[R-SIG-Mac] HELP

[BATES Timothy]

Just FYI, I've had a lot of students in the last 2-weeks reporting errors like this about other packages with dependencies that have their own dependencies.

The full chain doesn't appear to be being followed.

For some, on Macs I think, they also have the problem of install.packages() trying to install a source version which is newer than the current CRAN binary, and then failing because the required compiler isn't present.

Also, sometimes a package will get listed as being installed one or more times (this happened for "crayon" last week) , then not ending up installed at all.

There was a strange issue with the "stringi" package being missing despite multiple install.packages() calls by students to the top level package that ultimately wanted it.

Most solved by reading the error, manually installing (forcing binary if necessary) the missing package, and repeating until done.

Of course this leads to lots of "it's broken"  and "no... it didn't work" hassles :-)

Best, tim (currently happy the mac TZ warning has gone away in 3.4.4 :-) )


> On 20 Mar 2018, at 4:01 pm, Keith O'Hara <keith.ohara at nyu.edu> wrote:
> 
> The dependence seems to come from rms -> multcomp -> mvtnorm. 
> 
> Keith
> 
>> On Mar 20, 2018, at 11:44 AM, David Winsemius <dwinsemius at comcast.net> wrote:
>> 
>> 
>>> On Mar 20, 2018, at 8:35 AM, Keith O'Hara <keith.ohara at nyu.edu> wrote:
>>> 
>>> Did you try reinstalling the mvtnorm package?
>>> 
>>> Keith
>>> 
>>>> On Mar 19, 2018, at 10:56 PM, Yonabeth Nava de Escalante <yonabeth77 at gmail.com> wrote:
>>>> 
>>>> I am trying to install the ‘rms’ package in my Mac OS X Yosemite 10.10.5. When I try to load the library I get this error:
>>>> 
>>>>> library(rms)
>>>> Error in loadNamespace(i, c(lib.loc, .libPaths()), versionCheck = vI[[i]]) : 
>>>> there is no package called ‘mvtnorm’
>>>> Error: package or namespace load failed for ‘rms’
>>>> 
>>>> I tried installing the dependencies with 
>>>> 
>>>> install.packages("rms",dependencies = TRUE)
>>>> 
>>>> I also tried
>>>> 
>>>> install.packages("rms", repos=c("http://rstudio.org/_packages", "http://cran.rstudio.com”))
>>>> 
>>>> However, I keep getting the same error. 
>>>> 	[[alternative HTML version deleted]]
>> 
>> I can confirm this behavior. I load rms from startup in my .Rprofile file and if mvtrnorm is missing I also get:
>> 
>> Error: package or namespace load failed for ‘rms’ in loadNamespace(i, c(lib.loc, .libPaths()), versionCheck = vI[[i]]):
>> there is no package called ‘mvtnorm’
>> 
>> To me it signals a needt to let the package maintainer know about the issue.
>> 
>>> maintainer("rms")
>> [1] "Frank E Harrell Jr <f.harrell at vanderbilt.edu>"
>> 
>>>> 
>>>> _______________________________________________
>>>> R-SIG-Mac mailing list
>>>> R-SIG-Mac at r-project.org
>>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>>> 
>>> _______________________________________________
>>> R-SIG-Mac mailing list
>>> R-SIG-Mac at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>> 
>> David Winsemius
>> Alameda, CA, USA
>> 
>> 'Any technology distinguishable from magic is insufficiently advanced.'   -Gehm's Corollary to Clarke's Third Law
>> 
>> 
>> 
>> 
>> 
> 
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac


-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.