Hello everyone,
I am writing a program in HDF4. In one of my files, I have a few datasets
that have the same name, although in different groups. Currently, I have
some highly abstracted functions I have written that ultimately call the
HDF4 function SDnametoindex(). An SD name is passed into my function which
is passed to SDnametoindex which will select the appropriate dataset. The
problem I have now is that there are now multiple datasets with the same
name, and SDnametoindex will only select the first one. I need the ability
to specify exactly which dataset I want. I was wondering if there was a way
to give an absolute path to the dataset which would then give me the
appropriate index. As I understand it, SDnametoindex does not allow for
absolute paths.
I could use SDnametoindicies() to keep track of which of the same-name
datasets I've already visited, but changing much of my function would
require rewriting large portions of my entire program and I want to avoid
that if there is a simpler solution. My function is easy to use in that all
the caller needs to provide is the name of the dataset, not any dataset or
group identifiers. I would like to keep such ease of use if possible.
Thanks.
Regards,
Landon Clipp
Hi Landon,
What you encountered is the result of a deficiency in HDF4 and there is no way for your function to return the correct data set without having more information. You can see more information about this problem in the Know Problem section of the release notes, at:
https://support.hdfgroup.org/ftp/HDF/releases/HDF4.2.12/src/hdf-4.2.12-RELEASE.txt
So, adding more argument to your function is unavoidable, I'm afraid.
In HDF5, datasets can be specified with absolute paths, so you won't have this problem. Perhaps, converting the h4 files to h5 using the h4toh5 tool is an option. Unfortunately, you'll still need to rewrite your program.
Thank you,
Binh-Minh
ยทยทยท
________________________________
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org> on behalf of Landon Clipp <clipp2@illinois.edu>
Sent: Tuesday, November 22, 2016 3:59 PM
To: HDF Users Discussion List
Subject: [Hdf-forum] HDF4 Multiple datasets with same name
Hello everyone,
I am writing a program in HDF4. In one of my files, I have a few datasets that have the same name, although in different groups. Currently, I have some highly abstracted functions I have written that ultimately call the HDF4 function SDnametoindex(). An SD name is passed into my function which is passed to SDnametoindex which will select the appropriate dataset. The problem I have now is that there are now multiple datasets with the same name, and SDnametoindex will only select the first one. I need the ability to specify exactly which dataset I want. I was wondering if there was a way to give an absolute path to the dataset which would then give me the appropriate index. As I understand it, SDnametoindex does not allow for absolute paths.
I could use SDnametoindicies() to keep track of which of the same-name datasets I've already visited, but changing much of my function would require rewriting large portions of my entire program and I want to avoid that if there is a simpler solution. My function is easy to use in that all the caller needs to provide is the name of the dataset, not any dataset or group identifiers. I would like to keep such ease of use if possible. Thanks.
Regards,
Landon Clipp
