[Rcpp-devel] DataFrame::create() in endless loop

Hi,

thanks everyone for the help,
I used IntegerVector and friends, 
and changed from DataFrame to List.

Thanks,
Steffen

It's a memory problem from R.  It is trying to do garbage collection

(RunGenCollector) and getting hung up, probably because of

exhausting

one of the stacks in R.  Your std::vector<int> objects need to be

copied to R objects.  Is it possible to define them as

IntegerVector's

instead, in which case the Create method will (I think) just move

some

pointers around.

 

I had the same problem a while ago and the solution we?ve come up with
was to get rid of Rcpp::DataFrame and use Rcpp::XPtr instead.

It turned out that Rcpp::DataFrame is rather impractical for
sufficiently large data sets. 

As an alternative, Rcpp::XPtr is a little bit more complex to
implement but by far more reliable and efficient.

For more details see this
http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2012-January/003335.html and subsequent messages.

 

Hope this helps.

 

Cheers,

 

Slava Mazur

 

_______________________________________________
Rcpp-devel mailing list
Rcpp-devel at lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel

IPB Halle                    AG Massenspektrometrie & Bioinformatik
Dr. Steffen Neumann          http://www.IPB-Halle.DE
Weinberg 3                   http://msbi.bic-gh.de
06120 Halle                  Tel. +49 (0) 345 5582 - 1470
                                  +49 (0) 345 5582 - 0
sneumann(at)IPB-Halle.DE     Fax. +49 (0) 345 5582 - 1409

I used IntegerVector and friends, 
and changed from DataFrame to List.

Hello Steffen,
Do you mean you needed to do both those changes to get it to work? Or
either would have fixed your problem, but you decided to do both "just
in case"?

I had the same problem a while ago and the solution we?ve come up with
was to get rid of Rcpp::DataFrame and use Rcpp::XPtr instead.

(Ref:
http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2012-January/003345.html
)

XPtr is just for passing a handle to a block of bytes back to R, that it
can then pass back to your C++ code at a later stage, isn't it? I.e. not
a solution if you wanted to manipulate the data in R.
(I'm asking here in case I've misunderstood the scope of its purpose.)

Darren

Darren Cook, Software Researcher/Developer

http://dcook.org/work/ (About me and my work)
http://dcook.org/blogs.html (My blogs and articles)

| > I used IntegerVector and friends, 
| > and changed from DataFrame to List.
| 
| Hello Steffen,
| Do you mean you needed to do both those changes to get it to work? Or
| either would have fixed your problem, but you decided to do both "just
| in case"?
|

| >> I had the same problem a while ago and the solution we?ve come up with
| >> was to get rid of Rcpp::DataFrame and use Rcpp::XPtr instead.
| (Ref:
| http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2012-January/003345.html
| )
| 
| XPtr is just for passing a handle to a block of bytes back to R, that it
| can then pass back to your C++ code at a later stage, isn't it? I.e. not
| a solution if you wanted to manipulate the data in R.

Correct.

I think the main advantage of working with an IntegerVector is that you may
well save another copy, and dito with List vs DataFrame. Same diligent memory
profiling may reveal more.

Dirk

| (I'm asking here in case I've misunderstood the scope of its purpose.)
| 
| Darren
| 
| 
| -- 
| Darren Cook, Software Researcher/Developer
| 
| http://dcook.org/work/ (About me and my work)
| http://dcook.org/blogs.html (My blogs and articles)
| _______________________________________________
| Rcpp-devel mailing list
| Rcpp-devel at lists.r-forge.r-project.org
| https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel

"Outside of a dog, a book is a man's best friend. Inside of a dog, it is too
dark to read." -- Groucho Marx