Re: R, HDF5, Swarm output serialization - Part II

[Marcus G. Daniels]

>>>>> "NEN" == Norberto Eiji Nawa <address@hidden> writes:

NEN> Since the very beginning, I suspected that the cause of the
NEN> problem is actually the way I am serializing data from my
NEN> model. Each time the model outputs data, it builds a "compound
NEN> object", with one field to each variable that is saved, and
NEN> pushes it into an HDF5 archiver. For 2000 iterations, there are
NEN> 2000 objects, each one of them containing 7 to 10 float
NEN> variables. 

I made a new version that may fix the problem. In any case, it fixes
a very real bug, and is an important upgrade.

ftp://ftp.swarm.org/pub/swarm/src/testing/hdf5_1.2.tar.gz.

If you're lucky to have more time than me (or at least are in a newer
phase in your day), let me know if this new version helps at all.  If
not, I'll dig in to your examples.  The compound data type <-> data
frame handling is different code in the R/HDF5 module, so it is quite
plausible a bug is still lurking.  Let me know..


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.