I cannot solve this in osql, but I can provide some information that is
hopefully helpful.
1) how to solve in gremlin
g.v("12:0").out.out.path{it.getIdentity()}
relevant gremlin documentation is:
http://gremlindocs.com/#transform/path
g.v("12:0").out('edge').out('edge').path{it.getIdentity()}
will constrain going out to edges with label 'edge'.
2) related orientdb documentation:
a) there are several shortest path functions, and functions for getting all
traversed elements
http://www.orientechnologies.com/docs/last/orientdb.wiki/SQL-Functions.html#dijkstra
b) there is a "$path" context variable, I haven't managed to get it to work
tho...
http://www.orientechnologies.com/docs/last/orientdb.wiki/SQL-Traverse.html
c) in this post had a similar question
https://groups.google.com/forum/#!searchin/orient-database/path|sort:date/orient-database/bK-T3xt0fPo/UabnbtscHEgJ
the answer was:
The TRAVERSE command is not designed to find all paths between nodes. To
> prevent going into a loop it does not traverse nodes that has already been
> handled.
> To find all paths between vertexes you can use gremlin query or write your
> own function.
> Best regards,
> Artem Orobets
> Orient Technologies
> the Company behind OrientDB
hope this is helpful!
On Thursday, March 19, 2015 at 8:43:28 AM UTC-7, Mihai Pricop wrote:
>
> I want to query a vertex in OrientDB graph database to get the second
> level vertices, keeping the in between vertex id.
>
> The relationship would be represented as such:
>
> User <--edge--> User <--edge--> User
> (1) (2) (3)
>
> For User 1, I want to get all User 3 (distance of 2 edges), but I want to
> keep as a result all User 2 rids. The direction isn't important (will use
> both() method to traverse)
>
> An example:
>
> #10:1 <--edge--> #10:2
> #10:1 <--edge--> #10:3
> #10:2 <--edge--> #10:4
> #10:2 <--edge--> #10:5
> #10:3 <--edge--> #10:6
>
> For the example above, the query for #10:1 should return: (don't know how
> to represent this any better, but the query should return the object from
> the class User AND the rid from the intermediate User)
>
> User(#10:4), #10:2
> User(#10:5), #10:2
> User(#10:6), #10:3
>
> The query to find the vertices from the second level would look like: select
> expend(both('edge').both('edge')) from #10:2 (of course the query can be
> optimized with set function and other conditions in case of inter-relations
> of the output users). To this output, I would like to see an additional
> field (property / column, whatever you want to call it) where I can see the
> in between vertex of this kind of relationship.
>
> Hope somebody can help me with this, tell me if you need more information,
> or if you did not understand something. Thank you.
>
--
---
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.
