David Pursehouse, Ben Rohlfs, Alice Kober-Sotzek, Patrick Hiesel, Luca Milanesio
Online, April 21, 12:30 - 13:30 CEST
The next meeting will be held on May 5, 12:30 CEST.
Some new features and improvements can be included in the next forthcoming Gerrit project news:
As usual, we invite the community to propose any items that they think would be interesting.
Bazel v3 doesn't include the new version of errorprone, which would allow enabling more static analysis to the build and validation process of incoming Gerrit changes. David Pursehouse will continue the research with David Ostrovsky for understanding how to use a more recent version of errorprone and how to include in our Bazel build.
ElasticSearch has been flagged experimental for many releases. Marco Miller (Community Manager) has agreed to prepare and circulate a survey for understanding the current use of ElasticSearch and its desired version to be supported (see issue 40011610). Once the survey would be completed, the ESC can insert the ElasticSearch support in the Gerrit roadmap, with details of which version to consider and where the official readiness for production lies in the current or future Gerrit releases.
The most recent statistics of the reviewers plugin statistics have been published on Google docs. The contributions have only considered the non-trivial commits and do not want to define a qualitative analysis of the contributions made by different people (e.g. Dave Borowitz‘s contributions were just adaptations, reformatting and refactoring rather than the design of new functionalities). By looking at the fit for the reviewers plugin with the core Gerrit use-cases, it looks like the functionality makes sense and the current feature set seems like a nice addition to Gerrit. For the complete assessment, see the Issue 40010546. The current implementation and user-journeys need some refinement (e.g. The suggestion of reviewers don’t seem reasonable and the general behavior for WIP/private changes would need to be changed). From the look&feel perspective, there are minor issues with padding and the style can be improved. However, the issues are not a blocker and can be fixed easily afterwards. The code looks small and properly covered by integration tests. There are some
Thread.sleep() as a means of synchronization in the tests, therefore suboptimal but it is not a blocker and can be possibly fixed (see issue 40010545). Overall the issues found are currently blockers in making the reviewers a Gerrit core plugin. However, the final decision is postponed to when the issues raised will be resolved.
We discussed and finalized the release plan of Gerrit v3.2. A separate news item will provide all the details and dates.
More feedback is needed for the instance id proposal. The proposed change is very small, (almost a one-liner) so, if the design would get approval, it could be tentatively included in Gerrit v3.2.
There aren't major follow-ups or news on the other design proposals.
With Gerrit v3.2 approaching some of the items currently on the roadmap are most likely to be postponed to v3.3.
Once the Gerrit v3.2-RC0 will be cut, the updated roadmap will be published to the Gerrit home page with more details.
The issue is currently assigned to the ESC, however a follow-up with the CMs is needed to understand the possible involvement that we can have.
It was supposed to happen during the next forthcoming Gerrit Hackathon in April/May 2020, now cancelled because of Covid-19. A follow-up discussion needs to happen with the CMs to understand how this can be coordinated remotely with all the different time-zones to cover.
The discussion has not started yet, the focus will be possibly postponed to Gerrit v3.3.
There were no other issues that have updates or require attention.