Does ess (>= 5.3.10) now require R ?

Martin Maechler maechler at stat.math.ethz.ch
Thu Dec 4 09:31:39 CET 2008


Hi guys,

sorry to be late to all of these threads.

I've just now committed a patch that hopefully fixes the problem
that ess has become properly compilable only when at least one
version of R was available on the system.

Indeed, it's been mea culpa stemming from a minor reorganization
of 
   ess-versions-created, 
   ess-r-versions-created, 
   ess-s-versions-created

in ess-site.el
and I also have left there 
a three line comment with  "FIXME" about the subject.

Martin

>>>>> "StEgl" == Stephen Eglen <S.J.Eglen at damtp.cam.ac.uk>
>>>>>     on Wed, 03 Dec 2008 22:09:44 +0000 writes:

    StEgl> Stephen Eglen <S.J.Eglen at damtp.cam.ac.uk> wrote:
    >> To debug this on my ubuntu box, do I just need to remove R and then try
    >> to 'install ess package'?  If so, how do I 'install ess package'?  Sorry
    >> for the obvious question, but I assume its not just a case of 
    >> 'apt-get install ess' which is about as far as my regular debian
    >> knowledge goes!

    StEgl> and perhaps, even better, given the SVN copy of ESS in my home
    StEgl> directory, how can I get the packaging system to try to package that?
    StEgl> My current thinking on this bug is that the code may be fault-tolerant
    StEgl> but either the code is getting run in a funny order, or somehow
    StEgl> ess-r-versions-created isn't being defined as a variable when
    StEgl> ess-newest-R is run.  

    StEgl> Stephen

    StEgl> _______________________________________________
    StEgl> ESS-Debian mailing list
    StEgl> ESS-Debian at stat.math.ethz.ch
    StEgl> https://stat.ethz.ch/mailman/listinfo/ess-debian



More information about the ESS-Debian mailing list