#R folk. I'm having another go at running R in linux. I'm sure I'm doing something wrong loading CRAN modules. It appears to be compiling them, and intermittently failing with requests I apt-get stuff into debian.
I've been slowing clearing the failed/missing dependencies.
Should
install.packages("tidyverse")
be this hard? Is this normal? On windows it's a doddle.
No plan survives contact with reality, the Python dependencies are not happy with git in windows on work PC. I think it is fixed, but more compiling glacially moving along. I've had too much coffee for all this waiting.
And there's a Visual Studio dependency dead end because I am not admin.
Most software: I'll pretty much work for a decade.
Anything python: I'm sorry, this was compiled in a different phase of the moon by a morning person on a Tuesday. I require those conditions again.
@noodle Strategies for Windows and Linux are probably different, so I'd guess it is normal, sadly. Be nice if the tidyverse web site told you all the dependencies and you could install them all in one go. Maybe it does? Also 100% certain subtly different versions of these packages will screw you around one day. It happened for me with gdal - same R code, different results.
@greg Yup, after a 90min process making this work I was wondering about that. This is just for play* for now. RStudio on my work system remains mandatory, for many reasons. Thankfully I'm not allowed to take data home, so there's no urgency in making this robust.
@noodle "If you’re compiling from source, you can run pak::pkg_system_requirements("tidyverse"), to see the complete set of system packages needed on your machine."
https://tidyverse.tidyverse.org/
Might speed up the whole process?
@greg One sec, compiling pak dependencies so I can compile pak (not joking)
@greg
pak::pkg_system_requirements is depreciated since pak 0.6.0 use pak::pkg_sysreqs instead
Ooh, that generated an install script. Thank you! Tidyverse wasn't the only huge package I needed so this saves a lot of time and checking.
