Mapping of .NET objects to Lucene.Net Documents and vice versa.
Simply install the latest pre-release of the Lucene.Net.ObjectMapping NuGet package. It targets the recent pre-releases of the Lucene.Net and related packages.
Uses JSON Serialization to store the object in the Lucene.Net Document, and indexes each property (also nested properties) individually.
Any object that is JSON serializable (using Newtonsoft.Json) can be mapped to a Document, like this:
Document doc = myObject.ToDocument();Similarly, a Document created like called out above can be converted back into an object like this:
MyClass myObject = doc.ToObject<MyClass>();
// or
object myObject = doc.ToObject();Since the library stores the actual type of the object when creating a Document,
we can always reconstruct the same type of object when deserializing using
ToObject, provided that the types can be deserialized from JSON.
You can add your documents to your indices the usual way, or you can add an object to an IndexWriter like this:
myIndexWriter.Add(myObject);
// or
myIndexWriter.Add(myObject, myAnalyzer);Updating existing documents is just as easy. Like the UpdateDocument method on
the IndexWriter, it will delete the documents that match the given query and
then just add the new document. Use it like this:
myIndexWriter.Update(myObject, new TermQuery(new Term("Id", myObject.Id)));
// or
myIndexWriter.Update(myObject, new TermQuery(new Term("Id", myObject.Id)), myAnalyzer);Please note that it is not necessary to filter for the document type in your
query. The Update method does it automatically.
Search for Documents mapped from a specific class using the extensions to the
IndexSearcher, e.g. like this:
mySearcher.Search<MyClass>(myQuery, numResultsToReturn, mySort);There are other extensions to the IndexSearcher to, also non-generic ones in
case those are needed.
You can find more examples in the examples/ directory:
- DirectoryWeb - person directory of a fictional school