[Rd] Possible bug: file.exists() always returns TRUE for prn.us.txt

[Duncan Murdoch]

On 24/03/2018 6:16 AM, Joris Meys wrote:
> Dear all,
> 
> while preparing some exercises I came across some highly surprising
> behaviour of file.exists(). The specific value "prn.us.txt" always returns
> TRUE, even though that file is nowhere to be found on my system.

That's a Windows "bug", not an R bug.  Any name starting "prn" (upper or 
lowercase), followed by an extension (i.e. a dot and characters) is 
taken to be the DOS printer device.  According to Writing R Extensions, 
names starting with "‘con’, ‘prn’, ‘aux’, ‘clock$’, ‘nul’, ‘com1’ to 
‘com9’, and ‘lpt1’ to ‘lpt9' (possibly followed by extensions) are also 
bad.  You can Google "PRN filename in Windows" to find lots of people 
confused by this.  One page I get is

https://msdn.microsoft.com/en-us/library/aa365247(VS.85).aspx

but there's no guarantee that will work five minutes from now.

Duncan Murdoch

> 
> In a fresh R session 3.4.4 installed on Windows 10:
> 
>> grep("prn.us.txt", dir(recursive = TRUE))
> integer(0)
>> file.exists("prn.us.txt")
> [1] TRUE
>> file.exists("pnr.us.txt")
> [1] FALSE
>> file.exists("prn\\.us\\.txt")
> [1] FALSE
> 
>> sessionInfo()
> R version 3.4.4 (2018-03-15)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
> Running under: Windows 10 x64 (build 16299)
> 
> Matrix products: default
> 
> locale:
> [1] LC_COLLATE=English_United Kingdom.1252
> [2] LC_CTYPE=English_United Kingdom.1252
> [3] LC_MONETARY=English_United Kingdom.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United Kingdom.1252
> 
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
> 
> loaded via a namespace (and not attached):
> [1] compiler_3.4.4
> 
> This also happens in 3.4.3, 3.4.2 and 3.4.1 . It is confirmed by Roman
> Lustrik on his system as well :
> https://twitter.com/romunov/status/977486929380995072
> 
> I suspect this is a bug, or I must be missing something completely.
> 
> Cheers
> Joris
>