OData references for other conventions used for query.

Web API with Angular – Part 4 – Adding Angular Component and Service

2017-11-27T19:52:00.004+10:30

In other sections of these series we created a Web API using OData and Entity Framework and tested that on a sample data. We also installed angular-cli and created a project in our Visual Studio containing files generated by that installation. In this part we add a component and a service to the angular project. In another post we will make this new angular service use our Web API for it data.

Using Windows Explorer go to the location of angular-project then open a command window by pressing shift and doing a right click then select “Open PowerShell Window here”. In the window that comes up type

ng generate component person

This will generate a component called person which will be in a folder with that name. That folder will contain 4 files: person.component.ts, person.component.html, person.component.css and person.component.spec.ts. The last file is used for testing.

The content of person.component.ts is as shown below:

Note that the selector for this component is app-person that we will use in app.componenet.html later.

person.component.html at this stage is as shown below:

In the same folder type the following to generate person service

ng generate service person

This will generate two files in that folder: person.service.ts and person.service.spec.ts. The second file is used for testing.

person.service.ts will look like this:

After including these files into Visual studio the Solution Explorer looks like the following where newly added files are highlighted:

We modify person.service.ts by adding a getPerson() method to it:

To make person component use person service we need to modify it. We first import the service and then call its getPerson() method in ngOnInit(). Note the injection of the service in the constructor:

We need to register new component and service in app.mudule.ts as shown below

No if we modify person.component.html like below

<p>{{person}</p>

And add this to app.component.html:

<app-person></app-person>

Then the web page will look like this.

Note the text "Person Service returned This!" is coming from the person service.

In subsequent post we will change the person service to get its data from Web API instead of returning this fixed text.

Web API with Angular – Part 3 - OData with Entity Framework

2017-11-27T10:40:00.001+10:30

In Part 1 and 2 of these series of posts we created two projects in a Visual Studio solution one to provide data (Web API) and one to display the web page using angular-cli. In this post we create a web API controller using Open Data protocol (OData) and Entity Framework. By selecting this option we allow the front end to query the data using OData standards.

The first step is creating an Entity Framework model as an example. For this, we create Person model which we will store in Person.cs in the Model folder as shown below:

We need to include this model in ApplicationDbContext:

We seed the application with some data for testing. To do that first enable migration by typing the following command in the Command Window

enable-migration

Then

add-Migration “Initial”

This will create Migrations folder with config.cs file in it. That file has a method called Seed which we modify to seed one person to the database as shown below:

Next we create a controller by right clicking on the Controller folder and selecting Add -> Controller as shown below:

In the window that comes up we select “Web API 2 OData v3 Controller with actions using Entity Framework”:

Then in subsequent window we select Person as the model:

This will scaffold PeopleController.cs file. The top section of this file is shown below:

The comment in the auto-generated code indicates that for OData routing to take effect you need to update WebApiConfig.cs. Below I have applied that update:

I have also commented out the section that requires authentication to simplify testing at this stage.

There are different tools available to test Web API. One of these is Fiddler which is free. It can be used to create and run different HTTP requests to Web API and check the HTTP response. Please refer to http://www.tutorialsteacher.com/webapi/test-web-api to see how Fiddler can be used to Test Web API.

Below I have used Fiddler to send a GET odata/People request.

The result is highlighted below which shows the Web API for odata/people route works.

Now that we have created a sample Web API and tested it we can update the Angular application to use this Web API.

Web API with Angular – Part 2 - Installing angular-cli

2017-11-26T16:22:00.001+10:30

In Part 1 of this post we created a project to provide Web API for our solution. In this part we generate an angular-cli (Command Line Interface) project to include in the original solution. For downloading and installing angular-cli you need to have installed node.js and npm on your computer.

Using Windows Explorer go to where the newly created empty project is stored, An example is shown below:

While pressing Shift key, right click at this location and choose “Open PowerShell window here” as shown below:

Then type the following command:

npm install -g @angular/cli

when installation is complete type the follwing command. In this example angular-project is an arbitrary name for the project.

ng new angular-project

this will create a new project for you in angular-project folder. Change directory to that folder

cd angular-project

and type

npm install

This command would download and install all necessary files and packages.

In Visual Studio we include these files into the project we recently created. You do not need to include nodes_module folder.

To run this project at the angular-project folder type

ng serve

which will result in

As the message indicates the page is served at localhost:4200 as shown below:

This will provide the web page for us. We will use the WebAPI project only to get or store data for this web page.

Web API with Angular - Part 1 - Creating a Web API project

2017-11-26T14:26:00.001+10:30

Angular has become a widely used front end framework for modern web applications. It is an open-source platform with built in features such as dependency injection and declarative template. ASP.NET Web API provides a back end RESTful web interface for data access.

This post is about creating a Visual Studio project using Angular and Web API.

I will be using Visual Studio 2017 Community Edition in this exercise.

Creating a Web API project

Using Visual Studio create a new Single Page ASP.NET project with Web API as shown below:

Select New -> Project/Solution and the sellect ASP.NET Web Application:

Then select “Single Page Application” with MVC and Web API ticked as shown below:

This creates a template application that can run as shown below:

Note that by doing the above steps the project would also contain a Web API interface and Entity Framework. Entity Framework provides an abstraction layer for developers so they can work with data objects and not be concerned with how that data gets stored or retrieved form a storage entity such as SQL Server.

The default connection in this configuration is localDB as indicated in web.config.

The structure of the solution at this stage is as shown below:

The Controller folder contains the Web API end points.

Creating a placeholder project for Angular

The next step is to add another project to the solution which will work as a place holder for Angular code as shown below:

And then select

Then select an empty project:

At this stage the Solution Explorer will look like this:

In the next post we populate the new project with angular-cli.

Using Elastic Search in Visual Studio

2016-07-25T14:02:00.000+09:30

These days customers wan t to be able to search their document repository efficiently, like the way they search the web. The response should come quickly, most relevant results should come first and the source should not be limited to text documents.

If you were using SQL Server as the source of your data you could achieve some of these by using SQL Server Full-Text Search feature. It would create a full-text index on the tables containing data which you want to search. This would allow you to perform a text search on binary files such as Word documents or PDF files but it would still lack ranking of search results based on their relevance .

Elastic Search provides an alternative with built in scalability, redundancy and ranking in search results. Elastic runs as a service in your server and you can scale it out to use many other servers if you need to. Elastic by itself cannot search binary files (such as Word and PDF documents) but with Tika plugin you can extract meta data and text from these files and then feed those to Elastic for index and search.

Here I show how you can use Elastic in a .NET project in Visual Studio. First download and install Elastic . Since Elastic uses Java you need to have installed Java on your computer first. After installation configure ElasticSearch windows service to run automatically. To ensure ElasticSearch is running on your computer go to http://localhost:9200 (9200 is the port that is used by elastic search by default). You would see a JSON formatted response as shown below:

In Visual Studio create a ASP.NET MVC solution. I have called this solution ElasticSearch1 in this example. Install “Nest” NuGet Package (Tools -> “NuGet Package Manager” -> “Manage NuGet Packages for Solution” and Search for “Nest”). This package provides a high level interface to ElasticSearch.

In Models directory create a Person.cs file with this content:

In home controller add a reference to Nest then add the following to your code:

Create ElasticIndex action method as shown below

And ElsticSearch action method as shown below:

You need to create the corresponding view ElasticIndex.cshtml:

and ElasticSearch.cshtml:

You need to give access to these two action by modifying _Layout.cshtml and adding the following:

Now run the application, select Elastic Index from the menu first. Then select Elastic Search from menu. You will get the following:

This provides a basic platform to test different settings of Elastic Search.

Content Type for xls and xlsx files

2016-07-21T13:12:00.001+09:30

Exporting a web page data grid into Excel is a common functionality provided by some web applications. If these applications were written in old xls days they would have a line like:

Response.ContentType = "application/vnd.ms-excel";

with Excel 2007 and above when the file extension changed from xls to xlsx the Content Type is:

Response.ContentType = "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet";