Data.GraphViz.Algorithms

Defines various algorithms for use on DotRepr graphs. These are typically re-implementations of behaviour found in existing Graphviz tools but without the I/O requirement.

Canonicalisation Options

For simplicity, many algorithms end up using the canonicalisation functions to create the new DotGraph. CanonicaliseOptions allows you to configure how the output is generated.

data CanonicaliseOptions

Constructors

COpts

Fields

edgesInClusters :: Bool

Place edges in the clusters where their nodes are rather than in the top-level graph.

groupAttributes :: Bool

Put common Attributes as top-level GlobalAttributes.

Instances

Eq CanonicaliseOptions

Ord CanonicaliseOptions

Read CanonicaliseOptions

Show CanonicaliseOptions

defaultCanonOptions :: CanonicaliseOptions

dotLikeOptions :: CanonicaliseOptions

Options that are more like how dot -Tcanon works.

Canonicalisation

canonicalise :: DotRepr dg n => dg n -> DotGraph n

Implements similar functionality to dot -Tcanon. However, this method requires no IO and doesn't care about image locations, etc.

This function will create a single explicit definition for every node in the original graph and place it in the appropriate position in the cluster hierarchy. All edges are found in the deepest cluster that contains both nodes. Currently node and edge attributes are not grouped into global ones.

canonicaliseOptions :: DotRepr dg n => CanonicaliseOptions -> dg n -> DotGraph n

As with canonicalise, but allow custom CanonicaliseOptions.

Dealing with transitive edges

transitiveReduction :: DotRepr dg n => dg n -> DotGraph n

transitiveReductionOptions :: DotRepr dg n => CanonicaliseOptions -> dg n -> DotGraph n