Add common alias panel (-c) to all plotting functions

[weiji14]

Description of proposed changes

Used to advance to the selected subplot panel. See https://docs.generic-mapping-tools.org/6.1/gmt.html#c-full,
https://github.com/GenericMappingTools/gmt/blob/6.1.1/doc/rst/source/explain_-c_full.rst_, and https://docs.generic-mapping-tools.org/6.1/cookbook/options.html#selecting-subplot-panels-the-c-option. Needed for the subplot wrapper at #822 to resolve issue #20.

Questions:

1. Just confirming that we want to go with ax (as with matplotlib) rather than panel (used by GMT). See Wrap subplot using with statement #822 (comment). Ok, going with panel
2. Haven't tested to see if -c works on inset (https://docs.generic-mapping-tools.org/6.1/inset.html), i.e. putting an inset in a subplot panel. If it works, then I'll add the alias to the pygmt/src/inset.py file.

This can be merged before #822 to reduce the diff there.

Reminders

• Run make format and make check to make sure the code follows the style guide.
• Add tests for new features or tests that would have caught the bug that you're fixing.
• Add new public functions/methods/classes to doc/api/index.rst.
• Write detailed docstrings for all functions/methods.
• If adding new functionality, add an example to docstrings or tutorials.

Slash Commands

You can write slash commands (/command) in the first line of a comment to perform
specific operations. Supported slash commands are:

• /format: automatically format and lint the code
• /test-gmt-dev: run full tests on the latest GMT development version

[seisman]

• Just confirming that we want to go with ax (as with matplotlib) rather than panel (used by GMT). See #822 (comment).

I'm OK with both, although I prefer panel. When I learn matplotlib, I find it difficult to understand the differences between "axes" and "axis" (likely because I'm not a native English speaker). BTW, it seems Julia Plot.jl uses subplot (http://docs.juliaplots.org/latest/layouts/),

• Haven't tested to see if -c works on inset (https://docs.generic-mapping-tools.org/6.1/inset.html), i.e. putting an inset in a subplot panel. If it works, then I'll add the alias to the pygmt/src/inset.py file.

I think it doesn't work, to call inset, you need to have a larger map first.

[seisman]

I also find this (https://matplotlib.org/tutorials/intermediate/artists.html):

Most of you are probably familiar with the Subplot, which is just a special case of an Axes that lives on a regular rows by columns grid of Subplot instances.

If I understand it correctly, GMT's subplots also live on a regular rows by columns grid, so it should be called subplot, not axes, right?

[weiji14]

• Just confirming that we want to go with ax (as with matplotlib) rather than panel (used by GMT). See #822 (comment).

I'm OK with both, although I prefer panel. When I learn matplotlib, I find it difficult to understand the differences between "axes" and "axis" (likely because I'm not a native English speaker).

Oh, and just FYI, there's a Python library called Panel for making dashboards (i.e. interactive subplots). To be fair, I don't really get the "axes" and "axis" terminology used by matplotlib either, "axes" should be the plural of "axis" but they don't use it that way.

• Haven't tested to see if -c works on inset (https://docs.generic-mapping-tools.org/6.1/inset.html), i.e. putting an inset in a subplot panel. If it works, then I'll add the alias to the pygmt/src/inset.py file.

I think it doesn't work, to call inset, you need to have a larger map first.

Ok, will leave out inset for now then.

I also find this (https://matplotlib.org/tutorials/intermediate/artists.html):

Most of you are probably familiar with the Subplot, which is just a special case of an Axes that lives on a regular rows by columns grid of Subplot instances.

If I understand it correctly, GMT's subplots also live on a regular rows by columns grid, so it should be called subplot, not axes, right?

Not sure if it's a good idea to use 'subplot' as an alias for 'c' (if that's what you mean). We'll hit into the same problem as with def text(text=...) as in #808 (comment).

With all that said, I would still argue that using ax is the way to go, considering that both pandas.plot and xarray.plot has a kwarg called ax. It might be useful to use ax too (or perhaps confusing) to integrate with other Python libraries (c.f. #374, #366 (comment)) in the future.

[seisman]

Not sure if it's a good idea to use 'subplot' as an alias for 'c' (if that's what you mean). We'll hit into the same problem as with def text(text=...) as in #808 (comment).

Yes, I agree, subplot may be a bad idea.

With all that said, I would still argue that using ax is the way to go, considering that both pandas.plot and xarray.plot has a kwarg called ax. It might be useful to use ax too (or perhaps confusing) to integrate with other Python libraries (c.f. #374, #366 (comment)) in the future.

pandas.plot and xarray.plot have the ax kwarg because they are integrating well with matplotlib. However, pygmt can't insert a GMT figure to a matplot axes, right?

[willschlitzer]

Little late to the party, but my vote would be panel over ax. I understand the potential conflicts of panel and the familiarity of ax with other plotting packages, but I think panel makes more sense in this context, as the specific panel to be plotting on is being set, not just the axis/axes (even though they refer to the same thing).

[weiji14]

Little late to the party, but my vote would be panel over ax. I understand the potential conflicts of panel and the familiarity of ax with other plotting packages, but I think panel makes more sense in this context, as the specific panel to be plotting on is being set, not just the axis/axes (even though they refer to the same thing).

Ok, if we want to go with panel, then we should change fig.sca too (short for set current ax) in #822 to something else. fig.set_panel perhaps?

[willschlitzer]

Little late to the party, but my vote would be panel over ax. I understand the potential conflicts of panel and the familiarity of ax with other plotting packages, but I think panel makes more sense in this context, as the specific panel to be plotting on is being set, not just the axis/axes (even though they refer to the same thing).

Ok, if we want to go with panel, then we should change fig.sca too (short for set current ax) in #822 to something else. fig.set_panel perhaps?

I think set_panel works well. I particularly like that it doesn't use an acronym.

[seisman]

Little late to the party, but my vote would be panel over ax. I understand the potential conflicts of panel and the familiarity of ax with other plotting packages, but I think panel makes more sense in this context, as the specific panel to be plotting on is being set, not just the axis/axes (even though they refer to the same thing).

Ok, if we want to go with panel, then we should change fig.sca too (short for set current ax) in #822 to something else. fig.set_panel perhaps?

fig.set_panel looks good to me.

[weiji14]

Ok I've renamed ax -> panel. Will do the fig.set_panel rename later in #822, but just one comment/question.