Rather -0 on my side by experience cause this kind of tool is often abused, in particular using an IoC or a codebase with indirections like maven. One prerequisite being all the stack uses it - plexus, sisu, commons etc - Im not sure we can get the benefits without the downsides.
That said you can test it in a branch and see what's the status, only big constraint is to not deliver it in the distro but this is easy to do. Le lun. 12 févr. 2024 à 02:43, Elliotte Rusty Harold <elh...@ibiblio.org> a écrit : > I'd prefer not. Every additional dependency we add is a liability. > It's not clear we need something like this and even if we do, JSpecify > is only a few months old. This is only the latest of many efforts to > establish a standard for nullness annotations. I don't see why > JSpecify will be the one ring to rule them all. XKCD 927 applies: > > https://xkcd.com/927/ > > Really what needs to happen here is someone needs to reboot the JSR > 305 work in the JCP. I'm not sure why it failed the first time, but > perhaps it can be fixed. > > On Sun, Feb 11, 2024 at 9:28 PM Benjamin Marwell <bmarw...@apache.org> > wrote: > > > > Hello devs and plugin maintainers! > > > > Since JSR 305 (Nullable annotations etc.) did not get implemented > > officially, and Maven allows `null` in a lot of locations, I wanted to > > ask about your opinions about using jspecify: https://jspecify.dev/ > > > > JSpecify was created by the Java community (i.e. Java decs but also > > larger corporations like Google, if I am not mistaken. It was seen as > > needed since JSR 305 never materialized and probably never will. > > Google hat a Guava Situation [1] and decided to do something about it. > > > > JSpecify says it is safe to adopt it [2]. Although the current version > > is 0.3, it is more like 0.9. > > > > I honestly think that Maven would benefit from those annotations, as > > many recent libraries try to avoid `null` as best as they can. Younger > > developers may not be used to seeing nullable parameters, and even > > some of us don't know which parameters can be nulled (or not). > > > > Let me know what you think about this idea. > > > > My personal opinion: I am pretty confident jspecify will displace > > checker-qual as the "top dog" of nullness annotations in the future. I > > also want to give plugin authors a way to use null in their plugins, > > too, although we could make it an optional dependency (just > > annotations...). > > > > Looking forward to reading your opinions! > > > > - Ben > > > > 1: https://github.com/google/guava/issues/2960 > > 2: https://github.com/jspecify/jspecify/wiki/adoption > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org > > For additional commands, e-mail: dev-h...@maven.apache.org > > > > > -- > Elliotte Rusty Harold > elh...@ibiblio.org > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org > For additional commands, e-mail: dev-h...@maven.apache.org > >
