On Thu, Nov 21, 2013 at 6:40 PM, kcrisman <kcris...@gmail.com> wrote:
>
>
> On Thursday, November 21, 2013 3:08:44 PM UTC-5, François wrote:
>>
>> On Thu, 21 Nov 2013 09:54:10 Jason Grout wrote:
>> > On 11/21/13 9:35 AM, kcrisman wrote:
>> > > I love that the Sage cell supports it (there was a fascinating blog
>> > > post
>> > > about embedding R a year or so ago), ... it would be very, very
>> > > unfortunate if we were to drop support for this.
>> >
>> > Just to be clear: the Sage cell would *certainly* still have R (along
>> > with many other optional and even just plain python packages [1]).
>> >
>> Another option that could be interesting, sage doesn't use or link
>> against the R library so it is trivial to actual use a system wide R.
I don't believe this. Sage includes and builds rpy2, which is vastly
superior to our pexpect interface to R for working with nontrivial
data, and rpy2 links against the R library. Unfortunately, it would
(in my experience) be very difficult to get
import rpy2
rpy2.do_stuff
to work on all supported platforms if we don't build our own version of R.
-- William
>> It is actually easier to do that to link to an external ATLAS.
>> So we could have an advanced variable to use a system R and
>> possibly ask for a minimal version of R (and remove those doctest
>> that tests a specific version of R).
>> That would also solve at least one ticket that David Kirkby had about
>> R on solaris were the only way he could compile R was outside sage
>> with the solaris compiler instead of gcc.
>>
>
> I think that this is probably a good idea. If we only built R if we
> absolutely had to, that would be nice. It does actually use significant
> compile time. Naturally, for binaries we could still provide it (especially
> since at least some binary users definitely won't know about R, much less
> use it). How hard do you think it would be to properly modify the R stuff
> for this? (I.e., would it be more effort than its [1] worth?)
>
> [1] yes, its not it's
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/groups/opt_out.
--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/groups/opt_out.
