Gantt

README

Gantt is a python class to produce, well, Gantt charts. The charts are kept (very) simple, using a discreet time scale, unicolor bars and optional milesstones.

Background

Gantt charts are commonly used in project management. I wanted to make one myself and LibreProject was way too involved for the project in question. Hence I used OpenOffice Calc and made a plain horizontal bar chart. That took little time but was ok. Then I needed to add milestones (basically just a marker) to the chart. I completely failed doing this in OpenOffice (and still don't know how to do that ...).

Long story short: I don't know how to excel.

Basic usage

from gantt import Gantt
g = Gantt('./sample.json')
g.render()
g.show()                # or save w/ g.save('foo.png')

Or simply call runner.py

Data structure

All data is provided as a JSON structure that has to contain:

• a key packages containing an array of package definitions with label, end and start. Start and end values are discreet.
• a title string (may contain TeX, escaped)

{
  "packages": [
    { "label" : "WP 1-1",
      "start": 0,
      "end": 2,
      "milestones" : [2],
      "legend": "worker one"
    },
    { "label" : "WP 1-2",
      "start": 2,
      "end": 4,
      "milestones" : [3, 4]
    }
  ],
  "title" : " Sample GANTT for \\textbf{myProject}",
  "xlabel" : "time (weeks)",
  "xticks" : [2,4,6,8,10,12]
}

The milestones, colors and legend entry are optional as are the label for the x-axis and the definition of the tickmarks. The title may contain TeX, but make sure your system supports it. For

See sample.json for the data used to produce the image below.

Installation

The requirements are rather limited and can installed from the requirements file. I recommend using a virtual environment for that:

# virtualenv setup, recommended
python3 -m venv .venv
source .venv/bin/activate
# actual install for requirements
pip install -r requirements.txt
# to install dev dependencies too run
# pip install -r requirements-dev.txt

ToDo

• nicer data structure (JSON) :white_check_mark:
• dedicated class for packages :white_check_mark:
• dynamic TeX support :white_check_mark:
• add parameter object/dict for more control over colors etc :construction:

Screenshot

See sample.json for definition.

Supported Versions

Support is a bit much to say. The initial version was developed on 3.6 and worked just fine. With github-actions I can only test 3.7+ and this works fine (up to 3.12 currently). But basically, if you get numpy/matplotlib to run, all should be fine.