R help - Dec 2012 - print and cat not working with parallelised functions?

Dear R Community,

I am running R version 2.15.2 with package parallel version 2.15.2.
The problem is that cat and print do not produce any output. Also assigning
objects to the .GlobalEnv does not work. This makes it difficult for me to debug
code. This can be seen from the
following minimal working example:

library(parallel)
fun2 <- function(x) {
b <<- x; # try to export the object to the workspace
print(x); # try to print x
}

fun1 <- function(d) {
 cl <- makeCluster(getOption("cl.cores", detectCores()));
 parSapply(cl=cl, X=seq_along(d), FUN=fun2);
 stopCluster(cl=cl);
}
fun3 <- function(d) sapply(X=seq_along(d), FUN=fun2); 
fun1(d=1:5)
a <- fun3(d=1:5)

print only works when fun3 is called, that is when there is no parallelisation.
The
same is also true for the <<- assignment.
I am almost sure this is not a bug, but a feature, so I would like to ask you
for some explanation and also for some ideas how to handle debugging code when 
no printing or exporting objects to the workspace works.

Any suggestions will be appreciated.

Best regards,

Martin Ivanov

On 08.12.2012 21:04, Martin Ivanov wrote:
>   Dear R Community,
>
> I am running R version 2.15.2 with package parallel version 2.15.2.
> The problem is that cat and print do not produce any output. Also assigning
objects to the .GlobalEnv does not work. This makes it difficult for me to debug
code. This can be seen from the
> following minimal working example:
>

print: You are printing in the R client, not on the master
assign: You are assigning to the .GlobalEnv of the client, not the one 
of the master.

Best,
Uwe Ligges


> library(parallel)
> fun2 <- function(x) {
> b <<- x; # try to export the object to the workspace
> print(x); # try to print x
> }
>
> fun1 <- function(d) {
>   cl <- makeCluster(getOption("cl.cores", detectCores()));
>   parSapply(cl=cl, X=seq_along(d), FUN=fun2);
>   stopCluster(cl=cl);
> }
> fun3 <- function(d) sapply(X=seq_along(d), FUN=fun2);
> fun1(d=1:5)
> a <- fun3(d=1:5)
>
> print only works when fun3 is called, that is when there is no
parallelisation. The
> same is also true for the <<- assignment.
> I am almost sure this is not a bug, but a feature, so I would like to ask
you
> for some explanation and also for some ideas how to handle debugging code
when
> no printing or exporting objects to the workspace works.
>
> Any suggestions will be appreciated.
>
> Best regards,
>
> Martin Ivanov
>
> ______________________________________________
> R-help at 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.
>