RAPPOR implementation

components/rappor/bloom_filter.cc

Index: components/rappor/bloom_filter.cc
diff --git a/components/rappor/bloom_filter.cc b/components/rappor/bloom_filter.cc
new file mode 100644
index 0000000000000000000000000000000000000000..d3234ef31f3cde5421d1d6c9fbd89a2eb2e6f175
--- /dev/null
+++ b/components/rappor/bloom_filter.cc
@@ -0,0 +1,34 @@
+// 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/City.h"
+
+namespace rappor {
+
+BloomFilter::BloomFilter(uint32_t bytes_size,
+                         uint32_t hash_function_count,
+                         uint32_t hash_seed_offset)
+    : bytes_(bytes_size),
+      hash_function_count_(hash_function_count),
+      hash_seed_offset_(hash_seed_offset) {
+  DCHECK_GT(bytes_size, 0u);
+}
+
+BloomFilter::~BloomFilter() {}
+
+void BloomFilter::AddString(const std::string& str) {
+  for (size_t i = 0; i < hash_function_count_; ++i) {
+    uint32_t index =
+        CityHash64WithSeed(str.data(), str.size(), hash_seed_offset_ + i);

[Ilya Sherman, 2014/02/13 01:39:03]

Please document why CityHash is used here, rather than a more commonly used hash
function.  What are the constraints that are important for this application, on
which CityHash shines?

[Steven Holte, 2014/02/13 05:11:12]

Added a comment which says "because it's supported in Dremel".

+    // Note that the "bytes" are uint8_t, so they are always 8-bits.
+    uint32_t byte_index = (index / 8) % bytes_.size();
+    uint32_t bit_index = index % 8;
+    bytes_[byte_index] |= 1 << bit_index;
+  }
+}
+
+}  // namespace rappor