Summary:
Add a few keywords people search for when looking for this that won't otherwise
find this file.

Add a reference to the usage notes for Singleton which are a good description
of how you should operate SingletoThreadLocal as well.

Reviewed By: nbronson

Differential Revision: D5094598

fbshipit-source-id: 9980850805e1564e5c394af713e2fa17fe8844fe

Summary:
Test output before the fix is applied:
```
folly/experimental/Bits.h:247:59: runtime error: signed integer overflow: -2147483648 - 1 cannot be represented in type 'int'
     #0 folly/experimental/Bits.h:247 folly::Bits<...>::set(int*, unsigned long, unsigned long, int)
     #1 folly/experimental/test/BitsTest.cpp:228 std::enable_if<...>::type (anonymous namespace)::testSet<...>(unsigned char*, unsigned long, unsigned long, int)
     #2 folly/experimental/test/BitsTest.cpp:263 Bits_Boundaries_Test::TestBody()
    #17 folly/experimental/test/BitsTest.cpp:381 main
```

Reviewed By: philippv

Differential Revision: D5160789

fbshipit-source-id: 43f1926d58f1a5c019d4f8794d10a7a80a5c4749

Summary: All functions in this file are defined in `folly/fibers/FiberManager-inl.h`. There are no remaining references to this file.

Reviewed By: yfeldblum

Differential Revision: D5163160

fbshipit-source-id: 2fcb782a4de78f5c067610c876d0cf98f3b69e63

Summary:
Code may pass a callback which captures an object with a destructor which mutates through a stored reference, triggering heap-use-after-free or stack-use-after-scope.

```lang=c++
void performDataRace() {
  auto number = std::make_unique<int>(0);
  auto guard = folly::makeGuard([&number] { *number = 1; });
  folly::via(getSomeExecutor(), [guard = std::move(guard)]() mutable {}).wait();
  // data race - we may wake and destruct number before guard is destructed on the
  // executor thread, which is both stack-use-after-scope and heap-use-after-free!
}
```

We can avoid this condition by always destructing the provided functor before setting any result on the promise.

Retry at {D4982969}.

Reviewed By: andriigrynenko

Differential Revision: D5058750

fbshipit-source-id: 4d1d878b4889e5e6474941187f03de5fa84d3061

Summary:
The dirty hack I was using to link `folly_base` into `folly` doesn't work in newer versions of CMake :(
This refactors the makefile to not need to dirty hack by using an object library rule instead. It also makes the compiler version detection more generic.

Reviewed By: yfeldblum

Differential Revision: D5158919

fbshipit-source-id: cd052a5f58ed3d88c377c68cca07ca8497ca0c7a

Summary:
Callables that are passed as rvalues to makeFutureWith should be
executed as rvalues.
Generally callables that get captured by value should be executed
as rvalues, to allow passing e.g. folly::Partial objects that contain
move-only objects like unique_ptr or folly::Promise.
`collect` would move out of parameters passed as lvalues.

Reviewed By: fugalh

Differential Revision: D5120420

fbshipit-source-id: 1633c6b7e3fbb562f4d31e21d28c98b053de9912

Summary:
In mcrouter code, we would like a non-throwing alternative to `decodeVarint()`. There
are real, expected scenarios where only part of a serialized varint fits into a packet,
in which case mcrouter fails to decode the varint and throws.  In these scenarios,
throwing too many exceptions can lead to lock contention and degrade performance.

Reviewed By: yfeldblum

Differential Revision: D5153823

fbshipit-source-id: 138273af832903f0b04bee0bcacddd66b4274129

Summary:
Add `equals` interface to StringPiece (Range class). To support following case insensitive compare case:
```
folly::StringPiece a("hello");
if (a.equals("HELLO", folly::AsciiCaseInsensitive())) {
  // Do something.
}
```

Reviewed By: ot

Differential Revision: D5150495

fbshipit-source-id: 26816820f93959678f550768396f55293b5588cb

Summary:
[Folly] Fix `tryTo` to support conversion to enumerations.

`tryTo` should return `Expected<Tgt, ConversionCode>` instead of `Tgt`.

Reviewed By: yfeldblum

Differential Revision: D5144706

fbshipit-source-id: cd23f3cf75de7c5a26bc569f3cb47fff360f6e2a

Summary:
`#include` inside a namespace is disallowed when using clang modules. Here I
just unconditionally include them at top level, since they are available in all the
relevant platforms.

Reviewed By: yfeldblum, Orvid

Differential Revision: D5140355

fbshipit-source-id: bb87225c1d8f4ac64b90a7aff5f912c13e150e3a

Summary:
allow-large-files
Moving DestructorCheck from proxygen library to folly.

Reviewed By: djwatson

Differential Revision: D5110897

fbshipit-source-id: 02a6cf6336000e8c36484e75d2da820588baf2df

Summary: This has always been the design rule for what belongs in `folly/portability/`, but was never written down anywhere except in my head, so put it somewhere people can actually find it so I can free that space in my head for other things.

Reviewed By: yfeldblum

Differential Revision: D5140587

fbshipit-source-id: 1f09c17b7ccfff780e1825670a7ffaa75cd1b1d5

Summary:
In contrast to Container::insert(), for any STL container class, Container::emplace() does not have an overload that takes a Container::iterator as its first parameter. (The reason is to prevent ambiguity with respect to target class constructor parameters.) Instead, container classes have a separate emplace_hint() function.
Because of this separation, folly::emplace_iterator would fail to perform hinted emplacement in constructs where std::insert_iterator would perform hinted insertion.
This diff adds a new class, folly::hint_emplace_iterator(), and corresponding convenience function, folly::hint_emplacer(), which calls Container::emplace_hint() rather than Container::emplace().
It would have been trivial to copy&paste the existing folly::emplace_iterator class and simply replace the emplace() call with emplace_hint(), but I did not like the large amount of code repetition. So I decided to do some refactoring and move the emplace()/emplace_hint()/emplace_front()/emplace_back() calls into separate trait classes, and replace emplace_iterator()/hint_emplace_iterator()/front_emplace_iterator()/back_emplace_iterator() with suitable type aliases.

Reviewed By: yfeldblum

Differential Revision: D5097860

fbshipit-source-id: c0b733131a0d0d21fc0a8b08ad655142c6a41c19

Summary:
We were hitting a bug in CMake's Ninja generator that was causing it to incorrectly attempt to escape the closing quote on `_HAS_AUTO_PTR_ETC=1`, so remove the quotes, as they aren't actually needed.
This also makes a few quality-of-life changes to make building Folly with Ninja slightly easier.

Closes: https://github.com/facebook/folly/issues/604

Reviewed By: yfeldblum

Differential Revision: D5134585

fbshipit-source-id: 3816e53774f7963b9b4fd8ddf68b83ff7afdefe1

Summary:
Clearly someone noticed this was an issue before and fixed the other 2 overloads, but never fixed these :(

Specifically, ARM has unsigned chars, so this would be producing the wrong values there.

Reviewed By: yfeldblum, mzlee

Differential Revision: D5122992

fbshipit-source-id: eb71dea82c187c09c2c4496f1f7c99bd01db6dd0

Summary: `[[nodiscard]]` is the future. Let's start now.

Reviewed By: yfeldblum

Differential Revision: D5108297

fbshipit-source-id: c98f44af9e282616af92f9171516b6ea18e68c6d

Summary:
* Add streaming interface to the `ZstdCodec`
* Implement `ZstdCodec::doCompress()` and `ZstdCodec::doUncompress()` using the streaming interface.
  [fbgs CodecType::ZSTD](https://fburl.com/pr8chg64) and check that no caller requires thread-safety.

Reviewed By: yfeldblum

Differential Revision: D5026558

fbshipit-source-id: 61faa25c71f5aef06ca2d7e0700f43214353c650

Summary:
Adds a C-style streaming API to `folly/io/Compression.h`, with a zlib-esque interface.
Stacked diffs will add streaming support to zstd, zlib, gzip, lzma, lz4_frame, and automatic codecs.
This interface is targeting advanced users who are building higher level interfaces.
They can use this as a common base so they don't have to reimplement the same code for every codec.

Reviewed By: yfeldblum

Differential Revision: D5026332

fbshipit-source-id: e3abf1767b493c2fef153b895858a3a81b67d989

Summary:
This is generated by applying clang-tidy -checks=modernize-use-override to all .cpp files under folly.
It enforces the use of the keywords virtual, override and final in a way compliant to the style guide.

Reviewed By: yfeldblum

Differential Revision: D5108973

fbshipit-source-id: 61c55aa372feebe3b3be12a2705a04879aae57a4

Summary:
This class serves a similar use case to `ReadMostlySharedPtr`, but with two differences:

- It returns actual `shared_ptr`s, so it can be used as a drop-in replacement.
- Its memory utilization is constant, independent of the number of threads that use it.

Also, the code is much simpler. On the other hand, it is about 4x slower (but it still scales linearly in the number of concurrent threads acquiring the reference).

Reviewed By: djwatson

Differential Revision: D5104730

fbshipit-source-id: 8c18b635e0390394b06417b6df8b790e4bd2d90d

Summary:
`std::optional` allows heterogeneous comparisons, so `folly::Optional` should as well.
This allows numerical comparisons between different types, like `size_t` and `uint64_t`.

Fixes https://github.com/facebook/folly/issues/602.

Reviewed By: AsyncDBConnMarkedDownDBException

Differential Revision: D5110651

fbshipit-source-id: 34f3368283953033fbb2423ab30b04e38b5b7974

Summary:
[Folly] RFC: Embed `exception_wrapper` directly into `Try`.

Rather than storing it elsewhere on the heap. With `exception_wrapper` at 24 bytes on x64, it may now be small enough. However, it will expand the size of `Try<T>` for `sizeof(T) <= 16`, giving `Try` a new minimum size of 32 bytes on x64 instead of 16.

Reviewed By: ericniebler

Differential Revision: D5051436

fbshipit-source-id: 10d59686d64382c88d54340c97567eafb3e2f682

Summary: [Folly] Fix some old license headers.

Reviewed By: meyering

Differential Revision: D5110463

fbshipit-source-id: c7bb57194e0d33fbe1c6ddaa7d94437833b0cb0d

Add FOLLY_NODISCARD for [[nodiscard]] attribute when it exists, FOLLY_WARN_UNUSED_RESULT uses FOLLY_NODISCARD.

Summary: The `nodiscard` attribute was added to standard C++ in C++17. Prefer the standard formulation when it is available; otherwise, use `__attribute__((__warn_unused_result__))` on compilers that support the GNU extensions, and `_Check_return_` on MSVC. The old `FOLLY_WARN_UNUSED_RESULT` is now expands to `FOLLY_NODISCARD`.

Reviewed By: yfeldblum

Differential Revision: D5105137

fbshipit-source-id: 9aa22e81cd9f0b89f9343433aeae3ba365227ccb

Summary: This silences a few warnings under MSVC, and gets Folly closer to being able to compile with `-Wconversion` enabled.

Reviewed By: yfeldblum

Differential Revision: D5100686

fbshipit-source-id: e5e04c7aa143597e151a56a4d1f90dda26d0033b

Summary: Because just having a linter enforce it isn't good enough to catch everything.

Reviewed By: yfeldblum

Differential Revision: D5101828

fbshipit-source-id: a1bc482b37139a1eddeb93bc298596af950aa87d

Summary:
This makes it easier to consume Folly from a non-cmake project.
Note that this also requires making libevent, which was previously categorized as a semi-portable header into a non-portable header, meaning it can't be referenced directly. (libevent includes `Windows.h` in one of its headers -_-....)

Reviewed By: yfeldblum

Differential Revision: D5106051

fbshipit-source-id: 5ce2d4188c9036d6ab206af3036c8fd4b09516a4

Summary: To take advantage of read locks.

Reviewed By: siyengar

Differential Revision: D5106090

fbshipit-source-id: a32afd698e9204196aa3d23f21a7d41803b2eb66

Summary:
[Folly] Refactor `folly/gen/test/StringTest.cpp`.

Prefer to compare whole values once per test case, rather than sequentially comparing each part in turn.

Reviewed By: andrewjcg

Differential Revision: D5101467

fbshipit-source-id: 3d19d32bc6a5ecc5863bea76c4292935c775fe75

Summary: new method byLineFull will not trim the delimiter so that consumers can check if final line is ended up with delimiter or not.

Reviewed By: philippv, yfeldblum

Differential Revision: D5085371

fbshipit-source-id: 5045127ee11d008e3cd7d13d33bffad280fe0a7e

Summary:
Looks like we never tried to compile or run the PicoSpinLock under aarch64. There were two problems:

if sizeof(IntType) is 64, we try to do
  1 << 63
in lock and unlock. This doesn't compile because sizeof(1) is 32.

Also for try_lock, we were returning the data instead of checking for the locked bit.

Reviewed By: djwatson, yfeldblum

Differential Revision: D5091656

fbshipit-source-id: 4f81b8b32633e87b491548a1a24b9bcf20264063

Summary: A define was needed to be able to link when built against a shared-library version of gmock and gtest.

Reviewed By: yfeldblum

Differential Revision: D5100409

fbshipit-source-id: 9a4c9df20f682a25edde642c451476b2d1e1ae22

Summary: This includes adding options to toggle `/std:c++latest`, and `WIN32_LEAN_AND_MEAN`, along with giving a nice drop-down selector to `MSVC_FAVORED_ARCHITECTURE` when configured via the GUI, dropping `_WINSOCK_DEPRECATED_NO_WARNINGS` as it wasn't needed, and re-enabling warnings 4365 and 4775 within the scope of the standard library.

Reviewed By: yfeldblum

Differential Revision: D5100586

fbshipit-source-id: 87b20c6415391d4f0c75837cad5bc312910bf49f

Summary: MSVC needs these to be marked as `static` in order to be accessible within the lambdas without capturing them.

Reviewed By: yfeldblum

Differential Revision: D5099607

fbshipit-source-id: 0270e2191d504f74b1902789b662e1766b2056c4

Summary:
MSVC is being grumpy, and I want to continue to work from the preview build, so commit the workaround, as there's no significant difference.
Upstream bug report: https://developercommunity.visualstudio.com/content/problem/59614/compiler-crash-when-constexpr-constructing-stdchro.html

Reviewed By: yfeldblum

Differential Revision: D5099876

fbshipit-source-id: 64d5573d78b14339ff3177dd910a22787463ccd2

Summary: `std::hash` lives in `<functional>`, not `<xstddef>`.

Reviewed By: yfeldblum

Differential Revision: D5097468

fbshipit-source-id: d0684491d1d4b0f83f07ad837a66a915ce2ec8a6

Summary: 4.0 has been out for three months now, so it's pretty safe to remove them.

Reviewed By: Orvid, yfeldblum, mzlee

Differential Revision: D5096007

fbshipit-source-id: 595c3e86cd7cf1a1706f27f107fb0af8ab89a676

Summary: Use SafeAssert instead of assert.

Reviewed By: nbronson

Differential Revision: D5085575

fbshipit-source-id: daeff427b9b1fc5dff6ea986d16a70364acd2dd7

Summary: Name all repeating function threads for ease of debugging.

Reviewed By: yfeldblum

Differential Revision: D5065281

fbshipit-source-id: e875e654dfa644a265e44416baf5fbf23c9da434

Summary:
Previously we defined an assignment from std::string on all basic_fbstring
specialisations, but we don't actually want to do that because it's nonsense to
construct a basic_fbstring<char16_t> from an std::string. Equally it's not
nonsense to construct one of those from an std::u16string but the previous code
didn't allow it.

We now permit basic_fbstring<C, T, A1, S> to be constructed or assigned-to from
std::basic_string<C, T, A2>. The character type and traits must match but the
allocator is permitted to vary.

Background on my determination that the allocator type was unimportant and
could be disregarded: In part this is because C++17 made the same choice for
basic_string_view. Another factor was C++17's std::pmr::string (it's a
std::string with a different allocator) which I thought should be convertible
to fbstring in the same way as std::string.

Reviewed By: Gownta

Differential Revision: D5060569

fbshipit-source-id: f8984c528b76356240970c67916c58995d3f228d