How To Integrate Elasticsearch In ASP.NET Core

public class ArticleModel {
    public int Id {
        get;
        set;
    }
    public string Title {
        get;
        set;
    }
    public string Link {
        get;
        set;
    }
    public string Author {
        get;
        set;
    }
    public string AuthorLink {
        get;
        set;
    }
    public DateTime PublishedDate {
        get;
        set;
    }
}

{
    "Logging": {
        "LogLevel": {
            "Default": "Information",
            "Microsoft.AspNetCore": "Warning"
        }
    },
    "ElasticSettings": {
        "baseUrl": "https://localhost:9200/",
        "defaultIndex": "articles"
    },
    "AllowedHosts": "*"
}

public static class ElasticSearchExtension {
    public static void AddElasticSearch(this IServiceCollection services, IConfiguration configuration) {
        var baseUrl = configuration["ElasticSettings:baseUrl"];
        var index = configuration["ElasticSettings:defaultIndex"];
        var settings = new ConnectionSettings(new Uri(baseUrl ?? "")).PrettyJson().CertificateFingerprint("6b6a8c2ad2bc7b291a7363f7bb96a120b8de326914980c868c1c0bc6b3dc41fd").BasicAuthentication("elastic", "JbNb_unwrJy3W0OaZ07n").DefaultIndex(index);
        settings.EnableApiVersioningHeader();
        AddDefaultMappings(settings);
        var client = new ElasticClient(settings);
        services.AddSingleton < IElasticClient > (client);
        CreateIndex(client, index);
    }
    private static void AddDefaultMappings(ConnectionSettings settings) {
        settings.DefaultMappingFor < ArticleModel > (m => m.Ignore(p => p.Link).Ignore(p => p.AuthorLink));
    }
    private static void CreateIndex(IElasticClient client, string indexName) {
        var createIndexResponse = client.Indices.Create(indexName, index => index.Map < ArticleModel > (x => x.AutoMap()));
    }
}

In this code, in AddElasticSearch method you can see that basic options are configured. Username, password, and certificate footprint generated during Elasticsearch instance setup are used here. In AddDefaultMappings method, you can see Link and AuthorLink properties are ignored from the search. And CreateIndex is creating a mapping between the model class ArticleModel and the index from Elasticsearch instance.

Now let’s use this extension ElascitSearchExtension as middleware. For that open Program.cs in your project and use the extension like in this code

var builder = WebApplication.CreateBuilder(args);
builder.Services.AddElasticSearch(builder.Configuration);

Adding Documents in Elasticsearch

Here I will discuss 2 ways to add documents in Elasticsearch, add a single item and bulk import from a JSON file. Let’s take a scenario of an e-commerce site where you have thousands of products. Adding a single document can be used to add/update document when you add or update your product in your ecommerce site. And Bulk import can be used for rebuilding the indexing of all products.

This code will add new documents in the Elasticsearch index.

private readonly IElasticClient _elasticClient;
private readonly IWebHostEnvironment _hostingEnvironment;
public ArticleController(IElasticClient elasticClient, IWebHostEnvironment hostingEnvironment) {
    _elasticClient = elasticClient;
    _hostingEnvironment = hostingEnvironment;
}

[HttpPost]
[ValidateAntiForgeryToken]
public async Task < IActionResult > Create(ArticleModel model) {
    try {
        var article = new ArticleModel() {
            Id = 1,
                Title = model.Title,
                Link = model.Link,
                Author = model.Author,
                AuthorLink = model.AuthorLink,
                PublishedDate = DateTime.Now
        };
        await _elasticClient.IndexDocumentAsync(article);
        model = new ArticleModel();
    } catch (Exception ex) {}
    return View(model);
}

To update existing document you can use UpdateAsync method

await _elasticClient.UpdateAsync<ArticleModel>(article.Id, u => u
    .Index("articles")
    .Doc(article))

To delete a document you can use DeleteAsync method

await _elasticClient.DeleteAsync<ArticleModel>(article);

And bulk import can be done as in this code

[HttpPost]
[ValidateAntiForgeryToken]
public IActionResult Import() {
    try {
        var rootPath = _hostingEnvironment.ContentRootPath; //get the root path
        var fullPath = Path.Combine(rootPath, "articles.json"); //combine the root path with that of our json file inside mydata directory
        var jsonData = System.IO.File.ReadAllText(fullPath); //read all the content inside the file
        var articleList = JsonConvert.DeserializeObject < List < ArticleModel >> (jsonData);
        if (articleList != null) {
            foreach(var article in articleList) {
                _elasticClient.IndexDocumentAsync(article);
            }
        }
    } catch (Exception ex) {}
    return RedirectToAction("Index");
}

Searching Documents

This code search documents from Elasticsearch instance based on the input keywords.

[HttpGet]
public ActionResult Index(string keyword) {
    var articleList = new List < ArticleModel > ();
    if (!string.IsNullOrEmpty(keyword)) {
        articleList = GetSearch(keyword).ToList();
    }
    return View(articleList.AsEnumerable());
}
public IList < ArticleModel > GetSearch(string keyword) {
    var result = _elasticClient.SearchAsync < ArticleModel > (s => s.Query(q => q.QueryString(d => d.Query('*' + keyword + '*'))).Size(5000));
    var finalResult = result;
    var finalContent = finalResult.Result.Documents.ToList();
    return finalContent;
}

You can read more about the search here.

The output of the search looks like this in this screenshot

Conclusion

In this article, I discussed the basic introduction of Elasticsearch. I also discussed adding Elasticsearch middleware extension in ASP.NET Core, adding new documents, and searching documents. Elasticsearch has a lot of features that can not be covered in a single article. I hope this article will give you some idea about the use of Elasticsearch and put you in the right direction. You can download the source code from this article and give it a try.

Please feel free to ask in the comment section if you have any suggestions.

Thank you for reading the article.