ReSubPlot 1.0.0

ReSubPlot (Replot figures to SubPlots): Plot one or many existing figure objects into a subplots() and manage labels, titles, legends, alignments, etc.

ReSubPlot

ReSubPlot (Replot figures to SubPlots): Plot one or many existing figure objects into a subplots() and manage labels, titles, legends, alignments, etc. On top of this, the package can also be used to extract the legend from a Python Figure object and plot the figure and the label in two independent PDF files.

Installation

Use the package manager pip to install ReSubPlot (find the PyPi page here: https://pypi.org/project/ReSubPlot/).

pip install ReSubPlot

Install all the required packages (dependencies) from the requirements.txt file.

pip install -r requirements.txt

Place requirements.txt in the directory where you plan to run the command. If the file is in a different directory, specify its path, for example, path/to/requirements.txt.

Usage

Command line

The package can be used directly on the command line with the built-in CLI, by calling one of master or legend

Master plot

ReSubPlot master -f config_master.toml

When using the package this way, all the information is stored in the TOML file config_master.toml. It should respect the following structure:

# Configuration file for ReSubPlot.master_plot_from_toml

[pickle]
# path to the pickle where the list of list of Figure objects is saved
path = 'list_figs.pkl'

[sites]
# list of sites (string), there will be a column per site in the subplot result
list = ['Moosehide', 'Rockcreek', 'Sunnydale']

[pad]
# float that controls the padding between subplots, we suggest 0.03
pad  = 0.03

[save]
# Whether or not to save the master plot to a pdf
# If True: pass the name of the created pdf (e.g. path = 'master_plot')
# If False: pass None (path = None)
path = 'master_plot'

Importantly, the user needs to have saved all figures they want to have plotted on the master plot in a specific format in a pickle. The format of the pickle is a list of list of Python Figure objects. The list of lists should be in a (m * n) shape, e.g.

[[fig_1_1, ..., fig_1_n],
 [fig_2_1, ..., fig_2_n],
 ...
 [fig_m_1, ..., fig_m_n]]

The list of sites parameter should be a list of n string entries, one for each column of the master plot.

The output is a master plot with n columns (one per site), and m rows (one per type of plot).

Legend on and off

The second functionality of ReSubPlot is to strip the legend off of a Python figure. This can also be done from the command line, by calling

ReSubPlot legend -f config_legend.toml

When using the package this way, all the information is stored in the TOML file config_master.toml. It should respect the following structure:

# Configuration file for ReSubPlot.isolate_legend_from_toml

[pickle]
# path to the pickle where the Figure objects is saved
path = 'figure.pkl'


[save]
# Whether or not to save the master plot to a pdf
# If True: pass the name of the created pdf (e.g. path = 'master_plot')
# If False: pass None (path = None)
path = 'my_cool_figure'

This is straightforward, the pickle should be a figure object. It will then create 3 PDF plots:

• a plot with the legend (the original plot)
• the same plot without the legend
• a figure that only shows the legend and nothing else

Direct use of functions in Python

Master plot

The CLI master call corresponds to the function master_plot_from_toml(), hence, the equivalent call to what was described earlier is

from ReSubPlot.master_toml import master_plot_from_toml
master_plot_from_toml(config_toml_path)

Now, if the user has a list of list of figures (or matrix, in the shape (m * n)) called mat_fig, and a list of sites of length n (each entry is a string) called list_sites, then, the function master_plot() can be called directly in the following way

from ReSubPlot.plotting_funcs import master_plot
fig, tens = master_plot(mat_fig, list_sites)

The function will return:

• fig: the master plot figure, a Python figure object
• tens: a list of list of list... the outermost shape is (m * n), and each element is itself a list of integers which label the axis for a particular subplot (i,j). The (i,j) element of tens can be [17,18] for instance, and this means that the 17th and 18th axes of the figure are found in subplot (i,j) (in the case of a twinx() axis for instance).

Legend on and off

The CLI legend call corresponds to the function isolate_legend_from_toml(), hence, the equivalent call to what was described earlier is

from ReSubPlot.master_toml import isolate_legend_from_toml
isolate_legend_from_toml(config_toml_path)

Now, if the user has a figure (it is here called fig_no_legend) object in their python notebook, they can directly apply the function to it

from ReSubPlot.legends import isolate_legend
fig_with_legend, fig_no_legend, legend = isolate_legend(fig_no_legend, plot_name)

and it will return three objects:

• fig_with_legend is the original figure with the legend (we called it as fig_no_legend)
• fig_no_legend is now the original figure without the legend
• legend is the matplotlib legend object

Furthermore, 3 PDFs have been created.

Examples

The user can find some inspiration on how to use ReSubPlot by looking at the examples provided.

License

GNU GPLv3