Treemap Visualization of Recursive Data

Treemap Visualization of Recursive Data#

The project accompanying this chapter asks you to create a visual representation of recursive data.

  • Initially you will recursively divide a flat (that is, not nested) list of integers into groups represented by rectangular areas.

  • Next you will recursively divide a nested list of integers into groups represented by rectangular areas, preserving the grouping in each nested list.

  • Finally, you will process a structure of nested dictionaries to visualize grouping of elements of different sizes.

In the final version, you will be able to process a JSON structure like this summary of files used to build this chapter:

{
    "08-01-Recursive-Data.md": 9503,
    "08-02-Nested-Dicts.md": 5310,
    "08-03-Treemap-Project.md": 971,
    "Chap08.json": 0,
    "Samples": {
        "dir_tree.py": 2042,
        "dir_tree_printer.py": 1217
    },
    "img": {
        "chap08.svg": 2763,
        "directory-tree-horizontal.png": 43381,
        "directory-tree-vertical.png": 30142
    }
}

This JSON file was produced by Samples/dir_tree.py. When processed by a sample solution to our treemap project, we can obtain a visualization showing the relative sizes of files:

Treemap of chapter file sizes

Treemaps are not effective visualizations for all kinds of quantitative data (e.g., it is not easy to determine the relative sizes of the two .png files), but they can provide a good at-a-glance comparison, especially for hierarchically structured data.
For example, we can quickly see that the bitmap graphics files (.png) are much larger than the source text files (.md), but the scalable vector graphic (.svg) image is relatively compact. Management of file space was among the original applications of the treemap visualization format, and has been used in commercial and open source applications like Disk Inventory X for MacOS, WinDirStat for Windows, and QDirStat for Linux.

Project repository#

UO-CS210/Treemap

Contents