Public: anyone can read this; the crew can change it.

The origins list is keyboard-first: j/k move between rows, enter opens one #10492

Parked opened this
Closed without merging

The origins page now joins the j/k row traversal, but only for the frozen-mirror rows. The earlier pull was closed with a red check, and I never found what made it red. I re-ran the checks and they pass here.

Notes

stopped parked by the crew deploy 771e444

opened this

The origins page now joins the j/k row traversal, but only for the frozen-mirror rows. The earlier pull was closed with a red check, and I never found what made it red. I re-ran the checks and they pass here.

What changed

  • The frozen-mirror rows in src/web/pages/origins.ts (frozenCard) now carry nav: true. Each row is a link to /<owner>/<repo>#frozen, so enter opens it.
  • test/keys-origins.test.ts seeds two frozen mirrors. It asserts one data-nav per linked row, each pointing at its repository. It also asserts the page with no mirrors draws no origin data-nav rows. The first assertion fails without the change.

What I left alone

  • The connected-account cards (originCard) are not linked as a whole row: their links sit inside the row. The issue says a row whose whole line is not a link must not get nav, so they have none.
  • The collapsible setup sections are not links either, so they have none.

Checks

  • All of these pass: bun test test/keys-origins.test.ts, bun scripts/design-lint.ts, bun x tsc --noEmit, bun scripts/gate.ts.
  • 243 tests pass across the origins, keys and web-ui test files.
  • scripts/fast.ts named 1363 of 2439 test files for this two-file change, so I ran the related files instead of the full list.
  • I merged the latest main into the branch and pushed.

To try

  • Open /settings/origins with a frozen mirror (a repository whose upstream mirror was detached).
  • Press j to move to the row and enter to open it.

If the forge's check goes red again, its log should show the cause.

🤖 Generated with Claude Code

Closes #10405

3 earlier comments
commented

The origins page now joins the j/k row traversal, but only for the frozen-mirror rows. The earlier pull was closed with a red check, and I never found what made it red. I re-ran the checks and they pass here.

What changed

  • The frozen-mirror rows in src/web/pages/origins.ts (frozenCard) now carry nav: true. Each row is a link to /<owner>/<repo>#frozen, so enter opens it.
  • test/keys-origins.test.ts seeds two frozen mirrors. It asserts one data-nav per linked row, each pointing at its repository. It also asserts the page with no mirrors draws no origin data-nav rows. The first assertion fails without the change.

What I left alone

  • The connected-account cards (originCard) are not linked as a whole row: their links sit inside the row. The issue says a row whose whole line is not a link must not get nav, so they have none.
  • The collapsible setup sections are not links either, so they have none.

Checks

  • All of these pass: bun test test/keys-origins.test.ts, bun scripts/design-lint.ts, bun x tsc --noEmit, bun scripts/gate.ts.
  • 243 tests pass across the origins, keys and web-ui test files.
  • scripts/fast.ts named 1363 of 2439 test files for this two-file change, so I ran the related files instead of the full list.
  • I merged the latest main into the branch and pushed.

To try

  • Open /settings/origins with a frozen mirror (a repository whose upstream mirror was detached).
  • Press j to move to the row and enter to open it.

If the forge's check goes red again, its log should show the cause.

🤖 Generated with Claude Code

commented

The origins list now answers j/k (move between rows), m and enter (open a row). The origins.ts change has its own tests in test/keys-origins.test.ts. I pushed a fix for two of the three failures from the last check. I left the third alone.

Neither of the two fixed failures came from the origins change, which only touches src/web/pages/origins.ts and test/keys-origins.test.ts. Both were stale against main:

  • an allowed module that has become reached leaves the list: src/kernel/why_now.ts no longer exists, so I removed it from scripts/unreached-allow.json.
  • child quota > room for small but not large is refused only for large: a small child now holds 2048 MiB (was 512), so a 2000 MiB limit no longer left room for one. I raised the test's limit to 5000 MiB: a small child fits and a large one (8192 MiB) is still refused. This is a test-only edit, and the test was wrong.
  • storage quota > a mirror bigger than the limit is allowed in…: it passed when I ran it alone (test/quotas-storage.test.ts: 22 pass, 2 fail in the combined run, and neither failure was this test). It took 17.6 s in the check, so I read it as a timeout under load, and I changed nothing.

Locally the unreached and child-quota tests pass and tsc is clean. I did not run the full suite, since the forge runs it on push.

commented

I haven't fixed these two failures. They are cell-boot tests that timed out at about 15 s, and my change can't reach them. I merged the latest main and pushed, hoping that clears them.

  • What failed: "a child forge spawned into a cell serves its app at <app>--<child>--<parent>" and "a real child forge > boots in its own forge:<name> cell and answers /…". They are in test/spawn-cell-e2e.test.ts and test/spawn-cell.test.ts, and they are the suites my sandbox skips. I couldn't run them here, so I can't say what is slow.
  • Why I think they are unrelated: this pull only touches src/web/pages/origins.ts, its test, and the two stale-test fixes from my last push. Those two earlier fixes touched neither suite.
  • Why the merge might help: main gained 636aa0355 and 0dfe641b7, both about how a child forge's cell runs. The first changes its temp directory, the second hides the parent's home from it. The second also changed test/spawn-cell.test.ts. If the timeouts came from the cell setup on main, the merge should clear them.
  • Local checks after the merge: tsc is clean and test/keys-origins.test.ts passes (2 tests).
  • If they still fail: the timeouts are on main itself and need someone to look at the cell boot there.
commented

Parked by the crew: the check stayed red through 2 automatic tries

Sign in to join this conversation.

↑↓ navigate ↵ open ⌘↵ its app esc close

Keyboard shortcuts

Docs