AbstractDataGraph interface redesign

[jack-dunham]

This PR introduces a change to the DataGraphs interface. The following functions must now be overloaded for a graph to behave like a data graph:

underlying_graph(::AbstractDataGraph) = not_implemented()

has_vertex_data(::AbstractDataGraph, vertex) = not_implemented()
has_edge_data(::AbstractDataGraph, edge) = not_implemented()

get_vertex_data(::AbstractDataGraph, vertex) = not_implemented()
get_edge_data(::AbstractDataGraph, edge) = not_implemented()

set_vertex_data!(::AbstractDataGraph, data, vertex) = not_implemented()
set_edge_data!(::AbstractDataGraph, data, edge) = not_implemented()

unset_vertex_data!(::AbstractDataGraph, vertex) = not_implemented()
unset_edge_data!(::AbstractDataGraph, edge) = not_implemented()

The functions getindex , setindex! get! etc are all defined in terms of those. It is a bit more verbose than just having a function vertex_data etc that returns a mutable dictionary, but I think it is ultimately more generic. The return value of vertex_data etc is now constructed on the fly, so doing setindex!(vertex_data(g), val, vertex) will not modify the underlying graph.

To rename vertices, one must also overload:

GraphsExtensions.rename_vertices(f, ::AbstractDataGraph) = not_implemented()

The benefit of this change is that one no longer require the data on the vertices and edges to be stored as a mutable field; it can be generated on the fly. This then results in constant behavior when defining views of vertex/edge data. One can still use a mutable dictionary to store the data for a given concrete AbstractDataGraph subtype.

Other Changes

• AbstractDataGraph now subtypes AbstractNamedGraph. This greatly reduces required method overloads.

To do

• DataGraph allows an AbstractSimpleGraph as its underlying graph. This will result in mismatched vertices between the underlying graph vertices and the vertex data dictionary if one removes or adds a vertex from the data graph.
• Bump version number
• VertexDataView etc

[jack-dunham]

Closing as superseded by #55.