I have just sent that email to the help desk and realized that it might be better suited for the forum.
I have very recently started using HDF5 libraries with the Fortran interface. I am facing one issue I have not been able to solve, and I apologize in advance for the basic question I am about to ask.
I am trying to build a hdf5 file that has the following architecture (screenshot of the HDFView window):
I have very recently started using HDF5 libraries with the Fortran
interface.
I am facing one issue I have not been able to solve, and I apologize in
advance for the basic question I am about to ask.
I am trying to build a hdf5 file that has the following architecture
(screenshot of the HDFView window):
...
I am almost there except for the dataset 'SE_DATASET' which is attached
to the
group 'cycle-1'. As you can see, it has different columns: 'mass',
'radius',
'rho',etc...
The test code I wrote produces the following data:
...
It has the correct type except that the columns nametags are '0', '1'
and '2'.
So here is my question: how can I specify a name for these columns?
It is helpful to see the HDFView output for the two datasets.
I can see that the two datasets have different datatypes. The dataset you
created has a 64-bit floating point datatype, and the other dataset has a
compound datatype. (The text at the bottom of the HDFView window indicates
the datatype of the dataset.)
See this page for Fortran examples of creating datasets with different datatypes
(towards the end of the page):
thank you very much, I did not realize that.
I will have a look at the example you suggest.
Best,
Jean-Claude
···
On Jan 17, 2013, at 11:54 AM, Barbara Jones wrote:
Hello Jean-Claude,
I have very recently started using HDF5 libraries with the Fortran
interface.
I am facing one issue I have not been able to solve, and I apologize in
advance for the basic question I am about to ask.
I am trying to build a hdf5 file that has the following architecture
(screenshot of the HDFView window):
...
I am almost there except for the dataset 'SE_DATASET' which is attached
to the
group 'cycle-1'. As you can see, it has different columns: 'mass',
'radius',
'rho',etc...
The test code I wrote produces the following data:
...
It has the correct type except that the columns nametags are '0', '1'
and '2'.
So here is my question: how can I specify a name for these columns?
It is helpful to see the HDFView output for the two datasets.
I can see that the two datasets have different datatypes. The dataset you
created has a 64-bit floating point datatype, and the other dataset has a
compound datatype. (The text at the bottom of the HDFView window indicates
the datatype of the dataset.)
See this page for Fortran examples of creating datasets with different datatypes
(towards the end of the page):