New upstream version of RandomFields 3.1.36 fails to build from source

[Martin Schlather]

Dear Andreas,

although I have changed RU_sqrtPosDef to RU_sqrtPosDefFree
in direct.cc between the recent and the former version of 
RandomFields, I am puzzled as both functions are defined 
in exactly the same way in RandomFieldsUtils (in both the 
current 0.3.15 and the former version 0.3.8; see functions 
sqrtPosDef and sqrtPosDefFree in solve.cc and 
init_RandomFieldsUtils.* )

The file direct.cc is the alphabetically first one that 
uses an RU_* function, i.e. a function exported from
RandomFieldsUtils. Is there a miscommunication between
RandomFields and RandomFieldsUtils? 

Best,
Martin




On 08.01.2017 09:00, Andreas Tille wrote:
> Hi Martin,
>
> I intended to update the Debian package of randomfields but I've got the
> following error:
>
> ...
> g++ -I/usr/share/R/include -DNDEBUG   -I"/usr/lib/R/site-library/RandomFieldsUtils/include"   -fpic  -g -O2 -fdebug-prefix-map=/build/r-base-3.3.2=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -g  -c cubicsolver.cc -o cubicsolver.o
> g++ -I/usr/share/R/include -DNDEBUG   -I"/usr/lib/R/site-library/RandomFieldsUtils/include"   -fpic  -g -O2 -fdebug-prefix-map=/build/r-base-3.3.2=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -g  -c direct.cc -o direct.o
> direct.cc: In function 'int init_directGauss(cov_model*, gen_storage*)':
> direct.cc:196:52: error: 'RU_sqrtPosDefFree' was not declared in this scope
>    err = RU_sqrtPosDefFree(Cov, vdimtot, cov->Ssolve);
>                                                     ^
> /usr/lib/R/etc/Makeconf:139: recipe for target 'direct.o' failed
> make[1]: *** [direct.o] Error 1
> make[1]: Leaving directory '/build/r-cran-randomfields-3.1.36/src'
> make[1]: Entering directory '/build/r-cran-randomfields-3.1.36/src'
> make[1]: Leaving directory '/build/r-cran-randomfields-3.1.36/src'
> ERROR: compilation failed for package 'RandomFields'
>
>
> Do you have any idea what might be wrong here?
>
> Kind regards
>
>       Andreas.
>


-- 
Prof. Dr. Martin Schlather
Institut für Mathematik
Universität Mannheim
A5, 6, B 118
D-68131 Mannheim 

schlather at math.uni-mannheim.de  
https://ms.math.uni-mannheim.de
phone: +49 (0)621 181 2552