Liam Miller-Cushon

Results 378 comments of Liam Miller-Cushon

I'd like to suggest some additional changes: I updated the warnings about truncated reads of strings containing `\0` to mention `MemorySegment#getString(long, Charset, long)` (see discussion here: https://github.com/openjdk/jdk/pull/28043#discussion_r2549972828) I realized `StringSupport#copyBytes`...

> Also, if in the future we also enhance encoders to return the length of a string w/o encoding it, it might still be useful to have `copy` return the...

Thanks again for the reviews! I think the next step is to got through the CSR process (and wait for the JDK 27 branch to open). I made some updates...

I have tentatively renamed the parameter of `getString` from `length` to `byteLength`, in response to discussion in [the CSR](https://bugs.openjdk.org/browse/JDK-8372338)

I took a quick look at this and it seems like javac is dropping `@Override` annotations on record components before Error Prone sees them, e.g. for ```java public record R(@Override...

Sorry false alarm, it's possible to get them through `ClassSymbol#getRecordComponents` -> `RecordComponent.accessorMeth`

It would be be helpful to provide a summary in PRs and issues of what the goal is, instead of linking to discussions in other issues and PRs, it isn't...

> somehow picnic causing the spam and not being muted when the switch -XepDisableAllWarnings is on. From https://github.com/diffplug/spotless/issues/2745 and https://github.com/diffplug/spotless/pull/2754 it sounds like `-XepDisableAllWarnings` was not previously being used in...

Can you provide a minimal self-contained repro? Note that picnic is an extension of Error Prone, it is a separate project: https://error-prone.picnic.tech/

It isn't clear if this is a bug in Error Prone, or picnic, or a configuration issue with that specific build. If you are able to extract a minimal self-contained...