/usr/local/bin/gcc -bundle -flat_namespace -undefined suppress -L/usr/ local/lib -o kinship.so agfit6b.o bdsmatrix_index1.o bdsmatrix_index2.o bdsmatrix_index3.o bdsmatrix_prod.o bdsmatrix_prod2.o bdsmatrix_prod3.o chinv4.o chinv5.o cholesky4.o cholesky5.o chsolve4.o chsolve5.o coxfit6a.o coxfit6b.o coxfit6c.o gchol.o gchol_bds.o -F/Library/Frameworks/R.framework/.. -framework R /usr/bin/ld: multiple definitions of symbol _c6 agfit6b.o definition of _c6 in section (__DATA,__common) coxfit6a.o definition of _c6 in section (__DATA,__common) coxfit6b.o definition of _c6 in section (__DATA,__common) coxfit6c.o definition of _c6 in section (__DATA,__common) /usr/bin/ld: warning multiple definitions of symbol _signgam /Library/Frameworks/R.framework/../R.framework/R(lgamma.o) definition of _signgam /usr/lib/libSystem.dylib(gamma9.o) definition of _signgam collect2: ld returned 1 exit status make: *** [kinship.so] Error 1 ERROR: compilation failed for package 'kinship'
The usual problem: to link bundles (plug-ins) OS X uses flat namespaces, and by default you cannot have multiple symbol definitions if you use flat namespaces, because the dynamic linker would have no way of knowing which definition to use at run time. The proper way to fix this is to define your symbols only once and declare them extern in the other files. Those of you who want to use the package without waiting for the fix can use R CMD INSTALL in the terminal, which will break at the same place. Then copy the link statement to the clipboard, go to the src directory, and paste /usr/local/bin/gcc -bundle -flat_namespace -undefined suppress - Xlinker -m -L/usr/local/lib -o kinship.so agfit6b.o bdsmatrix_index1.o bdsmatrix_index2.o bdsmatrix_index3.o bdsmatrix_prod.o bdsmatrix_prod2.o bdsmatrix_prod3.o chinv4.o chinv5.o cholesky4.o cholesky5.o chsolve4.o chsolve5.o coxfit6a.o coxfit6b.o coxfit6c.o gchol.o gchol_bds.o -F/Library/Frameworks/ R.framework/.. -framework R INserting -Xlinker -m forces the link. Apple frowns on this, for obvious reasons. Then run R CMD INSTALL again. On Jan 21, 2006, at 12:18 , jing hua zhao wrote: > Dear list members, > > I was suggested to contact this list for information concerning the > kinship > package that I ported a while ago. It is said that the package > won't compile > properly but I have no means to check for it, although it appears > there is > no problem with the checking from CRAN. However, that was for a > version I > posted almost a year ago. It works with no problem on Sun Solaris, > Red Het > Linux and Windows that I usually use. > > Any indication and suggestions for fixing the problems, if any, > would be > greatly appreciated. I can also be contacted via my e-mail account > from work > [EMAIL PROTECTED] > > Many thanks in advance, > > > Jing Hua Zhao > > _______________________________________________ > R-SIG-Mac mailing list > [email protected] > https://stat.ethz.ch/mailman/listinfo/r-sig-mac > === Jan de Leeuw; Distinguished Professor and Chair, UCLA Department of Statistics; Editor: Journal of Multivariate Analysis, Journal of Statistical Software US mail: 8125 Math Sciences Bldg, Box 951554, Los Angeles, CA 90095-1554 phone (310)-825-9550; fax (310)-206-5658; email: [EMAIL PROTECTED] .mac: jdeleeuw ++++++ aim: deleeuwjan ++++++ skype: j_deleeuw homepages: http://gifi.stat.ucla.edu ++++++ http://www.cuddyvalley.org ------------------------------------------------------------------------ ------------------------- No matter where you go, there you are. --- Buckaroo Banzai http://gifi.stat.ucla.edu/sounds/nomatter.au _______________________________________________ R-SIG-Mac mailing list [email protected] https://stat.ethz.ch/mailman/listinfo/r-sig-mac
