Read on Twitter
this #rstats workshop was so much fun! participants were global 
, some joining at 4am or 10pm 
. everyone was so engaged and either asked great questions or contributed answers with me 
. some of the 
dev Q&A from the workshop in 
please feel free to add your 2Âą 
https://twitter.com/PipingHotData/status/1320381922363703298
, some joining at 4am or 10pm . everyone was so engaged and either asked great questions or contributed answers with me . some of the dev Q&A from the workshop in please feel free to add your 2Âą https://twitter.com/PipingHotData/status/1320381922363703298
1/
Q: When you execute usethis::create_package("C:/Users/Shannon.Pileggi/Desktop/ralph") Has the ralph folder already been previously created?
A: No, the ralph folder is created on the fly with the usethis::create_package function.
Q: When you execute usethis::create_package("C:/Users/Shannon.Pileggi/Desktop/ralph") Has the ralph folder already been previously created?
A: No, the ralph folder is created on the fly with the usethis::create_package function.
2/
Q: Are there other ways to create an R package outside of usethis::create_package?
A: File -> New Project -> New Directory -> R Package. You can then navigate to the directory where you want to save the package with âCreate project as a subdirectory of.â
Q: Are there other ways to create an R package outside of usethis::create_package?
A: File -> New Project -> New Directory -> R Package. You can then navigate to the directory where you want to save the package with âCreate project as a subdirectory of.â
3/
Q: What are considerations for naming packages?
A: See this SO post with more information and links. âThis should contain only (ASCII) letters, numbers and dot, have at least two characters and start with a letter and not end in a dot.â https://stackoverflow.com/questions/24201568/whats-a-good-r-package-name/24201910
Q: What are considerations for naming packages?
A: See this SO post with more information and links. âThis should contain only (ASCII) letters, numbers and dot, have at least two characters and start with a letter and not end in a dot.â https://stackoverflow.com/questions/24201568/whats-a-good-r-package-name/24201910
4/
Q: A comment from the workshop participants on licensingâŠ
A: If creating a package for your employer, check with them about their preferred licensing. You may need to list the company, and not a specific person, as the license holder.
Q: A comment from the workshop participants on licensingâŠ
A: If creating a package for your employer, check with them about their preferred licensing. You may need to list the company, and not a specific person, as the license holder.
5/Q:
You mentioned making sure examples actually runs bc of how devtools checks. Is it ok to use a sample dataset from the tidyverse without specifically including a line like library(tidyverse)? (As in, will it just work, or do we need to specify dependencies somewhere?)
You mentioned making sure examples actually runs bc of how devtools checks. Is it ok to use a sample dataset from the tidyverse without specifically including a line like library(tidyverse)? (As in, will it just work, or do we need to specify dependencies somewhere?)
5/A: For examples I would recommend using data from base R, toy data generated in your documentation, or data that accompanies your package (not covered in this workshop). For example, if we change our example to
compute_corr(data = diamonds, var1 = carat, var2 = price)
compute_corr(data = diamonds, var1 = carat, var2 = price)
5/A: We get the following error in devtools::check(), which indicates that we need to tell the package where the diamonds data set comes from, and appropriately declare dependencies.
6/
Q: How can you import packages that are not on CRAN? For example, I want to use a package from bioconductor or github.
A: See this thread! https://twitter.com/skyetetra/status/1329903827152080898?s=19
Q: How can you import packages that are not on CRAN? For example, I want to use a package from bioconductor or github.
A: See this thread! https://twitter.com/skyetetra/status/1329903827152080898?s=19
7/
Q: Can I import tidyverse as a whole instead of import broom, dplyr, etc. separately?
A: This is not recommended as tidyverse is not a proper package, it is a collection of packages. https://www.tidyverse.org/blog/2018/06/tidyverse-not-for-packages/
Q: Can I import tidyverse as a whole instead of import broom, dplyr, etc. separately?
A: This is not recommended as tidyverse is not a proper package, it is a collection of packages. https://www.tidyverse.org/blog/2018/06/tidyverse-not-for-packages/
8/Q: How is usethis::use_package() different from just putting a package in the Imports section of the DESCRIPTION file?
8/A: It isnât. From documentation: âuse_package() adds a CRAN package dependency to DESCRIPTION and offers a little advice about how to best use it.â It adds the name of your package in the DESCRIPTION file under Imports with proper indentation, and prints notes to the console.
9/
Q: Can we see the package ralph in the packages section?
A: Yes, after installation with either devtools::install() or devtools::install_github(), the ralph package should appear. However, you may need to restart R for it to refresh (Session -> Restart R).
Q: Can we see the package ralph in the packages section?
A: Yes, after installation with either devtools::install() or devtools::install_github(), the ralph package should appear. However, you may need to restart R for it to refresh (Session -> Restart R).
10/Q
Q: How do we add a new function in this package? Should we type in the bottom of the first one?
Q: How do we add a new function in this package? Should we type in the bottom of the first one?
10/A: usethis::use_r() will create a new R script for another function. Depending on your objectives, you can also include multiple functions with multiple Roxygen skeletons in a single script (see link below for an example). https://github.com/tidyverse/tidyr/blob/master/R/expand.R
10/A: I would defer to more advanced package developers to assess when that might be preferred. 
11/ big thanks to the awesome @RLadiesPhilly!!! 
