Re:Catching integer overflows with template metaprogramming ...
by winstonewert on Monday March 27, 2017 @12:56AM
(#13965277)
I'm unclear on how this would work out practically. Cases where my integers belong to a fixed range seems rare. I'd like to have seen more discussion of how it caught the bugs it did.
[ Parent
| on HN ]
Re:Catching integer overflows with template metaprogramming ...
by kentonv on Monday March 27, 2017 @01:24AM
(#13965370)
Cap'n Proto by its nature is doing a lot of work with integers that represent sizes, indices, offsets, addresses, etc. These tend to have well-defined ranges -- and unusually bad failure modes. So it works unusually well here.