On 05/23/2013 03:01 PM, Rodney Sparapani wrote:
possibly, just add ".cpp" to any file that does not already have it
before copying it to TEMP.
Actually, this latter suggestion is rather easy to implement
and, I believe, simplifies life for spaces in filenames, etc.
It adds 5 lines, but elimina
On 05/23/2013 09:26 AM, Hadley Wickham wrote:
You can't catch everything, but can you try ;) Given that people are
used to writing C++ with a variety of extensions, it seems like a
simple check here would be useful, especially given the otherwise
opaque nature of the error.
Since
Hadley
Hi!
> | So maybe sourceCpp could provide a warning if the extension is not in that
> list?
>
> Hmpd. I am not exactly sure why you advocate supporting .c or .f, .f90, .f95,
> .m and .mm none of which are likely to work _in any way_ with the parsing
> done and scaffolding added by sourceCpp. When you
On 23 May 2013 at 09:03, Hadley Wickham wrote:
| >> R governs the behaviour of R CMD based on the extension.
| >>
| >> The .cxx form was /never/ supported by R as far as I know.
| >
| > Indeed. According to "Package subdirectories" in Writing R Extensions
| > (http://cran.r-project.org/doc/manual
>> R governs the behaviour of R CMD based on the extension.
>>
>> The .cxx form was /never/ supported by R as far as I know.
>
> Indeed. According to "Package subdirectories" in Writing R Extensions
> (http://cran.r-project.org/doc/manuals/R-exts.html#Package-subdirectories)
> the only supported e
On 05/22/2013 08:41 PM, Dirk Eddelbuettel wrote:
R governs the behaviour of R CMD based on the extension.
The .cxx form was/never/ supported by R as far as I know.
And previously, I had not noticed that Rodney had used this extension; I had
repeatedly suggested he create smaller working examp
On Wed, May 22, 2013 at 6:41 PM, Dirk Eddelbuettel wrote:
> R governs the behaviour of R CMD based on the extension.
>
> The .cxx form was /never/ supported by R as far as I know.
>
Indeed. According to "Package subdirectories" in Writing R Extensions (
http://cran.r-project.org/doc/manuals/R-e
On 22 May 2013 at 17:59, JJ Allaire wrote:
| I also can build successfully on Ubuntu and OSX with the .cpp extension
however
| the .cxx extension fails. I think the file extension is the likely culprit.
R governs the behaviour of R CMD based on the extension.
The .cxx form was /never/ supporte
I also can build successfully on Ubuntu and OSX with the .cpp extension
however the .cxx extension fails. I think the file extension is the likely
culprit.
J.J.
On Wed, May 22, 2013 at 5:13 PM, Kevin Ushey wrote:
> Hi Rodney,
>
> I have the same problem as you on Mac OSX.
>
> I'm able to succe
I had mentioned earlier to you to maybe look into how TMP, TMPDIR, ... are
used or set.
My working example has
`.sourceCpp_44697_DLLInfo` <-
dyn.load('/tmp/Rtmpr8HLMP/sourcecpp_38624d1a5d0a/sourceCpp_73760.so')
which makes sense -- Rtmp* inside /tmp. Yours shows
`.sourceCpp_41979_DLLInfo`
Hi Rodney,
I have the same problem as you on Mac OSX.
I'm able to successfully compile the file through sourceCpp if I use a file
extension of .cpp rather than .cxx. Could this somehow be the reason why?
Let me know if that works for you.
-Kevin
On Wed, May 22, 2013 at 4:27 PM, Sparapani, Rod
And here is the failure on OS X ML...
> sourceCpp(verbose=TRUE, rebuild=TRUE, file="~/arma-sp.cxx")
Generated extern "C" functions
#include
RcppExport SEXP sourceCpp_41979_convertSparse(SEXP matSEXP) {
BEGIN_RCPP
Rcpp::RNGScope __r
On 05/22/2013 12:23 PM, Dirk Eddelbuettel wrote:
Methinks there may be something particular to your box.
Dirk
Possible, but I don't think it is likely. I found this issue
on the following OSes: RHEL 5, Oracle 6, CentOS 6 and OS X ML...
with various versions of GCC
--
Rodney Sparapani, PhD
And for completeness, here is the invocation that fails for Rodney:
R> sourceCpp(verbose=TRUE, rebuild=TRUE,
file="2012-12-25-armadillo-sparse-matrix.cpp")
Generated extern "C" functions
#include
RcppExport SEXP sourceCpp_44697_conv
Oops! I forgot the version info...
> sessionInfo()
R version 2.15.3 (2013-03-01)
Platform: x86_64-unknown-linux-gnu (64-bit)
locale:
[1] LC_CTYPE=en_US.ISO8859-1 LC_NUMERIC=C
[3] LC_TIME=en_US.ISO8859-1LC_COLLATE=en_US.ISO8859-1
[5] LC_MONETARY=en_US.ISO8859-1LC_MESSAGES=en
Rodney and I have been over this a bit, without resolution. At my urging, he
posted now on the list (as _everybody_ should: more eyeballs are in fact
better as you can tell from the fact that we still have not found this).
On my side, I just works. The second command is specific to my machine, b
16 matches
Mail list logo
