components/rappor/rappor_metric.cc

   1 // Copyright 2014 The Chromium Authors. All rights reserved.
   2 // Use of this source code is governed by a BSD-style license that can be
   3 // found in the LICENSE file.
   4
   5 #include "components/rappor/rappor_metric.h"
   6
   7 #include "base/logging.h"
   8
   9 namespace rappor {
  10
  11 RapporMetric::RapporMetric(const std::string& metric_name,
  12                            const RapporParameters& parameters,
  13                            int32_t cohort_seed)
  14     : metric_name_(metric_name),
  15       parameters_(parameters),
  16       bloom_filter_(parameters.bloom_filter_size_bytes,
  17                     parameters.bloom_filter_hash_function_count,
  18                     (cohort_seed % parameters.num_cohorts) *
  19                         parameters.bloom_filter_hash_function_count) {
  20   DCHECK_GE(cohort_seed, 0);
  21   DCHECK_LT(cohort_seed, RapporParameters::kMaxCohorts);
  22 }
  23
  24 RapporMetric::~RapporMetric() {}
  25
  26 void RapporMetric::AddSample(const std::string& str) {
  27   bloom_filter_.AddString(str);
  28 }
  29
  30 ByteVector RapporMetric::GetReport(const std::string& secret) const {
  31   // Generate a deterministically random mask of fake data using the
  32   // client's secret key + real data as a seed.  The inclusion of the secret
  33   // in the seed avoids correlations between real and fake data.
  34   // The seed isn't a human-readable string.
  35   const std::string personalization_string = metric_name_ +
  36       std::string(bytes().begin(), bytes().end());
  37   HmacByteVectorGenerator hmac_generator(bytes().size(), secret,
  38                                          personalization_string);
  39   const ByteVector fake_mask =
  40       hmac_generator.GetWeightedRandomByteVector(parameters().fake_prob);
  41   ByteVector fake_bits =
  42       hmac_generator.GetWeightedRandomByteVector(parameters().fake_one_prob);
  43
  44   // Redact most of the real data by replacing it with the fake data, hiding
  45   // and limiting the amount of information an individual client reports on.
  46   const ByteVector* fake_and_redacted_bits =
  47       ByteVectorMerge(fake_mask, bytes(), &fake_bits);
  48
  49   // Generate biased coin flips for each bit.
  50   ByteVectorGenerator coin_generator(bytes().size());
  51   const ByteVector zero_coins =
  52       coin_generator.GetWeightedRandomByteVector(parameters().zero_coin_prob);
  53   ByteVector one_coins =
  54       coin_generator.GetWeightedRandomByteVector(parameters().one_coin_prob);
  55
  56   // Create a randomized response report on the fake and redacted data, sending
  57   // the outcome of flipping a zero coin for the zero bits in that data, and of
  58   // flipping a one coin for the one bits in that data, as the final report.
  59   return *ByteVectorMerge(*fake_and_redacted_bits, zero_coins, &one_coins);
  60 }
  61
  62 }  // namespace rappor