The issue is caused by deleted `std::__detail::begin` declared in `bits/iterator_concepts.h`. This would be found by ADL, and because it is deleted, compilation would fail. This change makes it so that we SFINAE on `begin(std::declval<T>())` and `end(std::declval<T>())` being well-formed.
D:\vcpkg\toolsrc\include\catch2\catch.hpp(11285): warning C6330: 'char' passed as _Param_(1) when 'unsigned char' is required in call to 'isspace'.
D:\vcpkg\toolsrc\include\catch2\catch.hpp(11288): warning C6330: 'char' passed as _Param_(1) when 'unsigned char' is required in call to 'isspace'.
ISO/IEC 9899:2011:
"7.4 Character handling <ctype.h>"/1
[...] In all cases the argument is an int, the value of which shall be
representable as an unsigned char or shall equal the value of the macro
EOF. If the argument has any other value, the behavior is undefined.
This means if isspace was passed a character like ñ it could corrupt
memory without the static_cast to treat it as a positive value after
integral promotion (and C libraries commonly use the int index supplied
as a key into a table which result in out of bounds access if the
resulting int is negative).
Ideally, clang-tidy would be smart that if one alias of a warning
is suppressed, then the other one is suppressed as well, but as of
right now, it isn't. This means that for now we have to suppress
both aliases of this warning. Opened upstream issue to fix this:
https://bugs.llvm.org/show_bug.cgi?id=45859
Obviously, ideally clang-tidy would also not warn that we are calling
a vararg function when it is an unevaluated magic builtin, but that
also is not happening right now and I opened an issue for it:
https://bugs.llvm.org/show_bug.cgi?id=45860Closes#1921
Doing some benchmarking with ClangBuildAnalyzer suggests that
compiling Catch2's `SelfTest` spends 10% of the time instantiating
`std::unique_ptr` for some interface types required for registering
and running tests.
The lesser compilation overhead of `Catch::Detail::unique_ptr` should
significantly reduce that time.
The compiled implementation was also changed to use the custom impl,
to avoid having to convert between using `std::unique_ptr` and
`Catch::Detail::unique_ptr`. This will likely also improve the compile
times of the implementation, but that is less important than improving
compilation times of the user's TUs with tests.
This simplified variant supports only a subset of the functionality
in `std::unique_ptr<T>`. `Catch::Detail::unique_ptr<T>` only supports
single element pointer (no array support) with default deleter.
By removing the support for custom deleters, we also avoid requiring
significant machinery to support EBO, speeding up instantiations of
`unique_ptr<T>` significantly. Catch2 also currently does not need
to support `unique_ptr<T[]>`, so that is not supported either.
As far as I know, only a few users actually use it, but these changes
allow us to avoid including a surprising amount of code in the main
compilation path.
As part of `-Wdeprecated-copy-dtor` sweep, I noticed that Capturer's
copies are defaulted. Given that the class would likely break horribly
in the event of actual copy happening, they are now deleted.
This should improve the compilation times by decreasing the number
of TUs compiled, without making overly big TUs that would cause
problems with heavy-tailed compilation times.
There is one "combined TU" for the top level part, and each subpart,
except for Reporters, which currently do not have any trivial TUs.
