the algorithm of generating structures

Dear Axel,

I am confused about the algorithm of generating structures. Say, I construct a wurtzite-type lattice of four atoms. The maps code extends up the database of structures step by step, and 10 or 12 atoms/unit cell will turn up. Does it mean that the structures generated by the code are not necessarily the supercell of the lattice which I defined?
Also, I meet the same situation when I am constructing SQS. The input lattice is wurtzite while the outcome of MCSQS is orthorhombic.
I’m wondering can you give me some advice or references on the algorithm?
Thank you!

Generating 10 and 12/atom per unit cell out of a wurtzite would be a bug (that I have not seen before). Could you send me your input file so I can debug?

About your 2nd question: it is possible to generate orthorombic supercells out of a hexagonal primitive lattice. No problem here.

Dear Axel, the input files are in the attachment, and thanks a lot. I should like to express my appreciation.

I tried your input file. I only get 4,8,12, etc. atoms per supercell. There is no bug here.
Did you really get 10 atoms?
(12 is actually fine since 4x3=12)

I feel terrible sorry that I mistaked somewhere. I apologize for the inconvenience.

No problem - keep in mind that a "supercell" does not need to be a simple multiple along each axes. So any nb of atom of the form 4*n (in your case, since your primitive cell has 4 atoms) is fine.