Open royfrancis opened 6 years ago
Personally I think it is worth the extra effort to have a version without the executables (so purely have the plotting capabilities). I am detecting dynamically whether a library is available and if not available, download it automatically from biocLite or CRAN. Currently I have hardcoded to detect this library, as devtools does not automatically handle dependency management.
I get it might get more complex maintaining two builds, but is there a true usecase to run CLUMPP etc. directly from the R code? In my opininion, this package should be focussed on handling the output files only.
Sounds reasonable. I could create a version without executables for CRAN.
This is a question I get asked often and I am offering an explanation here.
I've been asked a few times about why this package is not on CRAN. And the answer is simply that the package contains windows and linux executables (clumpp/distruct) which is not allowed on CRAN under submission guidelines.
One possibility (the easy option) would be to have a version without these executables for CRAN. Another option (more work) is to re-write/implement the CLUMPP code in R. Personally, I don't think it's too much extra effort to install from GitHub anyway.
If you have any thoughts on this, feel free to comment here.