[R] Problem opening connections to removable hard drive

[Uwe Ligges]

I tried with R-2.4.1 and R-2.5.0 - it works for me.
Can you please try again with R-2.5.0?

Uwe Ligges

Elizabeth Purdom wrote:
> Hi,
> 
> I have my files and R session on a portable hard drive to be able to 
> work across computers since I have some extremely large files. I have 
> just started doing this and generally it works as I expect. But after a 
> while of using my R session, certain types of calls to open connections 
> on the hard drive don't work. But I experience no problems 
> opening/saving files on the portable hard drive using other programs.
> 
> In particular, if I explicitly call the entire folder path to the hard 
> drive, I can save an object to file, for example. But if I just want to 
> save it in the current working directory, without explicitly naming the 
> entire path, then I have problems (I show an example below). I am using 
> a package that assumes a certain file structure from the current working 
> directory in order to find/write output, which is why I want to be able 
> to rely on not explicitly calling the whole path. Furthermore, I was 
> experimenting just now, and when I changed the current directory to 
> another one, and then back to the removable hard drive, everything gets 
> 'reset' somehow and opening connections works again, just like when I 
> first started up my session.
> 
> I am on Windows XP, R 2.4.0.
> 
> Thanks,
> Elizabeth Purdom
> 
> example:
>  > getwd()
> [1] "I:/ICBP/Exon Array/Data"
> #explicit call to hard drive works...
>  > save(x,file="I:/ICBP/Exon Array/Data/tmp.rdata")
> #unexplicit call relying on knowing working directory doesn't...
>  > save(x,file="tmp.rdata")
> Error in gzfile(file, "wb") : unable to open connection
> In addition: Warning message:
> cannot open compressed file 'tmp.rdata'
> 
> ______________________________________________
> R-help at stat.math.ethz.ch 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.