Build issues with gfnff_api

[ER-Elivian]

I have got following error while building from source with cmake:

/home/USERNAME/crest-3.0.1/src/calculator/gfnff_api.F90:75:25:

   75 |       & ichrg=chrg,print=pr,iostat=io,iunit=iunit)
      |                         1
Error: Keyword argument ‘print’ at (1) is not in the procedure
/home/USERNAME/crest-3.0.1/src/calculator/gfnff_api.F90:80:25:

   80 |       & ichrg=chrg,print=pr,iostat=io,iunit=iunit)
      |                         1
Error: Keyword argument ‘print’ at (1) is not in the procedure

function gfnff_initialize (as I see from _deps/gfnff-src/src/gfnff_interface.f90) doesn't have "print" and "iunit" arguments)

After removing these arguments (which is improperly, but "iunit" doesn't described in gfnff_api.f90 and "pr" looks like "be verbose" flag). I've got many errors from tblite_api.f90 telling "Error: Missing actual argument for argument ‘error’" at new_gfn1_calculator, new_gfn2_calculator, new_ipea1_calculator and others

To Reproduce
I am using gfortran (Ubuntu 13.3.0-6ubuntu2~24.04.1) and gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0 to build. That issue arises both in crest 3.0.1 and 3.0.2

git clone https://github.com/crest-lab/crest.git
cd crest
git submodule init
git submodule update
git submodule update --remote
export FC=gfortran CC=gcc
cmake -B _build
make -C _build

[pprcht]

that is probably because I updated the code in gfnff, try switching into the gfnff subproject directory and checking out the older version with tag v0.0.1

[victorolet]

I couldn’t find a v0.0.1 tag in the gfnff submodule, so I just switched to v2.12

That sorted out the API mismatch with CREST.

I also hit a separate build issue caused by a really long string in solvtool.f90 (line 3161). It was triggering a line truncation / unterminated character constant error with strict compiler flags. Splitting/shortening that line fixed the build.

After that, everything compiled fine.