cancel
Showing results for 
Search instead for 
Did you mean: 

Head's Up! These forums are read-only. All users and content have migrated. Please join us at community.neo4j.com.

Neo4j Bloom not showing Virtual Nodes properly

Hi Team,

I want to use Virtual Nodes in Bloom. I have a very basic vnode statement that works in the Neo4j browser but not in bloom. This very simple query...

CALL apoc.create.vNode(["CG"],{GP:"Hello"}) Yield node as a RETURN *

...wlll show the label and Property correctly in the Neo 4j browser but in bloom, it can't identify the label and property. Is there a different syntax required in bloom?

In the first attachment (bloom1.png) notice how the browser shows Property as GP, which is correct, but in Bloom, it says no properties. I have refreshed the data, cleared the scene, restarted bloom, restarted the database, restarted Neo4j Desktop but it still never shows correctly.

In the 2nd attachment, (bloom2) I show the actual syntax I used in Bloom which is the same one as above.

Would appreciate any hints on how to make virtual nodes in Bloom.

Thanks and regards,
-robert

3 REPLIES 3

It looks like you are able to create the virtual node in Bloom, but since the CG label is not categorized in the Perspective, it can't be styled or captioned properly. Bloom uses the Perspective definition for all node and relationship styling. If you can have dummy nodes with the CG label in your data, then you can have Bloom pick up the label when creating a category and apply the styling.

Hi Anurag,
Thank you for that tip. Yes, it worked. By having the Labels and properties beforehand (via dummy nodes), then the styling shows up. My dilemma is that the Virtual nodes are created programmatically, on the fly, so I wouldn't know in advance what the labels may be. It would be not efficient if my script would have to create the physical nodes first before I create the vNodes just to display the styling. It would be great if a future release of Bloom can recognize this.

Bloom and vNodes are an extremely powerful combination and it really opens up new capabilities for aggregations, groupings, clusters, etc.

Thanks again Anurag

I second this, I am working with graph data science algorithms and would like to use clustering and centrality measures without having to write the data to the graph, this can be achieved via using virtual nodes but I can't figure how to visualize it in Neo4j Bloom.