the real trick is probably to just fix liten or to write my own dedupe program. this is not bad, though.

doop.py

#!/usr/bin/env python
"""

doop - a script for moving duplicates from a camera uploads folder from a report
generated by [liten](https://code.google.com/p/liten/)

    "DOOP? What's that?" -- philip j. fry

(c) 2014 marcos.a.ojeda <marcos@generic.cx>
MIT licensed (lol)
"""
import argparse
import csv
import hashlib
import itertools
import os
import os.path
import re

import unittest

def pick_dupe(path1, path2):
    dupe_re = re.compile(r'-(\d+)$')
    p1 = os.path.splitext(path1)[0]
    p2 = os.path.splitext(path2)[0]

    m1, m2 = dupe_re.search(p1), dupe_re.search(p2)
    if m1 or m2:
        if m1 and m2:
            # find biggest match
            if m1.group(1) > m2.group(1):
                return path1
            else:
                return path2
        else:
            if m1:
                return path1
            else:
                return path2
    return False

def dedupe(dest_dir, report_path):
    """deduplicates files in a liten report"""
    with open(report_path) as f:
        data = csv.DictReader(f)
        # since liten reports are like
        # [row1, row2, header, row4, row5, header, ... ]
        # this returns
        # [[row1, row2], [row4, row5], ...]
        filtered = itertools.compress(data, itertools.cycle([1,1,0]))
        paths = list(itertools.imap(lambda x: x['     Path'].strip(),
            filtered))
        pairs = zip(paths[::2], paths[1::2])

        for a, b in pairs:
            if not os.path.exists(a) or not os.path.exists(b):
                continue
            # confirm the files are dupes still
            md5 = lambda x: hashlib.md5(open(x, 'rb').read()).hexdigest()[:16]
            if md5(a) == md5(b):
                dupe = pick_dupe(a, b)
                print "moving %s to %s" % (os.path.basename(dupe), dest_dir)
                os.rename(dupe, os.path.join(dest_dir, os.path.basename(dupe)))


class PickTest(unittest.TestCase):
    # ridiculous, i know, but i get easily confused when doing these things
    def setUp(self):
        self.a = "foo.jpg"
        self.b = "foo.1.jpg"
        self.c = "foo.1-1.jpg"
        self.d = "foo.1-2.jpg"

    def test_no_dupe(self):
        self.assertFalse(pick_dupe(self.a, self.b))
        self.assertFalse(pick_dupe(self.b, self.a))

    def test_dupe(self):
        self.assertEqual(self.c, pick_dupe(self.b, self.c))
        self.assertEqual(self.c, pick_dupe(self.c, self.b))

    def test_two_dupes(self):
        self.assertEqual(self.d, pick_dupe(self.c, self.d))
        self.assertEqual(self.d, pick_dupe(self.d, self.c))


def main():
    parser = argparse.ArgumentParser(prog='doop',
        description='Deduplicate a dropbox camera uploads folder')

    parser.add_argument('report_path', type=str,
        help='The full path to the liten report')
    parser.add_argument('destination', type=str,
        help='A destination directory for duplicate files')

    args = parser.parse_args()

    dedupe(args.destination, args.report_path)

if __name__ == '__main__':
    main()