Plus one of the duties of package managers is to make sure that they build and 
pass CRAN checks. Doing that with a non standard environment seems a bit risky.

At the moment I can’t get anything to build on R-Devel which is making life 
difficult and I may raise a question if it still isn’t working when I’ve 
finished with the new experience that is online teaching.

> On 20 May 2020, at 10:49 am, David Winsemius <dwinsem...@comcast.net> wrote:
> 
> 
> On 5/19/20 1:46 PM, Yoshihiko Baba via R-SIG-Mac wrote:
>> Adelchi,
>> 
>> Why don’t you try Fink version of R?
>> 
>> Fink build mnormt on R 4.0 as well (although it is experimental).
>> https://github.com/fink/fink-distributions/issues/623
> 
> 
> This should be considered contrary to recommended practice: From the 
> page for macOS tools that was cited by the OP we see:
> 
> 
> "Although it is possible to compile R using tools from other package 
> managers such as Homebrew, MacPorts or Fink, such binaries are by 
> definition incompatible with macOS native libraries and applications. If 
> you choose one of those package managers, make sure you 
> compile/everything/using those tools including R and all packages and 
> libraries you intend to use."
> 
> 
> I think it should have been offered as advice to only be followed by 
> persons with great experience in compiling and building packages on Macs.
> 
> Best;
> 
> David
> 
>> 
>> Fink allows you to install different versions of R (3.6, 3.5, 3.4, 3.3, 3.2 
>> and 3.1).
>> 
>> For different versions, CRAN packages are separately installed 
>> (/sw/lib/R/3.6/site-library  and so on).
>> 
>> 
>>> 2020/05/19 22:45、Adelchi Azzalini <azzal...@stat.unipd.it>のメール:
>>> 
>>> Hi.
>>> In essence, my problem is that I cannot install a package which requires 
>>> Fortran-77 compilation on my Mac.
>>> 
>>> More in detail, this is my own package 'mnormt' which exists on CRAN since 
>>> 2006; it is currently at version 1.5-7. It has been developed in successive 
>>> versions on a Linux machine, where it compiles with no problems. The macOS 
>>> and MS-windows versions are created on CRAN, where it passes all pertaining 
>>> checks.
>>> 
>>> Now, in the new pandemic world, I need create a new version of the package, 
>>> but working from home, hence using my Mac laptop instead of the Linux 
>>> desktop as in the past. Unfortunately, compilation or loading does not 
>>> work. If I run
>>> 
>>> R CMD check mnormt_1.5-7.tar.gz
>>> 
>>> first I get a bunch of warning messages such as
>>> 
>>> Warning: Possible change of value in conversion from REAL(8) to INTEGER(4) 
>>> ....  [-Wconversion]
>>> Warning: ‘lxchng’ may be used uninitialized in this function 
>>> [-Wmaybe-uninitialized]
>>> Warning: Label 10 at (1) defined but not used [-Wunused-label]
>>> 
>>> and finally it crashes with
>>> 
>>> Error: package or namespace load failed for ‘mnormt’:
>>> .onLoad failed in loadNamespace() for 'mnormt', details:
>>>  call: dyn.load(file, DLLpath = DLLpath, ...)
>>>  error: unable to load shared object 
>>> '/Users/aa/SW-aa/Pkg-mnormt/mnormt.Rcheck/00LOCK-mnormt/00new/mnormt/libs/mnormt.so':
>>>  
>>> dlopen(/Users/aa/SW-aa/Pkg-mnormt/mnormt.Rcheck/00LOCK-mnormt/00new/mnormt/libs/mnormt.so,
>>>  6): Symbol not found: ___addtf3
>>>  Referenced from: /usr/local/lib/libquadmath.0.dylib
>>>  Expected in: /usr/local/lib/libgcc_s_x86_64.1.dylib
>>> in /usr/local/lib/libquadmath.0.dylib
>>> Error: loading failed
>>> 
>>> Needless to say, I have no idea what this "addtf3" is or it should be. 
>>> There is no such a symbol in the Fortran code.
>>> 
>>> I believe to have installed all ingredients indicated at 
>>> https://mac.r-project.org/tools/
>>> See the list below with details about my system and other installed tools.
>>> 
>>> Any suggestions for sorting this out?
>>> 
>>> Adelchi Azzalini
>>> 
>>> 
>>> ---------------------------------------------------
>>> Installation info:
>>> 
>>> R> R.version
>>>               _
>>> platform       x86_64-apple-darwin17.0
>>> arch           x86_64
>>> os             darwin17.0
>>> system         x86_64, darwin17.0
>>> status
>>> major          4
>>> minor          0.0
>>> year           2020
>>> month          04
>>> day            24
>>> svn rev        78286
>>> language       R
>>> version.string R version 4.0.0 (2020-04-24)
>>> nickname       Arbor Day
>>> 
>>> --- macOS Mojave 10.14.6
>>> 
>>> --- Xcode version 11.3.1
>>> 
>>> --- Fortran
>>> 
>>> [aa@mac-2013:~] gfortran -v
>>> Using built-in specs.
>>> COLLECT_GCC=gfortran
>>> COLLECT_LTO_WRAPPER=/usr/local/gfortran/libexec/gcc/x86_64-apple-darwin18/8.2.0/lto-wrapper
>>> Target: x86_64-apple-darwin18
>>> Configured with: ../gcc-8.2.0/configure --prefix=/usr/local/gfortran 
>>> --with-gmp=/Users/fx/devel/gcc/build_package/deps 
>>> --enable-languages=c,c++,fortran,objc,obj-c++ --build=x86_64-apple-darwin18 
>>> --disable-multilib --with-native-system-header-dir=/usr/include 
>>> --with-sysroot=/Library/Developer/CommandLineTools/SDKs/MacOSX.sdk
>>> Thread model: posix
>>> gcc version 8.2.0 (GCC)
>>> 
>>> [aa@mac-2013:~] echo $PATH
>>> .:/Users/aa/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin/:/usr/X11R6/bin:/opt/local/bin:/Library/TeX/texbin:/usr/local/gfortran/bin
>>> 
>>> -- other tools indicated at https://mac.r-project.org/tools/
>>>   xz-5.2.5/  is installed
>>>   PCRE2 10.34 built with --disable-jit
>>>   bzip2 Version 1.0.6 is installed
>>> 
>>> _______________________________________________
>>> R-SIG-Mac mailing list
>>> R-SIG-Mac@r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>> _______________________________________________
>> R-SIG-Mac mailing list
>> R-SIG-Mac@r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 
>       [[alternative HTML version deleted]]
> 
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac@r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 

_______________________________________________
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

Reply via email to