53

I would like to upgrade one R package to the newer version which is already available. I tried 

update.packages(c("R2jags"))

but it does nothing! No output on console, no error, nothing. I used the same syntax as for install.packages but perhaps I'm doing something wrong. I have been looking at ?update.packages but I was not been able to figure out how it works, where to specify the package(s) etc. There is no example. I also tried to update the package using install.packages to "install" it again but that says "Warning: package ‘R2jags’ is in use and will not be installed".

Tomas
  • 57,621
  • 49
  • 238
  • 373
  • 9
    If you read `?update.pacakges`, then you'll notice that the first argument to `update.packages` is **not** the package name, but the library location to check. If you only want to update one package, use `install.packages`. The warning you report tells you why it doesn't work: you have to stop using the package in all the R sessions that are using it. This is most easily accomplished by closing the offending R sessions. – Joshua Ulrich Jan 30 '14 at 16:18
  • ?Joshua I saw exactly what you say in the help - that's why I said I don't know where to specify packages. It's not very intuitive. What is `update.packages` for, then? – Tomas Jan 30 '14 at 16:27
  • The _Details_ section of `update.packages` tells you its purpose. – Joshua Ulrich Jan 30 '14 at 16:29
  • @MehdiNellen that's completely different question. I was asking about what function I should use to install the package. I don't need to override anything, no tricks, just to know the function. This question could stay here so other users find it easily by google. – Tomas Jan 30 '14 at 16:41
  • alright, I will remove it and leave the link as comment for anyone who is looking http://stackoverflow.com/questions/13163248/possible-to-override-the-blocking-of-a-packages-re-installation-after-it-has – Mehdi Nellen Jan 31 '14 at 08:15

4 Answers4

46

You can't do this I'm afraid, well, not with update.packages(). You need to call install.packages("R2jags") instead.

You can't install R2jags in the current session because you have already loaded the current version into the session. If you need to, save any objects you can't easily recreate, and quit out of R. Then start a new R session, immediately run install.packages("R2jags"), then once finished, load the package and reload in any previously saved objects. You could try to unload the package with:

detach(package:R2jags, unload = TRUE)

but it is quite complex to do this cleanly unless the package cleans up after itself.

update.packages() exists to update all outdated packages in a stated library location. That library location is given by the first argument (if not supplied it works on all known library locations for the current R session). Hence you were asking it the update the packages in library location R2jags which is most unlikely to exist on your R installation.

Gavin Simpson
  • 170,508
  • 25
  • 396
  • 453
  • Gavin, thanks!! So `install.packages` does it! But the `detach` thing, closing session etc. is not needed! I just started new R and installed it from that! Don't like to restart now... – Tomas Jan 30 '14 at 16:33
  • @Tomas Well obviously, but the impression you gave was that you wanted the new **R2jags** in the current session. Otherwise, what was the misunderstanding re the warning message you got? – Gavin Simpson Jan 30 '14 at 16:37
  • Gavin, I am happy to test the new package first. Well, the misunderstanding was that I intuitivelly expected that the `update.packages` function is the one that should work and `install.packages` won't do that. – Tomas Jan 30 '14 at 16:44
  • @Tomas Be careful, you can't *test* the new **R2jags**, *unless* you install it into a *new* library location. If you don't do that, the old version will only be available within the currently running session. There may also be issues with help files etc if you update the package (from another session) in the library location where the package was loaded from in the current session. – Gavin Simpson Jan 30 '14 at 16:47
27

Additionally, you can install RStudio and update all packages by going to the Tools menu and selecting Check for Package Updates.

americo
  • 1,013
  • 8
  • 17
2

update.packages(oldPkgs = "R2jags") will check updates only for that package and ask you if you want to update.

M4RT1NK4
  • 433
  • 1
  • 4
  • 9
0
# The following two commands remove any previously installed H2O packages for R.
if ("package:h2o" %in% search()) { detach("package:h2o", unload=TRUE) }
if ("h2o" %in% rownames(installed.packages())) { remove.packages("h2o") }

# Next, we download packages that H2O depends on.
pkgs <- c("RCurl","jsonlite")
for (pkg in pkgs) {
if (! (pkg %in% rownames(installed.packages()))) { install.packages(pkg) }
}

# Now we download, install and initialize the H2O package for R.
install.packages("h2o", type="source", repos="http://h2o-release.s3.amazonaws.com/h2o/rel-xia/2/R")

# Finally, let's load H2O and start up an H2O cluster
library(h2o)`enter code here`
h2o.init()
DJ6968
  • 109
  • 1
  • 2