Bug in poscar parser with single character/number in lines

[TyBalduf]

Discussed in #206

^{Originally posted by PeterBlaha June 9, 2023}
Dear developers,
I guess I found a problem in the subroutine vasp.f90, or, more specifically in subroutine parse_line, when reading a poscar file (all versions 3.x; the problem did not exist in version 2.x).

The parser fails when one has only one atom type in the cell and the line contains only one single character/digit in the first column like for an elemental solid or for elements with a single character (eg. diamond Carbon). I include a poscar file for diamond, which shows the problem even twice:
C # it cannot read the name of this element
1
0.0000000000 1.8184794535 1.8184794535
1.8184794535 0.0000000000 1.8184794535
1.8184794535 1.8184794535 0.0000000000
2 # it cannot read the "2"
cartesian
0.4546198634 0.4546198634 0.4546198634
3.1823390437 3.1823390437 3.1823390437

As soon as I move the "C" to " C" and "2" to " 2", it works.
I guess, the initialization of variable iend=1 in subroutine parse_line should be changed to zero.

Best regards
Peter

[e-kwsm]

Duplicate of grimme-lab/mctc-lib#60?