Use a different algorithm with the low entropy source for field trials.

chrome/common/metrics/entropy_provider.cc

Index: chrome/common/metrics/entropy_provider.cc
===================================================================
--- chrome/common/metrics/entropy_provider.cc	(revision 0)
+++ chrome/common/metrics/entropy_provider.cc	(revision 0)
@@ -0,0 +1,108 @@
+// Copyright (c) 2012 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 "chrome/common/metrics/entropy_provider.h"
+
+#include <algorithm>
+#include <limits>
+#include <vector>
+
+#include "base/logging.h"
+#include "base/rand_util.h"
+#include "base/sha1.h"
+#include "base/sys_byteorder.h"
+
+namespace internal {
+
+SeededRandGenerator::SeededRandGenerator(uint32 seed) {
+  mersenne_twister_.init_genrand(seed);
+}
+
+SeededRandGenerator::~SeededRandGenerator() {
+}
+
+uint32 SeededRandGenerator::operator()(uint32 range) {
+  // Based on base::RandGenerator().
+  DCHECK_GT(range, 0u);
+
+  // We must discard random results above this number, as they would
+  // make the random generator non-uniform (consider e.g. if
+  // MAX_UINT64 was 7 and |range| was 5, then a result of 1 would be twice
+  // as likely as a result of 3 or 4).

[SteveT, 2012/08/16 20:55:42]

nit: Looks like you can make better use of the 80 space line limit in the above
comment...

[Alexei Svitkine (slow), 2012/08/16 21:27:11]

This comment (and code) is copied out of base::RandGenerator(). I'd rather keep
them the same.

+  uint32 max_acceptable_value =
+      (std::numeric_limits<uint32>::max() / range) * range - 1;
+
+  uint32 value;
+  do {
+    value = mersenne_twister_.genrand_int32();
+  } while (value > max_acceptable_value);
+
+  return value % range;
+}
+
+uint32 HashName(const std::string& name) {
+  // SHA-1 is designed to produce a uniformly random spread in its output space,
+  // even for nearly-identical inputs.
+  unsigned char sha1_hash[base::kSHA1Length];
+  base::SHA1HashBytes(reinterpret_cast<const unsigned char*>(name.c_str()),
+                      name.size(),
+                      sha1_hash);
+
+  uint32 bits;
+  COMPILE_ASSERT(sizeof(bits) < sizeof(sha1_hash), need_more_data);
+  memcpy(&bits, sha1_hash, sizeof(bits));
+
+  return base::ByteSwapToLE32(bits);
+}
+
+}  // namespace internal
+
+SHA1EntropyProvider::SHA1EntropyProvider(const std::string& entropy_source)
+    : entropy_source_(entropy_source) {
+}
+
+SHA1EntropyProvider::~SHA1EntropyProvider() {
+}
+
+double SHA1EntropyProvider::GetEntropyForTrial(const std::string& trial_name) {
+  // SHA-1 is designed to produce a uniformly random spread in its output space,
+  // even for nearly-identical inputs, so it helps massage whatever client_id
+  // and trial_name we get into something with a uniform distribution, which
+  // is desirable so that we don't skew any part of the 0-100% spectrum.
+  std::string input(entropy_source_ + trial_name);
+  unsigned char sha1_hash[base::kSHA1Length];
+  base::SHA1HashBytes(reinterpret_cast<const unsigned char*>(input.c_str()),
+                      input.size(),
+                      sha1_hash);
+
+  uint64 bits;
+  COMPILE_ASSERT(sizeof(bits) < sizeof(sha1_hash), need_more_data);
+  memcpy(&bits, sha1_hash, sizeof(bits));
+  bits = base::ByteSwapToLE64(bits);
+
+  return base::BitsToOpenEndedUnitInterval(bits);
+}
+
+PermutedEntropyProvider::PermutedEntropyProvider(
+    uint16 low_entropy_source,
+    size_t low_entropy_source_max)
+        : low_entropy_source_(low_entropy_source),
+          low_entropy_source_max_(low_entropy_source_max) {
+}
+
+PermutedEntropyProvider::~PermutedEntropyProvider() {
+}
+
+double PermutedEntropyProvider::GetEntropyForTrial(
+    const std::string& trial_name) {
+  std::vector<uint16> mapping(low_entropy_source_max_);
+  for (size_t i = 0; i < mapping.size(); ++i)
+    mapping[i] = static_cast<uint16>(i);
+
+  internal::SeededRandGenerator generator(internal::HashName(trial_name));
+  std::random_shuffle(mapping.begin(), mapping.end(), generator);
+
+  return mapping[low_entropy_source_] /
+         static_cast<double>(low_entropy_source_max_);
+}