related query not returning all results, and returning unwanted results

[hpfast]

in my database I have France that liesIn Western Europe:

      {
        "name": "France",
        "dataset": "cshapes",
        "id": "cshapes/1520",
        "type": "hg:Country",
        "validSince": "1946-01-01",
        "validUntil": "2012-06-30",
        "geometryIndex": 0,
        "hairs": []
        "relations": {
          "hg:sameHgConcept": []
          "hg:liesIn": [
            {
              "@id": "oecd/WEUR"
            }
          ],
          "@id": "cshapes/1520"
        },
        "@id": "cshapes/1520"
      },

but I don't get this back from a query:

http://localhost:3001/search?geometry=false&related=hg:liesIn&related.id=oecd/WEUR

In fact, I get all kinds of results I don't want, which seem to be children of children of oecd/WEUR, (like Nigeria which liesIn United Kingdom (at one point) which liesIn WEUR).

what's going on? Is there a limit on the number of returns? and is the expansion (here?) indeed finding multi-hop relations?

[hpfast]

@wires is this a known issue or might my data be malformed?

[hpfast]

are you guys also getting this behaviour on your instance?

this is actually two things: If I search from liesIn Western Europe (this is in my installation), I am:

a) getting all the children of e.g. United Kingdom
b) not getting e.g. France

I managed to update my neo4j-plugin (see histograph/neo4j-plugin#8) and now b) is solved for me, but I am still getting too many results (a).

[wires]

thanks! I have to dive into this, forgot how this works exactly. There are several (semi-artificial) limits in place that can be causing this...

[mmmenno]

When I search for https://api.histograph.io/search?q=amsterdam&geometry=false&related=hg:liesIn&related.id=http://www.gemeentegeschiedenis.nl/departement/Zz I only find the municipality Amsterdam, as expected.

[hpfast]

ok, nevermind, I was still affected by b) (not getting France when searching for liesIn=oecd/WEUR.

I recompiled the neo4j-plugin with toDepth=1 instead of 2 (here) and then I nicely get France and all the other countries in W. Europe, presumably because it's not finding all their children and thus not running into the max limit before it gets to France. EDIT: still getting lots of children-of-children back, maybe that didn't do exactly what I think it did. But this is the first line of Java I've touched ever, so I'm not sure where to go from here yet :)