Hi guys, i'm saving json data in the database as a clob field, and indexing all json properties using FieldBridge:


I'm indexing using a custom FieldBridge, using the format:
myList.name
myList.active

If i query to: +name:test1 +active:true will return the object, but it's wrong because the "test1" it's false.

So how can i do the search work individually for each entry of json, do i have to create 3 different documents? How can i create new documents in the FieldBridge implementation?

Hello,
it's not possible to store multiple documents for a single object. This is usually done by mapping this as a real list of objects, and "inverting" the map so that each element of the list is an indexed element.

In your case as it's mapped as single blobl object you could apply some other "tricks".

What about mapping this with multiple fieldnames:
"name_active: test1"
"name_notactive: test2"

Then when you perform a Query you can decide to search for the name_active or name_notactive fields, or both.

Or, in case you don't need to search on non-active elements, to just skip indexing the ones which are not enabled.

I'm sorry but the Lucene index is not relational, so you have to be clever and think about this kind of tricks. We're thinking into hacking the Lucene format for version 4 to make it more relational-friendly.

My field it's really a String and i persist as a Clob in the database.

I had thought about the tricks, but the problem is i do not know the json format, users are free to save any data they want.

The "trick" is the only way I can go? Why I must have just one document per object?