Prof Brian Ripley wrote:
> Have you actually tried this recently? If you don't set R_EXE you get
> a fatal error:
>
> -- Making package base
> adding build stamp to DESCRIPTION
> /bin/sh: I failed to read the instructions to set R_EXE in MkRules:
> command not found
>
> An
Have you actually tried this recently? If you don't set R_EXE you get
a fatal error:
-- Making package base
adding build stamp to DESCRIPTION
/bin/sh: I failed to read the instructions to set R_EXE in MkRules:
command not found
And MkRules says
# path (possibly full path
ernesto wrote:
> Hi,
>
> I wrongly assumed it would use R in the cross compiling directory and I
> did not found any reference to it on documentation. Obviously my fault,
> as you say.
The cross-compile section of the admin manual did mention this:
"You also need a working copy of this version
Hi,
I wrongly assumed it would use R in the cross compiling directory and I
did not found any reference to it on documentation. Obviously my fault,
as you say.
Regards
EJ
Prof Brian Ripley wrote:
> You are pretty clearly not cross-compiling 'with R 2.5.0' (sic), as R
> 2.5.0 does contain Sys
You are pretty clearly not cross-compiling 'with R 2.5.0' (sic), as R
2.5.0 does contain Sys.setenv.
Looks like the R on Linux you are using is not R 2.5.0, and please
cross-check the instructions.
This has been tested.
On Mon, 30 Apr 2007, ernesto wrote:
> Hi,
>
> I'm cross compiling win R i
