When compiling on Arch linux, using the current gcc, the program fails to build. The problem has been described in a comment to an unofficial build-script of GraceGTK for Arch:
It seems that bad programming practice makes now trouble with gcc-10.
To remove the bug, I have concatenated loessf.f and supp.f into a single file called loessf.f90 and changed old FORTRAN fixed form in more modern Fortran90 free form.
This now produces warnings when compiling with the older gcc compiler of my Debian-9.6.
The trouble is that the second argument of subroutine ehg184 is declared as a rank 2 array but ehg184 is called with a scalar or a rank 1 array. Considering that ehg184 is used only to display messages, I have replaced the first type of calls by a direct use of the PRINT instruction and created a new ehg180 subroutine to deal with rank 1 arrays.
Now compilation with my Debian 9 no longer produces warnings.
I have pushed theses changes in the GIT repository (master branch).
Please, let me know if I have solved your problem.
Regards,
P. Vincent
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Vishnu,
It seems that bad programming practice makes now trouble with gcc-10.
To remove the bug, I have concatenated loessf.f and supp.f into a single file called loessf.f90 and changed old FORTRAN fixed form in more modern Fortran90 free form.
This now produces warnings when compiling with the older gcc compiler of my Debian-9.6.
The trouble is that the second argument of subroutine ehg184 is declared as a rank 2 array but ehg184 is called with a scalar or a rank 1 array. Considering that ehg184 is used only to display messages, I have replaced the first type of calls by a direct use of the PRINT instruction and created a new ehg180 subroutine to deal with rank 1 arrays.
Now compilation with my Debian 9 no longer produces warnings.
I have pushed theses changes in the GIT repository (master branch).
Please, let me know if I have solved your problem.
Regards,
P. Vincent
Hi VIncent,
There are still loads of Warnings that appear, but the code does compile.
Thanks a lot! Do you think you could make a minor release 1.0.3 for this?