[R] how to install R 3.0.1

[Uwe Ligges]

On 01.06.2013 15:58, Ranjan Maitra wrote:
> On Sat, 1 Jun 2013 14:58:22 +0200 Uwe Ligges
> <ligges at statistik.tu-dortmund.de> wrote:
>
>>
>>
>> On 01.06.2013 02:02, Ranjan Maitra wrote:
>>> On Fri, 31 May 2013 18:10:19 -0500 Robert Baer <rbaer at atsu.edu> wrote:
>>>
>>>> On 5/31/2013 11:59 AM, Ranjan Maitra wrote:
>>>>> Hi John,
>>>>>
>>>>> I suspect you may be missing a c()?
>>>>>
>>>>> On Fri, 31 May 2013 06:05:45 -0800 John Kane <jrkrideau at inbox.com>
>>>>> wrote:
>>>>>
>>>>>> paks  <-  install.packages( "Hmisc", "plyr")
>>>>> paks <- install.packages( c("Hmisc", "plyr"))
>>>>>
>>>>>
>>>>>>      install.packages(paks)
>>>>>> You can use the command  library()  to get a list of what is installed on your machine.
>>>>> Is it possible to get this as a vector of only the package names, or
>>>>> is post-processing the output of library() the only way out?
>>>> Do you mean something like:
>>>> paks = library()$results[,1]
>>>> save(paks, file = 'paks.RData')
>>>> Rob
>>>
>>> Exactly! Then
>>>
>>> install.packages(library()$results[,1])
>>>
>>> does all the updated installs in one shot.
>>
>>
>> Why do you want to reinstall all packages?
>>
>> I'd go for
>> udpate.packages(checkBuilt=TRUE)
>> if a reinstallation after an update of R is what you are aiming at.
>
> Actually, this did not work (on multiple architectures from 2.15.2 to
> 3.0.0 as well as from 3.0.0 to 3.0.1) for me in the sense that it only
> offered to update a very small subset packages, and that too, without
> their dependencies. Which resulted in a failure also for a large subset
> of the small subset of packages.
>
> As an example, consider the following:
>
>> update.packages(checkBuilt = TRUE)
>>
>
> (nothing to update, since I updated some things yesterday, and did a
> manual installation when the dependencies, eg lme4 comes to mind,
> failed)
>
>> installed.packages()[,"Built"]
>
> shows that AnalyzeFMRI was Built for 3.0.0.
>
> Perhaps not a big deal from 3.0.0 to 3.0.1

Right, it won't reinstall here since that is not necessary and the 
package sources were not changed, hence it safes you time and bandwidth.


> but was a big deal from
> 2.15.2 to 3.0.0.
>

Right, and it would have updated there....


> But after doing a
>
> ? install.packages("AnalyzeFMRI")
>
> I get that
>
> AnalyzeFMRI is Built for 3.0.1.


Yes, sure.



> Trying:
>
>> install.packages(installed.packages()[,1])
>
> I get that all are updated to 3.0.1, except for some, like XLConnectJars
> which are not available.


This is not a CRAN package and binaries may not be available in the 
repository you tried to get it from (Omegahat) or you have not selected 
Omegahat as a repository via setRepositories().
You can try to install from sources, though.

As I said before, replacing all your packages unconditionally wastes 
bandwidth. There are rarely situations where the other updates are 
really needed.

Best,
Uwe Ligges



> (I was surprised: it may be a bug, but is definitely a desirable
> feature that should be considered for inclusion.)
>
> Perhaps the better thing to do would have been to:
>
>> update.packages(installed.packages()[,1])
>
> but I did not think about that earlier.
>
> Many thanks and best wishes,
> Ranjan
>
>
>
>> Best,
>> Uwe Ligges
>>
>>
>>
>>> Thanks!
>>> Ranjan
>>>
>>> ____________________________________________________________
>>> GET FREE SMILEYS FOR YOUR IM & EMAIL - Learn more at http://www.inbox.com/smileys
>>> Works with AIM®, MSN® Messenger, Yahoo!® Messenger, ICQ®, Google Talk™ and most webmails
>>>
>>> ______________________________________________
>>> 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.
>>>
>>
>> ______________________________________________
>> 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.
>>
>
>