refactor: take the site chrome from the brand package - #27
Conversation
The bar, the links, the GitHub button and the footer are @origin89/brand 0.6.0's now. What stays is what this site adds: the product menu, how the bar collapses on a narrow window, and the amber the footer dot reports through --o89-chrome-status. The nav link rules were grouped with the menu button's, so only the link half moved; the button keeps its own.
|
Warning Review limit reachedNext included review available in 7 minutes. View limit detailsLimit details: You’ve used all 10 included reviews currently available. Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. Review configuration: ⚙️ Run configurationConfiguration used: Organization UI Review profile: QUIET Plan: Advanced Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (3)
Comment |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
Change
The floating navigation bar, the nav links, the GitHub button and the whole footer come from
@origin89/brand0.6.0 instead of being defined here.styles/chrome.cssgoes from 346 lines to 244.What stays is what this site adds to the shared chrome: the desktop product menu and its panel, how the bar collapses on a narrow window, and one line saying which state the footer dot reports — amber, through
--o89-chrome-status.The nav link rules were grouped with the menu button's (
.o89-nav > nav a, .o89-nav-menu > button), so only the link half moved. The button keeps its own copy.This site is the reference the package's chrome was taken from, so nothing here should look different.
Validation
just checkpasses (exit 0): lint, tests, type check, site build and Storybook build.The chrome is verified directly rather than through the screenshots, because the six the run photographs do not contain it — a crawl of all 55 stories finds
.o89-navand.o89-footerrendering in eight, none of them among the six. So both builds were served side by side and every element inside the bar and the footer compared across 29 computed properties plus::after, on a story that renders them. All 100 elements compute identically.The screenshot pass itself reports 55 stories with no page errors. Three of its six images differ from a build of
main, and those are the known flake rather than this change: two runs of the samemainbuild reproduce one of them exactly, and the chrome is absent from all six regardless. That flake is narrower than it was but not gone — see the note below.The flake fix in #26 is incomplete
That PR reported eight runs and 42 comparisons with zero differing pixels, which was true as measured in its worktree and has not held up. On
main,buddy-map-390still differs by 227 pixels between two runs of one build, with the same signature as before: part at the avatar, part in the map area. The image waiting removed most of it — five of six files were moving, now one is — but something in the map story is non-deterministic beyond images, and I have not found it yet. Worth its own pass rather than a claim that it is fixed.