Bug when using Al+++

[moro...@gmail.com]

Hello everyone,

I just encountered what seems like a strange bug. I was interested in the ionisation fractions of Aluminium, so I ran a network looking like this (simplified):

@var: T = Tgas
 
@format:idx,R,R,P,rate
1,H+,E,H,auto
2,He+,E,He,auto
3,He++,E,He+,auto
4,Al+,E,Al, auto
5,Al++,E,Al+, auto
6,Al+++,E,Al++, auto
 
@format:idx,R,R,P,P,P,rate
7,H,E,H+,E,E, auto
8,He,E,He+,E,E, auto
9,He+,E,He++,E,E, auto
10,Al,E,Al+,E,E, auto
11,Al+,E,Al++,E,E, auto
12,Al++,E,Al+++,E,E, auto

And ran krome with:

python3.8 krome -n networks/react_test_Al -useN -compact

And now at line 864 of krome_all.f90, there is

get_imass_sqrt(10) = 148793396170.d09375    !AL+++  

And this "d" in the number makes the compilation crash.

Is it possible that there is a typo somewhere in Krome, or did I mess up something?

Thank you for your help,

All the best,

Valentin

[tommaso grassi]

Dear Valentin,

thanks for using KROME and for spotting this bug.

I have just pushed a new version to the KROME repository with the bug fixed (commit d397881).

https://bitbucket.org/tgrassi/krome/commits/d3978814b9fc896eb71806d43f7dfd9fe3a2e7e0

Let me know if it works.

best,

-tg