xref: /aosp_15_r20/external/rappor/bin/sum_bits.py (revision 2abb31345f6c95944768b5222a9a5ed3fc68cc00) 
1#!/usr/bin/python
2#
3# Copyright 2014 Google Inc. All rights reserved.
4#
5# Licensed under the Apache License, Version 2.0 (the "License");
6# you may not use this file except in compliance with the License.
7# You may obtain a copy of the License at
8#
9#     http://www.apache.org/licenses/LICENSE-2.0
10#
11# Unless required by applicable law or agreed to in writing, software
12# distributed under the License is distributed on an "AS IS" BASIS,
13# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14# See the License for the specific language governing permissions and
15# limitations under the License.
16
17"""
18Read the RAPPOR'd values on stdin, and sum the bits to produce a Counting Bloom
19filter by cohort.  This can then be analyzed by R.
20"""
21
22import csv
23import sys
24
25import rappor
26
27
28def SumBits(params, stdin, stdout):
29  csv_in = csv.reader(stdin)
30  csv_out = csv.writer(stdout)
31
32  num_cohorts = params.num_cohorts
33  num_bloombits = params.num_bloombits
34
35  sums = [[0] * num_bloombits for _ in xrange(num_cohorts)]
36  num_reports = [0] * num_cohorts
37
38  for i, row in enumerate(csv_in):
39    try:
40      (user_id, cohort, unused_bloom, unused_prr, irr) = row
41    except ValueError:
42      raise RuntimeError('Error parsing row %r' % row)
43
44    if i == 0:
45      continue  # skip header
46
47    cohort = int(cohort)
48    num_reports[cohort] += 1
49
50    if not len(irr) == params.num_bloombits:
51      raise RuntimeError(
52          "Expected %d bits, got %r" % (params.num_bloombits, len(irr)))
53    for i, c in enumerate(irr):
54      bit_num = num_bloombits - i - 1  # e.g. char 0 = bit 15, char 15 = bit 0
55      if c == '1':
56        sums[cohort][bit_num] += 1
57      else:
58        if c != '0':
59          raise RuntimeError('Invalid IRR -- digits should be 0 or 1')
60
61  for cohort in xrange(num_cohorts):
62    # First column is the total number of reports in the cohort.
63    row = [num_reports[cohort]] + sums[cohort]
64    csv_out.writerow(row)
65
66
67def main(argv):
68  try:
69    filename = argv[1]
70  except IndexError:
71    raise RuntimeError('Usage: sum_bits.py <params file>')
72  with open(filename) as f:
73    try:
74      params = rappor.Params.from_csv(f)
75    except rappor.Error as e:
76      raise RuntimeError(e)
77
78  SumBits(params, sys.stdin, sys.stdout)
79
80
81if __name__ == '__main__':
82  try:
83    main(sys.argv)
84  except RuntimeError, e:
85    print >>sys.stderr, e.args[0]
86    sys.exit(1)
87