Paths for downloading data from openadas seem to be out of date in some cases

[eldond]

I was attempting
aurora.atomic.get_atom_data('C', ['pls']), which failed because it attempted to download https://open.adas.ac.uk/download/adf11//pls/pls_C_14.dat, which doesn't exist. I looked up PLS for carbon manually at open.adas.ac.uk and found that the filename is pls96_c.dat.

I changed

files["C"]["pls"] = "pls_C_14.dat"

to

files["C"]["pls"] = "pls96_c.dat"

on line 229 of adas_files.py and then aurora.atomic.get_atom_data('C', **['pls']) completed without an error.

[fsciortino]

Hi @eldond , thanks for opening the issue. The default there is for a better file while is not yet publicly available through ADAS. On GA servers, if Aurora is correctly set up (@odstrcilt could advise here), the default files should be available. We could update the defaults to openadas files, but some do not exist publicly or the public ones are pretty bad...

[odstrcilt]

Some files can be found by setting the variable
AURORA_ADAS_DIR to /fusion/projects/omfit-results/aurora_dir/

pls are SXR filtered radiation files, which are specific for every diagnostic. Number 14 is for DIII-D SXR diagnostic with 127um Be filter and number 15 is for 12.7um Be filter. These files are located here
/fusion/projects/codes/strahl/public/atomdat/newdat/

I guess, pls96_c.dat is a file from JET, which had 250 and 350um Be filters. They cannot be used on DIII-D.

IMPRAD should load the files from the right location.

[eldond]

Yes, if I run on iris, it does work properly. When I tried running on my laptop, it didn't work because I didn't have a copy of the files. Maybe an error message could explain limitations of openadas that you just mentioned?

[odstrcilt]

I have tried to make the error message a bit more useful
#92