[Bioc-devel] dyn.load() problem with RHDF5 package.

[Robert Gentleman]

Hi Mark,


Mark.Oberfield at noaa.gov wrote:
> All,
> 
> I am hoping to use R to analyze climatological datasets packaged in
> HDF5 format.  After consulting the R FAQ and various documentation
> provided, I downloaded the HDF5 package and successfully
> installed it. Unfortunately, I get an error after downloading the
> file from this web site:
> 
> Error in
> hdf5load("http://www.mdl.nws.noaa.gov/~avnfps/data/hdf5/KGUY.hd5") :
>   can't handle hdf class 10
> 
   You should bring that up with that maintainer, and on R-devel.

> Okay. Turn to Plan B
> 
> R-project web page also makes references to RHDF5 package on the
> Bioconductor web site.  After much searching and googling, I find
> ver. 1.6.0, which is the latest I can find.  Compilation fails on
> hdf_dataset.c and hdf_subassign.c due to missing prototype information
> for the vectorSubscript() routine.  I get that fixed and install rhdf5
> package into R. Now when I attempt to load it, I get the following
> error:
> 
>> library(rhdf5)
> Error in dyn.load(file, ...) :
>   unable to load shared library
> '/some/directory/R-2.6.2/library/rhdf5/libs/rhdf5.so':
>   /some/directory/R-2.6.2/library/rhdf5/libs/rhdf5.so: undefined symbol:
> Rf_mat2indsub
> Error in library(rhdf5) : .First.lib failed for 'rhdf5'
> 
> Well, for some reason, libR.so isn't built which contains the routine
> subscript.c which has the desired function. So I get that created and
> placed into $R_HOME/lib so that LD_LIBRARY_PATH technique will work.
> 
> $ nm -AP libR.so | grep Rf_mat2indsub
> libR.so: Rf_mat2indsub t 001b12d8 00000241
> 
> Unfortunately, even with the new library, libR.so, I get the same
> result as before.
> 
> This as far as I got with RHDF5 package -- can't get the libR.so
> library (which I had to build manually) to load.
> 
> Why is this so difficult?  If RHDF5 package is obsolete (or not
> maintained), why doesn't the Bioconductor and R pages state that
> somewhere or stop making references to it at least.  I web searched
> for RHDF5 and examined the various "hits" (and the R and Bioconductor
> mailing lists) and did not come across anything to suggest this in
> writing. It would a matter of courtesy for the former RHDF5 
> maintainers to say that it (or any package) is depreciated for the
> record, so others do not waste their time trying to get it to work.

   The rhdf5 package is only in BioC devel, not in release, and some 
minor checking of the build reports would have clearly indicated that it 
does not currently build.  We have not had the resources recently (nor 
any practical uses) to keep up with HDF5 development. Nor have we 
completely abandoned the notion that the code might be resurrected, or 
may be useful to someone who does want to make it work with current 
releases.

I am not sure how you found it, but Googling for rhd5 took me to:
http://bioconductor.fhcrc.org/checkResults/2.2/bioc-LATEST/rhdf5/liverpool-buildsrc.html
which I hope is sufficiently explicit as to suggest that it is not working.

   You might look at Netcdf, as it seems to both be getting more use and 
to have in its most recent release I think can also handle HDF5, of 
course the R packages may not have kept up with that development.
http://www.unidata.ucar.edu/software/netcdf/netcdf-4/


  best wishes
    Robert




> 
> However, if RHDF5 package *not* obsolete and is being maintained, then
> can someone tell me how these library functions can get resolved? Or
> point me to a more up-to-date version that is compatible with R 2.6.2?
> 
> Thanks!
> 
> mgo
> 
> _______________________________________________
> Bioc-devel at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
> 

-- 
Robert Gentleman, PhD
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M2-B876
PO Box 19024
Seattle, Washington 98109-1024
206-667-7700
rgentlem at fhcrc.org