e.g. we may introduce abstract bases like "loop" that the grammar doesn't care about in order to model is-a concepts that might make refactorings more expressive. This is less natural in a grammar-like idiom.

In D89995#2359635, @mehdi_amini wrote:

@gribozavr2 the paragraph now starts with "If shortly after landing a commit, a developer expresses concerns ..." ; WDYT?

Fair enough. But I don't think that Clang developers just copied the implementation from GCC or from MSVC.

Generally, I think this patch reflects the existing expectation in the community, when the post-commit feedback is provided promptly.

I am having a hard time to accept "this is how it is implemented in our fork" as a technical argument. Besides, I am not sure how could the Clang community benefit about being backward compatible with a specialized fork and thus making superfluous complications.

Could you add tests that verify the pairing of elements and delimiters?

Any chance of adding a test?

In D89544#2334494, @lebedev.ri wrote:

What reasoning was used in the patch that added that wording?

Thank you! LGTM assuming you only moved the tests into separate cpp files (I didn't verify that the text is exactly the same).

Unfortunately I see this change causing ASan failures in llvm-objcopy/MachO/strip-all.test:

Google C++ style guide does not have this rule anymore. Thanks for the cleanup!

This introduces the new swift_name attribute that allows annotating
interfaces with an alternate spelling for Swift. This is used as part
of the importing mechanism to allow interfaces to be imported with a new

In D87683#2274883, @martong wrote:

In D87683#2274663, @baloghadamsoftware wrote:

In D87683#2274569, @njames93 wrote:

Please fix the test case first, can't call operator new(unsigned long, void*) with an argument of type void*
The other failures the pre merge bot detected can safely be disregarded

Placement new is defined per standard:

void* operator new  ( std::size_t count, void* ptr );

Here, the problem is that size_t is unsigned long on Linux and it seems that it is unsigned long long on Windows. How should I overcome this?

In CSA, to overcome the platform problem, it is common to add a target triple for the RUN line:
https://github.com/llvm/llvm-project/blob/master/clang/test/Analysis/std-c-library-functions.c#L14

Sorry, I didn't realize you don't have commit access. Pushed this patch to git, thanks for the contribution!

LGTM!

Please also add tests for calls to constructors without arguments and calls to implicit constructors 1 argument, as requested in https://reviews.llvm.org/D86700.