You mentioned the phypepat.html tutorial in your question. The phypepat.html tutorial shows how to write by pattern in Parallel HDF5.
By "chunk" did you mean "by pattern", or are you looking at another tutorial example?
I took the example in this tutorial, made it a 'serial' example, and added code at the end that re-opens the file and reads
by pattern. See the attached example. (Note that reading and writing are very similar!)
Using hyperslab selection can be confusing. We have a very simple example in the HDF5 Tutorial that you can play
around with to get a feel for how changing the offset/count/stride/block affects what gets written. See this topic in the
"Learning the Basics" tutorial:
I attached the h5_subset.c example from this tutorial.
There is also an example, h5ex_hyper.c, on this page, that you may want to look at:
If you still have questions or would like to see more examples, just let me know! I'd be happy to help.
hyperslab-pat.c (3.98 KB)
h5_subset.c (5.32 KB)
The HDF Group Helpdesk
From: Hdf-forum [mailto:firstname.lastname@example.org] On Behalf Of Maxime Boissonneault
Sent: Tuesday, February 03, 2015 3:28 PM
To: HDF Users Discussion List
Subject: [Hdf-forum] Question for reading data by pattern
I am trying to read data from an HDF5 file, by chunk. I wrote it following the tutorial here :
However, I have trouble reading the data the same way. Any chance someone has the corresponding example for reading by chunk ?
Analyste de calcul - Calcul Québec, Université Laval Ph. D. en physique
Hdf-forum is for HDF software users discussion.