Non-point spatial data still supported?

Hi, using spatial views, I’m able to work with points, lines, polygons - all the GeoJSON data types. In the documentation, it now has a note:

Note: In Couchbase Server 6.0+, Spatial Views are no longer supported.

I’ve looked into the Search service geospatial query type, but the index seems to only support points.

a geospatial query must reference an index that applies the geopoint type mapping to the field containing the latitude-longitude coordinate pair.

Does Couchbase still support more than points?

@pjoyce It is fantastic to know that you are using our Couchbase product. Please look here https://docs.couchbase.com/server/current/release-notes/relnotes.html#release-651

Additional details can be found here https://docs.couchbase.com/server/current/fts/fts-geospatial-queries.html
Hope this helps

@raju thanks for your response.
My second quote in the original question is from the second link you supplied.
Specifically the first sentence in section Creating a Geospatial Index https://docs.couchbase.com/server/current/fts/fts-geospatial-queries.html#creating_a_geospatial_index says

To be successful, a geospatial query must reference an index that applies the geopoint type mapping to the field containing the latitude-longitude coordinate pair.

Note “pair” singular. Hence my question: is non-point data still supported?

In case you’re not familiar with GeoJSON, the RFC is the reference RFC 7946 - The GeoJSON Format
If you look in Appendix A you can see all the different types of geometries that Couchbase spatial views supported.

Hi @pjoyce,

FTS only supports geopoint indexing and queries on them like bounded polygon/rectangle or point radius.
If you throw some light into your exact functional requirement, it would be very informative for us to prioritise them for FTS backlog…

You wanted to index polygon shapes and run overlapping queries on them?

Thanks,

Hi Sreekanth
Thanks for your reply

I’ve done some more investigation, and found this is where your current geopoint functionality is implemented
github.com/blevesearch/bleve/tree/master/geo

I see it has this note in the README.md

First, all of this geo code is a Go adaptation of the [Lucene 5.3.2 sandbox geo support]

Checking what Lucerne implemented I can see that it only indexed points as well.

By comparison, Elastisearch handles “arbitrary geo shapes such as rectangles or polygons” Geoshape field type | Elasticsearch Guide [8.11] | Elastic

Also, the old Couchbase 5.5 spatial views supported all the GeoJSON types:

So the requirement is to provide the same level of spatial functionality that spatial views provided, since these are no longer supported.

Regards, Paul
PS: sorry - I had to remove a few hyperlinks because a “new user” can only put two in a post

1 Like

Hi @sreeks , I didn’t @ you in my reply, so in case you’ve missed it I’m tagging you here.
Is there enough information in my reply for your backlog?
Thanks.