search

web-platform-dx / web-features

Exploring how to present Web platform features adoptability
https://web-platform-dx.github.io/web-features/
Apache License 2.0
365 stars 71 forks source link

Baseline status regressions should be more obvious on review #1947

Open ddbeck opened 2 weeks ago

ddbeck commented 2 weeks ago

It'd be nice if PRs that change a feature's Baseline status from high or low to false get more scrutiny. I'm imagining a GitHub Actions workflow that leaves a comment or appends to the PR description when this happens.

I didn't notice that popover regressed when I upgraded BCD in https://github.com/web-platform-dx/web-features/pull/1797. Discovered via https://github.com/GoogleChrome/webstatus.dev/issues/739

ddbeck commented 2 weeks ago

The calculation is correct with respect to the data, so it's not a bug. But I believe this is the first time that we've moved a recent feature's status backwards solely on a BCD upgrade, which changes many files. Without a little more attention, we might allow an unintended status change to take place (e.g., from a later addition being tagged in BCD).

jcscottiii commented 2 weeks ago

Hey @ddbeck!

Thoughts on this one too?

https://github.com/GoogleChrome/webstatus.dev/issues/740

ddbeck commented 2 weeks ago

Thoughts on this one too?

GoogleChrome/webstatus.dev#740

I've updated that issue with more information and a link to my related BCD PR: https://github.com/mdn/browser-compat-data/pull/24720

Also, for anyone else following along with regards to the inciting popover feature. I shared my commentary (see https://github.com/GoogleChrome/webstatus.dev/issues/739#issuecomment-2407080876) with the WebDX chat and—so far—there's agreement that popover's status is correct. I think that one will stand, but explicitly affirming that fact before releasing an update would've been nice.