Question about Espresso write_input() implementation

[hironori-kondo]

In espresso.py, I see that write_input() has different functionality based on the choice of binary. For example, the phonon package relies on write_espresso_ph() as written in ASE.

Why does the pw.x binary go to ASE's write() instead of its Espresso/pw-specific write_espresso_in() in the io/espresso.py file, where write_espresso_ph() is?

[Andrew-S-Rosen]

@nerdwarrior: Thank you for your question.

Admittedly, I am not sure! I am not an Espresso user myself, and the original implementation was made by @tomdemeyere.

If you think there are advantages for using write_espresso_in, please feel to propose a change. That said, does ASE's write() not default to write_espresso_in()?

[tomdemeyere]

ASE write will use write_espresso_in, I would say that it does not really matter? The espresso ph is not supported by the write function, we have to call it directly

[hironori-kondo]

I missed that---makes sense! Thank you!

[tomdemeyere]

No worries, if you have any other question about the espresso interface do not hesitate to ask!