Description
In a json file I have a 2d array
"myarr": [
[2.240000e+04, 0.000000e+00],
[1.250000e+02, 1.361000e+02]
],
the trailing comma is because I have other entries afterwards - and anyway my problem occurs at ccompilation time, NOT runtime, so it should not be a json-syntax problem.
I compile with ifort and I get the following error
error #8486: There is no matching specific subroutine for this type bound generic subroutine call. [GET]
call config%get('myarr', myarr, found)
"config" is a perfectly valid json_file object, and reading all the 1D arrays works perfectly. "found" is a LOGICAL.
In my fortran code I declare
real, allocatable, dimension(:, :) :: myarr
I tried also double precision, same error. Is my Intel compiler too old? It reads 2023.1.0.x, it is quite new in fact. It's difficult for me to change the compiler because that's part of a larger code which takes advantage of the intel compiler features. ChatGPT is not helpful, stackoverflow neither. What am I doing wrong?
Note that if I declare
real, allocatable, dimension(:) :: myarr
the compilation error vanishes, but a simple
write(*, *) myarr(1)
gives a segmentation fault.
write(*, *) SHAPE(myarr)
outputs
0
If instead I declare
double precision, allocatable, dimension(:) :: myarr
then the SHAPE is 2.