Fix provenance of figures of several R diagnostics

[bouweandela]

Description

Several R diagnostic scripts made use of the plot_file attribute of the provenance entry of the NetCDF file to log provenance. Unfortunately, this does not work. This pull request fixes the problem by writing a separate provenance record for the plots.

To pass the lintr checks, I also had to modify some variable names to follow the lowercase character/snake case convention from the R style guide and I needed an extra dependency to get litnr to run from pre-commit.

---

Before you get started

• ☝ Create an issue to discuss what you are going to do

Checklist

It is the responsibility of the author to make sure the pull request is ready to review. The icons indicate whether the item will be subject to the 🛠 Technical or 🧪 Scientific review.

• 🛠 This pull request has a descriptive title
• 🛠 Code is written according to the code quality guidelines
• 🛠 Tests run successfully
• 🛠 The list of authors is up to date
• 🛠 All checks below this pull request were successful

New or updated recipe/diagnostic

• 🧪 Recipe runs successfully
• 🛠 Provenance information has been added

---

To help with the number of pull requests:

• 🙏 We kindly ask you to review two other open pull requests in this repository

[bouweandela]

@katjaweigel Would you be willing to review this, since you know some R?

[katjaweigel]

ok, I'll have a look.

[katjaweigel]

@esmvalbot Please run recipe_spei.yml

[esmvalbot]

Since @katjaweigel asked, ESMValBot will run recipe recipe_spei.yml as soon as possible, output will be generated here

[esmvalbot]

ESMValBot is happy to report recipe recipe_spei.yml ran OK, output has been generated here

[katjaweigel]

@bouweandela Looks fine for me, thanks! But there are several other R diagnostics, which would need the same change, so far I found:
esmvaltool/diag_scripts/pv_capacityfactor/pv_capacity_factor.R
and
esmvaltool/diag_scripts/magic_bsc/capacity_factor.R
there it could be solved about the same way.
For esmvaltool/diag_scripts/hyint/*.R it is probably somewhat more difficult.
But probably I should approve this one and then open a new PR for others?

[bouweandela]

I'll have a look, thanks for bringing those to my attention. My R is not good enough to know that there are multiple ways to add keys to a list, so I only looked for the string "plot_file" ;-)

[bouweandela]

@katjaweigel I updated all occurrences of plot_file I could find in the R diagnostics. At first glance, it looks like the provenance of hyint is implemented without using the problematic plot_file key, but maybe I'm missing something there?

[valeriupredoi]

seems like the newly requested R package has made an already very slow conda env update to go even slower, dis not gud, amigo 🦙

[katjaweigel]

@bouweandela I think hyint uses functions stored in other scripts (e.g. hyint_plot_trends.R) to do the plots, but they don't write any provenance. But I think it is possible (or already part of the data given back), to get the file name of the figure back from these functions, to write the provenance file for all figures where the other provenance file is written. But I'm not really sure about this and have to check it in more detail.

[bouweandela]

@esmvalbot Please run recipe_hyint.yml

[esmvalbot]

Since @bouweandela asked, ESMValBot will run recipe recipe_hyint.yml as soon as possible, output will be generated here

[bouweandela]

From just looking at the code of the hyint diagnostics, provenance looked OK, but let's try running the recipes to be sure.

[esmvalbot]

ESMValBot is sorry to report it failed to run recipe recipe_hyint.yml: exit is 1, output has been generated here

[valeriupredoi]

OK by me from a purely technical (environment/testing) point of view, but please merge only when the sciencey bit is reviewed 👍

[bouweandela]

There is no science bit to this pull request, it's only a bugfix for the provenance recording procedure.

[zklaus]

@bouweandela, fair enough. Could you merge the latest main to check the right tests?

[bouweandela]

Done