feat(ui-mode): update individual snapshots

[Devin Rousso (dcrousso)]

reviewing image diffs currently requires rerunning tests with snapshot updates enabled

add a button for each diff that copies the actual image over the expected snapshot

fixes #42393

[github-actions]

Hi, I'm the Playwright bot and I took a first look at the CI failures here.

🟢 CI is clear — the one failure is a pre-existing flake

The single failure, [firefox] › mcp/cli-core.spec.ts:123 › select, is a known flake: the same test failed on the same mcp-windows-latest-firefox bot on an unrelated push SHA (610977b), and timed out on chromium/chrome on other SHAs/PRs. This PR only touches UI-mode snapshot-update code, which the MCP select tool doesn't exercise.

Details

Overall: nothing in this diff reaches the failing paths. The PR adds a "Save actual as expected" button in trace-viewer plus an updateSnapshot test-server endpoint (testServer.ts, testServerInterface.ts, attachmentsTab.tsx, uiModeTraceView.tsx, workbench.tsx). None of that is loaded by the MCP CLI or the library/page tests below.

Pre-existing flake / infra

• [firefox] › mcp/cli-core.spec.ts:123 › select — pre-existing flake. In the test-results DB this test has failed/timed out on SHAs this PR can't be responsible for: firefox on push SHA 610977b (run 30020951381), chromium timedOut on push SHA 3e14efd (run 30034115836), and chrome timedOut on PR feat(trace): load trace files independently, produce a single context entry #42350 (run 32489690957). Firefox alone is green in 722 of 724 runs. The MCP browser_select path is untouched by this PR.

Not failures (no action)

• The "tests 1" report lists 3 flaky, 0 failed — within-run flakes rescued on retry, all with long flake histories (e.g. firefox-page › page-goto.spec.ts:90 › should work with Cross-Origin-Opener-Policy rescued 302 times; page-event-request.spec.ts:181 › ...COOP... rescued 373 times; chromium.spec.ts:436 › Service Worker (advanced) rescued 239 times). Nothing to triage.

I'm a first pass, not the final word — but I couldn't find any failure that points back at this change.

_{Triaged by the Playwright bot - agent run}

[Dmitry Gozman (dgozman)]

There have been multiple attempts at this functionality (some examples: #12707, #14883, #24310, #28415) that never landed, so I'd recommend to first research what went wrong with previous attempts.

[Pavel Feldman (pavelfeldman)]

Requesting a demo of this one, unclear from the code what is happening and which scenarios this supports.

[Devin Rousso (dcrousso)]

before

after

[Dmitry Gozman (dgozman)]

I think we agreed to flash a green checkmark instead of updating the text?

[Dmitry Gozman (dgozman)]

What's going on here? Why don't we have correct attachment paths? Why do we only call before a single setModel(), and not before all the others?

[Devin Rousso (dcrousso)]

ive reworked this

[github-actions]

Test results for "tests 1"

5 flaky

⚠️ [chromium-library] › library/video.spec.ts:736 › screencast › should work with video+trace `@realtime-time-library-chromium-linux`
⚠️ [chromium-library] › library/video.spec.ts:736 › screencast › should work with video+trace `@chromium-ubuntu-22.04-node22`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-ubuntu-22.04-node20`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@firefox-ubuntu-22.04-node20`
⚠️ [firefox-page] › page/page-event-request.spec.ts:181 › should return response body when Cross-Origin-Opener-Policy is set `@firefox-ubuntu-22.04-node20`

51277 passed, 1241 skipped

---

Merge workflow run.

[github-actions]

Test results for "MCP"

8305 passed, 1367 skipped

---

Merge workflow run.