count or sum unix command

count

#!/usr/bin/env python
"""Count or sum, while uniquing rows, without full sort of data

By using --key-fields, can also show example row that has some particular fields.

(This was much simpler when it just counted!)
"""

import sys
import argparse


def slice_type(x):
    if '-' in x:
        start, _, stop = x.partition('-')
        return slice(int(start) - 1, int(stop))
    return slice(int(x) - 1, int(x))


ap = argparse.ArgumentParser()
ap.add_argument('-k', '--key-fields', default=slice(None, None), type=slice_type, help='Key field or contiguous fields for tab separated input')
ap.add_argument('-s', '--sum-field', type=int, help='Perform summation over this field')
args = ap.parse_args()
key_fields = args.key_fields
sum_field = getattr(args, 'sum_field', None)
if sum_field is not None:
    sum_field -= 1

res = {}
total = 0

for l in sys.stdin:
    fields = l.rstrip('\n\r').split('\t')
    key = tuple(fields[key_fields])
    if sum_field is None:
        val = 1
    else:
        val = fields[sum_field]
        if '.' in val:
            val = float(val)
        else:
            val = int(val)
    total += val
    if key in res:
        val += res[key][1]
    res[key] = (l, val)
print(res)

for k in sorted(res.iterkeys()):
    sys.stdout.write('{1}\t{0}'.format(*res[k]))
    
sys.stderr.write('Total\t{}\n'.format(total))