sink() does not seem to release the file - R-help

Tue, Jan 2, 2001 2:57 AM #

I run Win2000, and when I try to write to a file and later use it
it seems that R 1.02.0 has a firm grip on it:

_              
platform i386-pc-mingw32
arch     x86            
os       Win32          
system   x86, Win32     
status                  
major    1              
minor    2.0            
year     2000           
month    12             
day      15             
language R

some text

Volume in drive N is Local Disk
 Volume Serial Number is 1C5E-3A70

 Directory of N:\Bendix\Steno\JaSv\iddm

02-01-2001  11:33                    9 foo
               1 File(s)              9 bytes
               0 Dir(s)  172.857.344.000 bytes free

N:\Bendix\Steno\JaSv\iddm\foo
The process cannot access the file because it is being used by another
process.


However, when I fire up my old R 1.01.1 (which is not installed under
Win2000,
but merely a verbatim copy from my old NT4 machine), I get:

_         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status             
major    1         
minor    1.1       
year     2000      
month    August    
day      15        
language R

any text

Volume in drive N is Local Disk
 Volume Serial Number is 1C5E-3A70

 Directory of N:\Bendix\Steno\JaSv\iddm

02-01-2001  11:37                    8 oof
               1 File(s)              8 bytes
               0 Dir(s)  172.857.339.904 bytes free

Volume in drive N is Local Disk
 Volume Serial Number is 1C5E-3A70

 Directory of N:\Bendix\Steno\JaSv\iddm

File Not Found
Warning message: 
cmd execution failed with error code 1 in: shell("dir oo*")

Is this a new bug or a new facility??

I have a vague recollection that a similar problem was also present in an
older version of R, too.
The workaround in R 1.2 is to:

and then use oof for the relevant purpose.

The background for this is that I use R to write a small SAS-program
that writes a SAS-dataset in a read.table format, that I then read into
R. But when strted from withn R, SAS complains that the program file just 
sink()ed by R is in use (apparently by R). 

Any one interested can get the rather clumsy sas.get.R and the corresponding
%out SAS-macro from me on request. Requires that SAS is on your macine 
as well. No bells and whistles at all.

Bendix Carstensen

----------------------
Bendix Carstensen
Senior Statistician
Steno Diabetes Centre
Niels Steensens Vej 2
DK-2820 Gentofte
Denmark
tel: +45 44 43 87 38
mob: +45 28 25 87 38
fax: +45 44 43 73 13
bxc at novo.dk
www.biostat.ku.dk/~bxc
----------------------




-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Peter Dalgaard

Tue, Jan 2, 2001 3:32 AM #

"BXC (Bendix Carstensen)" <bxc at novonordisk.com> writes:

[Proof deleted]

This was on the R-devel list 2 weeks ago:

On Tue, 19 Dec 2000, Prof Brian D Ripley wrote:

O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Brian Ripley

Tue, Jan 2, 2001 3:33 AM #

To add to that, this behaviour has been changed in the R-patched version
(and hence for 1.2.1): from NEWS/BUG FIXES

    o   Explicitly close files opened by sink() for OSes (such as
        Windows) that benefit from it.

In other words, sink() will close the connection for you.

On 2 Jan 2001, Peter Dalgaard BSA wrote:

"BXC (Bendix Carstensen)" <bxc at novonordisk.com> writes:

I run Win2000, and when I try to write to a file and later use it
it seems that R 1.02.0 has a firm grip on it:

[Proof deleted]

This was on the R-devel list 2 weeks ago:

On Tue, 19 Dec 2000, Prof Brian D Ripley wrote:

On Tue, 19 Dec 2000 Setzer.Woodrow at epamail.epa.gov wrote:

When I use sink(filename) to save the results of the run, the file
does not seem to be closed or released back to the OS after I issue
the "sink()" command.  Here is a baby example:

sink works on connections, not files, as from 1.2.0.

sink("test.txt")
tdta <- data.frame(x = x <- 0:10,y = 2 * x + rnorm(11))

print(summary(lm(y ~ x, data=tdta)))
sink()

At this point (even after refreshing the Explorer window) the filesize of
"test.txt" is listed as 0 bytes, and as long as Rgui is running, I
cannot delete the file.  Emacs cannot read the file (but Notepad can).  This happens both in the gui
and when the commands are in a file run via "source()".

Yes, but what is the bug exactly?  The connection is still open so you can
switch output back to it.  If you want different behaviour, you can open a
connection and close it yourself.  As in

zz <- file("test.txt", "w")
sink(zz)
tdta <- data.frame(x = x <- 0:10,y = 2 * x + rnorm(11))
print(summary(lm(y ~ x, data=tdta)))
sink()
close(zz)

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._