I too would love to see VESTA able to handle Quantum Espresso inputs and outputs.I see this thread is now a bit old, so possibly this is not so easy.
In order to ease development, it may make sense to make VESTA work with just the triclinic case to start with. This would enable all crystals. (It has a slightly different format for each crystal system).
For triclinic you simply set:
ibrav=4,
celldm(1) = a in bohr (0.529177249 angstroms),
celldm(2) = b/a,
celldm(3) = c/a,
celldm(4) = cos(bc),
celldm(5) = cos(ac),
celldm(6) = cos(ab),
nat = # of atoms (Fe2O3 would be 5)
ntyp = # of types of atoms (Fe2O3 would be 2)
and then use crystal coordinates for atomic positions, where x,y,z are in units of a,b,c. For example an atom in the middle of the unit cell is at coordinate 0.5, 0.5, 0.5:
ATOMIC_POSITIONS crystal
Fe x y z
O x y z
etc.
If we just had that it would be a miracle!
The ability to read the output files would be the next step, but even being able to write the input files would be a huge boon.
Thanks,
Zack