find duplicates in a (sorted) tsv by multiple keys, example usage mysql -Be 'select * from table order by field3,field4'|dedup.py field3,field4

dedup.py

#!/usr/bin/python

import sys

header = None
keys = []
firstmatch = None

for line in sys.stdin.readlines():
    s = line[:-1].split('\t')
    if not header:
        header = s
        for a in sys.argv[1].split(','):
            keys.append(header.index(a))
        continue
    if firstmatch:
        equal = True
        for key in keys:
            if (firstmatch[key] != s[key]):
                equal = False
                break
        if equal:
            print s[0]

    firstmatch = s