Export of internal Abseil changes.

--
7a6ff16a85beb730c172d5d25cf1b5e1be885c56 by Laramie Leavitt <lar@google.com>:

Internal change.

PiperOrigin-RevId: 254454546

--
ff8f9bafaefc26d451f576ea4a06d150aed63f6f by Andy Soffer <asoffer@google.com>:

Internal changes

PiperOrigin-RevId: 254451562

--
deefc5b651b479ce36f0b4ef203e119c0c8936f2 by CJ Johnson <johnsoncj@google.com>:

Account for subtracting unsigned values from the size of InlinedVector

PiperOrigin-RevId: 254450625

--
3c677316a27bcadc17e41957c809ca472d5fef14 by Andy Soffer <asoffer@google.com>:

Add C++17's std::make_from_tuple to absl/utility/utility.h

PiperOrigin-RevId: 254411573

--
4ee3536a918830eeec402a28fc31a62c7c90b940 by CJ Johnson <johnsoncj@google.com>:

Adds benchmark for the rest of the InlinedVector public API

PiperOrigin-RevId: 254408378

--
e5a21a00700ee83498ff1efbf649169756463ee4 by CJ Johnson <johnsoncj@google.com>:

Updates the definition of InlinedVector::shrink_to_fit() to be exception safe and adds exception safety tests for it.

PiperOrigin-RevId: 254401387

--
2ea82e72b86d82d78b4e4712a63a55981b53c64b by Laramie Leavitt <lar@google.com>:

Use absl::InsecureBitGen in place of std::mt19937
in tests absl/random/...distribution_test.cc

PiperOrigin-RevId: 254289444

--
fa099e02c413a7ffda732415e8105cad26a90337 by Andy Soffer <asoffer@google.com>:

Internal changes

PiperOrigin-RevId: 254286334

--
ce34b7f36933b30cfa35b9c9a5697a792b5666e4 by Andy Soffer <asoffer@google.com>:

Internal changes

PiperOrigin-RevId: 254273059

--
6f9c473da7c2090c2e85a37c5f00622e8a912a89 by Jorg Brown <jorg@google.com>:

Change absl::container_internal::CompressedTuple to instantiate its
internal Storage class with the name of the type it's holding, rather
than the name of the Tuple.  This is not an externally-visible change,
other than less compiler memory is used and less debug information is
generated.

PiperOrigin-RevId: 254269285

--
8bd3c186bf2fc0c55d8a2dd6f28a5327502c9fba by Andy Soffer <asoffer@google.com>:

Adding short-hand IntervalClosed for IntervalClosedClosed and IntervalOpen for
IntervalOpenOpen.

PiperOrigin-RevId: 254252419

--
ea957f99b6a04fccd42aa05605605f3b44b1ecfd by Abseil Team <absl-team@google.com>:

Do not directly use __SIZEOF_INT128__.

In order to avoid linker errors when building with clang-cl (__fixunsdfti, __udivti3 and __fixunssfti are undefined), this CL uses ABSL_HAVE_INTRINSIC_INT128 which is not defined for clang-cl.

PiperOrigin-RevId: 254250739

--
89ab385cd26b34d64130bce856253aaba96d2345 by Andy Soffer <asoffer@google.com>:

Internal changes

PiperOrigin-RevId: 254242321

--
cffc793d93eca6d6bdf7de733847b6ab4a255ae9 by CJ Johnson <johnsoncj@google.com>:

Adds benchmark for InlinedVector::reserve(size_type)

PiperOrigin-RevId: 254199226

--
c90c7a9fa3c8f0c9d5114036979548b055ea2f2a by Gennadiy Rozental <rogeeff@google.com>:

Import of CCTZ from GitHub.

PiperOrigin-RevId: 254072387

--
c4c388beae016c9570ab54ffa1d52660e4a85b7b by Laramie Leavitt <lar@google.com>:

Internal cleanup.

PiperOrigin-RevId: 254062381

--
d3c992e221cc74e5372d0c8fa410170b6a43c062 by Tom Manshreck <shreck@google.com>:

Update distributions.h to Abseil standards

PiperOrigin-RevId: 254054946

--
d15ad0035c34ef11b14fadc5a4a2d3ec415f5518 by CJ Johnson <johnsoncj@google.com>:

Removes functions with only one caller from the implementation details of InlinedVector by manually inlining the definitions

PiperOrigin-RevId: 254005427

--
2f37e807efc3a8ef1f4b539bdd379917d4151520 by Andy Soffer <asoffer@google.com>:

Initial release of Abseil Random

PiperOrigin-RevId: 253999861

--
24ed1694b6430791d781ed533a8f8ccf6cac5856 by CJ Johnson <johnsoncj@google.com>:

Updates the definition of InlinedVector::assign(...)/InlinedVector::operator=(...) to new, exception-safe implementations with exception safety tests to boot

PiperOrigin-RevId: 253993691

--
5613d95f5a7e34a535cfaeadce801441e990843e by CJ Johnson <johnsoncj@google.com>:

Adds benchmarks for InlinedVector::shrink_to_fit()

PiperOrigin-RevId: 253989647

--
2a96ddfdac40bbb8cb6a7f1aeab90917067c6e63 by Abseil Team <absl-team@google.com>:

Initial release of Abseil Random

PiperOrigin-RevId: 253927497

--
bf1aff8fc9ffa921ad74643e9525ecf25b0d8dc1 by Andy Soffer <asoffer@google.com>:

Initial release of Abseil Random

PiperOrigin-RevId: 253920512

--
bfc03f4a3dcda3cf3a4b84bdb84cda24e3394f41 by Laramie Leavitt <lar@google.com>:

Internal change.

PiperOrigin-RevId: 253886486

--
05036cfcc078ca7c5f581a00dfb0daed568cbb69 by Eric Fiselier <ericwf@google.com>:

Don't include `winsock2.h` because it drags in `windows.h` and friends,
and they define awful macros like OPAQUE, ERROR, and more. This has the
potential to break abseil users.

Instead we only forward declare `timeval` and require Windows users
include `winsock2.h` themselves. This is both inconsistent and poor QoI, but so
including 'windows.h' is bad too.

PiperOrigin-RevId: 253852615
GitOrigin-RevId: 7a6ff16a85beb730c172d5d25cf1b5e1be885c56
Change-Id: Icd6aff87da26f29ec8915da856f051129987cef6

1 files changed, 246 insertions, 0 deletions

diff --git a/absl/random/discrete_distribution_test.cc b/absl/random/discrete_distribution_test.cc
new file mode 100644
index 000000000000..7296f0ac226a
--- /dev/null
+++ b/absl/random/discrete_distribution_test.cc
@@ -0,0 +1,246 @@
+// Copyright 2017 The Abseil Authors.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      https://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "absl/random/discrete_distribution.h"
+
+#include <cmath>
+#include <cstddef>
+#include <cstdint>
+#include <iterator>
+#include <numeric>
+#include <random>
+#include <sstream>
+#include <string>
+#include <vector>
+
+#include "gmock/gmock.h"
+#include "gtest/gtest.h"
+#include "absl/base/internal/raw_logging.h"
+#include "absl/random/internal/chi_square.h"
+#include "absl/random/internal/distribution_test_util.h"
+#include "absl/random/internal/sequence_urbg.h"
+#include "absl/random/random.h"
+#include "absl/strings/str_cat.h"
+#include "absl/strings/strip.h"
+
+namespace {
+
+template <typename IntType>
+class DiscreteDistributionTypeTest : public ::testing::Test {};
+
+using IntTypes = ::testing::Types<int8_t, uint8_t, int16_t, uint16_t, int32_t,
+                                  uint32_t, int64_t, uint64_t>;
+TYPED_TEST_SUITE(DiscreteDistributionTypeTest, IntTypes);
+
+TYPED_TEST(DiscreteDistributionTypeTest, ParamSerializeTest) {
+  using param_type =
+      typename absl::discrete_distribution<TypeParam>::param_type;
+
+  absl::discrete_distribution<TypeParam> empty;
+  EXPECT_THAT(empty.probabilities(), testing::ElementsAre(1.0));
+
+  absl::discrete_distribution<TypeParam> before({1.0, 2.0, 1.0});
+
+  // Validate that the probabilities sum to 1.0. We picked values which
+  // can be represented exactly to avoid floating-point roundoff error.
+  double s = 0;
+  for (const auto& x : before.probabilities()) {
+    s += x;
+  }
+  EXPECT_EQ(s, 1.0);
+  EXPECT_THAT(before.probabilities(), testing::ElementsAre(0.25, 0.5, 0.25));
+
+  // Validate the same data via an initializer list.
+  {
+    std::vector<double> data({1.0, 2.0, 1.0});
+
+    absl::discrete_distribution<TypeParam> via_param{
+        param_type(std::begin(data), std::end(data))};
+
+    EXPECT_EQ(via_param, before);
+  }
+
+  std::stringstream ss;
+  ss << before;
+  absl::discrete_distribution<TypeParam> after;
+
+  EXPECT_NE(before, after);
+
+  ss >> after;
+
+  EXPECT_EQ(before, after);
+}
+
+TYPED_TEST(DiscreteDistributionTypeTest, Constructor) {
+  auto fn = [](double x) { return x; };
+  {
+    absl::discrete_distribution<int> unary(0, 1.0, 9.0, fn);
+    EXPECT_THAT(unary.probabilities(), testing::ElementsAre(1.0));
+  }
+
+  {
+    absl::discrete_distribution<int> unary(2, 1.0, 9.0, fn);
+    // => fn(1.0 + 0 * 4 + 2) => 3
+    // => fn(1.0 + 1 * 4 + 2) => 7
+    EXPECT_THAT(unary.probabilities(), testing::ElementsAre(0.3, 0.7));
+  }
+}
+
+TEST(DiscreteDistributionTest, InitDiscreteDistribution) {
+  using testing::Pair;
+
+  {
+    std::vector<double> p({1.0, 2.0, 3.0});
+    std::vector<std::pair<double, size_t>> q =
+        absl::random_internal::InitDiscreteDistribution(&p);
+
+    EXPECT_THAT(p, testing::ElementsAre(1 / 6.0, 2 / 6.0, 3 / 6.0));
+
+    // Each bucket is p=1/3, so bucket 0 will send half it's traffic
+    // to bucket 2, while the rest will retain all of their traffic.
+    EXPECT_THAT(q, testing::ElementsAre(Pair(0.5, 2),  //
+                                        Pair(1.0, 1),  //
+                                        Pair(1.0, 2)));
+  }
+
+  {
+    std::vector<double> p({1.0, 2.0, 3.0, 5.0, 2.0});
+
+    std::vector<std::pair<double, size_t>> q =
+        absl::random_internal::InitDiscreteDistribution(&p);
+
+    EXPECT_THAT(p, testing::ElementsAre(1 / 13.0, 2 / 13.0, 3 / 13.0, 5 / 13.0,
+                                        2 / 13.0));
+
+    // A more complex bucketing solution: Each bucket has p=0.2
+    // So buckets 0, 1, 4 will send their alternate traffic elsewhere, which
+    // happens to be bucket 3.
+    // However, summing up that alternate traffic gives bucket 3 too much
+    // traffic, so it will send some traffic to bucket 2.
+    constexpr double b0 = 1.0 / 13.0 / 0.2;
+    constexpr double b1 = 2.0 / 13.0 / 0.2;
+    constexpr double b3 = (5.0 / 13.0 / 0.2) - ((1 - b0) + (1 - b1) + (1 - b1));
+
+    EXPECT_THAT(q, testing::ElementsAre(Pair(b0, 3),   //
+                                        Pair(b1, 3),   //
+                                        Pair(1.0, 2),  //
+                                        Pair(b3, 2),   //
+                                        Pair(b1, 3)));
+  }
+}
+
+TEST(DiscreteDistributionTest, ChiSquaredTest50) {
+  using absl::random_internal::kChiSquared;
+
+  constexpr size_t kTrials = 10000;
+  constexpr int kBuckets = 50;  // inclusive, so actally +1
+
+  // 1-in-100000 threshold, but remember, there are about 8 tests
+  // in this file. And the test could fail for other reasons.
+  // Empirically validated with --runs_per_test=10000.
+  const int kThreshold =
+      absl::random_internal::ChiSquareValue(kBuckets, 0.99999);
+
+  std::vector<double> weights(kBuckets, 0);
+  std::iota(std::begin(weights), std::end(weights), 1);
+  absl::discrete_distribution<int> dist(std::begin(weights), std::end(weights));
+
+  absl::InsecureBitGen rng;
+
+  std::vector<int32_t> counts(kBuckets, 0);
+  for (size_t i = 0; i < kTrials; i++) {
+    auto x = dist(rng);
+    counts[x]++;
+  }
+
+  // Scale weights.
+  double sum = 0;
+  for (double x : weights) {
+    sum += x;
+  }
+  for (double& x : weights) {
+    x = kTrials * (x / sum);
+  }
+
+  double chi_square =
+      absl::random_internal::ChiSquare(std::begin(counts), std::end(counts),
+                                       std::begin(weights), std::end(weights));
+
+  if (chi_square > kThreshold) {
+    double p_value =
+        absl::random_internal::ChiSquarePValue(chi_square, kBuckets);
+
+    // Chi-squared test failed. Output does not appear to be uniform.
+    std::string msg;
+    for (size_t i = 0; i < counts.size(); i++) {
+      absl::StrAppend(&msg, i, ": ", counts[i], " vs ", weights[i], "\n");
+    }
+    absl::StrAppend(&msg, kChiSquared, " p-value ", p_value, "\n");
+    absl::StrAppend(&msg, "High ", kChiSquared, " value: ", chi_square, " > ",
+                    kThreshold);
+    ABSL_RAW_LOG(INFO, "%s", msg.c_str());
+    FAIL() << msg;
+  }
+}
+
+TEST(DiscreteDistributionTest, StabilityTest) {
+  // absl::discrete_distribution stabilitiy relies on
+  // absl::uniform_int_distribution and absl::bernoulli_distribution.
+  absl::random_internal::sequence_urbg urbg(
+      {0x0003eb76f6f7f755ull, 0xFFCEA50FDB2F953Bull, 0xC332DDEFBE6C5AA5ull,
+       0x6558218568AB9702ull, 0x2AEF7DAD5B6E2F84ull, 0x1521B62829076170ull,
+       0xECDD4775619F1510ull, 0x13CCA830EB61BD96ull, 0x0334FE1EAA0363CFull,
+       0xB5735C904C70A239ull, 0xD59E9E0BCBAADE14ull, 0xEECC86BC60622CA7ull});
+
+  std::vector<int> output(6);
+
+  {
+    absl::discrete_distribution<int32_t> dist({1.0, 2.0, 3.0, 5.0, 2.0});
+    EXPECT_EQ(0, dist.min());
+    EXPECT_EQ(4, dist.max());
+    for (auto& v : output) {
+      v = dist(urbg);
+    }
+    EXPECT_EQ(12, urbg.invocations());
+  }
+
+  // With 12 calls to urbg, each call into discrete_distribution consumes
+  // precisely 2 values: one for the uniform call, and a second for the
+  // bernoulli.
+  //
+  // Given the alt mapping: 0=>3, 1=>3, 2=>2, 3=>2, 4=>3, we can
+  //
+  // uniform:      443210143131
+  // bernoulli: b0 000011100101
+  // bernoulli: b1 001111101101
+  // bernoulli: b2 111111111111
+  // bernoulli: b3 001111101111
+  // bernoulli: b4 001111101101
+  // ...
+  EXPECT_THAT(output, testing::ElementsAre(3, 3, 1, 3, 3, 3));
+
+  {
+    urbg.reset();
+    absl::discrete_distribution<int64_t> dist({1.0, 2.0, 3.0, 5.0, 2.0});
+    EXPECT_EQ(0, dist.min());
+    EXPECT_EQ(4, dist.max());
+    for (auto& v : output) {
+      v = dist(urbg);
+    }
+    EXPECT_EQ(12, urbg.invocations());
+  }
+  EXPECT_THAT(output, testing::ElementsAre(3, 3, 0, 3, 0, 4));
+}
+
+}  // namespace