A Wednesday 17 October 2007, Francesc Altet escrigué: > A Wednesday 17 October 2007, Darren Dale escrigué: > > > Mmm, I'm curious about which datatype should be causing this. > > > Can you try applying the attached patch and send me the output? > > [snip] > > > datatype--> f > > datatype--> <---- > > Edatatype--> f <---- > > Well, this seems to correspond to a complex type. Apparently the go > and back trip to/from a netcdf3 file for complex types doesn't work > as it should on your machine.
Yesterday I responded while in a hurry, so I've double checked this. The expected datatype here was 'c' that certainly does not correspond to 'complex' but to 'string' datatype. So, it seems that some problem should be expected on your machine when converting native NetCDF3 files with 'string' types on it to HDF5 via tables.netcdf3 module. > The fact is that the netcdf3 module is becoming more and more > dangerous as the Numeric package is ageing. Perhaps we should start > to think in deprecate it (unless the original contributor, Jeff > Whitaker, would like to contribute a complete replacement of Numeric > by NumPy for it). Well, I was somewhat wrong here too :(. Most of the netcdf3 module has been migrated to use NumPy internally and the only place where Numeric is required is when using the tables.netcdf3.h5tonc and tables.netcdf3.nctoh5 functions (the native NetCDF3 files <==> HDF5 conversor). So, if you are not requering doing this conversion, your installation should be fairly safe. > > The manual does not indicate that Numeric is required for netcdf3. > > You are right. I've added a ticket on this: > http://www.pytables.org/trac/ticket/124 Now that I think more about this, perhaps it is not worth mentioning that you need Numeric as it is implicit if you are going to use Scientific.IO.NetCDF (it seems that it still needs Numeric, although I thought that the author was thinking to replace it by NumPy). See more details at: http://www.pytables.org/trac/ticket/124 Cheers, -- >0,0< Francesc Altet http://www.carabos.com/ V V Cárabos Coop. V. Enjoy Data "-" ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Pytables-users mailing list Pytables-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/pytables-users
