Skip to content

R2.12 for Mac

17 messages · David Winsemius, Ivan Calandra, Brian Ripley +4 more

#
Dear users,

I've seen a few problems on this list related to R2.12 on Mac. It looks 
to me that the Mac release is not really stable yet, or is it? Maybe 
it's just a few instances and most of you have no problems at all with it.

Therefore, I've kept my R2.11 on my Mac. Would you advise to upgrade or 
to wait some time? And if I should/can upgrade, do you have any 
recommendations regarding how to do it correctly (that are not included 
in the FAQ)?

Thanks for the advice!
Ivan
#
On Nov 15, 2010, at 5:24 AM, Ivan Calandra wrote:

            
I installed it yesterday and had three "moments" when I thought things  
had gone wrong, none of which proved to be enduring problems.  During  
the install process using the downloaded bundle from the ATT site I  
noticed that the progress bar had stopped for what seemed to be an  
unreasonable period and realized that I had a running instance of R.  
Quit()-ting appeared to allow progress to resume. (So my first bit of  
potentially glaringly obvious advice would be to shutdown all  
instances of R before the install. I couldn't find this bit of advice  
on the Install directions but it is certainly possible that the  
authors thought to too obvious to mention.)

The next puzzle was when I checked the installation of the R.app and  
R64.app GUI's. I assumed (wrongly as I eventually figured out) that  
they would have a more recent date than 2009 when viewed in the  
Applications folder. It turned out that they should (still) be version  
1.35 and that the new ones looked just like my old ones and work just  
fine.

The third "moment(s)" was when reinstalling package rms and dependent  
packages. I first tried installing the binary versions from my CRAN  
mirror and nothing seemed to be happening in hte time frame to which I  
have been accustomed. I "Force quit". I then restarted and installed  
from source. There was a very long period of no console activity, but  
I couldn't see any burning need to "Force quit" and I went away and  
came back 10 minutes later, to find the compliation process begun and  
then watch it run throught o completion and notice of success. Looking  
up at the intial portion of the console massages I discovered that  
about 20 packages were installed, so my initial impatience cause me to  
prematurely interupt what was probably a normal pause for the packages  
to be downloaded.

Normal behavior seems to have been restored after a few temporary  
moments of cluelessness on my part.
#
David, thank you for your answer.
If I understand you right, it looks like one should be patient during 
the installation process!

When I installed packages for R2.10, I ended up installing the 
dependencies manually one by one or else it was always installing many 
unnecessary packages (I guess they are unnecessary because I've never 
installed all of them on Windows), making the process last for ages. 
Looks like it still persists with R2.12

Regarding the packages, I thought that this website 
http://cran.r-project.org/web/checks/check_summary.html provided an 
up-to-date check of the packages, but I'm not sure it does. The packages 
were only tested with the pre-release for MacOSX and some packages (even 
the basics like plyr and reshape) show errors (which I don't really 
understand). Maybe I should just try out or contact the maintainer of 
the packages I need.

And since you've upgraded, you don't have any speed problem, as reported 
I think twice on the list?

Regards,
Ivan


Le 11/15/2010 15:35, David Winsemius a ?crit :

  
    
#
Hi,

On Mon, Nov 15, 2010 at 5:24 AM, Ivan Calandra
<ivan.calandra at uni-hamburg.de> wrote:
I've been using R-2.12 since day 0 and don't have any issues to report.

I've also stopped using R from the GUI for a long time now (I only run
it via the terminal, or through emacs/terminal), so if there are
problems with using that for an extended period of time, I wouldn't
know (I'm guessing there aren't, though).
Keep in mind that you can keep R-2.11 on your machine *and* install
R-2.12. If for some reason you think R-2.12 isn't stable enough for
you, you can switch back to R-2.11 with Rswitch, which you can find
here:

http://r.research.att.com/#other

-steve
#
On Mon, 15 Nov 2010, Steve Lianoglou wrote:

            
I second that (and since I do quite a lot of development on my MacBook 
Air, I was using pre-2.12.0 from ca day -30).

  
    
#
On 15-11-2010, at 11:24, Ivan Calandra wrote:

            
Installing R and updating installed packages went smoothly on both computers (iMac,MacBook Pro).

I haven't encountered any problems using R.
I use both the GUI and the R bundle of TextMate (latest version is 1.5.10 and all is well).

Berend
#
On 15-11-2010, at 20:45, steven mosher wrote:

            
/sw/...   is a Fink thing.
Somewhere in your .profile  Fink "things" are being set.
When you open a terminal and do "set" what do you get?
What does echo $PATH show?
etc.,...


I did an otool  -L on /Library/Frameworks/R.framework/Versions/2.12/Resources/lib/libR.dylib l and got this

libR.dylib:
        /Library/Frameworks/R.framework/Versions/2.12/Resources/lib/libR.dylib (compatibility version 2.12.0, current version 2.12.0)
        /Library/Frameworks/R.framework/Versions/2.12/Resources/lib/libRblas.dylib (compatibility version 0.0.0, current version 0.0.0)
        /Library/Frameworks/R.framework/Versions/2.12/Resources/lib/libgfortran.2.dylib (compatibility version 3.0.0, current version 3.0.0)
        /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 476.19.0)
        /Library/Frameworks/R.framework/Versions/2.12/Resources/lib/libreadline.5.2.dylib (compatibility version 5.0.0, current version 5.2.0)
        /usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.3)
        /usr/lib/libicucore.A.dylib (compatibility version 1.0.0, current version 36.0.0)
        /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 111.1.5)
        /usr/lib/libiconv.2.dylib (compatibility version 7.0.0, current version 7.0.0)
        /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0)


Berend
#
IF you are using a CRAN binary version of R, it is strange how fink is
involved at all.  You must have a messed up system

My best guess is that DYLD_LIBRARY_PATH or LD_LIBRARY_PATH has been
set (you can do
  echo $LD_LIBRARY_PATH
in the terminal).  Unset those variables, and proceed.

Kasper
On Mon, Nov 15, 2010 at 3:20 PM, Berend Hasselman <bhh at xs4all.nl> wrote:
#
Hi,
On Mon, Nov 15, 2010 at 4:51 PM, steven mosher <moshersteven at gmail.com> wrote:
`echo` in lowercase, not capitals. But anyway, we already see the
problem from your previous email. I guess this was what was in your
~/.profile file, or something?
Your DYLD_LIBRARY_PATH is, in fact, hosing you.

Comment out that line.

Or, you can temporary move /sw to /sw2 to get fink out of the way for
now, and see what is the what.

-steve
#
On Mon, Nov 15, 2010 at 5:13 PM, Steve Lianoglou
<mailinglist.honeypot at gmail.com> wrote:
This line screws up your entire system.  You sometimes see people on
various email lists recommend doing stuff like this, but this is a
specific place where OS X is very different from Linux.  In principle,
this could also mess with your other applications.

However, you also need to figure out how that got set in your GUI
(check by doing
Sys.getenv("DYLD_LIBRARY_PATH")
) and you will need to fix that as well in case fixing .profile does
not fix the GUI.

Kasper
#
Yeah remove that

But I don't even get why you are doing this.  On a mac (at least with
a newer version of the OS like Tiger or newer) RCurl works out of the
box.

So what you do it
  remove offending line in bashrc
  do a grep in you home folder to check you don't have other offending
lines including either LD_LIBRARY_PATH or DYLD_LIBRARYPATH
  cd /etc and do a recursive grep for LD_LIBRARY_PATH, and see if it
hits anything.
  restart Terminal
  start R and verify it works

Now, after this you start the _GUI_ and do the Sys.getenv thing.  The
GUI works differently from terminal R in this case.  If the offending
value is still there, you might want to have a look in
  .MacOS/environment.plist
There might be a similar line somewhere

Finally, write 100 times "I will not do random things to my system I
read on random websites".  Ok, that is a little tough, but you do
realize this is all your own doing, right.

Kasper
On Mon, Nov 15, 2010 at 7:11 PM, steven mosher <moshersteven at gmail.com> wrote: