Chris Povirk
Chris Povirk
@sdeleuze pointed out in our meeting that we should explicitly call out some common differences from the semantics of other annotations and tools, like those discussed in [his question on...
We've already observed that people are going to see this project and point us to https://xkcd.com/927/. And while we have a lot of docs floating around for our requirements, our...
(I don't think we've settled on a _format_ for stubs / external annotations, so I'll just follow the Checker Framework's lead and write Java sources -- even though it's possible...
Following #101's earlier proposal of a meta-annotation: For nullness, we don't assign meaning to annotations in certain locations, like `class @Nullable Foo {}` and `class Foo extends @Nullable Bar`. Perhaps...
I discovered in https://github.com/protocolbuffers/protobuf/issues/7827#issuecomment-676563016 that Animal Sniffer does not fail the build for dangerous covariant return types. This means that, if we build a Guava release with JDK 11 (which...
IIRC, we currently consider such an appearance of `@foo` to be an invalid tag (or, if it is a recognized one like `@return`, maybe a valid tag? or maybe never...
The background is that we run this code through a compiler that complains about unused variables. As I see it, it doesn't really make sense for the compiler to complain...
See summary in https://github.com/jspecify/jspecify/issues/43#issuecomment-881061588, with discussion starting in https://github.com/jspecify/jspecify/issues/43#issuecomment-700277489. It might be that all we can do is to encourage tools to report an error if compiling code that uses...
like https://github.com/jspecify/jspecify/pull/187#issuecomment-873183875 but nicer
The nullness annotations for those either aren't known to javac at all or aren't propagated by it into the places we'd want in the bytecode. This isn't a catastrophe: The...