Lázaro Clapp

Just as a note, whatever instructions we end up giving, we should make sure that our sample projects use as similar a setup as we can, while still pulling the...

Hi @talios, First of all, really sorry this bug report went without response for almost a full month, entirely my fault. Unfortunately, I just tried to repro the issue, and...

This sounds doable, but only with a custom handler, since library models are not currently expressive enough to reason across multiple chained calls like that (e.g. `isNotNull()` performs the nullness...

We are actively working on this, actually! 😉 Just to set expectations, though: the general version of "take an annotated codebase and add annotations in all the right places" is...

This is way trickier than it seems at first, see #305 : > One potential issue I see with this change, is that, while it is likely to work for...

If we do implement the access-path capable version of #96 , then this can be probably handled by the same code.

Note that the issue as reported for the original code here is not that those fields can be set to `null` but that they aren't guaranteed to be set to...

> Yes, I'd also like to see the error message you're getting. > > Also, note that there is no easy way for Error Prone / NullAway to analyze the...

Worth trying and seeing what the performance impact actually is. Did we ever measure this and found it was unacceptable and hence that default behavior?

Hey @vinothsvk92, do you have a repro case for this? There are tests currently in NullAway that should fail if that option weren't working at all, so we need more...