[R] help with updating to R2.7

Thu May 8 23:22:25 CEST 2008

Disclaimer: I don't use R on Windows much myself. But: I can't help  
noticing such threads occurring once in a while on r-help.

Why not simply define some folder to host one's personal library  
*outside* of the main R library? As explained in ?Startup, one can put  
something like

	R_LIBS="<path_to_personal_library>:${R_LIBS}"

in a .Renviron file located in his/her HOME directory (see http://cran.r-project.org/bin/windows/base/rw-FAQ.html#What-are-HOME-and-working-directories_003f 
  for information on this). Then, install.packages() will, well,  
install packages in this personal folder and R updates do not require  
shuffling  around of the library. At most an update.packages() once in  
a while.

I have been using this scheme for a long time with much success (and  
peace of mind) on Linux and OS X.

Hope this helps.

Le jeu. 8 mai à 15:02, John C Frain a écrit :

> When I upgrade in Windows from say 2.6.2 tp 2.7.0 I do the following
>
> 1. Install 2.7.0 in a new directory
>
> 2 Rename the library subdirectory in the new version from library to  
> library2
>
> 3 Copy the library subdirectory in 2.6.2 to 2.7.0
>
> 4 Copy the contents of library2 to the transferred library.  This
> re-installs the updated base packages.
>
> 5 Within R run update.packages(checkBuilt = TRUE) and you will be
> presented with a list of packages and be given the option of updating
> each package.
>
> This procedure is easy to implement.  I have used it for several
> updates and it has always worked well
>
> Best Regards
>
> John.
>
> 2008/5/8 Duncan Murdoch <murdoch at stats.uwo.ca>:
>> On 5/8/2008 10:40 AM, ravi wrote:
>>>
>>> Hi,
>>> Ouch! That really hurt. But I get the point.
>>> Here's what I did now. I copied all the package folders from R2.6,  
>>> except
>>> for the R.css file, and copied them into the R2.7 folder.
>>
>> That's your problem.  You've hosed the 2.7 libraries.
>>
>> You need to reinstall 2.7.  Only copy package folders from 2.6 if  
>> they don't
>> exist in 2.7.  If they already exist there, they've already been  
>> updated.
>>
>> The older packages may not work in 2.7, but once you run the
>> update.packages() function, you should get the latest versions.
>>
>> Duncan Murdoch
>>
>>> In the process, I overwrote the common files that came with the
>>> installation of R2.7.
>>> Here's the output that I obtained in R2.7 :
>>>>
>>>> library()
>>>
>>> Error in unique.default(ans) :  2 arguments passed  
>>> to .Internal(unique)
>>> which requires 3
>>>>
>>>> update.packages(checkBuilt=TRUE, ask=FALSE)
>>>
>>> Error: could not find function "update.packages"
>>>
>>> My intention with my previous mail was to check and obtain  
>>> confirmation on
>>> details like what I should do with the R.css file.
>>> But I realise that I should have then restricted myself to that  
>>> single
>>> question. That way, I would not be wasting the valuable time of  
>>> the R list
>>> experts, from whom I have received a lot of help previously.
>>> After reading the FAQ once again, I am guessing that the problem  
>>> could be
>>> a case of the internet downloading functions failing.
>>> As suggested in the FAQ, I tried the following commands :
>>>>
>>>> "path_to_R\bin\Rgui.exe" http_proxy=http://gannet/  
>>>> http_proxy_user=ask
>>>
>>> Error: unexpected symbol in ""path_to_R\bin\Rgui.exe" http_proxy"
>>>>
>>>> "path_to_R\bin\Rgui.exe" http_proxy=http://user:pass@gannet:80/
>>>
>>> Error: unexpected symbol in ""path_to_R\bin\Rgui.exe" http_proxy"
>>> Should I run the above commands in R, or in the windows command  
>>> window?
>>>
>>> Thanking You,
>>> Ravi
>>>
>>> ----- Original Message ----
>>> From: Duncan Murdoch <murdoch at stats.uwo.ca>
>>> To: ravi <rv15i at yahoo.se>
>>> Cc: r-help at r-project.org
>>> Sent: Thursday, 8 May, 2008 2:57:07 PM
>>> Subject: Re: [R] help with updating to R2.7
>>>
>>> On 5/8/2008 8:30 AM, ravi wrote:
>>>>
>>>> I know that it would be best if I reproduced the exact error  
>>>> messages,
>>>> but I have tried so many different things now and have lost track  
>>>> of the
>>>> exact error messages at each stage.
>>>
>>> This is not a reasonable request.  Rather than trying those two  
>>> methods
>>> one more time, and recording what goes wrong, you expect someone  
>>> on the list
>>> to do it for you?  If the advice was in the FAQ, presumably it  
>>> works for
>>> others, so the reason it didn't work for you is likely that you
>>> misinterpreted part of it, or your system has a unique problem,  
>>> etc.  The
>>> *only* way someone could diagnose that would be to see the error  
>>> messages.
>>>
>>> You need to use some common sense when you're asking for help.
>>>
>>> Duncan Murdoch
>>>
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> -- 
> John C Frain
> Trinity College Dublin
> Dublin 2
> Ireland
> www.tcd.ie/Economics/staff/frainj/home.html
> mailto:frainj at tcd.ie
> mailto:frainj at gmail.com
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

---
   Vincent Goulet, Associate Professor
   École d'actuariat
   Université Laval, Québec
   Vincent.Goulet at act.ulaval.ca   http://vgoulet.act.ulaval.ca