RAPPOR implementation

components/rappor/bloom_filter.cc

+// Copyright 2014 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "components/rappor/bloom_filter.h"
+
+#include "base/logging.h"
+#include "third_party/smhasher/src/MurmurHash3.h"
+
+namespace {
+
+// Distinct seeds are used to create unique hash functions for the Bloom filter.
+// These seeds were chosen at random, and then checked to ensure that when
+// used in 128 bit bloom filter that each bit can be set many elements in the
+// Alexa 10000.
+const uint32_t kHashSeeds[] = {0xd123957d, 0x6752fc9b, 0xcb6a0102, 0x1a82ea95};
+
+uint32_t MurmurHash3String(const std::string& str, uint32_t seed) {
+  uint32_t output = 0;
+  // This function is optimized for x86_32, but should work on any platform.
+  MurmurHash3_x86_32(str.data(), str.size(), seed, &output);
+  return output;
+}
+
+}  // namespace
+
+namespace rappor {
+
+BloomFilter::BloomFilter(uint32_t bytes_size, uint32_t hash_function_count)
+    : bytes_(bytes_size), hash_function_count_(hash_function_count) {
+  DCHECK_GT(bytes_size, 0u);
+  DCHECK_LE(hash_function_count, arraysize(kHashSeeds));
+}
+
+BloomFilter::~BloomFilter() {}
+
+void BloomFilter::AddString(const std::string& str) {
+  for (size_t i = 0; i < hash_function_count_; ++i) {
+    uint32_t index = MurmurHash3String(str, kHashSeeds[i]);
+    uint32_t byte_index = (index / 8) % bytes_.size();
+    uint32_t bit_index = index % 8;
+    bytes_[byte_index] |= 1 << bit_index;
+  }
+}
+
+void BloomFilter::AddStrings(const std::vector<std::string>& strings) {
+  for (size_t i = 0; i < strings.size(); ++i)
+    AddString(strings[i]);
+}
+
+}  // namespace rappor