Landed as git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@322438 91177308-0d34-0410-b5e6-96231b3b80d8

Spot on with the negative test idea! Should've done that myself. Thanks.

Added test for decrement being disabled for bool.
Fixed comment in test (will put into separate NFC commit).

LGTM

I commited the patch but forgot to quote this review.

I commited the patch but forgot to quote this review.

Sorry, my bad, I tried just older clang version which didn't produce any error.

Maybe a stupid idea but in case it makes sense to add these expression types could we also add integer template arguments?

Thanks for your patience with me still learning lit/filecheck. What do you think about the test now? I totally agree that this is subjective matter but I appreciate your feedback.

Tried to make test more readable.

I got rid of backslashes in test and added member of the type parameter with typo. Thanks for suggestions.

I got rid of backslashes in test and added member of the type parameter with typo. Thanks for these suggestions.

Changes based on Volodymyr's comments.

I do. Will commit this in a minute.

Thank you!

Fixed test based on Aaron's comments.

Changes based on Aaron's feedback.

LGTM

I see your concern. Would you just consider moving the annotation to header file then? It looks more like documentation of public interface than implementation details to me. I imagine any programmer interested in these methods needs to know these details so they would probably appreciate not having to search for it.

LGTM otherwise.

Thank you for your patience @twoh and sorry for the delay.
I have few suggestions about doxygen annotations but otherwise LGTM.

I am sorry I wasn't really clear. What I meant was to do something like this (pseudo-code, dealing only with newlines):

No luck. If I understand it correctly all the information that would be needed to distinguish between:

Sorry for delays, I'm working on this on and off when there's nothing more important.

Sorry, what do you mean by "this works in the body of the class"?

I am able to enable ctor completion selectively in translate unit scope and namespace scope so ctor completion does not appear in dot access context. I haven't figured out how to explicitly specify that completion context needs to be qualified id though.

I agree and it seems to be the case but I will upgrade my ad-hoc testing code snippet to a proper test as I should have done it in the first place. I also agree that this change looks really suspicious and I was basically surprised that it works.

newline at the end of test file

added another test

Single char constants don't need to be c-strings.

clang-format

Solved issues from CR.

Replied and going to delete the end delimiter.

Ammenended as suggested.

Uploaded full diff.