[R-sig-ME] Something broken in lme4 after update?

Douglas Bates bates at stat.wisc.edu
Thu Dec 18 00:49:20 CET 2008


The lme4 package has to be recompiled after the Matrix package was
updated.  Martin recently released a new version of lme4, 0.999375-29,
which depends on the version 0.999375-17 of the Matrix package.

On Wed, Dec 17, 2008 at 1:11 PM, Daniel Ezra Johnson
<danielezrajohnson at gmail.com> wrote:
> Hello,
>
> Sorry if the solution to this was already posted...
>
> I'm getting this error:
>
>> library(lme4)
> Error in dyn.load(file, DLLpath = DLLpath, ...) :
>  function 'cholmod_start' not provided by package 'Matrix'
> Error: package/namespace load failed for 'lme4'
>
> trying to install lme4 version 0.999375-27 on top of Matrix
> 0.999375-17 (both from CRAN, not using R-Forge).
>
> This is with R for Mac OS X 2.8.0.
>
> Any suggestions would be appreciated.
>
> Thanks,
> Daniel
>
> On Thu, Nov 27, 2008 at 3:39 PM, Douglas Bates <bates at stat.wisc.edu> wrote:
>> Is anyone else encountering this problem?  It looks like Mac OS X and
>> apparently a problem with the Matrix package declarations.  It is a
>> trivial-enough problem, a declaration of an argument as "const char*"
>> in one case and "char*" in another but I don't know which installed
>> version of the Matrix package this occurs in.
>>
>> On Thu, Nov 27, 2008 at 8:11 AM, Nick Isaac <njbisaac at googlemail.com> wrote:
>>>
>>>>> The issue is a bit more subtle than that.  You need to reinstall the
>>>>> lme4 package after changing the version of the Matrix package.
>>>>> There is a change in the internal code in the Matrix package between
>>>>> -16 and -17.  The -16 version requires two versions of every function
>>>>> in the CHOLMOD, AMD and COLAMD libraries whereas the -17 version is
>>>>> back to only one version.  However, lme4 must be reinstalled (in
>>>>> particular, the C code must be recompiled) after a change in the
>>>>> Matrix package.
>>>
>>> On 28 Oct 2008, at 10:12, ONKELINX, Thierry wrote:
>>>>
>>>> Dear Douglas,
>>>>
>>>> Reinstalling lme4 from R-forge after removing it from the harddisk
>>>> worked.
>>>>
>>>> Thank you,
>>>>
>>>> Thierry
>>>>
>>>>
>>>
>>>
>>> I have come up against the same issue as Thierry, but Prof Bates' solution
>>> does not work for me. The problem seems to be at:
>>>
>>> Matrix_stubs.c:494: error: conflicting types for 'M_R_cholmod_error'
>>> cholmod.h:798: error: previous declaration of 'M_R_cholmod_error' was here
>>>
>>> I have tried removing both lme4 and Matrix in a different order and by
>>> various means (deleting the folder in library vs remove.packages()). Most
>>> recently, I removed both lme4 and Matrix, emptied the trash then reinstalled
>>> Matrix from r-forge before quitting R and restarting my machine. On
>>> restarting R I tried:
>>>
>>>>
>>>> install.packages("lme4",type="source",repos="http://r-forge.r-project.org")
>>>
>>> A full transcript of the session is attached.
>>>
>>> The attached is a full session history from the most recent attempt. I would
>>> be grateful for any suggestions.
>>> Best wishes, Nick
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>> sessionInfo()
>>> R version 2.8.0 (2008-10-20)
>>> i386-apple-darwin8.11.1
>>>
>>> locale:
>>> en_GB.UTF-8/en_GB.UTF-8/C/C/en_GB.UTF-8/en_GB.UTF-8
>>>
>>> attached base packages:
>>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>>
>>> other attached packages:
>>> [1] Matrix_0.999375-17 lattice_0.17-15
>>>
>>> loaded via a namespace (and not attached):
>>> [1] grid_2.8.0  tools_2.8.0
>>>
>>>>
>>>> install.packages("lme4",type="source",repos="http://r-forge.r-project.org")
>>>
>>>>
>>>
>>>
>>
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>




More information about the R-sig-mixed-models mailing list