atlas for R1.1.1 and R1.2 different?
Bingo, before I tried to install R1.2.0 I installed integrate2.2.2 which indeed needed atlas2-dev. R1.1.1 went smoothly, and also did R1.2.0, until I reinstalled integrate2.2.2 which - I did not know - was already shipped with R1.2.0. Marcus ps. to Peter Daalgard: I installed R1.2.0 and R1.1.1 in different directories and even removed the old installation before doing a make install again...
Kurt Hornik wrote:
I suspect that your R 1.1.1 is from a .deb file which requires package atlas2, and the R 1.2.0 is compiled from source but cannot find atlas (look at config.log) because package atlas2-dev is not installed. Bingo?
Peter Dalgaard BSA wrote:
Marcus Eger <marcus.eger at physik.uni-marburg.de> writes:
I guess I know where the problem comes from... I reinstalled integrate2.2.2 not knowing that integrate was already contained in R1.2.0. Doing a make install of R1.2.0 solved the problem.
Er? It isn't...
I did install.packages('integrate') from inside R. However, if you
were loading an older .so file, which was linked against atlas into an
R which wasn't, you might get into trouble.... Conversely, make
install may cause you to install R on top of your old R 1.1 and get
you your *previous* integrate.so. So I'd try reinstalling the
integrate package before claiming success!
-- +------------------------------------------------------- | Marcus Eger +-------------------------------------------------------- | E-Mail: eger.m at gmx.de (NEW) | marcus.eger at physik.uni-marburg.de (OLD) | WWW: http://neuro.physik.uni-marburg.de/~eger (NEW) +-------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: https://stat.ethz.ch/pipermail/r-help/attachments/20001114/da7c57df/attachment.html