Supporting a tool written in R

Dear all,

I probably made a mistake and developed a tool in R.

Obviously, when R version updates, many packages stop working just because.

So, instead of install.packages I try to use the particular version of the packages (old one). It depends on other packages, who can not be installed using an old R version. I was trying to install them from sources - after 10th step I gave up - it is not a good way to install packages for sure.

The tool is supposed to be used in clinics = no docker is allowed.

What are the other ways to provide an installation of R packages, given that many people use old R versions (say, starting from 3.2) and are not going to update it?

devtools also can not be installed using old R since it depends on other packages that have the same problem.

UPD: solution: https://github.com/imgag/megSAP/blob/master/data/install_deps_clincnv.R , but before devtools has to be installed via Ubuntu package manager (not from R).

sudo apt-get install -y r-cran-devtools

• 804 views • link

updated 3.2 years ago by Ram

• written 6.7 years ago by German.M.Demidov

Hello kuckunniwid!

We believe that this post does not fit the main topic of this site.

Not a bioinformatics question. Better ask on stackoverflow.com

For this reason we have closed your question. This allows us to keep the site focused on the topics that the community can help with.

If you disagree please tell us why in a reply below, we'll be happy to talk about it.

Cheers!

• 0 views • link

written 6.7 years ago by Nicolas Rosewick

wanted to remove the question since it does not belong - did not work =( thanks, yes, the package is solving a bioinformatics problem, but the question still does not belong to biostars, I agree

• 0 views • link

written 6.7 years ago by German.M.Demidov

1 answer

As Nicolas said, this is more appropriate for stackoverflow (you'll get answers from a more appropriate audience there). Having said that, I should point out that the earliest version of R that you can guarantee your package works with is the version you develop/test on. In general, there's no guarantee of backwards compatibility in R (or python for that matter). Requiring users to upgrade R to some moderately recent version is a fair requirement.

BTW, docker or singularity containers tend to be preferred in clinical situations.

• 0 views • link

written 6.7 years ago by Devon Ryan

In general, there's no guarantee of backwards compatibility in R (or python for that matter).

I had no bad experience with python in this sense - I use quite basic functions/packages so the high-level updates usually don't disrupt the work.

Requiring users to upgrade R to some moderately recent version is a fair requirement.

It is fair, I totally agree, but they don't. These are clinics and updating R for old versions of Ubuntu is the thing they do not do, and the updating their Ubuntu version is something impossible.

BTW, docker or singularity containers tend to be preferred in clinical situations.

Docker is not allowed due to security threats...I should try singularity, thanks =) but then I'll have to convince our IT team that it is safe...

• 0 views • link

written 6.7 years ago by German.M.Demidov