tvl-depot/absl/container/flat_hash_set.h

494 lines
18 KiB
C
Raw Normal View History

// Copyright 2018 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.
//
// -----------------------------------------------------------------------------
// File: flat_hash_set.h
// -----------------------------------------------------------------------------
//
// An `absl::flat_hash_set<T>` is an unordered associative container designed to
// be a more efficient replacement for `std::unordered_set`. Like
// `unordered_set`, search, insertion, and deletion of set elements can be done
// as an `O(1)` operation. However, `flat_hash_set` (and other unordered
// associative containers known as the collection of Abseil "Swiss tables")
// contain other optimizations that result in both memory and computation
// advantages.
//
// In most cases, your default choice for a hash set should be a set of type
// `flat_hash_set`.
#ifndef ABSL_CONTAINER_FLAT_HASH_SET_H_
#define ABSL_CONTAINER_FLAT_HASH_SET_H_
#include <type_traits>
#include <utility>
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
#include "absl/algorithm/container.h"
#include "absl/base/macros.h"
#include "absl/container/internal/container_memory.h"
#include "absl/container/internal/hash_function_defaults.h" // IWYU pragma: export
#include "absl/container/internal/raw_hash_set.h" // IWYU pragma: export
#include "absl/memory/memory.h"
namespace absl {
namespace container_internal {
template <typename T>
struct FlatHashSetPolicy;
} // namespace container_internal
// -----------------------------------------------------------------------------
// absl::flat_hash_set
// -----------------------------------------------------------------------------
//
// An `absl::flat_hash_set<T>` is an unordered associative container which has
// been optimized for both speed and memory footprint in most common use cases.
// Its interface is similar to that of `std::unordered_set<T>` with the
// following notable differences:
//
// * Requires keys that are CopyConstructible
// * Supports heterogeneous lookup, through `find()`, `operator[]()` and
// `insert()`, provided that the set is provided a compatible heterogeneous
// hashing function and equality operator.
// * Invalidates any references and pointers to elements within the table after
// `rehash()`.
// * Contains a `capacity()` member function indicating the number of element
// slots (open, deleted, and empty) within the hash set.
// * Returns `void` from the `erase(iterator)` overload.
//
// By default, `flat_hash_set` uses the `absl::Hash` hashing framework. All
// fundamental and Abseil types that support the `absl::Hash` framework have a
// compatible equality operator for comparing insertions into `flat_hash_map`.
// If your type is not yet supported by the `absl::Hash` framework, see
// absl/hash/hash.h for information on extending Abseil hashing to user-defined
// types.
//
// NOTE: A `flat_hash_set` stores its keys directly inside its implementation
// array to avoid memory indirection. Because a `flat_hash_set` is designed to
// move data when rehashed, set keys will not retain pointer stability. If you
// require pointer stability, consider using
// `absl::flat_hash_set<std::unique_ptr<T>>`. If your type is not moveable and
// you require pointer stability, consider `absl::node_hash_set` instead.
//
// Example:
//
// // Create a flat hash set of three strings
// absl::flat_hash_set<std::string> ducks =
// {"huey", "dewey", "louie"};
//
// // Insert a new element into the flat hash set
// ducks.insert("donald");
//
// // Force a rehash of the flat hash set
// ducks.rehash(0);
//
// // See if "dewey" is present
// if (ducks.contains("dewey")) {
// std::cout << "We found dewey!" << std::endl;
// }
template <class T, class Hash = absl::container_internal::hash_default_hash<T>,
class Eq = absl::container_internal::hash_default_eq<T>,
class Allocator = std::allocator<T>>
class flat_hash_set
: public absl::container_internal::raw_hash_set<
absl::container_internal::FlatHashSetPolicy<T>, Hash, Eq, Allocator> {
using Base = typename flat_hash_set::raw_hash_set;
public:
Export of internal Abseil changes. -- ee19e203eca970ff88e8f25ce4e19c32e143b988 by Jon Cohen <cohenjon@google.com>: Exception safety testing no longer uses absl::optional PiperOrigin-RevId: 220336204 -- 460666eb0b316a8b4aeedc589644d53b05251bd1 by Derek Mauro <dmauro@google.com>: Rework SwissTable SSE2 support - Use SSE2 on MSVC when available https://github.com/abseil/abseil-cpp/issues/210 - Emulate _mm_cmpgt_epi8 with other SSE2 instructions when using -funsigned-char under GCC https://github.com/abseil/abseil-cpp/issues/209 PiperOrigin-RevId: 220312351 -- 1f4318ecedf8d539b7b698eb803d613ad6b69278 by Abseil Team <absl-team@google.com>: Change CollectPerfectRatios to use 10 trials to smooth out the outliers in the sample. PiperOrigin-RevId: 220286579 -- 6755abc2673553a7f578bb29c6e9ca8d991bc9c8 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 220274307 -- 8645b6187329ebf0aaf3c2de2888ba44466cd879 by Abseil Team <absl-team@google.com>: * #endif for a header guard should reference the guard macro in a comment PiperOrigin-RevId: 220206868 -- 3987a7ad11319230910931cd2468b60b3fd1b85c by Gennadiy Civil <misterg@google.com>: Internal Change PiperOrigin-RevId: 220136674 -- cc908c1db2ee0d4523dc813e33f600583bb986c5 by Abseil Team <absl-team@google.com>: absl: fix backoff logic in SpinLockWait There are 3 bugs in loop variable handling: 1. It starts with 0, but AbslInternalSpinLockDelay ignores loop == 0. So it does not actually wait when it should. 2. loop is incremented after successful state changes, but it should not (why would be increase backoff delay after that?). 3. loop is incremented after CAS failures, but it should not (why would be increase backoff delay after that?). Use the same handling of loop as used in SpinLock. PiperOrigin-RevId: 220136079 -- a0a1c6ef5910ebd28e07215d7df03cc0da0b3eed by Abseil Team <absl-team@google.com>: absl: relax unnecessarily strong memory ordering in SpinLock::SlowLock We don't need to acquire visibility over anything when setting kSpinLockSleeper. Replace the confusing and unnecessarily strong memory order with relaxed. PiperOrigin-RevId: 220023380 -- c50858b51af28b9fca1a62616324f85f3e84ea74 by Tom Manshreck <shreck@google.com>: Update comments in flat_hash_map, node_hash_{set, map} and the containers developer guide PiperOrigin-RevId: 219938692 -- e87b7d1a5f61e165b1c44d3b16d8d967197cdfce by CJ Johnson <johnsoncj@google.com>: Rearranges the public methods of InlinedVector and cleans up the comments PiperOrigin-RevId: 219896257 -- f3234c466f792e0fc4bfd21fc7919dba5e679375 by CJ Johnson <johnsoncj@google.com>: Adds branch prediction to exceptional early exit cases of inlined vector's API PiperOrigin-RevId: 219887173 -- 4dfccf1a81ca0425912d3da25a8470f78c532ce4 by CJ Johnson <johnsoncj@google.com>: Fixes the InlinedVector public interface to use the allocator type references instead of assuming the type Also cleans up some cruft in formatting and comments PiperOrigin-RevId: 219878876 -- 4bb6a2b892abb10bd6a424db7e94ed8640802470 by Tom Manshreck <shreck@google.com>: Add comments on constructor and assignment operator support to flat_hash_set PiperOrigin-RevId: 219825338 -- c23f973e2f7f4feea0da36bf8a9c3f8a8954bb74 by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 219823847 GitOrigin-RevId: ee19e203eca970ff88e8f25ce4e19c32e143b988 Change-Id: I288c927ca481dc57340420dbb4c278a05cf15e83
2018-11-06 22:01:25 +01:00
// Constructors and Assignment Operators
//
// A flat_hash_set supports the same overload set as `std::unordered_map`
// for construction and assignment:
//
// * Default constructor
//
// // No allocation for the table's elements is made.
// absl::flat_hash_set<std::string> set1;
//
// * Initializer List constructor
//
// absl::flat_hash_set<std::string> set2 =
// {{"huey"}, {"dewey"}, {"louie"},};
//
// * Copy constructor
//
// absl::flat_hash_set<std::string> set3(set2);
//
// * Copy assignment operator
//
// // Hash functor and Comparator are copied as well
// absl::flat_hash_set<std::string> set4;
// set4 = set3;
//
// * Move constructor
//
// // Move is guaranteed efficient
// absl::flat_hash_set<std::string> set5(std::move(set4));
//
// * Move assignment operator
//
// // May be efficient if allocators are compatible
// absl::flat_hash_set<std::string> set6;
// set6 = std::move(set5);
//
// * Range constructor
//
// std::vector<std::string> v = {"a", "b"};
// absl::flat_hash_set<std::string> set7(v.begin(), v.end());
flat_hash_set() {}
using Base::Base;
// flat_hash_set::begin()
//
// Returns an iterator to the beginning of the `flat_hash_set`.
using Base::begin;
// flat_hash_set::cbegin()
//
// Returns a const iterator to the beginning of the `flat_hash_set`.
using Base::cbegin;
// flat_hash_set::cend()
//
// Returns a const iterator to the end of the `flat_hash_set`.
using Base::cend;
// flat_hash_set::end()
//
// Returns an iterator to the end of the `flat_hash_set`.
using Base::end;
// flat_hash_set::capacity()
//
// Returns the number of element slots (assigned, deleted, and empty)
// available within the `flat_hash_set`.
//
// NOTE: this member function is particular to `absl::flat_hash_set` and is
// not provided in the `std::unordered_map` API.
using Base::capacity;
// flat_hash_set::empty()
//
// Returns whether or not the `flat_hash_set` is empty.
using Base::empty;
// flat_hash_set::max_size()
//
// Returns the largest theoretical possible number of elements within a
// `flat_hash_set` under current memory constraints. This value can be thought
// of the largest value of `std::distance(begin(), end())` for a
// `flat_hash_set<T>`.
using Base::max_size;
// flat_hash_set::size()
//
// Returns the number of elements currently within the `flat_hash_set`.
using Base::size;
// flat_hash_set::clear()
//
// Removes all elements from the `flat_hash_set`. Invalidates any references,
// pointers, or iterators referring to contained elements.
//
// NOTE: this operation may shrink the underlying buffer. To avoid shrinking
// the underlying buffer call `erase(begin(), end())`.
using Base::clear;
// flat_hash_set::erase()
//
// Erases elements within the `flat_hash_set`. Erasing does not trigger a
// rehash. Overloads are listed below.
//
// void erase(const_iterator pos):
//
// Erases the element at `position` of the `flat_hash_set`, returning
// `void`.
//
Export of internal Abseil changes. -- 8b7c3bc2fb69608e9b2389b1be0b0de840a4c59d by Derek Mauro <dmauro@google.com>: Set correct flags for clang-cl. https://github.com/abseil/abseil-cpp/pull/278 clang-cl produce binaries with MSVC ABI and wants to be as flag-compatible with pure MSVC as possible, so this leads to all sorts of weird cases. clang-cl alias /Wall as clang's -Weverything which is way too verbose, so it needs /W3 like pure MSVC. clang-cl only understand GCC style warning flags (-W[no]blah) and just silent drop MSVC style warning flags (/wd[num]). clang-cl needs MSVC define flags since it is consuming the same header files as pure MSVC. CMake set CMAKE_CXX_COMPILER_ID as Clang when clang-cl is detected, so need extra if (MSVC) to differentiate it. We are not doing clang-cl specialization in Bazel as currently there is no reliable way to detect clang-cl in Bazel.. Other changes: Add ABSL_ prefix to variable names to avoid name collision in CMake. PiperOrigin-RevId: 239841297 -- add96c3fc067d5c7b6f016d2ba74725a443a185e by CJ Johnson <johnsoncj@google.com>: Eventually Storage will need to refer to the type `absl::InlinedVector<...>*`. This can be done via a forward declaration. However, doing so would move the defaulted allocator template parameter to the forward declaration and thus inside an internal file. Instead of doing that, this change gives Storage access to the template and it's parameters so the complete type can be formed without including it. PiperOrigin-RevId: 239811298 -- b5f5279f1b13b09cae5c745597d64ea1efab146b by CJ Johnson <johnsoncj@google.com>: Simplify/cleanup the benchmark tests for InlinedVector PiperOrigin-RevId: 239805767 -- f5991e51b43b13a0ae95025474071f5039a33d27 by Matt Calabrese <calabrese@google.com>: Update the internal-only IsSwappable traits to be nested inside of namespace absl so that the script to add inline namespaces for LTS releases works with the implementation. PiperOrigin-RevId: 239622024 -- d1cb234dc5706f033ad56f4eb16d94ac5da80d52 by Abseil Team <absl-team@google.com>: Mutex: fix tsan annotations This fixes 2 bugs: 1. We call cond directly in Mutex::AwaitCommon without using EvalConditionAnnotated. As the result we call into user code ignoring synchronization, miss synchronization and report false positives later. Use EvalConditionAnnotated to call cond as we should. 2. We call Mutex invariant ignoring synchronization. Result is the same: we miss synchronization and report false positive races later. Reuse EvalConditionAnnotated to call mutex invariant too. PiperOrigin-RevId: 239583878 -- 52295e4922a9b408fa0dd03d27bc91ccc6645cd7 by Abseil Team <absl-team@google.com>: Clarify how to obtain the same behavior as std::unordered_map::erase if need be. PiperOrigin-RevId: 239549513 -- 6e76e68ed084fd1247981dbb92677ce8e563b0ec by Jon Cohen <cohenjon@google.com>: Avoid the -S -B form of `cmake` since it's only supported starting in CMake 3.13 PiperOrigin-RevId: 239473143 GitOrigin-RevId: 8b7c3bc2fb69608e9b2389b1be0b0de840a4c59d Change-Id: Ib6d356fa1a7435260273df991e65df4149bd5861
2019-03-22 20:20:05 +01:00
// NOTE: returning `void` in this case is different than that of STL
// containers in general and `std::unordered_set` in particular (which
// return an iterator to the element following the erased element). If that
// iterator is needed, simply post increment the iterator:
//
// set.erase(it++);
//
// iterator erase(const_iterator first, const_iterator last):
//
// Erases the elements in the open interval [`first`, `last`), returning an
// iterator pointing to `last`.
//
// size_type erase(const key_type& key):
//
// Erases the element with the matching key, if it exists.
using Base::erase;
// flat_hash_set::insert()
//
// Inserts an element of the specified value into the `flat_hash_set`,
// returning an iterator pointing to the newly inserted element, provided that
// an element with the given key does not already exist. If rehashing occurs
// due to the insertion, all iterators are invalidated. Overloads are listed
// below.
//
// std::pair<iterator,bool> insert(const T& value):
//
// Inserts a value into the `flat_hash_set`. Returns a pair consisting of an
// iterator to the inserted element (or to the element that prevented the
// insertion) and a bool denoting whether the insertion took place.
//
// std::pair<iterator,bool> insert(T&& value):
//
// Inserts a moveable value into the `flat_hash_set`. Returns a pair
// consisting of an iterator to the inserted element (or to the element that
// prevented the insertion) and a bool denoting whether the insertion took
// place.
//
// iterator insert(const_iterator hint, const T& value):
// iterator insert(const_iterator hint, T&& value):
//
// Inserts a value, using the position of `hint` as a non-binding suggestion
// for where to begin the insertion search. Returns an iterator to the
// inserted element, or to the existing element that prevented the
// insertion.
//
Export of internal Abseil changes. -- 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 by Derek Mauro <dmauro@google.com>: Fix the internal namespacing in unaligned_access.h PiperOrigin-RevId: 215434506 -- 17d4400aebf025a230690fc1c7a968ef8d85bbba by Eric Fiselier <ericwf@google.com>: gtest depends on the GCC extension allowing variadic macros to be passed a empty parameter pack for ..., but LLVM diagnoses this as a GNU extension. This patch suppresses the warning when building the absl tests. PiperOrigin-RevId: 215426161 -- f2c49dde23a9f445b9de963f1bbe840ebb568b30 by Eric Fiselier <ericwf@google.com>: Use EXPECT_DEATH_IF_SUPPORTED instead of EXPECT_DEATH. This avoids breaking the test when gtest doesn't support death tests. PiperOrigin-RevId: 215423849 -- cd687c1e121709603f4fc3726b534f6a9c52cc89 by Eric Fiselier <ericwf@google.com>: Disable LLVM's -Wmissing-variable-declarations in tests. GCC's configuration already disables this via -Wno-missing-declarations, this change makes LLVM do the same. The warning would otherwise flag most tests which use ABSL_FLAG. PiperOrigin-RevId: 215407429 -- d14098824c84e3a8c8f6fb920e0335fb48fe2010 by Eric Fiselier <ericwf@google.com>: Fix local variable shadowing in city hash implementation. PiperOrigin-RevId: 215407249 -- 4b5e140ba743f0d231790a26c49083abb4329e2c by Abseil Team <absl-team@google.com>: Make raw_hash_set::reserve 2X fast when reserve doesn't do any allocation. Make raw_hash_set::reserve ~1% faster when reserve does some (128~4k) allocation. PiperOrigin-RevId: 215348727 -- 461161e65e04b801480aa117af2534c594654ccf by Eric Fiselier <ericwf@google.com>: Internal change PiperOrigin-RevId: 215272283 -- 50413ae31ad3d3a177257416acd8ede47a17bff2 by Eric Fiselier <ericwf@google.com>: Internal Change PiperOrigin-RevId: 215233183 -- 477be54c43d61019a8fe4e190e340eb52737d383 by Abseil Team <absl-team@google.com>: Clarify misleading comment on ABSL_ATTRIBUTE_UNUSED PiperOrigin-RevId: 215185496 -- 2cafa2b5287507d3a946682aee9ab13af6d471c9 by Matt Kulukundis <kfm@google.com>: Add support for absl::Hash to various absl in types. PiperOrigin-RevId: 215039569 -- 082248901991aa3d29be0ea3689c7f213cf0fd83 by Derek Mauro <dmauro@google.com>: Remove an instance of HAS_GLOBAL_STRING from hash_function_defaults.h PiperOrigin-RevId: 214989094 -- b929f61907f0786a6133e3a9d7287e339c0a0acb by Derek Mauro <dmauro@google.com>: Internal import of Github #174 Fix code snippet in comment https://github.com/abseil/abseil-cpp/pull/174 PiperOrigin-RevId: 214958849 -- f2c5e829eca11c352e121f56eefbf87083305023 by Derek Mauro <dmauro@google.com>: Internal import of GitHub #173 Fix CMake build for absl::container. https://github.com/abseil/abseil-cpp/pull/173 PiperOrigin-RevId: 214957796 -- d704f860f9fddafb99e34e6c5032e49f73874e10 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 214828181 GitOrigin-RevId: 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 Change-Id: I551de2b1ba0cbaf6856cd5959358cf6651179dea
2018-10-02 21:09:18 +02:00
// void insert(InputIterator first, InputIterator last):
//
// Inserts a range of values [`first`, `last`).
//
// NOTE: Although the STL does not specify which element may be inserted if
// multiple keys compare equivalently, for `flat_hash_set` we guarantee the
// first match is inserted.
//
Export of internal Abseil changes. -- 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 by Derek Mauro <dmauro@google.com>: Fix the internal namespacing in unaligned_access.h PiperOrigin-RevId: 215434506 -- 17d4400aebf025a230690fc1c7a968ef8d85bbba by Eric Fiselier <ericwf@google.com>: gtest depends on the GCC extension allowing variadic macros to be passed a empty parameter pack for ..., but LLVM diagnoses this as a GNU extension. This patch suppresses the warning when building the absl tests. PiperOrigin-RevId: 215426161 -- f2c49dde23a9f445b9de963f1bbe840ebb568b30 by Eric Fiselier <ericwf@google.com>: Use EXPECT_DEATH_IF_SUPPORTED instead of EXPECT_DEATH. This avoids breaking the test when gtest doesn't support death tests. PiperOrigin-RevId: 215423849 -- cd687c1e121709603f4fc3726b534f6a9c52cc89 by Eric Fiselier <ericwf@google.com>: Disable LLVM's -Wmissing-variable-declarations in tests. GCC's configuration already disables this via -Wno-missing-declarations, this change makes LLVM do the same. The warning would otherwise flag most tests which use ABSL_FLAG. PiperOrigin-RevId: 215407429 -- d14098824c84e3a8c8f6fb920e0335fb48fe2010 by Eric Fiselier <ericwf@google.com>: Fix local variable shadowing in city hash implementation. PiperOrigin-RevId: 215407249 -- 4b5e140ba743f0d231790a26c49083abb4329e2c by Abseil Team <absl-team@google.com>: Make raw_hash_set::reserve 2X fast when reserve doesn't do any allocation. Make raw_hash_set::reserve ~1% faster when reserve does some (128~4k) allocation. PiperOrigin-RevId: 215348727 -- 461161e65e04b801480aa117af2534c594654ccf by Eric Fiselier <ericwf@google.com>: Internal change PiperOrigin-RevId: 215272283 -- 50413ae31ad3d3a177257416acd8ede47a17bff2 by Eric Fiselier <ericwf@google.com>: Internal Change PiperOrigin-RevId: 215233183 -- 477be54c43d61019a8fe4e190e340eb52737d383 by Abseil Team <absl-team@google.com>: Clarify misleading comment on ABSL_ATTRIBUTE_UNUSED PiperOrigin-RevId: 215185496 -- 2cafa2b5287507d3a946682aee9ab13af6d471c9 by Matt Kulukundis <kfm@google.com>: Add support for absl::Hash to various absl in types. PiperOrigin-RevId: 215039569 -- 082248901991aa3d29be0ea3689c7f213cf0fd83 by Derek Mauro <dmauro@google.com>: Remove an instance of HAS_GLOBAL_STRING from hash_function_defaults.h PiperOrigin-RevId: 214989094 -- b929f61907f0786a6133e3a9d7287e339c0a0acb by Derek Mauro <dmauro@google.com>: Internal import of Github #174 Fix code snippet in comment https://github.com/abseil/abseil-cpp/pull/174 PiperOrigin-RevId: 214958849 -- f2c5e829eca11c352e121f56eefbf87083305023 by Derek Mauro <dmauro@google.com>: Internal import of GitHub #173 Fix CMake build for absl::container. https://github.com/abseil/abseil-cpp/pull/173 PiperOrigin-RevId: 214957796 -- d704f860f9fddafb99e34e6c5032e49f73874e10 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 214828181 GitOrigin-RevId: 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 Change-Id: I551de2b1ba0cbaf6856cd5959358cf6651179dea
2018-10-02 21:09:18 +02:00
// void insert(std::initializer_list<T> ilist):
//
// Inserts the elements within the initializer list `ilist`.
//
// NOTE: Although the STL does not specify which element may be inserted if
// multiple keys compare equivalently within the initializer list, for
// `flat_hash_set` we guarantee the first match is inserted.
using Base::insert;
// flat_hash_set::emplace()
//
// Inserts an element of the specified value by constructing it in-place
// within the `flat_hash_set`, provided that no element with the given key
// already exists.
//
// The element may be constructed even if there already is an element with the
// key in the container, in which case the newly constructed element will be
Export of internal Abseil changes. -- cd076f55c1fa600131f6dda392533dfe61679fc0 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 224008762 -- e05f62b01286d51044ff86ec6ef565749b9faf82 by Abseil Team <absl-team@google.com>: Create a pow10() test helper function to compute guaranteed-precise double values of 10^x. Not all standard libraries ship bit-accurate pow() functions, causing tests to fail that rely on expected values generated by it. PiperOrigin-RevId: 223883762 -- fd88e5e3f7ab80f7f5df9fd1488cd58b4573be69 by Abseil Team <absl-team@google.com>: Remove some absl:: qualifications to work around inline namespace bugs on MSVC 2015. PiperOrigin-RevId: 223869642 -- 6276cfff969d596edd36a2bbaba65ee045808903 by Abseil Team <absl-team@google.com>: Update absl/memory/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223854224 -- 359de9afc7a34c975fd3e0cbc52afd96637d97bd by Chris Kennelly <ckennelly@google.com>: Mark spinlock_benchmark_common as alwayslink = 1. PiperOrigin-RevId: 223844536 -- 450cd8cbe2789a6d54ed1eb87170259bb334f8b9 by Abseil Team <absl-team@google.com>: Support .* (pointer-to-member dereference) expressions in demangle.cc. PiperOrigin-RevId: 223826797 -- 772ca92179c3634f3e31a80bbc272ed8022e3572 by Abseil Team <absl-team@google.com>: Fix misspellings in absl::variant comments and replace a ' with a `. PiperOrigin-RevId: 223807911 -- 35dcdc2fbf299d195658aac101887f6dcad1de2f by Abseil Team <absl-team@google.com>: Bug fix in CMakeLists.txt file (SRCS --> HDRS). The compressed_tuple header-only library is being defined with the SRCS parameter instead of the HDRS parameter and this has been observed to cause some builds on some platforms to attempt to create a static library from it which fails since there are no .cc sources. PiperOrigin-RevId: 223805367 -- 4a57a3d2045bb137c0c97958e45ce425190b8d3e by Chris Kennelly <ckennelly@google.com>: Add test that absl::make_unique value initializes memory. PiperOrigin-RevId: 223801819 -- dfe8289d7f4dcc6bb568a26aaf192a89e896bdfd by Chris Kennelly <ckennelly@google.com>: SpinLock: Use exchange to avoid missing wakeups. The default fast path for SpinLock::Unlock does not use an atomic. If the SpinLock becomes contended while we are unlocking between lockword_.load and lockword_.store, we will fail to wake up the new waiter. This can cause unexpected latency spikes. PiperOrigin-RevId: 223800369 -- 9b9d35df786482f0016f77dd31691eff81503d23 by Abseil Team <absl-team@google.com>: Update absl/hash/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223755819 -- c2014e2704b87e7cdce2d2a0287c7e2397752296 by Abseil Team <absl-team@google.com>: Update absl/debugging/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223751986 -- d83a4e09126400e3fd80645cb03ee558f532271e by Derek Mauro <dmauro@google.com>: Cleanup synchronization benchmarks. PiperOrigin-RevId: 223589416 -- fad140b473586531b5b12843f942ec27dfcf5e93 by CJ Johnson <johnsoncj@google.com>: Makes unifies the order of forward_iterator and input_iterator overloads PiperOrigin-RevId: 223580660 -- 6cd7c96faa7cc5f79f574e35a1b13837ef187d05 by Abseil Team <absl-team@google.com>: Internal Change. PiperOrigin-RevId: 223561629 -- bd2e545356b0f548af0e3c14bb2f7f0e712e49d0 by Shaindel Schwartz <shaindel@google.com>: Remove misleading comments. try_emplace() does not exist for the hash_set containers. PiperOrigin-RevId: 223543089 -- 0cd380a53b587eb7aacc4003a4a3bbb6c78d7c10 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 223512551 -- 7156dfee599cb72e9adddfe0e6ae07a95ddf10bb by Greg Miller <jgm@google.com>: Fixes UB that would result from constructing, multiplying, or dividing a Duration with a double "NaN" value. This CL changes the absl::Duration *implementation* to return an InfiniteDuration value that has the same sign as the given NaN. PiperOrigin-RevId: 223407499 -- 196b7d18609958267951882baf7f9429e49bcafa by CJ Johnson <johnsoncj@google.com>: Addresses NVCC+MSVC compilation bug where `inlined_capacity()` was not considered valid in constexpr PiperOrigin-RevId: 223397718 GitOrigin-RevId: cd076f55c1fa600131f6dda392533dfe61679fc0 Change-Id: I5423ca6470f661a7c6f73aa8fee49990446f157f
2018-12-04 20:01:12 +01:00
// destroyed immediately.
//
// If rehashing occurs due to the insertion, all iterators are invalidated.
using Base::emplace;
// flat_hash_set::emplace_hint()
//
// Inserts an element of the specified value by constructing it in-place
// within the `flat_hash_set`, using the position of `hint` as a non-binding
// suggestion for where to begin the insertion search, and only inserts
// provided that no element with the given key already exists.
//
// The element may be constructed even if there already is an element with the
// key in the container, in which case the newly constructed element will be
Export of internal Abseil changes. -- cd076f55c1fa600131f6dda392533dfe61679fc0 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 224008762 -- e05f62b01286d51044ff86ec6ef565749b9faf82 by Abseil Team <absl-team@google.com>: Create a pow10() test helper function to compute guaranteed-precise double values of 10^x. Not all standard libraries ship bit-accurate pow() functions, causing tests to fail that rely on expected values generated by it. PiperOrigin-RevId: 223883762 -- fd88e5e3f7ab80f7f5df9fd1488cd58b4573be69 by Abseil Team <absl-team@google.com>: Remove some absl:: qualifications to work around inline namespace bugs on MSVC 2015. PiperOrigin-RevId: 223869642 -- 6276cfff969d596edd36a2bbaba65ee045808903 by Abseil Team <absl-team@google.com>: Update absl/memory/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223854224 -- 359de9afc7a34c975fd3e0cbc52afd96637d97bd by Chris Kennelly <ckennelly@google.com>: Mark spinlock_benchmark_common as alwayslink = 1. PiperOrigin-RevId: 223844536 -- 450cd8cbe2789a6d54ed1eb87170259bb334f8b9 by Abseil Team <absl-team@google.com>: Support .* (pointer-to-member dereference) expressions in demangle.cc. PiperOrigin-RevId: 223826797 -- 772ca92179c3634f3e31a80bbc272ed8022e3572 by Abseil Team <absl-team@google.com>: Fix misspellings in absl::variant comments and replace a ' with a `. PiperOrigin-RevId: 223807911 -- 35dcdc2fbf299d195658aac101887f6dcad1de2f by Abseil Team <absl-team@google.com>: Bug fix in CMakeLists.txt file (SRCS --> HDRS). The compressed_tuple header-only library is being defined with the SRCS parameter instead of the HDRS parameter and this has been observed to cause some builds on some platforms to attempt to create a static library from it which fails since there are no .cc sources. PiperOrigin-RevId: 223805367 -- 4a57a3d2045bb137c0c97958e45ce425190b8d3e by Chris Kennelly <ckennelly@google.com>: Add test that absl::make_unique value initializes memory. PiperOrigin-RevId: 223801819 -- dfe8289d7f4dcc6bb568a26aaf192a89e896bdfd by Chris Kennelly <ckennelly@google.com>: SpinLock: Use exchange to avoid missing wakeups. The default fast path for SpinLock::Unlock does not use an atomic. If the SpinLock becomes contended while we are unlocking between lockword_.load and lockword_.store, we will fail to wake up the new waiter. This can cause unexpected latency spikes. PiperOrigin-RevId: 223800369 -- 9b9d35df786482f0016f77dd31691eff81503d23 by Abseil Team <absl-team@google.com>: Update absl/hash/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223755819 -- c2014e2704b87e7cdce2d2a0287c7e2397752296 by Abseil Team <absl-team@google.com>: Update absl/debugging/CMakeLists.txt to use new functions i.e. absl_cc_(library|test) PiperOrigin-RevId: 223751986 -- d83a4e09126400e3fd80645cb03ee558f532271e by Derek Mauro <dmauro@google.com>: Cleanup synchronization benchmarks. PiperOrigin-RevId: 223589416 -- fad140b473586531b5b12843f942ec27dfcf5e93 by CJ Johnson <johnsoncj@google.com>: Makes unifies the order of forward_iterator and input_iterator overloads PiperOrigin-RevId: 223580660 -- 6cd7c96faa7cc5f79f574e35a1b13837ef187d05 by Abseil Team <absl-team@google.com>: Internal Change. PiperOrigin-RevId: 223561629 -- bd2e545356b0f548af0e3c14bb2f7f0e712e49d0 by Shaindel Schwartz <shaindel@google.com>: Remove misleading comments. try_emplace() does not exist for the hash_set containers. PiperOrigin-RevId: 223543089 -- 0cd380a53b587eb7aacc4003a4a3bbb6c78d7c10 by Derek Mauro <dmauro@google.com>: Internal change PiperOrigin-RevId: 223512551 -- 7156dfee599cb72e9adddfe0e6ae07a95ddf10bb by Greg Miller <jgm@google.com>: Fixes UB that would result from constructing, multiplying, or dividing a Duration with a double "NaN" value. This CL changes the absl::Duration *implementation* to return an InfiniteDuration value that has the same sign as the given NaN. PiperOrigin-RevId: 223407499 -- 196b7d18609958267951882baf7f9429e49bcafa by CJ Johnson <johnsoncj@google.com>: Addresses NVCC+MSVC compilation bug where `inlined_capacity()` was not considered valid in constexpr PiperOrigin-RevId: 223397718 GitOrigin-RevId: cd076f55c1fa600131f6dda392533dfe61679fc0 Change-Id: I5423ca6470f661a7c6f73aa8fee49990446f157f
2018-12-04 20:01:12 +01:00
// destroyed immediately.
//
// If rehashing occurs due to the insertion, all iterators are invalidated.
using Base::emplace_hint;
// flat_hash_set::extract()
//
// Extracts the indicated element, erasing it in the process, and returns it
// as a C++17-compatible node handle. Overloads are listed below.
//
// node_type extract(const_iterator position):
//
// Extracts the element at the indicated position and returns a node handle
// owning that extracted data.
//
// node_type extract(const key_type& x):
//
// Extracts the element with the key matching the passed key value and
// returns a node handle owning that extracted data. If the `flat_hash_set`
// does not contain an element with a matching key, this function returns an
// empty node handle.
using Base::extract;
// flat_hash_set::merge()
//
// Extracts elements from a given `source` flat hash map into this
// `flat_hash_set`. If the destination `flat_hash_set` already contains an
// element with an equivalent key, that element is not extracted.
using Base::merge;
// flat_hash_set::swap(flat_hash_set& other)
//
// Exchanges the contents of this `flat_hash_set` with those of the `other`
// flat hash map, avoiding invocation of any move, copy, or swap operations on
// individual elements.
//
// All iterators and references on the `flat_hash_set` remain valid, excepting
// for the past-the-end iterator, which is invalidated.
//
// `swap()` requires that the flat hash set's hashing and key equivalence
// functions be Swappable, and are exchaged using unqualified calls to
// non-member `swap()`. If the map's allocator has
// `std::allocator_traits<allocator_type>::propagate_on_container_swap::value`
// set to `true`, the allocators are also exchanged using an unqualified call
// to non-member `swap()`; otherwise, the allocators are not swapped.
using Base::swap;
// flat_hash_set::rehash(count)
//
// Rehashes the `flat_hash_set`, setting the number of slots to be at least
// the passed value. If the new number of slots increases the load factor more
// than the current maximum load factor
// (`count` < `size()` / `max_load_factor()`), then the new number of slots
// will be at least `size()` / `max_load_factor()`.
//
// To force a rehash, pass rehash(0).
//
// NOTE: unlike behavior in `std::unordered_set`, references are also
// invalidated upon a `rehash()`.
using Base::rehash;
// flat_hash_set::reserve(count)
//
// Sets the number of slots in the `flat_hash_set` to the number needed to
// accommodate at least `count` total elements without exceeding the current
// maximum load factor, and may rehash the container if needed.
using Base::reserve;
// flat_hash_set::contains()
//
// Determines whether an element comparing equal to the given `key` exists
// within the `flat_hash_set`, returning `true` if so or `false` otherwise.
using Base::contains;
// flat_hash_set::count(const Key& key) const
//
// Returns the number of elements comparing equal to the given `key` within
// the `flat_hash_set`. note that this function will return either `1` or `0`
// since duplicate elements are not allowed within a `flat_hash_set`.
using Base::count;
// flat_hash_set::equal_range()
//
// Returns a closed range [first, last], defined by a `std::pair` of two
// iterators, containing all elements with the passed key in the
// `flat_hash_set`.
using Base::equal_range;
// flat_hash_set::find()
//
// Finds an element with the passed `key` within the `flat_hash_set`.
using Base::find;
// flat_hash_set::bucket_count()
//
// Returns the number of "buckets" within the `flat_hash_set`. Note that
// because a flat hash map contains all elements within its internal storage,
// this value simply equals the current capacity of the `flat_hash_set`.
using Base::bucket_count;
// flat_hash_set::load_factor()
//
// Returns the current load factor of the `flat_hash_set` (the average number
// of slots occupied with a value within the hash map).
using Base::load_factor;
// flat_hash_set::max_load_factor()
//
// Manages the maximum load factor of the `flat_hash_set`. Overloads are
// listed below.
//
// float flat_hash_set::max_load_factor()
//
// Returns the current maximum load factor of the `flat_hash_set`.
//
// void flat_hash_set::max_load_factor(float ml)
//
// Sets the maximum load factor of the `flat_hash_set` to the passed value.
//
// NOTE: This overload is provided only for API compatibility with the STL;
// `flat_hash_set` will ignore any set load factor and manage its rehashing
// internally as an implementation detail.
using Base::max_load_factor;
// flat_hash_set::get_allocator()
//
// Returns the allocator function associated with this `flat_hash_set`.
using Base::get_allocator;
// flat_hash_set::hash_function()
//
// Returns the hashing function used to hash the keys within this
// `flat_hash_set`.
using Base::hash_function;
// flat_hash_set::key_eq()
//
// Returns the function used for comparing keys equality.
using Base::key_eq;
};
namespace container_internal {
template <class T>
struct FlatHashSetPolicy {
using slot_type = T;
using key_type = T;
using init_type = T;
using constant_iterators = std::true_type;
template <class Allocator, class... Args>
static void construct(Allocator* alloc, slot_type* slot, Args&&... args) {
absl::allocator_traits<Allocator>::construct(*alloc, slot,
std::forward<Args>(args)...);
}
template <class Allocator>
static void destroy(Allocator* alloc, slot_type* slot) {
absl::allocator_traits<Allocator>::destroy(*alloc, slot);
}
template <class Allocator>
static void transfer(Allocator* alloc, slot_type* new_slot,
slot_type* old_slot) {
construct(alloc, new_slot, std::move(*old_slot));
destroy(alloc, old_slot);
}
static T& element(slot_type* slot) { return *slot; }
template <class F, class... Args>
static decltype(absl::container_internal::DecomposeValue(
std::declval<F>(), std::declval<Args>()...))
apply(F&& f, Args&&... args) {
return absl::container_internal::DecomposeValue(
std::forward<F>(f), std::forward<Args>(args)...);
}
static size_t space_used(const T*) { return 0; }
};
} // namespace container_internal
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
namespace container_algorithm_internal {
// Specialization of trait in absl/algorithm/container.h
template <class Key, class Hash, class KeyEqual, class Allocator>
struct IsUnorderedContainer<absl::flat_hash_set<Key, Hash, KeyEqual, Allocator>>
: std::true_type {};
} // namespace container_algorithm_internal
} // namespace absl
Export of internal Abseil changes. -- 07575526242a8e1275ac4223a3d2822795f46569 by CJ Johnson <johnsoncj@google.com>: Comment cleanup on InlinedVector PiperOrigin-RevId: 221322176 -- 49a5e643f85e34d53c41f5e6cc33357c55c9115d by Matt Kulukundis <kfm@google.com>: Internal cleanup PiperOrigin-RevId: 221309185 -- bb35be87ec9c74244b7d902e7e7d2d33ab139d76 by Abseil Team <absl-team@google.com>: Fix typo in comment. PiperOrigin-RevId: 221145354 -- afd4d7c106919708004e06aeea068a57c28aec44 by Derek Mauro <dmauro@google.com>: Update the debugging log message in CallOnceImpl() PiperOrigin-RevId: 221103254 -- 0b9dace8b88113777bf26a6d38f9bc0bcaf053a1 by Abseil Team <absl-team@google.com>: Workaround an MSVC 2015 bug in compile-time initialization. PiperOrigin-RevId: 220871483 -- ea0a3854511ed26beab827e5a5113766b334db86 by Marek Gilbert <mcg@google.com>: Fix ABSL_HAVE_THREAD_LOCAL when compiling for iOS 8 with Xcode 10. Xcode 10 has moved the check for thread_local to a link time, so clang reports __has_feature(cxx_thread_local) but then linking fails with messages like this: ld: targeted OS version does not support use of thread local variables PiperOrigin-RevId: 220815885 -- 485b6876c158c3dcf37eb32d7e512242d5d4ecc6 by Greg Falcon <gfalcon@google.com>: Make the absl::c_set_xxxx() algorithms refuse to compile when passed an unordered collection from std:: or absl::. These algorithms operate on sorted sequences; passing an unordered container to them is nearly certainly a bug. This change is technically an API break, but it only breaks incorrect code. We could try to be more clever and detect unordered collections from other libraries, but false positives will break legal code, and this would constitute an API break Abseil cannot afford. PiperOrigin-RevId: 220794190 -- c47cff7f9cc70a4c1604eee0131af552f40e46d6 by Jon Cohen <cohenjon@google.com>: MSVC 2017's STL throws a Structured Exception (not a C++ exception, essentially equivalent to SIGSEGV) when variant::emplace calls a throwing constructor when using the debug multithreaded MSVC runtime DLL. This manifests in dbg mode in Bazel builds. Disable tests which trigger this bug. It's impossible to specifically pull out MSVC 2017 -dbg modes because there's no way for Bazel to know when version of MSVC is being used -- you tell Bazel the directory where the MSVC tools live, not which version of MSVC tools to use. Thus the best we can do is switch on _DEBUG, which is set whenever the debug runtime is selected with the /MDd build flag, as in Bazel -dbg modes. See https://msdn.microsoft.com/en-us/library/b0084kay.aspx ctrl-f "_DEBUG" PiperOrigin-RevId: 220706161 -- 43993d4af309d92f4ebff38391dcc245f154ecc7 by Shaindel Schwartz <shaindel@google.com>: Internal change PiperOrigin-RevId: 220688429 -- 2448802972dcc261af153af464f2b022ef54a2a9 by Abseil Team <absl-team@google.com>: Speed up operator* for uint128 in WIN64. PiperOrigin-RevId: 220678790 -- 7b376403dd05ba10152fb52e40b29d8af79b58bb by Abseil Team <absl-team@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220654834 -- ae08af58111c3f838b8d4de25f501c3559c86002 by Abseil Team <absl-team@google.com>: CMake: Add absl_cc_test function PiperOrigin-RevId: 220603940 GitOrigin-RevId: 07575526242a8e1275ac4223a3d2822795f46569 Change-Id: Iba7f53eb394c8a9de564582a976793f9bb0596d9
2018-11-13 22:22:00 +01:00
#endif // ABSL_CONTAINER_FLAT_HASH_SET_H_