vizlint 0.2.0

Lint Python visualizations for common misleading patterns.

vizlint

Lint Python charts for common misleading or low-quality patterns.

Install

pip install vizlint
pip install "vizlint[mpl]"

Quickstart (Matplotlib)

import matplotlib.pyplot as plt
import vizlint as vl

fig, ax = plt.subplots()
ax.bar(["A", "B"], [101, 103])
ax.set_ylim(100, 104)

report = vl.lint(fig)
print(report.summary())

Example output

For the truncated bar chart above, report.summary() might look like:

vizlint report:
- WARN [bar_zero_baseline] Bar chart y-axis does not include zero, which can exaggerate differences between bars. Hint: Consider starting the y-axis at zero or switch charts if you need to zoom in.

Depending on your chart, you might also see warnings such as axis_labels_missing or axis_range_overexpanded alongside other issues.

Current checks

• bar_zero_baseline (warning): Flags bar charts whose y-axis range does not include zero, which can exaggerate differences between bars.
• axis_labels_missing (warning): Warns when the x-axis, y-axis, or both are unlabeled, making the chart harder to interpret.
• axis_range_overexpanded (warning): Warns when the y-axis span is far larger than the data range and the data floats in the middle, which can visually minimize changes.

CLI

vizlint path/to/script_that_makes_plots.py

By default the CLI uses matplotlib's Agg backend so it runs cleanly in headless CI. If you want to silence specific checks without editing code, pass --disable with the rule's function name:

vizlint my_script.py --disable axis_labels_missing --disable bar_zero_baseline