2010/8/13, Rodel Piano <[email protected]>:
> Hi,
>
> Has anyone here Tried to select a record from a compound dataset using a
> keyword.
>
> I know about hyperslabs, but what i want is something like
>
> Select * from table1 where field1='test'

This is not currently supported by HDF5 itself, but there are several
tools floating around that allows to do that following different
techniques.  The important thing here is not having to load the
complete dataset in memory (if you want to deal with really large
tables, that is).

The approach that follows PyTables (disclaimer: I'm the main author)
is to read an hyperslab (normally it should fit in CPU cache), and
apply the selection for it.  Consolidate the outcome.  Repeat this
operation for all the interesting hyperslabs, and you are done.

Hope this helps,

-- 
Francesc Alted

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org

Reply via email to