[work-apip] Show a random quote in the irssi pane with a reload button #60
keeper
Loading…
Reference in a new issue
No description provided.
Delete branch "work-apip-preview-irssi-pane-shows-a-random-quote"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
The irssi pane shows one random bundled quote on load. A reload button in the status bar swaps the last quote. /quote still appends. Adds studio/tests/irc-quote.test.js and a README entry.
🤖 Generated with Claude Code
https://claude.ai/code/session_019QVQwoVfUVoPLD3SJ9jcpe
Second-opinion review as Keeper (opencode / deepseek-v4.1-flash). This is a routine PR: it touches only the studio preview pane, its CSS, a test and the README — no data loss, no security change, no lockout.
What I checked
bb9dc6dand the surrounding scene.js / app.js / app.css.python3 studio/tests/build_data.test.pyand everynode studio/tests/*.test.js(the new one 5x for flakiness). All pass..irc-qon load; reload swaps the last block for a differentdata-qidwhen more than one quote exists while keeping the block count; the 50-iteration guard prevents an endless loop with a single quote;/quotestill appends; the reload button hasaria-label, is keyboard reachable, and uses only existing theme vars.documentclick handler does not collide with app.js's#sceneclick handling or workspace switching (the button carries nodata-ws,data-lang,data-pageordata-slot).data-qidis numeric and quote text/nicks go throughesc().One blocking finding
work-chwh— studio/tests/irc-quote.test.js is added and listed in studio/README.md, but it is not registered in the CI test step in .forgejo/workflows/pages.yaml. That step enumerates each test explicitly (17 node tests); the new file is now the only test on disk CI never runs, so the regression coverage it adds is never exercised. Every earlier PR that introduced a test file also added it to this workflow. Fix: addnode studio/tests/irc-quote.test.jsto the Tests step next to quotes.test.js.Fix
work-chwh, push to the same branch, close it, and runkeeper-reviewagain.Second-opinion review as Keeper (opencode / deepseek-v4.1-flash), on PR head
614f922.Routine PR: it touches only the studio preview pane (studio/js/scene.js), a little CSS, a new test, the README and one CI step. No data loss, no security change, no lockout. No
review:humanlabel on the bead/epic, no human requested reviewer, and the repo has no AGENTS.md/CLAUDE.md asking for one.What I checked
d883071): scene.js, app.css, studio/tests/irc-quote.test.js, README.md, .forgejo/workflows/pages.yaml.node studio/tests/*.test.js(18 files, including the new irc-quote test) andpython3 studio/tests/build_data.test.py. All pass. CItestjob on614f922is green.work-chwhis fixed:node studio/tests/irc-quote.test.jsis now registered in the Tests step of .forgejo/workflows/pages.yaml next to quotes.test.js, and the bead is closed..irc-qon load; reload swaps only the last block for a differentdata-qidwhile keeping the block count;/quotestill appends;MF.quotes.length > 1guard plus the 50-iteration cap prevents an endless loop with a single quote; the reload button istype=button, keyboard reachable, hasaria-label/title, and uses only existing theme vars.documentclick handler does not collide with app.js's#sceneclick handling: the button carries nodata-ws/data-lang/data-page/data-slot, so at most the enclosing window gets focused.data-qidis numeric and quote text/nicks still go throughesc().No blocking findings left. Merging as Keeper.
Second-opinion review as Keeper (opencode / deepseek-v4.1-flash), on PR head
614f922. Routine PR; earlier blocking finding work-chwh is fixed and CI is green. No blocking findings left. Merging as Keeper.