Repository navigation
test: fix the tests that fail on the sandbox branch itself - #69
Merged
t3-code-sandboxes[bot] merged 6 commits intoNov 15, 2026
Merged
t3-code-sandboxes[bot] merged 6 commits into
t3-code-sandboxes[bot] merged 6 commits into
Conversation
…emd and TUN On a host with systemd running and a tailscaled unit, the script took the systemd branch and restarted the real unit instead of the stubbed daemon, so it reported ERROR start. Stub systemctl in the harness and route the TUN check through the root prefix so the test pins it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The panel now checks whether the current thread is a sandbox mirror, which reads modelSelection.instanceId. The contracts require modelSelection on every thread shell, so the fixture, not the code, was incomplete. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
b703713 made useThreadComposerState read sandboxStartQueue from ./threads, but the test's ./threads mock never exported it, so 9 of 11 cases threw on render. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The renders project (only run through scripts/render-metrics.ts, not CI) failed app-stream-assistant-turn (MarkdownParagraph 59, expected 30) and app-switch-thread (SidebarThreadRow 4, expected 2). Both fail identically on the upstream base the fork sits on (065aca4) and pass on upstream main; the sandbox work did not cause them. - Streaming re-rendered each finished Markdown block once more when the next block arrived, because its `cache` prop flipped. Upstream dd50613 (pingdotgg#33794, Lakshmi Tanmay) renders once and remembers the element in an effect. - Every sidebar row re-rendered on thread navigation through the `openPullRequestsInRightPanel` prop and an unstable `startThreadRename`. Upstream f968db1 (pingdotgg#33435, Dominik Zabcik) reads the flag at click time and stabilises the callback; bisected as the first upstream commit that passes app-switch-thread. (cherry picked from commit dd50613) (cherry picked from commit f968db1) Co-Authored-By: Lakshmi Tanmay <lakshmi@voltcrash.com> Co-Authored-By: Dominik Žabčík <35886301+dominikzabcik@users.noreply.github.com> Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ingdotgg#33061) Co-authored-by: Julius Marminge <julius0216@outlook.com> (cherry picked from commit 294a841) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
t3-code-sandboxes
Bot
changed the title
This was referenced Nov 15, 2026
…test/sandbox-flaky-fixtures # Conflicts: # apps/web/test/appScenarios.renders.tsx
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem. Some tests fail on
feat/sandbox-environmentsitself, independent of the sandbox PR stack (#60–#68). Some come from stale fixtures and mocks, two from upstream render regressions the fork still carries, one from a test race, and the rest from the machine running them.Fix.
sandboxSetup.test.ts): the harness stubbedsudo,tailscaleandtailscaledbut notsystemctl, so on a host running systemd the script restarted the realtailscaled. The TUN check also used a fixed/dev/net/tun. The test now stubssystemctland provides its own TUN under the script'sroot, andtailscalePathsgainstun(/dev/net/tunwith the default root, so production is unchanged).modelSelection, hidden by a cast../threadsmock didn't exportsandboxStartQueue, which the hook reads since the sandbox start-queue work.ServerBrowser.test.ts(8) andProjectionStore.test.ts(1) need Node 26 (Uint8Array.fromBase64,StatementSync.stat), as pinned inenginesand used by CI. They pass under Node 26.12.1.scripts/render-metrics.ts.Test plan
sandboxSetup.test.tstailscale case andThreadRelationshipsControl.remote-parent.browser.test.tsxpass (failed on the base)thread-task-permissions.test.ts11/11 (base: 2/11)node scripts/render-metrics.ts run --runs 116/16 (base: 2 failed);ChatMarkdownandmarkdownRenderertests 41/41AcpSessionManagementSection.browser.test.tsx14/14ServerBrowserandProjectionStore151/151 under Node 26.12.1vp run check:changed origin/feat/sandbox-environmentsClaude Opus 5.5 via Claude Code.
🤖 Generated with Claude Code