Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Piotr P. Karwasz
Hi Gary, On 2.12.2024 20:28, Gary D. Gregory wrote: Just on the CP for one app tooling stack: org.jetbrains.annotations.NotNull org.jetbrains.annotations.Nullable com.sun.istack.NotNull com.sun.istack.Nullable org.springframework.lang.NonNull org.springframework.lang.Nullable jakarta.annotation

Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Gary D. Gregory
Just on the CP for one app tooling stack: org.jetbrains.annotations.NotNull org.jetbrains.annotations.Nullable com.sun.istack.NotNull com.sun.istack.Nullable org.springframework.lang.NonNull org.springframework.lang.Nullable jakarta.annotation.Nonnull jakarta.annotation.Nullable javax.annotation.N

Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Matt Sicker
Yeah, now I’m noticing even more, particularly from library-specific versions of the annotations. I’m glad we decided to use this library directly rather than introduce another set of annotations! > On Dec 2, 2024, at 13:28, Gary D. Gregory wrote: > > Just on the CP for one app tooling stack:

Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Matt Sicker
Yeah, that’s an annoying problem we have in Spinnaker. We have the javax.annotation one, we have Lombok’s annotation, there is the javax.validation one, the jakarta.validation update, and more nullability annotations from various static code analysis tools. Lombok is one of the things that look

Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Matt Sicker
I’m alright with the JSpecify dependency assuming it has accumulated the momentum we expected. Also, I thought annotations didn’t have to exist at runtime anyways as long as you weren’t using reflection on them. > On Nov 22, 2024, at 03:52, Piotr P. Karwasz wrote: > > Hi all, > > On 19.10.202

Re: Add JSpecify as `compile` dependency

2024-12-02 Thread Gary Gregory
The main issue with these annotations is that there are so many providers. It's so bad that tooling just now look for names like @NonNull _without considering the package name_, so there is no limit as to how much you end up with :-( On Mon, Dec 2, 2024 at 1:58 PM Matt Sicker wrote: > > I’m alrig

Re: Problems creating custom appender using JPMS service in 3.0.0-beta3

2024-12-02 Thread Matt Sicker
Yes, it could make sense to replace PluginService stuff with things based on CFIPP. I developed the DI system somewhat independently from the plugin system because of the need to support components from the API, and the plugin system was reimplemented in terms of the DI system, so any simplifica