RAPPOR implementation

components/rappor/byte_vector_utils.cc

Index: components/rappor/byte_vector_utils.cc
diff --git a/components/rappor/byte_vector_utils.cc b/components/rappor/byte_vector_utils.cc
new file mode 100644
index 0000000000000000000000000000000000000000..846227f9daad0008d688b5bbc464c599ba5993c5
--- /dev/null
+++ b/components/rappor/byte_vector_utils.cc
@@ -0,0 +1,161 @@
+// 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/byte_vector_utils.h"
+
+#include <string>
+
+#include "base/logging.h"
+#include "base/rand_util.h"
+#include "base/strings/string_number_conversions.h"
+#include "crypto/random.h"
+
+namespace rappor {
+
+ByteVector* ByteVectorOr(const ByteVector& lhs, ByteVector* rhs) {
+  DCHECK_EQ(lhs.size(), rhs->size());
+  for (size_t i = 0, len = lhs.size(); i < len; ++i) {
+    (*rhs)[i] = lhs[i] | (*rhs)[i];
+  }
+  return rhs;
+}
+
+ByteVector* ByteVectorMerge(const ByteVector& mask,
+                            const ByteVector& lhs,
+                            ByteVector* rhs) {
+  DCHECK_EQ(lhs.size(), rhs->size());
+  for (size_t i = 0, len = lhs.size(); i < len; ++i) {
+    (*rhs)[i] = (lhs[i] & ~mask[i]) | ((*rhs)[i] & mask[i]);
+  }
+  return rhs;
+}
+
+int CountBits(const ByteVector& vector) {
+  int bit_count = 0;
+  for(size_t i = 0; i < vector.size(); ++i) {

[Alexei Svitkine (slow), 2014/02/12 18:13:05]

Nit: space after for

[Steven Holte, 2014/02/12 22:28:50]

Done.

+    uint8_t byte = vector[i];
+    for (int j = 0; j < 8 ; ++j) {
+      if (byte & (1 << j))
+        bit_count++;
+    }
+  }
+  return bit_count;
+}
+
+ByteVectorGenerator::ByteVectorGenerator(size_t byte_count)
+    : byte_count_(byte_count) {}
+
+ByteVectorGenerator::~ByteVectorGenerator() {}
+
+ByteVector ByteVectorGenerator::GetRandomByteVector() {
+  ByteVector bytes(byte_count_);
+  crypto::RandBytes(&bytes[0], bytes.size());
+  return bytes;
+}
+
+ByteVector ByteVectorGenerator::GetWeightedRandomByteVector(
+    Probability probability) {
+  ByteVector bytes = GetRandomByteVector();
+  switch (probability) {
+    case PROBABILITY_75:
+      return *ByteVectorOr(GetRandomByteVector(), &bytes);
+    case PROBABILITY_50:
+      return bytes;
+  }
+  NOTREACHED();
+  return bytes;
+}
+
+HmacByteVectorGenerator::HmacByteVectorGenerator(
+    size_t byte_count,
+    const std::string& entropy_input,
+    const std::string& personalization_string)
+    : ByteVectorGenerator(byte_count),
+      hmac_(crypto::HMAC::SHA256),
+      value_(hmac_.DigestLength(), 0x01),
+      requested_bytes_(0) {
+  // HMAC_DRBG Instantiate Process
+  // 1. seed_material = entropy_input + nonce + personalization_string
+  // Note: We are using the 8.6.7 interpretation, where the entropy_input and
+  // nonce are acquired at the same time from the same source.
+  DCHECK_EQ(entropy_input.size(), kEntropyInputSize);

[Alexei Svitkine (slow), 2014/02/12 18:13:05]

Nit: expected value should be on the left

[Steven Holte, 2014/02/12 22:28:50]

Done.

+  std::string seed_material(entropy_input + personalization_string);
+  // 2. Key = 0x00 00...00
+  ByteVector key(hmac_.DigestLength(), 0x00);
+  // 3. V = 0x01 01...01
+  // (value_ in initializer list)
+
+  // 4. (Key, V) = HMAC_DRBG_Update(seed_material, Key, V)
+  Update(seed_material, key);
+}
+
+HmacByteVectorGenerator::~HmacByteVectorGenerator() {}
+
+const size_t HmacByteVectorGenerator::kEntropyInputSize = (128 / 8) * 3 / 2;
+
+// static
+std::string HmacByteVectorGenerator::GenerateEntropyInput() {
+  return base::RandBytesAsString(kEntropyInputSize);
+}
+
+void HmacByteVectorGenerator::Update(const std::string& provided_data,
+                                     const ByteVector& key1) {
+  std::string value1(value_.begin(), value_.end());
+  // HMAC_DRBG Update Process
+  // 1. K = HMAC(K, V || 0x00 || provided_data)
+  ByteVector key2(hmac_.DigestLength());
+  crypto::HMAC hmac1(crypto::HMAC::SHA256);
+  if (!hmac1.Init(std::string(key1.begin(), key1.end())))

[Alexei Svitkine (slow), 2014/02/12 18:13:05]

I'm not super happy with the amount of copying of data going on in this code
(note just this line, in general).

For example, this line copies the |key1| byte vector into a string. I think you
can make a string piece out of |key1| by doing:

base::StringPiece(key1.data(), key1.size())

[Steven Holte, 2014/02/12 22:28:50]

Using StringPiece.

+    NOTREACHED();
+  if (!hmac1.Sign(value1 + char(0x00) + provided_data, &key2[0], key2.size()))
+    NOTREACHED();
+  // 2. V = HMAC(K, V)
+  crypto::HMAC hmac2(crypto::HMAC::SHA256);
+  if (!hmac2.Init(std::string(key2.begin(), key2.end())))
+    NOTREACHED();
+  if (!hmac2.Sign(value1, &value_[0], value_.size()))
+    NOTREACHED();
+  // 3. If (provided_data = Null), then return K and V.
+  DCHECK_GE(provided_data.size(), kEntropyInputSize);
+  // 4. K = HMAC(K, V || 0x01 || provided_data)
+  std::string value2(value_.begin(), value_.end());
+  ByteVector key3(hmac_.DigestLength());
+  if (!hmac2.Sign(value2 + char(0x01) + provided_data, &key3[0], key3.size()))

[Alexei Svitkine (slow), 2014/02/12 18:13:05]

This logic seems very similar to what you do for the about (1 vs. 2). Can you
make a helper function for this in the anon namespace, so the logic isn't
repeated (with the different char value being a param)?

[Steven Holte, 2014/02/12 22:28:50]

Added helper function for K = HMAC(K, data)

[Alexei Svitkine (slow), 2014/02/12 22:42:36]

Thanks, this is much cleaner!

+    NOTREACHED();
+  // 5. V = HMAC(K, V)
+  if (!hmac_.Init(std::string(key3.begin(), key3.end())))
+    NOTREACHED();
+  if (!hmac_.Sign(value2, &value_[0], value_.size()))
+    NOTREACHED();
+}
+
+ByteVector HmacByteVectorGenerator::GetRandomByteVector() {
+  ByteVector bytes(byte_count_);
+  uint8_t* data = bytes.data();
+  size_t digest_length = hmac_.DigestLength();

[Alexei Svitkine (slow), 2014/02/12 18:13:05]

Nit: const

I'd also put this at the top of the function, so bytes, data and bytes_to_go are
together.

[Steven Holte, 2014/02/12 22:28:50]

Done.

+  size_t bytes_to_go = byte_count_;
+  while (bytes_to_go > 0) {
+    size_t requested_byte_in_digest = requested_bytes_ % digest_length;
+    if (requested_byte_in_digest == 0) {
+      // Do step 4.1 of the HMAC_DRBG Generate Process for more bits.
+      // V = HMAC(Key, V)
+      if (!hmac_.Sign(std::string(value_.begin(), value_.end()),
+                      &value_[0], value_.size())) {
+        NOTREACHED();
+      }
+    }
+    size_t n = std::min(bytes_to_go,
+                        digest_length - requested_byte_in_digest);
+    memcpy(data, &value_[requested_byte_in_digest], n);
+    data += n;
+    bytes_to_go -= n;
+    requested_bytes_ += n;
+    // Check max_number_of_bits_per_request from 10.1 Table 2
+    // max_number_of_bits_per_request == 2^19 bits == 2^16 bytes
+    DCHECK_LT(requested_bytes_, 1U << 16);
+  }
+  return bytes;
+}
+
+}  // namespace rappor