Small issues with vignette raised by Pat in email

NEON-biodiversity / Ostats

O-statistics (community pairwise niche overlap statistics)

https://neon-biodiversity.github.io/Ostats/

Other

7 stars 2 forks source link

Small issues with vignette raised by Pat in email #18

Closed qdread closed 3 years ago

qdread commented 3 years ago

@billspat I just took the liberty of pasting some stuff you wrote in your email into some issues on here. Thanks again for the feedback.

While the " stringsAsFactors" option is only needed for R version < 4.0 some folks are old computers that can't upgrade. If your package doesn't require R 4, then I would leave this in.
~~I noticed in the vignette that a new column "log_weight" is created when the data set already has a "logweight" column - are the different bases? Is the new column necessary?~~
I did find a few other issues in the example code (Roxygen mainly) when I ran the 'check package' in Rstudio but you are probably aware of those.

qdread commented 3 years ago

I fixed a few of the issues that come up on package check (e16f4b4) but mainly those involved with the source code, not the vignette example code yet. So that still needs to be done

qdread commented 3 years ago

Also yes, the logweight column in the original dataset was log base e. The newly calculated column has log base 10. It won't change the results at all but the log base 10 is more interpretable I think. So no need to change that.

qdread commented 3 years ago

All right, I fixed the vignette so that it no longer depends on dplyr! It still depends on ggplot2, so I just went ahead and added ggplot2 to the import dependencies of the entire package (since it's needed for Ostats_plot() anyway). I also had to add gridExtra to the dependencies b/c of the grid.arrange function we call. See commit 7b2580d.

Basically the only thing left here are the issues which we will see anyway when we run devtools::check()

qdread commented 3 years ago

I cleaned up the vast majority of the issues that appear when you run devtools::check() so I'm closing this issue. See commit c963cf5. (The outstanding errors from R CMD CHECK should be resolved when other issues are resolved.)