[R] ScieViews installer

Philippe Grosjean phgrosjean at sciviews.org
Sun May 8 08:42:44 CEST 2005


Prof Brian Ripley wrote:
> On Sat, 7 May 2005, Erich Neuwirth wrote:
> 
>> Thank you for your help.
>> The library directory contained some very strangely named
>> subdirectories, file3132 and similarly.
>> After removing these and vereything related to SciViews
>> I was able to install SciViews.
>> Looking back I seem to remember that form time to time
>> these strange directories appear and I do not understand
>> where they come from.
>> Could it be a timing issue, i.e. Windows trying to remove
>> a directory too early? I remember similar problems with
>> temporary files not being deletable by Excel.
> 
> 
> No, on NT a Windows atomic file rename is used.  Neither Duncan nor I 
> have ever encountered these: they seem to result from a Windows OS bug.

This happens occasionally when R is not able to unzip a file because it 
is corrupted. In such a case, you are better to reload the file.
Best,

Philippe Grosjean

P.S.: note that the SciViews bundle is not compatible yet with 
internationalization introduced in R 2.1.0. I am currently working on it.



>> Prof Brian Ripley wrote:
>>
>>> On Sat, 7 May 2005, Erich Neuwirth wrote:
>>>
>>>> bundle 'SciViews' successfully unpacked and MD5 sums checked
>>>> Error in sprintf(gettext("unable to move temp installation '%d' to
>>>> '%s'"),  :
>>>>        use format %s for character objects
>>>>
>>>> how can I solve this problem?
>>>
>>>
>>>
>>> Well, use R-patched where that exact problem goes away, but that is not
>>> the problem that needs solving.
>>>
>>> The underlying cause is a problem on your Windows file system.  Do you
>>> have a version of SciViews left over that Windows cannot remove?
>>>
>>> I've just tested this, and it works for me even in 2.1.0.
>>>
>>
>> -- 
>> Erich Neuwirth, University of Vienna
>> Faculty of Computer Science
>> Didactic Center for Computer Science
>> Visit our SunSITE at http://sunsite.univie.ac.at
>> Phone: +43-1-4277-39464 Fax: +43-1-4277-39459
>>
>>
>




More information about the R-help mailing list