Removing HTML tags

In most situations where content to be indexed contains HTML tags, remove the tags before indexing. If you do not, search results return HTML markup.

Example of removing HTML tags from a specific RemoveHtmlTagsWhenIndexing attribute found in the EPiServer.Find.Json namespace:

using EPiServer.Find.Json;

public class WithStringProperty {
  public string Title {
    get;
    set;
  }

  [RemoveHtmlTagsWhenIndexing]
  public string Content {
    get;
    set;
  }
}

You also can customize the Client conventions to remove HTML tags from string fields:

client.Conventions.ForInstancesOf<object>()
  .FieldsOfType<string>().StripHtml();

To remove HTML tags from a specific field when indexing a particular type, use the ForType and Field methods:

client.Conventions.ForType<BlogPost>()
  .Field(x => x.Content).StripHtml();

The StripHtml method also performs HTML decoding. The goals are to index the text that users see when viewing the page and to be able to find that content.

For example, Optimizely Search & Navigation stores the Swedish text Jag gillar äpplen as Jag gillar äpplen and decodes it back when indexing. A user can find the text using a query like äpplen.