On 09/03/2011 2:34 PM, jim holtman wrote:
I see something similar on my system.  Running your command line, I
can see the command window flashup and then disappear.

We have had issues in the past with setting the input file handle for external commands to an empty stream. Then cmd
reads from the stream, immediately hits EOF, and quits.

It should do this consistently, so the fact that some people are seeing inconsistent behaviour looks like a bug to me. But whether it is a bug in R or a bug in Windows is hard to determine.

I'm not going to work on this in the near term (I keep getting friendly reminders of other problems I've promised to work on), but if anyone else feels like debugging it, I would suggest looking at how the file handles are handled.

Duncan Murdoch


>  sessionInfo()
R version 2.12.1 (2010-12-16)
Platform: i386-pc-mingw32/i386 (32-bit)

locale:
[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United
States.1252
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252

attached base packages:
[1] tcltk     stats     grDevices utils     datasets  graphics
methods   base

other attached packages:
  [1] RColorBrewer_1.0-2    sqldf_0.3-5           chron_2.3-39
gsubfn_0.5-5
  [5] proto_0.3-8           RSQLite.extfuns_0.0.1 RSQLite_0.9-4
DBI_0.2-5
  [9] zoo_1.6-4             plyr_1.4

loaded via a namespace (and not attached):
[1] grid_2.12.1     lattice_0.19-17 tools_2.12.1
>


On Wed, Mar 9, 2011 at 12:14 PM, Ralph Olsson
<ralphols...@btinternet.com>  wrote:
>  Hello,
>
>  I work for a company in which a number of employees use R. Many of them like 
to run executables via the system function in such a way that the output of that 
executable is displayed in a separate window. To give an example of the behavior 
they require, the following command can be run in R GUI:
>
>  system("cmd",invisible=FALSE,show.output.on.console=FALSE)
>
>  The result is that a command prompt opens up and can be used interactively 
until it is closed. We have never had a problem with this command running R on 
Windows XP machines.
>
>  Recently a number of us have switched to PCs running Windows 7. Under this 
operating system the above command often, if not always, fails to open a window. 
Some people can successfully run the command in one R session but then fail to get 
a window if a second R session is started. Others cannot get a window to open at 
all. We've tested the command with R 2.12.1, 2.12.1 patched, 2.12.2 and 2.12.2 
patched (all in both 32 and 64 bit). All of these exhibit the same problem.
>
>  I raised the issue as a bug 
(https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14516), but this was promptly closed 
as being "not reproducible". Since we are having issues running the above command 
on all of our windows 7 PCs but none of our windows XP PCs, and since whoever closed the bug 
clearly had no issues running the command on his or her Windows 7 PC, I'm not really sure 
what to do next in terms of trying to resolve the problem.
>
>  Can anyone shed any light on this issue? Is anyone able to reproduce it 
(only takes 30 seconds, just copy the above line into RGui on Windows 7 and 
confirm whether a command prompt opens; then, with the first session still open, 
try the same in a second RGui window)?
>
>  Any help or feedback would be appreciated,
>
>  Cheers,
>
>  Ralph
>          [[alternative HTML version deleted]]
>
>  ______________________________________________
>  R-help@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@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.

Reply via email to