views:

53

answers:

1

Let's say I have 2 instance of a class called 'Animal'.

Animal has 3 fields: Name, Age, and Type

The name field is nullable, so before I insert an instance of Animal as a Lucene indexed document, I check if Animal.Name == null, and if it does, I do not insert it as a field in my document. If I were to retrieve all animals, I would see that the Name field does not exist and I can set its value to null.

However, there may be situations where I want to say "Get me all animals that do not have a name specified yet." In this situation I want to retrieve all Lucene.NET documents from my animal index that do not contain the Name field.

Is there an easy way to do this with Lucene.NET? I want to stay away from having to perform some sort of hack to check if my name field has a value of 'null'.

+2  A: 

I believe you can do this with Solr, but not with Lucene directly, so it's not possible with Lucene.Net.

Here's two workarounds, which are not that bad:

  1. For items with a NULL value in the field, add a custom string like __NULL__ or similar instead of omitting the field. This would be searchable.
  2. For items with a NULL value in the field, add a field which will not be present on the items with a value. Eg. EMPTY_FIELD = "no". This can be used in a filter.

Hope this helps you a bit on the way.

Mikael Svenson
Your solution of just performing a hack with a reserved keyword will work. Now I just need to figure out why special characters are not properly handled when I perform a search.I created another post to find out why I am getting such strange behavior with special characters. http://stackoverflow.com/questions/2732987/lucene-and-special-characters
Brandon