renjin
renjin copied to clipboard
update some stats package functions to fix org.bioconductor:affy
GNU R version 3.3.1
@psolaimani is this still relevant?
the problem was that some packages were failing due to difference in optim
function which has been manually implemented in java and probably has changed since in GNU R. The goal was to get the C optim
code transpiled like we did with C code in packages. But this came to halt when I learned that transpiling C code of core functions is not similar to what happen with packages.
SETCADR
and related functions are in the meantime implemented, so this can be removed.
.External2
was used somewhere in functions that use optim
and still needs to be implemented.
I'm not sure if transpiling optim
was a good idea from the begining. I think we can better update the current implementation to match the current GNU R version in its formals and return.
What do you mean by
transpiling C code of core functions is not similar to what happen with packages.
?