Liam Miller-Cushon

I pushed a release that merges `error_prone_type_annotations` into `error_prone_annotations`, so it should be possible to just depend on `error_prone_annotations` to avoid the split package issue: https://github.com/google/error-prone/releases/tag/v2.37.0

> in general, they'd like to be on the newest version of Guava This is generally how I was thinking about it. I can try to publish a new release...

I cut a release that downgrades to Guava 33.4.0: https://github.com/google/error-prone/releases/tag/v2.39.0

The latest releases of https://github.com/googleapis/java-cloud-bom have updated to Guava v33.5.0, I am going to update Error Prone's version to match in the next release

@ascopes that looks unrelated. `ImmutableMap$Builder#buildOrThrow` was added in [Guava 31.0](https://github.com/google/guava/releases/tag/v31.0), that crash means there's an older version of Guava on the same processorpath as Error Prone

Thanks for the bug and the investigation! Presumably the difference in JDK 23 is due to the new support for [markdown doc comments](https://openjdk.org/jeps/467).

I think it's probably a deliberate change on the javac side, to be able to process the entire `///` comment as a single token instead of multiple line comments. Your...

I agree the difference between `FormatStringAnnotation` and `FormatString` is surprising and seems arbitrary. When they were created, there were difference use-cases. `FormatStringAnnotation` was deliberately stricter, and intended to be used...

This is currently intentional, although there's some ongoing internal discussion about what the Google Java Style Guide should say about text blocks, and the 'outdenting' behaviour may change. I think...

I am going to close this out for now. The 'automatic deindent' behaviour is being revisited, and I will update #1087 when that happens.