问题
I use R 2.13.1 and have unsuccessfully tried to load the package "plyr 1.6" in R. I have manually installed it into a directory "~/R/library". My code is:
.libPaths("~/R/library")
library(plyr)
I get the message:
Error in library(plyr) : 'plyr' is not a valid installed package
It works fine with other packages ("chron", "zoo", "ismev", "Lmoments"), but not for the "plyr" package, and I have no idea what is goin on. I have tried installing and loading earlier versions of "plyr", but with the same result.
I appreciate any help a lot since I am stuck! Regards Sisse
回答1:
This isn't an answer to manually installing plyr. This is more answer about why you have to manually install in the first place.
I suspect your CRAN mirror is improperly set. To check, type
options("repos")[[1]][1]
This should return something like:
> options("repos")[[1]][1]
CRAN
"http://streaming.stat.iastate.edu/CRAN"
>
try setting your repo to a different mirror like this:
options(repos="http://streaming.stat.iastate.edu/CRAN")
or use any other mirror of your choice.
Then try loading plyr:
install.packages("plyr")
library("plyr")
and let us know what happens.
回答2:
Might want to have a look at ?install.packages. It makes it very straightforward to install packages from CRAN. As simple as install.packages(pkgs="plyr").
回答3:
Unpacking the tar.gz file only works if the package contains only R code. plyr uses external code, which isn't compiled if you just extract the source to a library.
Use install.packages with repos = NULL to install from the source file.
install.packages("path/to/the/file/plyr_1.6.tar.gz", repos = NULL)
In the long term, the best solution is to fix your internet issues. If you're on a corporate network, speak to your network admin because they are likely blocking the traffic.
来源：https://stackoverflow.com/questions/7258639/problem-loading-the-plyr-package