Skip to content

4.3.1 fails to load

2 messages · Carl Witthoft, Duncan Murdoch

#
Problem fixed but not solved.
On previous updates to R, I always copied all the packages (that aren't 
part of the base installation) from the previous version's Library 
folder to the new version's Library folder, and then updated when I 
actually use them.
The fix for my problem was to run "Update All" before trying to load any 
library.
The problem remains not-solved, because (prior to updating everything), 
any attempt to run

 >>  library(any_package)

would cause  failure similar to those I posted previously.  So, while 
I'm reasonably confident that R 4.3.1 doesn't play nicely with 4.2 - 
level libraries, I most definitely cannot point to any particular 
library.  The deepest level of the traceback report for  ggplot2 was

1: dyn.load(file, DLLpath = DLLpath, ...)

but when trying to load gmp, get the traceback quoted below.


carl
On 7/22/23 8:09 PM, r-sig-mac-request at r-project.org wrote:

  
    
#
On 23/07/2023 11:51 a.m., Carl Witthoft wrote:
You should run

  update.packages(ask = FALSE, checkBuilt = TRUE)

to make sure all packages are properly installed for 4.3.1.  In general, 
installing a package for x.y.z should let it work in x.y.(z+1), but not 
in x.(y+1).z'.  Errors like that are what "checkBuilt" is looking for. 
I don't know what "Update All" is doing; that's some front end calling R.

Duncan Murdoch