package scaffolding

[wibeasley]

As I've expressed with a few tweaks:

I'd (like to) modify some of the package's structure to more closely resemble the approach described in Hadley's R Packages book (and the tidyverse style guide). The first reason is because I think it's a good approach and has worked well for me in the past. A second reason is that this approach is documented better and more coherently than any other approach --therefore it will be easier for someone else to modify or become maintainer in the future.

But this won't be a major overhaul. Most of the important elements of the existing package follow this book's approach.

The changes come in two sets: the little guys that come before adding a test suite (issues #4 & #29) and those that come after the test suite (which includes some of the recently accumulating issues like #33, #35, & #37).

Before test suite

• Rproj

• update Roxygen

• updates to .Rbuildignore and .gitignore files.

• move daily development off the master and to the dev branch

• update person entries in DESCRIPTION file

• semantic versioning for dev versions (e.g., x.x.x.9001)

• remove date from DESCRIPTION file

• (re)connect codecov which currently reads 0%. (@pdurbin, as an admin of this GitHub org, you'll probably see my request for additional permissions granted to Codecov. I requested it only for this single public repo.)

  (click the image for the live staus)

After test suite and functional changes

• pkgdown site pkgdown site #72
• comply with some of lintr's flags. lintr flags #41
• Maybe add to CRAN's Reproducible Research view

I'm flexible if anyone has other ideas.

[pdurbin]

my request for additional permissions granted to Codecov

Yes, you may have seen that @whorka approved this request earlier today. Thanks for working on the tests! 🎉

[kuriwaki]

This tidyverse style package scaffolding seems largely complete. The issues in the create_* functions will get dealt with using the same scaffolding. Thanks @wibeasley !!