Tony Murphy
Tony Murphy
looks ok. in general, i would usually check in in main first, then back port to all necessary branches
I'm not comfortable approving this review without the addition of non-deterministic stress tests. These have proven to be essential at finding complex interactions between clients where eventual consistency doesn't hold....
> My intention was to do them in separate PRs, after "setting the stage up" with this one. Just trying to keep them small and easy to review (and to...
added a bunch of comments around how we can minimize the surface area impact here, which seems like a good plan to side step the concerns here, and give us...
i worry about testability and maintainability here. it been our stance that telemetry is for diagnostic only, and there are no guarantees on stability. I'd worry if we make other...
what was causing the batches to be valid? were they regenerated with reconnect?
a thought i had on this area, could we augment the reconnect farm to hit it? that test is pretty simplistic right now.
there is another place we do a similar costly walk: packages\dds\matrix\src\permutationvector.ts would be good to ensure we can support that case as well.
we do use api extractor already, but it doesn't quite tell you if a change is breaking or not. just that the surface area changed, which can be ok in...
In general, most of the tooling i've seen depends on the developer to annotate changes in some way. This is how beachball, semantic-version, and yarn2 all work. From those annotations...