Replace strings in input files¶
The “pam –replace” is a fantastic option which we encourage you to use it. It replaces any string in input files by user defined string.
**DIRAC .WAVE FUNCTION **HAMILTONIAN .hamiltonian **WAVE FUNCTION .SCF *END OF
DIRAC . . C 1 10. 1 Ne 0.0 0.0 0.0 LARGE BASIS basis FINISH
now run it with:
pam --replace hamiltonian=LEVY-LEBLOND --replace basis=cc-pVDZ
it will write to the output file:
you get the point: you can now run whole tables with just two input files!
A nice application of this feature is to make a PES scan, written as a script:
for r in `seq 1.00 0.05 2.00`; do pam --replace distance=$r .. done
Transfer renamed files¶
The pam scripts enables putting/getting renamed files. Thanks to this feature we are able to save DIRAC working files under own original names and ensure their renaming upon moving them to and from the scratch directory.
This feature is great for running many calculation at the same time without risking that files will be overwritten by concurrent runs.
Here we extract the DFCOEF file from the scratch directory and save it as DFCOEF.tl2.v3z in the home directory:
pam --get "DFCOEF=DFCOEF.tl2.v3z"
In next run we copy files to the scratch directory (where calculations occur) with their proper names:
pam --put "DFCOEF.tl2.v3z=DFCOEF" pam --copy="/home/milias/my_scratch/Tl2_fscc02/MDCINT.18corr_el.v3z=MDCINT"