Re: [R-pkg-devel] R CMD check crash for Ubuntu 14.04 R-3.2.1 only

2015-07-04 Thread Charles Geyer
Your "mocking" phrase in unknown to me. I am not sure where to trace that down. Read the bash script for check? Read tools:::.check_packages? Both are clear as mud to me. Also I just tried debugging by print statements. That does not work. It leaves the C called just before the crash. Putti

Re: [R-pkg-devel] R CMD check crash for Ubuntu 14.04 R-3.2.1 only

2015-07-04 Thread Dirk Eddelbuettel
Agreed on divide-and-conquer. There is no other way. Valgridn et al may move some code to registers for subtle changes. On 4 July 2015 at 10:20, Charles Geyer wrote: | I should add a more direct question.  When a crash occurs ONLY when running R | CMD check, how does one debug that?  Usually one

Re: [R-pkg-devel] R CMD check crash for Ubuntu 14.04 R-3.2.1 only

2015-07-04 Thread Charles Geyer
I should add a more direct question. When a crash occurs ONLY when running R CMD check, how does one debug that? Usually one would say that R CMD check --use-valgrind is the answer. But it only segfaults when --use-vagrind is not used. So somehow we have to debug R CMD check itself. How? On S

Re: [R-pkg-devel] R CMD check crash for Ubuntu 14.04 R-3.2.1 only

2015-07-04 Thread Charles Geyer
The diff for bar.Rout is OK. The solution is the same (the two matrices have the same rows and so determine the same convex polyhedron). Why it picks different orders of rows on different machines, I have no idea. I also saw the pages of warnings from gcc -W -Wall -Wextra but they are all about

Re: [R-pkg-devel] Saving binary files in a package

2015-07-04 Thread Hadley Wickham
On Sat, Jul 4, 2015 at 10:00 AM, Dirk Eddelbuettel wrote: > > On 4 July 2015 at 09:12, Thierry Onkelinx wrote: > | You have no guarantee that the user has write access to those folders. So > | you might want to rethink your strategy. > > Seconded. And I am not aware of a good cross-platform soluti

Re: [R-pkg-devel] R CMD check crash for Ubuntu 14.04 R-3.2.1 only

2015-07-04 Thread Dirk Eddelbuettel
Charles, I took a look based on your tar ball. When I use the normal 'loud' setting I default to for g++ (ie -pedantic -Wall) or clang++ I get more than a page full of errors. I just tried it using the UBSAN-via-clang instrumented R I provide via a docker container (and for which I showed usage

Re: [R-pkg-devel] Saving binary files in a package

2015-07-04 Thread Dirk Eddelbuettel
On 4 July 2015 at 09:12, Thierry Onkelinx wrote: | You have no guarantee that the user has write access to those folders. So | you might want to rethink your strategy. Seconded. And I am not aware of a good cross-platform solution so your best bet may be to require the user to specify a filename.

Re: [R-pkg-devel] Saving binary files in a package

2015-07-04 Thread Thierry Onkelinx
You have no guarantee that the user has write access to those folders. So you might want to rethink your strategy. ​ ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance Klinieks