[Rcpp-devel] Building/linking trouble with cxxfunction()

[Christian Gunning]

On Wed, Feb 16, 2011 at 1:21 PM,
<rcpp-devel-request at r-forge.wu-wien.ac.at> wrote:
>
> Yes, of course.  I'm just wondering if there's a way to do the social
> engineering better, taking advantage of other community members for the
> "easy" stuff, and freeing you to concentrate on aspects you might find
> more rewarding.  Closed, purpose-built lists don't really have a good way
> to get people who start out casually involved to become more intimately
> involved.

I'm going to chime in here and suggest that Rcpp's approach has
focused "social engineering" largely on documentation, testing (see
unit tests in the source), and examples. As it stands, the Rcpp manual
and vignette collection is significant -- I frequently google "cran
rcpp" to grab one of them. Most list questions that have been asked >2
are addressed in either the FAQ or the quickref.  Maybe a bit on
Makefile is warranted.

R.e. S4 -- a good place to start is here:
http://www.r-bloggers.com/resources-for-s4-classes-and-methods/

R.e. original problem
"
  dlopen(/var/folders/TL/TL+wXnanH5uhWm4RtUrrjE+++TM/-Tmp-//RtmpAA9upc/file2d4606aa.so,
6): Symbol not found: __Z8svdLAS2AP4smatl
"

Have you gotten a minimal example up and running without reference to
dgCMatrix?  Looking at your stackoverflow post, I'm having a hard time
seeing inside your C++ code.  Quite a while ago, I ran into some
"Symbol not found: __Z***" problems and solved them (don't remember
why or how), but I don't recall any issues with external libraries.
So, you might want to check that you're solving the correct problem by
starting off with a simpler one :)

-christian
-- 
A man, a plan, a cat, a ham, a yak, a yam, a hat, a canal – Panama!