Do you intend to require R >= 3.3?

alexpghayes / safepredict

Consistent prediction following tidymodels principles

https://alexpghayes.github.io/safepredict/

Other

17 stars 1 forks source link

Do you intend to require R >= 3.3? #14

Open maurolepore opened 5 years ago

maurolepore commented 5 years ago

Just wondering safepredict really requires R >= 3.3. Maybe check if the build passes on travis using R 3.2 and R 3.1 (the goal of tidyverse packages)?

This discussion might be helpul https://community.rstudio.com/t/determining-which-version-of-r-to-depend-on/4396/10

alexpghayes commented 5 years ago

I haven't thought about this much to be honest. I probably will based on experiences with broom. Will consider more fully when I next sit down to develop safepredict more seriously again.

maurolepore commented 5 years ago

👍 The issue comes up when some developer depends on your package. Tidyverse packages aim to work with R >= 3.1 and I guess many developers may try to aim for that too. For example, I'm aiming for R >= 3.2, so I wanted to import or suggest safepredict I will get errors during check.