Docs

Adjacency List

The Adjacency List format is so-named because it contains all of your data in one Excel sheet — it looks like a single huge list. You write an entity in each row; the columns contain the attributes of the entity.

Picture of annotated adjacency list

Required and highly-recommended columns

Mandatory columns These are the bare miniumum columns you need to have so that Rhumbl knows what entities exist in your data

  • id Required
  • type Required

Highly-recommended columns

  • name because mouseover node labels default to using name and non-nodal entities (aka parents of nodes) use only the name attribute. If you don't have this field, you won't be able to see any labels on non-nodal entities.

  • short name because node labels default to using short name.
  • rel::parent::[name of relationship] Every entity except the root entity must have one and only one parent. If you have an entity without a parent, this will make the layout of your map look strange.

Relationship columns

Relationship columns Relationship columns must be named in these patterns:

  • Directed relationships must be written as rel::dir::[name of relationship], e.g. rel::dir::studied at
  • Undirected relationships must be written as rel::undir::[name of relationship], e.g. rel::dir::is similar to
  • Parent relationships must be written as rel::parent::[name of relationship], e.g. rel::dir::belongs to
  • You can name your relationships however you would like, but they must not contain colons (:).
  • Entries in relationship columns are semi-colon (;) separated, for example: x; y; z

Other Attributes

Other attribute columns You can put any attribute you want. However, Rhumbl recognizes these columns as "special" columns:

  • Links: If there are links in the columns, Rhumbl will automatically recognize these as links and will display them as clickable.
  • Node coloring The column names fill:: and stroke:: is reserved for individual node coloring. Read more on node coloring here.
  • Node sizing The column names radius:: is reserved for individual node sizing. Read more on node sizing here.
  • Node or not? By default, Rhumbl will guess which entities are nodes or groups. However, you can override this behavior by specifying a column called is_node::, and filling in TRUE/FALSE.