gmx dump --sys no longer lists all atoms in system
## Summary The -h output for `gmx dump` includes: ``` -[no]sys (no) List the atoms and bonded interactions for the whole system instead of for each molecule type ``` `gmx dump`'s `-sys` flag behavior seems to have changed between the 2019 and 2024 versions for the following file: [gmx.tpr](/uploads/eaa8954a8a710837331e722a11544705/gmx.tpr). It no longer lists atoms for the whole system, and instead only does so for the molecule type. I think the new behavior isn't consistent with the flag's description. ## Exact steps to reproduce ### Using gromacs 2019 from bioconda: Conda package info: `gromacs 2019.1 h04f5b5a_2 bioconda` running `gmx dump -s gmx.tpr -sys` yields an atom section that lists all atoms in the whole system ``` topology: name="Clusters of 2 water molecules extracted from liquid, solid, and gas phase" atoms: atom (8): atom[ 0]={type= 0, typeB= 0, ptype= Atom, m= 1.59994e+01, q= 0.00000e+00, mB= 1.59994e+01, qB= 0.00000e+00, resind= 0, atomnumber= 8} atom[ 1]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 0, atomnumber= 1} atom[ 2]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 0, atomnumber= 1} atom[ 3]={type= 2, typeB= 2, ptype= VSite, m= 0.00000e+00, q=-5.40000e-01, mB= 0.00000e+00, qB=-5.40000e-01, resind= 0, atomnumber= 0} atom[ 4]={type= 0, typeB= 0, ptype= Atom, m= 1.59994e+01, q= 0.00000e+00, mB= 1.59994e+01, qB= 0.00000e+00, resind= 1, atomnumber= 8} atom[ 5]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 1, atomnumber= 1} atom[ 6]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 1, atomnumber= 1} atom[ 7]={type= 2, typeB= 2, ptype= VSite, m= 0.00000e+00, q=-5.40000e-01, mB= 0.00000e+00, qB=-5.40000e-01, resind= 1, atomnumber= 0} ``` ### Using gromacs 2024 from conda-forge: Conda package info `gromacs 2024.2 nompi_h9c65464_104 conda-forge` running `gmx dump -s gmx.tpr -sys` yields an atom section that only lists the atoms for each molecule type ``` cmap moltype (0): name="SOL" atoms: atom (4): atom[ 0]={type= 0, typeB= 0, ptype= Atom, m= 1.59994e+01, q= 0.00000e+00, mB= 1.59994e+01, qB= 0.00000e+00, resind= 0, atomnumber= 8} atom[ 1]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 0, atomnumber= 1} atom[ 2]={type= 1, typeB= 1, ptype= Atom, m= 1.00800e+00, q= 1.02000e+00, mB= 1.00800e+00, qB= 1.02000e+00, resind= 0, atomnumber= 1} atom[ 3]={type= 2, typeB= 2, ptype= VSite, m= 0.00000e+00, q=-5.40000e-01, mB= 0.00000e+00, qB=-5.40000e-01, resind= 0, atomnumber= 0} ``` ## If this is a bug, (1) what happens, and (2) what did you expect to happen? I expected the output to include all 8 particles in the system instead of just the 4 from a single instance of the molecule type. ## Relevant input files, logs and/or screenshots The tpr file above was generated using the 2019 gromacs package, since that can be processed by both 2019 and 2024 `gmx dump`. Here's an equivalent one generated for the same system by the 2024 package in case it's useful: [gmx_2024.tpr](/uploads/46ee1312fae43634629ab46961b8893a/gmx_2024.tpr)
issue