Create a .csv file containing the distance matrix of an optimized molecule from a Gaussian output file.

matrixparse

#!/usr/bin/env bash
# Generates a standardized table of bond lengths and bond angles
# from a gaussian .log file.

if [[ $# > 0 ]]
then
    # Find distance Matrix
    awk '/^ *1 *2 *3 *4 *5 *$/ {p=1}; p; /Stoich/ {p=0}' $1 |
    # Extract lines with distance data
    egrep '^[ \t]*[0-9]+[ \t]*[A-Z][ \t]*([-+]?[0-9]*\.?[0-9]+[ \t]*)' |
    # Build 2D distance array with awk
    awk '
BEGIN {
    old = 0
    offset = 0
}
{
    col = 0
    current = $1
    if (current < old){
        offset = offset + 5
    }
    if (current == 1){
        offset = 0
    }
    while (col++ < NF-2) {
          distances[current][col+offset] = $(col+2)
    }
    old = current
}
END{ for(i in distances){
    for(j=1; j <= length(distances[i]); j++){
        {printf "%f,", distances[i][j]}
    }
    print ""
    }
}
' | awk '{ print length, $0 }' | sort -n | awk '{$1=""; print $0}' #Sort by line length
else
    echo 'Enter a Gaussian .log file to parse'
fi