Additional customisation options

[eshwen]

Some small, additional customisation options could be useful (or used as default):

• Option for hatched error bar on background MC instead of filled (in main plot)
• Alter the range of the data/MC sub-plot (e.g., 0.5 - 1.5)

[eshwen]

Chopping off the additional whitespace on the x-axis would be nice by default (i.e., when x-limits are not set explicitly). Examples are shown below.

Current default x-limits (the HT range in the dataframes was 200-2000):

Preferred default x-limits:

The addition of overflow bins would also be nice, perhaps as an option. On the x-axis, they could be given a width equal to the other bins if all the bins are the same width, or if it's variable width it can adopt the bin width of the previous bin or something

[benkrikler]

Good suggestions. Happy to have PRs to address them!

For the last point, the matplotlib function that we need is: https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.margins.html

[eshwen]

Good suggestions. Happy to have PRs to address them!

For the last point, the matplotlib function that we need is: https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.margins.html

Could we get this implemented either by default or a config entry or something? Since it's matplotlib > 3 (so only applies to Python 3), I'm not sure how straightforward it would be

[benkrikler]

I think I failed to mention this properly in the last PR I added, but the newest version of plotter should implement this already. If instead of passing a 2-tuple of floats to the limits section for an axis, you give a float followed by a percent symbol, that float will be used to set the margins of the axis. So I think giving 0% might achieve what you want. E.g.

limits: 
    y: [1e-2, 1e7]
    x: 0%