Unify fcidump 1e and 0e integral spacings with 2e integrals

[brianz98]

Description

The fcidump.py file writes out 2e integrals by calling fcidump_helper.cc, in the %28.20E%4d%4d%4d%4d (cpp) format, and in fcidump.py the additional 1- and 0- electron integrals are written in the {:29.20E} {:4d} {:4d} {:4d} {:4d} (Python) format. The whitespace in between the Python format string causes the 1/0e integrals fields to be 1 wider than the 2e integrals, causing problems in reading the file. E.g. in Fortran the 2e integrals can be read normally by READ(*,'(1X,G28.20E2,4I4)') Z,I,J,K,L, but will fail to read the 1/0e integrals properly.

Todos

• Update fcidump.py
• Update fcidump test outputs

Checklist

• Tests added for any new features
• All or relevant fraction of full tests run

Status

• Ready for review
• Ready for merge

[loriab]

good catch, thank you!

(for anyone else looking for the spacing changes, go to the end of the intdump.ref files.)

[robertodr]

Good catch! The tests read in the FCIDUMP file without relying on the format and that's why we missed this.