Skip to content

SP 1.0-12 install error

6 messages · Alex Mandel, Michael Sumner, Edzer Pebesma

Original
Alex Mandel
# 
I was getting an error today installing/upgrading sp.

Error in setClass("CRS", slots = c(projargs = "character"), prototype =
list(projargs = character(1)),  :
  unused argument(s) (slots = c(projargs = "character"))
Error : unable to load R code in package ?sp?
ERROR: lazy loading failed for package ?sp?

I tried uninstalling/reinstalling, using a different mirror, doing it
system-wide vs just for my user etc... Then I noticed that 1.0-12 was
released only a few days ago, so I downloaded the archived version of
1.0-11 and the install works.
http://cran.r-project.org/src/contrib/Archive/sp/

R is 2.15.3 on Ubuntu 12.04

Is there a bug in the package or does the new sp require R>3?

Thanks,
Alex
Edzer Pebesma
# 
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thanks, you are right, sp 1.0-12 depends on R 3.0.0 but claims it
doesn't.

Ubuntu users are very well served by binary updates of R releases, see
for instance http://cran.at.r-project.org/bin/linux/ubuntu/

If the list considers this as an annoyance, let me know, it's not a
big deal to convert back to pre 3.0.0 syntax (for some time).
On 09/03/2013 09:40 PM, Alex Mandel wrote:
- -- 
Edzer Pebesma
Institute for Geoinformatics (ifgi), University of M?nster
Heisenbergstra?e 2, 48149 M?nster, Germany. Phone: +49 251
83 33081, Fax: +49 251 8339763 http://ifgi.uni-muenster.de
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJSJwzrAAoJEM1OCHCtOnfxMgwIAIeJHUmZ3bz1ohNA4ys8poMd
kAxLHohEbsXvHAVBRSUg2MJf5wLarHTouPA8SVM6LS48u7S/fTPeHs70C2kPPmHN
Pw6jQIzGwb7z3XdauF5VD9fZYEKApvbJVCfQoEPp8JoqXPvB1zXudxRRU9lQ5yMU
MmrREzhhFKvd/21hhgBSho1KxLszYyVRwEtG60xEfVXhHVHdnCSrEF1eYnF1iiWU
w/wk/zBp57rhGvLBaS0kCmduZICjVgxK/ZUhmm6yfR8MbtOcGjpr+uAEYZ03te7U
dDiganJW1lEEvwj0toU1OnOs+elutk3K4GBXh96oHoXgqyPt58NMlA8qhGa5VBg=
=SFJP
-----END PGP SIGNATURE-----
Alex Mandel
# 
I ran into trouble because I needed to add Rgdal (someone had forgotten
to install it) to an existing 2.15 server and it wanted sp newer than
1.0 but I had 0.9.x. Upgrading this particular machine to 3 was too
risky and would involve verifying all the other running packages could
be upgraded.

For many of my machines I do track the CRAN debs for R updates but not
on deployed servers where an accidental upgrade could break R web
services that are running or long standing analysis code being used by
researchers.

So yes it would be good if I had gotten the warning in R or saw it in
some news or changelog somewhere. I don't think there's any particular
need to revert as long as people are aware of the issue and the way to
work around it.


Thanks,
Alex
On 09/04/2013 03:35 AM, Edzer Pebesma wrote:
Michael Sumner
# 
I noticed this in an RStudio server instance, which was running R
2.15.3, I had to download each package .tar.gz and install
individually to avoid the 'unused argument(s) (slots = c(projargs =
"character"))' problem.

I seemed to need older versions of maptools, rgdal, rgeos, sp and
raster to get everything working there. It might be worth getting a
formal listing of this if many people are affected. This is what I
used, I just went back a few months in the archive, I was not careful
to check when the change was made:

 "maptools_0.8-24.tar.gz"
"raster_2.1-12.tar.gz"
"rgdal_0.8-7.tar.gz"
"rgeos_0.2-16.tar.gz"
"sp_1.0-8.tar.gz"
On Thu, Sep 5, 2013 at 4:38 AM, Alex Mandel <tech_dev at wildintellect.com> wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Alex Mandel
    

    

      
      #
    

  


  

      
I only had to manually install sp_1.0-11.tar.gz
After that install.packages worked fine using web sources for rgdal and
raster. I did not try rgeos or maptools.

Wonder why you had to manually do all of those.

Thanks,
Alex

        
On 09/04/2013 03:48 PM, Michael Sumner wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Edzer Pebesma
    

    

      
      #
    

  


  

      
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

sp_1.0-13, now on CRAN, should work with R >= 2.14 again.

        
On 09/05/2013 01:28 AM, Alex Mandel wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        
- -- 
Edzer Pebesma
Institute for Geoinformatics (ifgi), University of M?nster
Heisenbergstra?e 2, 48149 M?nster, Germany. Phone: +49 251
83 33081, Fax: +49 251 8339763 http://ifgi.uni-muenster.de
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJSKCfkAAoJEM1OCHCtOnfxH+AH+QE/DDfZIECVw1ACIljWYlrJ
5I+Wgj3FqfYYLWaXuRCNZDvsCU1+8SddD+cEWoz6Len2YI1GXBOhtNj0eBL3Ns85
WuhPI/Ahg0DuEwG6Fl+j3ER89/G9Rt5d/PfT4iszVRV8YVCr3Ra4JzStCuc03B3B
1i4gq7E1teDSzEVGtT+FEEHMmnMAmtCNnvNMEY2kjpWkg7jnRQ6UE7M0JpdeyQCP
xS6dhWzu37UxWBufCqkeMq2zF9mlUvvT9SwT01tkIPCtfn9/P7U8oChqUhTUj76v
jSoQMLEWhzTgbJE4eXDIv0BJjSW5QDbsTaN5l2PdviP+0vm3un4nMYKOJEhfZ84=
=Wdj8
-----END PGP SIGNATURE-----