[LLVMdev] New type of smart pointer for LLVM

[Duncan P. N. Exon Smith]

> On Oct 2, 2014, at 2:43 AM, Anton Yartsev <anton.yartsev at gmail.com> wrote:
> 
>> I'm still concerned this isn't the right direction, and that we instead want a more formal "maybe owning pointer". The specific use case you've designed for here, in my experience, doesn't seem to come up often enough to justify a custom ADT - but the more general tool of sometimes-owning smart pointer seems common enough (& problematic enough) to warrant a generic data structure (and such a structure would also be appliable to the TGParser use case where this came up).
> David, could you, please, clarify the concept of the "maybe owning pointer"?

I think the idea is something like the compromise I suggested [1] as an
alternative to misusing `std::unique_ptr<>`.  But it's even better to
design away "maybe-owning pointers" entirely.

[1]: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20140811/230727.html