Carl Dehlin

I like the idea of being able to further examine the exception. If this is to be implemented I think the second version with `CHECK_THROWS_WITH_PREDICATE` is more viable since it's...

I see, that makes this feature more important. Here is a workaround draft ```cpp #define DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN #include struct PosixException : std::exception { int code; PosixException(int code) : code{code} {} };...

Yes I think it is a good feature, and this issue clearly demonstrate the need for it. It might take some time before that happens so I always try to...

Just replace `static` with `thread_local`? Might be worth scanning the rest of the code base for non-reentrant functions.

@opencv-alalek This is not a duplicate of https://github.com/opencv/opencv/issues/9336 . That (low quality, no repro code) issue talks about nested (i.e. vertical) parallelization, this issue is about horizontal.

Here is minimal reproducing example (MRE). It starts by allocating an big 20k x 20k image, and then performs two image up-samplings in parallel using lanczos4 interpolation (to maximize processing...

Ah sry, ofc you cant replace `static` with `thread_local`, brain fart on my side. This issue might require more effort to solve that I thought. Not experienced enough with the...

I would do as suggested above. You can also implement a custom wrapper type around float that have the desired non-standard behavior for equals. BUT, I am very curious why...

This issue is tricky. Changing minimum required version to 3.5 breaks downstream consumers who use 3.0 ... 3.4 . As noted in the deprecation warning, one can whitelist maximum allowed...

That works except for version 4.0 which is still incompatible. Thinking a bit more about this, is it possible to maintain =4.0 simultaneously? What changes to the cmake files are...