Adds a BoxPlot plot type

[Dando18]

No description provided.

[Copilot]

Pull Request Overview

This PR adds a new BoxPlot class to the pssgplot library, providing functionality for creating box plot visualizations with customization options including hatching patterns, labels, legends, and axis formatting. The implementation includes an animation method stub that is currently not implemented.

Key Changes:

• New BoxPlot class with comprehensive plotting options and seaborn integration
• Support for hatching patterns, error bars, labels, and legend customization
• Animation method placeholder (not yet implemented)

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
pssgplot/boxplot.py	Introduces the complete BoxPlot class with plotting and animation methods
pssgplot/init.py	Exports the new BoxPlot class for public API access

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

The label positioning logic uses get_width() and get_height() which are appropriate for bar plots, but box plots use different patch geometries. For box plots, you should use the patch's bounding box or specific box plot coordinates to position labels correctly.

Suggested change

	for p in self.ax.patches:
	if p.get_width() <= 0:
	continue
	self.ax.annotate(
	label_fmt.format(p.get_height()),
	(p.get_x() + p.get_width() / 2., p.get_height()),
	ha='center',
	va='center',
	xytext=(0, 10),
	textcoords='offset points',
	fontsize=label_fontsize
	)
	# Find median lines and annotate at their positions
	medians = [line for line in self.ax.lines if line.get_linestyle() == '-']
	for median in medians:
	xdata = median.get_xdata()
	ydata = median.get_ydata()
	# The median line is horizontal, so ydata[0] == ydata[1]
	x = (xdata[0] + xdata[1]) / 2.
	y = ydata[0]
	self.ax.annotate(
	label_fmt.format(y),
	(x, y),
	ha='center',
	va='center',
	xytext=(0, 10),
	textcoords='offset points',
	fontsize=label_fontsize
	)

[Copilot]

This code is unreachable because the function raises NotImplementedError at line 156 before this validation. Either remove the unreachable code or move the NotImplementedError to after the validation checks.

Suggested change

	if not self.ax or not self.fig:
	raise ValueError("Plot must be created before animating.")
	self.ax.set_visible(False)
	if by == 'column':
	frames = len(self.data[self.x].unique())
	group_size = 1
	elif by == 'hue' and 'hue' in self.kwargs:
	frames = len(self.data[self.kwargs['hue']].unique())
	group_size = len(self.data[self.x].unique())
	else:
	raise ValueError("Invalid value for 'by'. Must be 'column' or 'hue' with 'hue' in kwargs.")
	# check if save_dir exists; if not, create it; if it does, error if it's not empty
	if not os.path.exists(save_dir):
	os.makedirs(save_dir)
	#elif os.listdir(save_dir):
	# raise ValueError(f"Directory {save_dir} is not empty.")
	# save each frame as frame_format; and also save a gif
	for i in range(frames):
	self.ax.set_visible(True)
	for j, bar in enumerate(self.ax.patches):
	if j+1 <= (i+1)*group_size:
	#if j <= i:
	bar.set_visible(True)
	else:
	bar.set_visible(False)
	self.fig.savefig(os.path.join(save_dir, f"frame_{i}.{frame_format}"), **kwargs)

[Copilot]

Accessing self.ax.get_legend() without checking if a legend exists will raise an AttributeError if no legend was created. This code path is only reached when hatch=True, but the legend is only created when legend=True (line 98). Add a check: if 'hue' in kwargs and self.ax.get_legend() is not None: