This program helps with the creation of visualizations of todo lists. It works as a simple filter between taskwarrior and graphviz.
The code is now on https://pypi.org and can be installed via
python3 -m pip install task2dot
if you prefer to work with installed software, compared to interpreted source code. Otherwise execution is performed via python3 task2dot/task2dot.py.
At the command line write
task export | task2dot | dot -Tsvg > test.svg
With this, all todo items that you have ever created are fed into
task2dot
. Without any arguments, it just translates the export
into a format that is suitable for graphviz
whose output is saved
in the file test.svg
. We can expect this to result in an insanely
messy network graphics.
For clarity, I will not show the dot command and the output redirection into a file in any the following code snippets. Note that those have to be added for obtaining useful visualizations.
When exporting data from taskwarrior
one has to explicitely state
that one only wants to export pending tasks:
task status:pending export | task2dot
Tasks, tags and projects are nodes as well as user defined attributes!
User defined attributes are supported if the task configuration file is
~/.taskrc
or can be found in the environment variable $TASKRC
.
Edges are task dependencies and all connections from tasks to their project, all of their tags as well as the values of their user defined attributes.
Note that if all node with all of its connections are shown in a graph with a sufficiently many tasks, the resulting graph will become overwhelmingly messy. Therefore, the following exclusion mechanisms are available.
To exclude a specific node write
task status:pending export | task2dot -node
Then there will be no node with the content 'node' in the output graph.
So why is this useful? If you try to implement Kanban or something
similar you are very likely to have a certain tag or a user
defined attribute, like todo
far too often for having it in a
graph visualization. Almost all tasks would be connected to it via
edges, which is useless. Also, if you export taskwarrior data,
which is filtered with a specific tag
will cause the resulting
graph having a lot of connections to that tag. So the following
visualization would be useful:
task status:pending +work | task2dot -work
A specific type of node can be excluded by using two hyphens. For example, not showing any project nodes looks like this:
task status:pending export | task2dot --project
Or not showing any tags:
task status:pending export | task2dot --tags
In my workflow, paths and emails ids are attached to tasks, so I need to write
task export | task2dot --path --email
When converting tasks with annotations to a graph, by default the
annotations are concatenated and added to the task description. To
turn of annotations altogether, you can use task2dot --annotation
.
It is also possible to exclude certain connections also by using double hyphen. Let's get rid of all connections from tasks to tags:
task export | task2dot --task-tags
It is possible to add additional edges than what taskwarrior exports directly. One could for example add edges between projects and tags because they are connected by tasks that have both. If one then removes the tasks one can look at a graph that shows us which 'actions' are needed for certain projects, if the tags represent 'actions'. Similarly to edge exclusion, we use ++node1-node2 to add additional edges.
task export | task2dot ++tags-project --task
To adjust the colors to your needs, you will need to provide a file
~/.task2dotrc.json
in your home directory. It contains a json string
with the following format:
{
"nodes": {
"tags": {
"shape": "egg",
"fillcolor": "#000000",
"penwidth": "2pt",
"style": "filled",
"fontcolor": "white",
"color": "white"
},
"outcome": {
"shape": "egg",
"fillcolor": "#000000",
"penwidth": "2pt",
"style": "filled",
"fontcolor": "white",
"color": "white"
},
"people": {
"shape": "octagon",
"style": "filled",
"fillcolor": "#dd9900",
"color": "white"
},
"project": {
"shape": "diamond",
"penwidth": "2pt",
"color": "#22ff22",
"fontcolor": "#ffffff",
"style": "filled",
"fillcolor": "#115500"
},
"task": {
"shape": "box",
"color": "white",
"fontcolor": "white",
"style": "rounded,filled",
"fillcolor": "#222299",
"fontsize": "16"
},
"default": {
"shape": "box",
"color": "white",
"fontcolor": "white",
"style": "rounded,filled",
"fillcolor": "#222299",
"fontsize": "16"
},
"annotation": {
"shape": "note",
"color": "white",
"fontcolor": "white",
"style": "filled",
"fillcolor": "#111155"
},
"state": {
"shape": "circle",
"color": "white",
"fontcolor": "white"
}
},
"edges": {
"default": {
"color": "white"
},
"task-tags": {
"color": "white",
"style": "dotted",
"penwidth": "5pt",
"arrowsize": "0.1"
},
"project-project": {
"color": "#22ff22",
"penwidth": "2pt",
"arrowhead": "odot",
"arrowtail": "odot"
},
"task-project": {
"color": "#aa2211"
}
},
"graph": {
"layout": "fdp",
"splines": "ortho",
"size": "30,30",
"bgcolor": "#111519"
}
}
The parameters that are set in the configuration directly correspond to the graphviz (see graphviz) settings available for nodes, edges and the graph as a whole. The example configuration shown here corresponds to the color scheme used for the example graphics referred to from within this README file.
If you want to provide color settings for your own user defined attribute, just
add a property to the nodes
property with the name of your attribute and supply the parameters that shall deviate from the default
configuration.