[Rd] problem in compililng c code using R CMD SHLIB
Prof Brian Ripley
ripley at stats.ox.ac.uk
Wed Jun 6 20:08:01 CEST 2012
On 06/06/2012 18:12, Joshua Ulrich wrote:
> On Wed, Jun 6, 2012 at 8:20 AM, Simon Urbanek
> <simon.urbanek at r-project.org> wrote:
>> On Jun 6, 2012, at 6:16 AM, Arathy Ram wrote:
>>> Hello all,
>>> I was trying to compile a simple C program hello.c using R CMD SHLIB
>>> MY R software residing in C:\Program Files\R\R-2.15.0, and also I have
>>> downloaded the Rtools 2.15.0 and it is in C:\Rtools, MY file(hello.c) is in
>>> I have followed the procedures given below
>>> 1. In a new CMD.exe window I have set the path as
>>> c:\ Rtools\bin; C:\Rtools\MinGW\bin;C:\Program Files\R\R-2.15.0\bin
>>> 2. I have changed the directory to E:\ where my file(hello.c) is residing
>>> E:\>cd \R_dir
>>> then I have given the command like this
>>> E:\>cd \R_dir> R CMD SHLIB hello.c
>>> Then I am getting the following message
>>> cygwin warning:
>>> MS-DOS style path detected: E:/PROGRA~1/R/R-214~1.0/etc/
>>> Preferred POSIX equivalent is:
>>> CYGWIN environment variable option "nodosfilewarning" turns off this
>>> Consult the user's guide for more details about POSIX paths:
>>> Even though I have gone through several discussions in the same topic I am
>>> not able to resolve my problem. It would be a great help if somebody knows
>>> what exactly the problem and its solution and favorable suggestions.
>> There is no problem in what you posted - all you posted is merely
>> a note from cygwin that it prefers different path style, but it's
perfectly ok and doesn't affect anything.
Or put another way, the problem is a user read failure: see below.
>> So what exactly is the problem you're having?
> Probably has something to do with PATH not being set correctly because
> the Cygwin warning points to R-2.14.0 on E:, not R-2.15.0 on C:.
But as for the note, it *is* discussed in the appropriate manual. Search
for 'nodosfilewarning' in
You most likely will need to set that environment variable.
See what the posting guide has to say about doing your own homework.
The author of the docs you discourteously ignored (and much else you are
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
More information about the R-devel