Note

Click here to download the full example code or to run this example in your browser via Binder

Identifying function names in a script¶

Work in progress

This feature has not been fully ported to mkdocs, see #10

This demonstrates how Mkdocs-Gallery identifies function names to figure out which functions are called in the script and to which module do they belong.

# Code source: Óscar Nájera
# License: BSD 3 clause

import os  # noqa, analysis:ignore
import matplotlib.pyplot as plt
from mkdocs_gallery.backreferences import identify_names
from mkdocs_gallery.py_source_parser import split_code_and_text_blocks

filename = os.__file__.replace('.pyc', '.py')
_, script_blocks = split_code_and_text_blocks(filename)
names = identify_names(script_blocks)
figheight = len(names) + .5

fontsize = 12.5

Mkdocs-Gallery examines both the executed code itself, as well as the documentation blocks (such as this one, or the top-level one), to find backreferences. This means that by writing numpy.sin and numpy.exp here, a backreference will be created even though they are not explicitly used in the code. This is useful in particular when functions return classes -- if you add them to the documented blocks of examples that use them, they will be shown in the backreferences.

Also note that global variables of the script have intersphinx references added to them automatically (e.g., fig and fig.text below).

fig = plt.figure(figsize=(7.5, 8))

for i, (name, obj) in enumerate(names.items()):
    fig.text(0.55, (float(len(names)) - 0.5 - i) / figheight,
             name,
             ha="right",
             size=fontsize,
             transform=fig.transFigure,
             bbox=dict(boxstyle='square', fc="w", ec="k"))
    fig.text(0.6, (float(len(names)) - 0.5 - i) / figheight,
             obj[0]["module"],
             ha="left",
             size=fontsize,
             transform=fig.transFigure,
             bbox=dict(boxstyle='larrow,pad=0.1', fc="w", ec="k"))

plt.draw()