Release 1.0.0

[ro0NL]

would be nice to have 1.x stable

Major version zero (0.y.z) is for initial development. Anything MAY change at any time. The public API SHOULD NOT be considered stable. (https://semver.org/)

[glennmcewan]

It would be great to have an idea of what's left on the roadmap for v1 even if there's no time estimate; we're looking to move away from Trikoder as soon as this is marked as stable.

[ro0NL]

@glennmcewan it runs in prod, it's stable.

cc @chalasr

[djsunderland]

@ro0NL can you share what kind of production environment(s) it is running on?

We are looking to move our SaaS financial services CRM to this and therefore its mission critical it functions at scale.

[chalasr]

This builds on libraries your CRM certainly relies already, it does not do anything fancy with them by default and have extension points to customize its behavior at pretty much every part that comes with perf implications (storage), so it's really difficult to imagine it not running at scale.

The only concerns for which 1.0 wasn't tagged yet are architectural, mostly opinionated concerns. With ~150K installs/month, 3M installs overall and 3 years since the first pre-release, I think enough time has passed and we're good for a stable release that I'd be safe installing on any enterprise project.

[ajgarlag]

Before tagging v1.0, I'd like to propose a PR to rename the NullUser class. I think this class should be renamed to ClientUser or ClientCredentialsUser and that the getUserIdentifier method should return the client ID. Returning an empty string has been deprecated since Symfony 7.2 (symfony/symfony#58007). What do you think?

Update: See #217