On non-Windows platforms, there is no standard way to get the hash of an
item -- before this PR, what we did was check for the existence of a few
common utility names (shasum, sha1, sha256, sha512), and then call that
utility on a file we created containing the contents we wish to hash.
This PR adds internal hashers for sha1, sha256, and sha512, and
standardizes the interface to allow anyone to implement hashers in the
future.
These hashers are not extremely optimized, so it's likely that in the
future we could get more optimized, but for now we just call out to
BCryptHasher on Windows, since it's standard and easy to use (and about
2x faster for sha1 and sha256, and 1.5x faster for sha512). However,
they are reasonably fast for being unoptimized. I attempted a few minor
optimizations, which actually made the code slower! So as of right now,
it's implemented as just a basic conversion of the code on Wikipedia to
C++. I have tested these on the standard NIST test vectors (and those
test vectors are located in vcpkg-test/hash.cpp).
The VS2015 standard library requires, in debug mode, a comparison
operator on `T × U` and `U × T` to also be a comparison operator on
`T × T` and on `U × U`, and so in
vcpkg::Install::install_package::intersection_compare, I've added two
new `operator()` overloads which take those respectively, on VS2015.
Also, `[nodiscard]` was added to somewhere in `vcpkg/base/strings.h`,
which gives a warning in VS2015 -- thus, I added the `vcpkg/pragmas.h`
include, since that fixes the warning.
I added benchmarks to measure how fast the parallel remove_all code was
-- it turns out, about 3x slower than stdfs::remove_all. Since this was
the case, I removed all of the parallelism and rewrote it serially, and
ended up about 30% faster than stdfs::remove_all (in addition to
supporting symlinks).
In addition, I did the following three orthogonal changes:
- simplified the work queue, basing it on Billy O'Neal's idea
- Fix warnings on older versions of compilers in tests, by splitting
the pragmas out of pch.h.
- Ran clang-format on some files
In fixing up remove_all, the following changes were made:
- On Windows, regular symlinks and directory symlinks are distinct;
as an example, to remove directory symlinks (and junctions, for that
matter), one must use RemoveDirectory. Only on Windows, I added new
`file_type` and `file_status` types, with `file_type` including a new
`directory_symlink` enumerator, and `file_status` being exactly the
same as the old one except using the new `file_type`. On Unix, I
didn't make that change since they don't make a distinction.
- I added new `symlink_status` and `status` functions which use the
new `file_status` on Windows.
- I made `Filesystem::exists` call `fs::exists(status(p))`, as opposed
to the old version which called `stdfs::exists` directly.
- Added benchmarks to `vcpkg-test/files.cpp`. They test the
performance of `remove_all` on small directories (~20 files), with
symlinks and without, and on large directories (~2000 files), with
symlinks and without.
* [vcpkg] Modify Filesystem::remove and Filesystem::rename to not throw.
* [.gitignore] Ignore new VS2019 CMake integration default location
* [.gitignore] Ignore CMakeSettings.json in toolsrc
* [vcpkg] Time external processes called with System::cmd_execute
* [vcpkg] Work around VS2019 CMake bug
* [vcpkg] Fix several unused variable warnings.
* [vcpkg] Improve error handling in vcpkg::Files::Filesystem
Always require either std::error_code or LineInfo to print better errors.
* [vcpkg] Fixup missing return value.
Drive by fix: silence warnings in tests.
* [vcpkg] Fix exiting in error_code overload
Drive by fixes for /analyze with VS2019