To Do

[stuppie]

To do

• Remove all properties that return strings from the properties submenu and move them into the Node Table
• Properly store lists of ID in node table
• Make sure the edge relationships from a query lookup match what you get from the properties lookup (Query->compound->get proteins is wrong)
• Change the name of edges to: (subject) <-> (relationship/predicate) <-> (object)
• Lookup needs to lookup the name also
• Figure out how to apply layout only to new nodes
• Add a linkout to wikidata for each node
• Populate an "inverse query" submenu with all properties that link to the object (inverse of the properties submenu)
• Query the number of items a node's property links to and place that in the property query submenu

Extra Features

• Add a feature that allows you to find nodes that link an arbitrary selection of nodes
• Allow customization of node style by type
• Allow addition of new node types (by specifying unique ID properties for them)

[stuppie]

Populate an "inverse query" submenu with all properties that link to the object (inverse of the properties submenu) Not sure this is the most efficient way to do this, but it works...

SELECT * WHERE {
  {
    SELECT distinct ?item ?itemLabel ?prop ?propLabel ?vals (datatype (?vals) AS ?type) WHERE {
      hint:Query hint:optimizer "None" .
      values ?item {wd:Q17861702 }
      ?item ?p ?vals .
      ?prop wikibase:directClaim ?p .
      SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
    }
  } UNION {
      hint:Query hint:optimizer "None" .
      values ?vals {wd:Q17861702 }
      ?item ?p ?vals .
      ?prop wikibase:directClaim ?p .
      SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
  }
}