PVer4: RICE decode bytes to list of uint32 or 4-byte hash prefixes

components/safe_browsing_db/v4_rice.cc

+// Copyright 2016 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 "base/logging.h"
+#include "base/strings/stringprintf.h"
+#include "components/safe_browsing_db/v4_rice.h"
+
+using google::protobuf::RepeatedField;
+
+namespace safe_browsing {
+
+namespace {
+
+const unsigned int kMaxBitIndex = 8 * sizeof(uint32_t);
+
+void GetBytesFromUInt32(uint32_t word, char* bytes) {
+  const size_t mask = 0xFF;
+  bytes[0] = (char)(word & mask);
+  bytes[1] = (char)((word >> 8) & mask);
+  bytes[2] = (char)((word >> 16) & mask);
+  bytes[3] = (char)((word >> 24) & mask);
+}
+
+}  // namespace
+
+// static
+V4DecodeResult V4RiceDecoder::DecodeIntegers(const uint32_t first_value,
+                                             const int32_t rice_parameter,
+                                             const int32_t num_entries,
+                                             const std::string& encoded_data,
+                                             RepeatedField<uint32_t>* out) {
+  DCHECK(out);
+
+  if (num_entries < 0) {
+    return NUM_ENTRIES_NEGATIVE_FAILURE;

[palmer, 2016/07/27 23:39:40]

Could this be a DCHECK (in addition to a production check as now)?

[vakh (use Gerrit instead), 2016/07/28 07:26:31]

Done.

+  }
+
+  out->Reserve(num_entries + 1);
+  uint32_t last_value = first_value;
+  out->Add(last_value);
+  if (num_entries > 0) {

[palmer, 2016/07/27 23:39:40]

Nit: You could save a level of indentation by putting something like this:

    if (num_entries == 0)
      return DECODE_SUCCESS;

But, up to you.

[vakh (use Gerrit instead), 2016/07/28 07:26:31]

Done.

+    V4RiceDecoder decoder(rice_parameter, num_entries, encoded_data);
+    while (decoder.HasAnotherValue()) {
+      uint32_t offset;
+      V4DecodeResult result = decoder.GetNextValue(&offset);
+      if (result != DECODE_SUCCESS) {
+        return result;
+      }
+      last_value += offset;

[palmer, 2016/07/27 23:39:40]

Could this arithmetic overflow? (Does it matter if it does?)

[vakh (use Gerrit instead), 2016/07/28 07:26:31]

Unexpected. Added a DCHECK.

+      out->Add(last_value);
+    }
+  }
+  return DECODE_SUCCESS;
+}
+
+// static
+V4DecodeResult V4RiceDecoder::DecodeBytes(const uint32_t first_value,
+                                          const int32_t rice_parameter,
+                                          const int32_t num_entries,
+                                          const std::string& encoded_data,
+                                          std::string* out) {
+  DCHECK(out);
+
+  if (num_entries < 0) {
+    return NUM_ENTRIES_NEGATIVE_FAILURE;
+  }
+
+  out->reserve((num_entries + 1) * 4);
+  char bytes[4];
+  uint32_t last_value = first_value;
+  GetBytesFromUInt32(last_value, bytes);
+  out->append(bytes, 4);
+  if (num_entries > 0) {
+    V4RiceDecoder decoder(rice_parameter, num_entries, encoded_data);
+    while (decoder.HasAnotherValue()) {
+      uint32_t offset;
+      V4DecodeResult result = decoder.GetNextValue(&offset);
+      if (result != DECODE_SUCCESS) {
+        return result;
+      }
+      last_value += offset;
+
+      GetBytesFromUInt32(last_value, bytes);
+      out->append(bytes, 4);
+    }
+  }
+  return DECODE_SUCCESS;
+}
+
+V4RiceDecoder::V4RiceDecoder(const int rice_parameter,
+                             const int num_entries,
+                             const std::string& encoded_data)
+    : rice_parameter_(rice_parameter),
+      num_entries_(num_entries),
+      data_(encoded_data),
+      current_word_(0) {
+  DCHECK_LE(0, num_entries_);
+  DCHECK_LE(2u, rice_parameter_);
+  DCHECK_GE(28u, rice_parameter_);
+
+  data_byte_index_ = 0;
+  current_word_bit_index_ = kMaxBitIndex;
+}
+
+V4RiceDecoder::~V4RiceDecoder() {}
+
+bool V4RiceDecoder::HasAnotherValue() const {
+  return num_entries_ > 0;
+}
+
+V4DecodeResult V4RiceDecoder::GetNextValue(uint32_t* value) {
+  if (!HasAnotherValue()) {
+    return DECODE_NO_MORE_ENTRIES_FAILURE;
+  }
+
+  V4DecodeResult result;
+  uint32_t q = 0;
+  uint32_t bit;
+  do {
+    result = GetNextBits(1, &bit);
+    if (result != DECODE_SUCCESS) {
+      return result;
+    }
+    q += bit;
+  } while (bit);
+  uint32_t r = 0;
+  result = GetNextBits(rice_parameter_, &r);
+  if (result != DECODE_SUCCESS) {
+    return result;
+  }
+
+  *value = (q << rice_parameter_) + r;
+  num_entries_--;
+  return DECODE_SUCCESS;
+}
+
+V4DecodeResult V4RiceDecoder::GetNextWord(uint32_t* word) {
+  if (data_byte_index_ >= data_.size()) {
+    return DECODE_RAN_OUT_OF_BITS_FAILURE;
+  }
+
+  const size_t mask = 0xFF;
+  *word = (data_[data_byte_index_] & mask);
+  data_byte_index_++;
+  current_word_bit_index_ = 0;
+
+  if (data_byte_index_ < data_.size()) {
+    *word |= ((data_[data_byte_index_] & mask) << 8);
+    data_byte_index_++;
+
+    if (data_byte_index_ < data_.size()) {
+      *word |= ((data_[data_byte_index_] & mask) << 16);
+      data_byte_index_++;
+
+      if (data_byte_index_ < data_.size()) {
+        *word |= ((data_[data_byte_index_] & mask) << 24);
+        data_byte_index_++;
+      }
+    }
+  }
+
+  *word = *word & 0xFFFFFFFF;

[palmer, 2016/07/27 23:39:40]

Is this line necessary? (Maybe it is, I'm not sure.)

[vakh (use Gerrit instead), 2016/07/28 07:26:31]

Done. You're right. I had implemented this method slightly differently earlier
and it needed that mask at the end. Not anymore.

+  return DECODE_SUCCESS;
+}
+
+V4DecodeResult V4RiceDecoder::GetNextBits(unsigned int num_requested_bits,
+                                          uint32_t* x) {
+  if (num_requested_bits > kMaxBitIndex) {
+    NOTREACHED();
+    return DECODE_REQUESTED_TOO_MANY_BITS_FAILURE;
+  }
+
+  if (current_word_bit_index_ == kMaxBitIndex) {
+    V4DecodeResult result = GetNextWord(&current_word_);
+    if (result != DECODE_SUCCESS) {
+      return result;
+    }
+  }
+
+  unsigned int num_bits_left_in_current_word =
+      kMaxBitIndex - current_word_bit_index_;
+  if (num_bits_left_in_current_word >= num_requested_bits) {
+    // All the bits that we need are in current_word_

[palmer, 2016/07/27 23:39:40]

Nit:

    // All the bits that we need are in |current_word_|.

[vakh (use Gerrit instead), 2016/07/28 07:26:31]

Done.

+    GetBitsFromCurrentWord(num_requested_bits, x);
+  } else {
+    // |current_word_| contains fewer bits than we need so we store the current
+    // value of |current_word_| in |lower|, then read in a new |current_word_|,
+    // and then pick the remaining bits from the new value of |current_word_|.
+    uint32_t lower = current_word_;
+    // Bits we need after we read all of |current_word_|
+    unsigned int num_bits_from_next_word =
+        num_requested_bits - num_bits_left_in_current_word;
+    uint32_t upper;
+    V4DecodeResult result = GetNextWord(&current_word_);
+    if (result != DECODE_SUCCESS) {
+      return result;
+    }
+    GetBitsFromCurrentWord(num_bits_from_next_word, &upper);
+    *x = (upper << (num_requested_bits - num_bits_from_next_word)) | lower;
+  }
+  return DECODE_SUCCESS;
+}
+
+void V4RiceDecoder::GetBitsFromCurrentWord(unsigned int num_requested_bits,
+                                           uint32_t* x) {
+  uint32_t mask = 0xFFFFFFFF >> (kMaxBitIndex - num_requested_bits);
+  *x = current_word_ & mask;
+  current_word_ = current_word_ >> num_requested_bits;
+  current_word_bit_index_ += num_requested_bits;
+};
+
+std::string V4RiceDecoder::DebugString() const {
+  return base::StringPrintf(
+      "current_word_: %x; data_byte_index_; %x, "
+      "current_word_bit_index_: %x; rice_parameter_: %x",
+      current_word_, data_byte_index_, current_word_bit_index_,
+      rice_parameter_);
+}
+
+std::ostream& operator<<(std::ostream& os, const V4RiceDecoder& rice_decoder) {
+  os << rice_decoder.DebugString();
+  return os;
+}
+
+}  // namespace safe_browsing