[Rcpp-devel] Fwd: Rcpp and rpy2
Jonas Rauch
jonas.rauch at googlemail.com
Mon Aug 15 15:52:52 CEST 2011
Sorry, I don't have a small example. I could come up with one, but that
would be purely artificial.
I am using Rcpp and rpy2 so I can use OpenOpt, a python optimization
framework with interfaces to a ton of different open source and proprietary
nonlinear and linear solvers. My objective function is very expensive to
evaluate and is implemented in R using Rcpp and various other R libraries.
Using OpenOpt allows to try a lot of different solvers without having to
change my implementation.
On Mon, Aug 15, 2011 at 2:59 PM, Dirk Eddelbuettel <edd at debian.org> wrote:
>
> On 15 August 2011 at 10:00, Jonas Rauch wrote:
> | Just to close this up:
> | I found out that my Rcpp installation must have occured when my R was not
> built
> | with shared lib. Thus Rcpp was not linked to libR. Just reinstalling Rcpp
> from
> | source solved the problem.
>
> That makes sense, and thank you for closing the loop by reporting back. If
> you have a small self-contained example of rpy2 and Rcpp, can you send it
> my
> way? It may make sense to add this to the Rcpp_FAQ vignette.
>
> Dirk
>
> | Regards,
> | Jonas
> |
> | On Fri, Aug 12, 2011 at 3:16 PM, Dirk Eddelbuettel <edd at debian.org>
> wrote:
> |
> |
> | On 12 August 2011 at 14:33, Jonas Rauch wrote:
> | | Hi Dirk and Romain,
> | |
> | | Aparrently libR is not loaded by rpy2, at least not in a way such
> that
> | Rcpp can
> | | access it. I have solved the problem by adding -lR to the linker
> flags
> | when
> | | building Rcpp. Since for normal use of Rcpp from R this does not
> make a
> | | difference, would you mind adding this to the Makevars file in a
> future
> | version
> | | of Rcpp? Or does this lead to problems in case R is build without
> | libR.so?
> |
> | I would be somewhat hesitant to do so as it may affect all users of
> Rcpp.
> |
> | Rcpp is for getting code into R, and we can take libR as given as R
> is
> | always
> | around when loading Rcpp. I think the problem is really with the
> linker
> | and
> | passing symbols through from one end to the other.
> |
> | Other views, anyone?
> |
> | Dirk
> |
> |
> | | Best regards,
> | | Jonas
> | |
> | | ---------- Forwarded message ----------
> | | From: Jonas Rauch <jonas.rauch at googlemail.com>
> | | Date: Fri, Aug 12, 2011 at 10:36 AM
> | | Subject: Rcpp and rpy2
> | | To: rcpp-devel <rcpp-devel at r-forge.wu-wien.ac.at>
> | |
> | |
> | | Is Rcpp compatible with rpy2?
> | | I would like to call a very complex R function from python. My
> function
> | depends
> | | on code implemented with Rcpp. When I try to load Rcpp from python
> via
> | |
> | | import rpy2.robjects as robjects
> | | robjects.r('library(Rcpp)')
> | |
> | | I get the following:
> | |
> | | Error in dyn.load(file, DLLpath = DLLpath, ...) :
> | | unable to load shared object '/home/jonas/R/
> | x86_64-unknown-linux-gnu-library
> | | /2.13/Rcpp/libs/Rcpp.so':
> | |
> /home/jonas/R/x86_64-unknown-linux-gnu-library/2.13/Rcpp/libs/Rcpp.so:
> | | undefined symbol: R_ClassSymbol
> | | Error: package/namespace load failed for 'Rcpp'
> | | Traceback (most recent call last):
> | | File "RcppTest.py", line 2, in <module>
> | | robjects.r('library(Rcpp)')
> | | File "/usr/local/lib/python2.7/dist-packages/
> | |
> rpy2-2.2.2dev_20110811-py2.7-linux-x86_64.egg/rpy2/robjects/__init__.py",
> | line
> | | 225, in __call__
> | | res = self.eval(p)
> | | File "/usr/local/lib/python2.7/dist-packages/
> | | rpy2-2.2.2dev_20110811-py2.7-linux-x86_64.egg/rpy2/robjects/
> | functions.py", line
> | | 82, in __call__
> | | return super(SignatureTranslatedFunction, self).__call__(*args,
> | | **kwargs)
> | | File "/usr/local/lib/python2.7/dist-packages/
> | | rpy2-2.2.2dev_20110811-py2.7-linux-x86_64.egg/rpy2/robjects/
> | functions.py", line
> | | 34, in __call__
> | | res = super(Function, self).__call__(*new_args, **new_kwargs)
> | | rpy2.rinterface.RRuntimeError: Error: package/namespace load failed
> for
> | 'Rcpp'
> | |
> | | From what I understand R_ClassSymbol is supplied by libR.so, which
> should
> | by
> | | loaded by rpy2 anyway in order to work at all, or am I totally
> wrong
> | here?
> | |
> | | Running python 2.7.1, rpy2-2.2.2, R 2.13.0
> | | I would appreciate any help.
> | |
> | | Regards,
> | | Jonas
> | |
> | |
> | |
> ----------------------------------------------------------------------
> | | _______________________________________________
> | | 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
> |
> | --
> | Two new Rcpp master classes for R and C++ integration scheduled for
> | New York (Sep 24) and San Francisco (Oct 8), more details are at
> | http://dirk.eddelbuettel.com/blog/2011/08/04#
> | rcpp_classes_2011-09_and_2011-10
> | http://www.revolutionanalytics.com/products/training/public/
> | rcpp-master-class.php
> |
> |
>
> --
> Two new Rcpp master classes for R and C++ integration scheduled for
> New York (Sep 24) and San Francisco (Oct 8), more details are at
>
> http://dirk.eddelbuettel.com/blog/2011/08/04#rcpp_classes_2011-09_and_2011-10
>
> http://www.revolutionanalytics.com/products/training/public/rcpp-master-class.php
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20110815/5dbf4622/attachment-0001.htm>
More information about the Rcpp-devel
mailing list