Compilation Error : Attribute comes to real to complex (vice versa) in Riksfr.f90

[Camillever]

Dear Mister Gallovic,

I'm a french student working on a Python tool generating simulated ground motion by stochastic method and developed recently by a PhD student in France.
In this tool, the source model is generated by your RIKsfr fortran code.

I don't know if you can help me with my issue (about the second command) but I had somehow some errors, refering to the code source mainly, while compiling.

Files in my folder RIK, initially : Time_2d.c ; crustal.dat ; RIKsrf.in ; RIKsrf.f90.

Commands :

• The first one :
  gcc -c -Wall -DNO-IEEE_INFINITY Time_2d.c
  Error :
  
  'INFINITY' in Time_2d.c is also a attribute of math.h in fortran, there is an incoherence for the program.

Resolution : I decided to change all 'INFINITY' with a 'INFINITY_PERSO'. The error no longer occurs when executing the command. Time_2d.o is created.

• The second one :
  gfortran -Ofast -fopenmp -o RIKsrf2 RIKsrf.f90 Time_2d.o
  Error :
  

The second warning warns that the type of one of the arguments of the functions fourn and rlft3 of RIKsrf.f90 is not good. We expect a complex number while we provide a real number (or vice versa).

I checked into the source code of RIFsfr.f90 and the warning seems right and also a little bit incoherent. I don't want to modify the source code because I absolutely do not master what you have developed in your code and I'm just a casual fortran user.

Versions:
gcc version 7.4
gfortran version 8.3

Thank you

Best regards

Camille

[Camillever]

I tried also with a older version of gfortran (version 7). Maybe it's less responsive to this type of error, I guessed.
But unfortunately, I just get the same error about the source code of .f90 with the second command.

Note: I can't install intel fort compiler on my device because of a proxy at my work place, and compile with $ifort.

[fgallovic]

Dear Camille,
Thank you for using the code.
I am aware of the warnings. They do not affect the code functioning, so I consider them unimportant.
Best regards,
Frantisek