[R-sig-ME] Release Dates lme4 Version

Vincent Dorie vjd4 at nyu.edu
Sun Mar 16 21:03:25 CET 2014


I wrote that in a rush so I'm likely mistaken, but when I tried to test backwards compatibility using 2.15.3 the packages weren't available. Specifically the lme4 version for macs on CRAN was  < 1.0. It looks as if OS X users don't have the same depth of backwards compatibility out of the box, although I suppose that could depend on the mirror.

Anyhow, moral may or may not be: just upgrade everything.

On Mar 16, 2014, at 3:11 PM, Ben Bolker wrote:

> 
>  Hmmm. As Brian Ripley pointed out to me offline,
> 
>> Not so.  RcppEigen (> 0.3.1.2.3) appears to mean >= 0.3.2.0, which
> depends on Rcpp (>= 0.10.5) and R (>= 2.15.1) (also the dependence of
> Rcpp 0.10.5).
> 
>  I just double-checked this myself and (as usual) he seems to be right.
> 
>  Vince, maybe you were looking at RcppEigen 0.3.2.1.1 instead?
> 
>  Rolf, where were you looking on CRAN for RcppEigen?
> 
> * http://cran.r-project.org/web/packages/RcppEigen/index.html shows
> version 0.3.2.1.1
> * The Windows binary available for 2.15 is 0.3.2.0.2:
> http://cran.r-project.org/bin/windows/contrib/2.15/RcppEigen_0.3.2.0.2.zip
> 
>  cheers
>    Ben Bolker
> 
> 
> On 14-03-14 07:27 PM, Vincent Dorie wrote:
>> As far as I can tell, there is a bit of understating when it comes to
>> version numbers for R itself. RcppEigen 0.3.1.2.1 depends on Rcp
>> 0.11.0, which in turn depends on R 3.0.0. So even though just about
>> every package in the lme4 chain claims to work on 2.15 or so, the
>> backwards compatibility has been broken.
>> 
>> On Mar 14, 2014, at 7:16 PM, Rolf Turner wrote:
>> 
>>> On 15/03/14 07:11, Ben Bolker wrote:
>>>> Viechtbauer Wolfgang (STAT <wolfgang.viechtbauer at ...> writes:
>>>> 
>>>>> 
>>>>> Great! Looking forward to the new version on CRAN.
>>>>> 
>>>>> Best, Wolfgang
>>>> 
>>>> New version (1.1-5) is on CRAN now; binaries gradually being
>>>> built/ propagating to mirrors.  See 
>>>> https://github.com/lme4/lme4/blob/master/inst/NEWS.Rd for changes
>>>> in this release ...
>>> 
>>> Just tried to install version 1.1-5 (tar ball obtained from
>>> Austrian mirror).  It complains that  "‘RcppEigen’ 0.3.1.2.1 is
>>> being loaded, but 0.3.1.2.3 is required".
>>> 
>>> But the latest version of RcppEigen that I can find on CRAN is good
>>> ol' 0.3.1.2.1.
>>> 
>>> So do I have to go somewhere else to get 0.3.1.2.3?  That's a bit
>>> bewildering for us bunnies.
>>> 
>>> cheers,
>>> 
>>> Rolf Turner
>>> 
>>> _______________________________________________ 
>>> R-sig-mixed-models at r-project.org mailing list 
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>> 
>> _______________________________________________ 
>> 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