[vishnuv688]

What & why

Part of #510 and #494. Stacked on #514.

The hosted player always runs the newest reader, while people open zips written by whatever adapter version recorded them. This commits four real trace zips, one per adapter, from the writer as it was before #508, with a test that the current reader still yields what each one held. A reader change that breaks an older trace now fails pnpm test.

Type of change

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Polish (an improvement to an existing feature)
• Breaking change (existing behavior changes for users)
• Documentation
• Internal (build, CI, dependencies, tooling)

Packages touched

• shared (types and contracts)
• core (framework-agnostic capture/reporting)
• elements (published element/snapshot API — @wdio/elements)
• service (WebdriverIO adapter)
• nightwatch-devtools (Nightwatch adapter)
• selenium-devtools (Selenium adapter)
• selenium-devtools-py (Selenium Python adapter)
• backend (server)
• app (UI)
• script (page-injected runtime)
• trace (Trace mode)

Capability matrix

• packages/conformance/src/features.ts updated, and tables regenerated with pnpm docs:capabilities, if this adds or changes user-facing behaviour

Not needed: tests only.

Notes for reviewers

Fixtures

Fixture	Original	Committed	Covers
wdio-mocha-2026-10-28-14977b93	261,400 B	19,762 B	the 13-action spec
selenium-mocha-2026-10-28-14977b93	159,706 B	19,098 B	a failed wait, a failed node:assert, JPEG screencast frames beside PNG ones
nightwatch-bdd-2026-10-28-14977b93	376,379 B	29,946 B	waitForElementVisible rows, a failed native assert
python-pytest-2026-10-28-14977b93	804,136 B	55,080 B	three browser sessions in one zip

From the examples-smoke artifact of CI run 40774299701 (commit 14977b93, 2026-10-28). The two earlier artifacts available wrote the same format, so they add nothing a reader could break separately. The whole fixture directory is 164 KB.

Reduction

• reduce-fixture.ts, committed beside the fixtures so the next one is made the same way, keeps every entry, in order and under its name, and changes only images. Each image resource becomes a 1x1 image of the same format carrying its own resource name; each inline data:image payload in .trace streams becomes a 1x1 image; every other entry keeps its bytes. Reducing the same source twice gives byte-identical zips.
• This does not weaken the test: the reader never decodes an image. The test reads each frame's embedded resource name back, so a frame paired with the wrong resource still fails.
• Each manifest is counted from the zip's raw entries, not through the reader, so a reader bug cannot write itself into the expected values. Breaking the reader on purpose (dropping network entries, reversing frame order) failed seven of the tests.

Sensitive data

Checked every non-image entry: only CI runner paths, the local fixture host and the-internet's public demo credentials. The original screenshots are gone.

Why this leaves #510 open

These cover one format: the one written until #508. Still open in #510: a fixture for each later format change as it ships, and a foreign trace, for which no zip was available.

Testing

pnpm build, pnpm test (3337 passed) and pnpm lint are green, on top of #508 and #513, so the current reader opens zips written before both.

Screenshots / recordings

Not applicable: no UI change.

🤖 Generated with Claude Code