Patrick O'Connell

A related issue is that even if you manage to construct a `Box`, AFAICT it's not possible to call it because it can't be moved from: ```rust fn f(b: bumpalo::boxed::Box`...

As a workaround, you can use a predefined macro to document it as unconditionally explicit: ``` MACRO_EXPANSION = YES EXPAND_ONLY_PREDEF = YES PREDEFINED = explicit(condition)=explicit ``` This at least avoids...

[This page](https://isocpp.org/wiki/faq/assignment-operators) contradicts the rule: > Self-assignment is not valid for move assignment. I agree with the above arguments against C.65, the `swap(a, a)` claim is wrong and it breaks...

@Snektron Hey, sorry for the delay on this! I haven't started on the implementation but did finally look into how these functions can be classified. Seems like the most reliable...

@Snektron Ping, waiting for your thoughts on this whenever you get the chance: > IMO it's maybe worth forgetting about this and just leaving the whitelist. It's not necessarily pretty...

No problem! First off, I don't want to take up too much of your time with this so if you don't feel like responding in detail then that's totally fine...

A future extension could add a possible error code (e.g. some of the "maintenance" extensions modify existing functions like this), and given that it returns a `VkResult`, Khronos might think...