I guess it is not possible.
Maybe someone else have more reliable information than me but I'm pretty
sure it doesn't index fields recursively.
I have a similar scenario: very complex embedded documents, variably deep
and composed by many different data types. They are inside graph nodes and
I need to search over them with full text and the powerful expressivity of
lucene library but I can't do that directly within OrientDB.
So I use fulltext lucene indexes natively just for flat fields (or maps and
lists) and I relay on ElasticSearch as external tool for more complex
documents.
Of course you have to reindex your documents externally but having an hash
id comes really useful.
ElasticSearch is able to index documents recursively but its major
competitor Solr is not.
Both of them are able to accomplish the same goal. It's matter of strategy
about data representation. With Solr you have to make your multilevel
documents flat composing appropriate field names. As a key/value store, for
instance.
Using that very strategy you can do the same with native OrientDB lucene
index but in that case I guess it's much better to keep things simple.
My 2 cents.
On Sunday, January 24, 2016 at 12:42:26 PM UTC+1, Rajesh Vaya wrote:
>
> I have a property of type "embedded" and I have created a LUCENE index on
> that property. But it only searches the first level of embedded document
>
>
> e.g. the property is called "places" and the data stored is in the
> structure "country.state.city"
>
>
> USA:{
> FLORIDA:{
> MIAMI: "Some TEXT"
> }
> }
>
> I need to be able to search on "Some TEXT" which is a city level (3rd
> level)
>
>
> Any suggestions on how to run the query to check at city level?
>
>
>
>
>
--
---
You received this message because you are subscribed to the Google Groups
"OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
