Return data slots

[damianooldoni]

This PR has the goal to solve #84 by changing the output returned by indicator_total_year and indicator_introduction_year functions. Instead of returning a plot (with or without facting) now the functions return a list with three slots:

1. plot slot containing the plot as before
2. data_top_graph, the data.frame used in the top graph, i.e. the global plot
3. data_facet_graph, the data.frame used in the facet graph, i.e. the plot below the global one, showing the faceting

As this update can break code, this is a major change and so I bumped the version to 2.0.0.

Some minor changes:

1. Improved documentation: typo, removing warnings while reading example text files
2. using pkgname::func_name for some functions (see Adapt Roxygen documentation by using pkgname::func_name everywhere #86)
3. Replace superseded dplyr functions such rename_at() or mutate_all()
4. Replace deprecated dplyr function distinct_()

[damianooldoni]

@SanderDevisscher: there are strange issues while installing dependencies on Win and macOS machines in automatic checks (see the GitHub actions in the box). However this shouldn't be a problem for you to review the PR. So, do it whenever you want it. Meanwhile, I will try to solve these check related issues.

[damianooldoni]

Temp solution: win and mac operative systems removed from automatic checks.

[SanderDevisscher]

@damianooldoni I think you currently export the wrong dataframe in the data_top_graph slot for indicator_total_year. It now contains these columns year , key, kingdom & first_observed. Can you check this ?

[damianooldoni]

Function indicator_total_year() returns the internal defined df_extended in slot data_top_graph, see L227 and L200.
This data.frame is used in L180(https://github.com/trias-project/trias/blob/return-data-slots/R/indicator_total_year.R#L180) to generate top_graph.

What I see is that I use ggplot2::geom_line(stat = "count") for the top graph, while for facet graph I group first and then I use ggplot2::geom_line(stat = "identity"). I suppose you want data_top_graph looking something like:

year	n
1900	2
1902	5
...	...

Is this what you want?

[SanderDevisscher]

Function indicator_total_year() returns the internal defined df_extended in slot data_top_graph, see L227 and L200. This data.frame is used in L180(https://github.com/trias-project/trias/blob/return-data-slots/R/indicator_total_year.R#L180) to generate top_graph.

What I see is that I use ggplot2::geom_line(stat = "count") for the top graph, while for facet graph I group first and then I use ggplot2::geom_line(stat = "identity"). I suppose you want data_top_graph looking something like:

year n
1900 2
1902 5
... ...
Is this what you want?

that is exactly what I want ;-)

[SanderDevisscher]

@damianooldoni should I wait for all checks to finish before merging or does it not matter ?

[damianooldoni]

Well, a good rule is to wait for it. Just to be 100% sure. But, if you have urgent need, merge. Still, it hsould not take so much time more.

[SanderDevisscher]

I'll wait