[Rd] (PR#7100) and #7102: list.files fails with all.files = TRUE

ligges at statistik.uni-dortmund.de ligges at statistik.uni-dortmund.de
Fri Jul 16 19:23:21 CEST 2004


rpeng at jhsph.edu wrote:

> I get different behavior on Linux:
> 
> system("mkdir newdir")
> dir("newdir", all.files = TRUE, recursive = TRUE)  ## R hangs

The point is that all.files=TRUE implies returning also "." and ".."!
Hence the recursion uses to be infinite (from "." to "." to "." ......).

If recursiv = TRUE, the function should never look into "." and "..", 
except of the first step of the recursion.

Uwe Ligges


> ## After hitting Ctrl-C
> There were 50 or more warnings (use warnings() to see the first 50)
>  > dir("newdir", all.files = TRUE, recursive = TRUE)
> character(0)
> Warning message:
> list.files: newdir is not a readable directory
> 
> After hitting Ctrl-C the first time, dir() doesn't hang but it 
> produces the warning.
> 
>  > version
>           _
> platform i686-pc-linux-gnu
> arch     i686
> os       linux-gnu
> system   i686, linux-gnu
> status
> major    1
> minor    9.1
> year     2004
> month    06
> day      21
> language R
> 
> -roger
> 
> 
> joehl at gmx.de wrote:
> 
>>>list.files("c:/tmp", all.files = TRUE, recursive = TRUE)
>>
>>Error in list.files("c:/tmp", all.files = TRUE, recursive = TRUE) : 
>>        directory/folder path name too long
>>
>>
>>
>>>version
>>
>>         _              
>>platform i386-pc-mingw32
>>arch     i386           
>>os       mingw32        
>>system   i386, mingw32  
>>status                  
>>major    1              
>>minor    9.1            
>>year     2004           
>>month    06             
>>day      21             
>>language R 
>>
>>
>>Best
>>
>>
>>Jens Oehlschlägel
>>
>>______________________________________________
>>R-devel at stat.math.ethz.ch mailing list
>>https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
>>
> 
> 
> ______________________________________________
> R-devel at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-devel



More information about the R-devel mailing list