csvdiff.py

import csv, os, sys


"""
A bash script I also looked at for a while, but ended up being less useful than just a filtered view of the CSVs.

function diff-clients() {
	xsv select '!"Icon","Last seen","Channel","Signal"' "$1" | xsv sort -s MAC -o /tmp/before.csv
	xsv select '!"Icon","Last seen","Channel","Signal"' "$2" | xsv sort -s MAC -o /tmp/after.csv
	csv-diff --key MAC /tmp/before.csv /tmp/after.csv
	xsv table /tmp/before.csv -o /tmp/before.txt
	xsv table /tmp/after.csv -o /tmp/after.txt
	delta -s /tmp/before.txt /tmp/after.txt
}
"""

"""
Usage: python3 csvdiff.py <path to before csv> <path to after csv> [--added]

Output is sent to stdout. By default, output entries present in "before" but missing from "after".
Pass `--added` to reverse this and output entries present in "after" that weren't in "before".

"""
def main():
    before_path = sys.argv[1]
    after_path = sys.argv[2]
    show_added = "--added" in sys.argv

    with open(before_path, "r") as before_file, open(after_path, "r") as after_file:
        before_rows = list(csv.DictReader(before_file))
        after_rows = list(csv.DictReader(after_file))

    writer = csv.DictWriter(sys.stdout, before_rows[0].keys())
    writer.writeheader()

    if show_added:
        before_macs = {row["MAC"] for row in before_rows}
        added = [row for row in after_rows if row["MAC"] not in before_macs]
        writer.writerows(added)
    else:
        after_macs = {row["MAC"] for row in after_rows}
        missing = [row for row in before_rows if row["MAC"] not in after_macs]
        writer.writerows(missing)


if __name__ == "__main__":
    main()