[R-SIG-Mac] Flushing the buffer?

[Rolf Turner]

Please reply to me directly since I don't subscribe to this list.

I am using an Imac, to which I am relatively new, and have  
encountered somewhat puzzling behaviour.
I wrote a script to source in order to do a simulation, and put in a  
``cat'' statement in a for-loop to
let me know how things were progressing.  The strange behaviour was  
that the output from the
``cat()s'' seemed to be buffered, and did not appear on the screen  
until the whole procedure
completed.  Which is pretty  useless.

The funny thing is I'm pretty sure that I did exactly the same thing  
last week, and the buffering
behaviour did not occur.  I don't *think* I changed any settings, but  
something must have changed.
Can anyone suggest how I might change things back or otherwise over- 
ride the buffering.
I tried putting in a call to flush.console(), but that has no effect.

The output from sessionInfo() is as follows:

R version 2.5.0 (2007-04-23)
i386-apple-darwin8.9.1

locale:
C

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

other attached packages:
    misc
"0.0-1"

One other thing:  If I replace the cat() statements by print()  
statements, then the buffering
does not seem to occur.  This strikes me as very weird.

Here is a toy example:

for(r in 1:20) {
         X <- sum(rnorm(1e6))
         cat(r,"")
}
cat("\n")

The buffering phenomenon occurs whether I put the forgoing code in a  
file and source that file,
or type the code directly into the console.

Any ideas?  Thanks.

				cheers,

					Rolf Turner

######################################################################
Attention:\ This e-mail message is privileged and confidenti...{{dropped}}