see on GitHub

Dynamic fields

While strongly typed entities are well processed by LINQ expressions, some scenarios demand the use of dynamic properties. To support searching in object graphs they cannot have their entire structure declared upfront. RavenDB exposes low-level API for creating fields from within index definitions.

With this feature, you can search for documents even with properties which were created on the fly. For example, consider a Product object that is declared as follows:

public class Product
	public string Id { get; set; }
	public List<Attribute> Attributes { get; set; }

public class Attribute
	public string Name { get; set; }
	public string Value { get; set; }

This way, properties such as color, size, weight, and the like are added only to those products of which they are indeed valid properties. However, while they are easily stored, they cannot be easily searched on.

This is where dynamic fields come in. With the following index definition, RavenDB will index the attribute value under the attribute name in its own field:

public class Products_ByAttribute : AbstractIndexCreationTask<Product>
	public class Result
		public string Color { get; set; }

	public Products_ByAttribute()
		Map = products => from p in products
					select new
						_ = p.Attributes
							.Select(attribute => 
								CreateField(attribute.Name, attribute.Value, false, true))

The underscore used for defining the field name in the Map object is just a convention, you can use any field name instead, but since we just want to call the CreateField method and we are not interested in the field value, we are using _ as a convention to reflect that.

The call to CreateField(...) will generate index fields based on the properties in the provided collection, without creating any field with the name specified there, hence the underscore.

Obviously, this index can have more attributes defined in it for indexing, just like any other ordinary index.


Field options like FieldStorage.No and FieldIndexing.Analyzed are configurable with dynamic fields as well.

When the index is created, we can easily look for documents using the attribute name as a field to look on, as if it was a real object property.

IList<Product> results = session
	.Query<Products_ByAttribute.Result, Products_ByAttribute>()
	.Where(x => x.Color == "Red")
IList<Product> results = session
	.DocumentQuery<Product, Products_ByAttribute>()
	.WhereEquals("Color", "Red")

This will also work for numeric values, so range queries or searches with numeric operators like WhereGreaterThan() for DocumentQuery will work as well.