wag | word profile generator out of diverse data resources | Natural Language Processing library

You can use map_dfc rather than map_df, as it will bind the columns.

We may use across - loop over the numeric column, get the rank and create new column names by adding a suffix in .names

The problem is, that you are adding a new scene.input.on('pointermove',...) event-handler in the pet function, on each click.

I would only change the code abit (look below), this should prevent generating too many particles and too many event-handlers (too many event-handlers could harm performance, so be careful, when adding them).

Here is how I would modify the Code:
(I stripped out and added some stuff to make the demo, easier to understand and shorter. And also that the snippet can be executed, without errors/warnings)

The main changes are marked and explained in the code, with comments

You don't need a function for this. We can just use a list and splice it into mutate with !!!:

You can't make a string negative. -"mpg" causes an error. (- is a unary operator, as it only takes one argument. -x tries to make x negative, but x must be numeric or you will get this error. There are a handful of functions that allow you to use - on a column name like base::subset and dplyr::select, but they only work because of special handling under the hood.)

If you want to remove the column from the original data, assigning NULL works well:

This is far from a perfect answer and is based on trial and error while poking the ggplot object... but it gets you he labels from layout to data:

This works; does it solve your problem?

You could add footnote_as_chunk = T:

One option to achieve your desired result would be to make use of the .data and the .env pronouns from rlang:

• .data will pick columns from the data
• .env will pick variables from the environment

See e.g. this blog post