Skip to content

test(android): retain original helper diagnostics - #4648

Merged
thymikee merged 2 commits into
0.22from
t3/android-helper-transport-failure
Nov 17, 2026
Merged

thymikee merged 2 commits into
0.22from
t3/android-helper-transport-failure

Conversation

@thymikee

@thymikee thymikee commented Nov 17, 2026 •

Copy link
Copy Markdown
Member

Summary

Retain the original Android smoke preflight and catalog snapshot helper diagnostics, metadata, warnings, and request correlation when the persistent-session assertion fails. Session acquisition diagnostics distinguish unavailable and reused paths; evidence writing cannot change the scenario verdict. The change touches 10 files, including the shared test harness. Related to #4636.

This improves diagnostic visibility only. The original incident cause remains unknown. The strict persistent-session assertion, budgets, fallback, and retry policies are unchanged.

Validation

  • Base: 7ccde11cc409d1973d7e26199af6bb75cfa2db82; tested head: b517ddc2d4a96a58d0e24508a93be92d5e04a255.
  • pnpm check:affected --run --base 7ccde11cc409d1973d7e26199af6bb75cfa2db82 --head b517ddc2d4a96a58d0e24508a93be92d5e04a255: exact-commit receipt, 9 passed, 8 skipped, 1 deduplicated. Full format, quick checks, and focused diagnostic red/green tests passed.
  • Independent source review covered a0f99a69b6; root cold-reviewed the follow-up through the tested head and accepted its regressions.
  • Hosted results at this head: 13 successful, one skipped and one neutral, including Android smoke and Coverage success. No review threads are open. Sentinel reports no open findings and one hidden below its configured threshold. The prior incident cause remains unknown; this diagnostic change does not establish a repair.

Autofix with [code]smith
Open in [code]smith · Need help on this PR? Tag @codesmith-bot with what you need.

View guided diff Turn on auto-fix

@github-actions

Copy link
Copy Markdown
Contributor

Size Report

Metric Base Current Diff
Installed (including dependencies) 6.15 MB 6.15 MB +400 B
Package (unpacked) 6.14 MB 6.15 MB +400 B
Package (download) 1.86 MB 1.86 MB +75 B

Startup median (7 runs, lower is better):

Scenario Base Current Diff
CLI --version 33.7 ms 37.2 ms +3.5 ms
CLI --help 61.4 ms 75.2 ms +13.9 ms

@apex-sentinel-staging

Copy link
Copy Markdown

Apex Sentinel · ✅ Approve

No open findings.

1 finding below minor hidden (user_settings.min_severity).

Reviewed with this repository's own Sentinel configuration (945b1f0d018f).

💬 Talking to Sentinel

In a finding's thread, reply with:

Command What it does
@apex-sentinel-staging note <context> Adds context; Sentinel re-checks the cited code
@apex-sentinel-staging pivot <why> Says the finding is wrong; Sentinel re-checks the cited code
@apex-sentinel-staging like Marks the finding as useful
@apex-sentinel-staging dislike <why> Marks it as not useful, so it is not raised again

On this conversation, comment @apex-sentinel-staging review to run a full review once again.

@thymikee

Copy link
Copy Markdown
Member Author

This PR is ready at b517ddc. I found no problems in the code, and all 15 checks pass, including Android smoke, which runs the changed route (preflight and catalog snapshots through captureDiagnosticSnapshot with --debug). There are no conflicts.

Not blocking, and you can take or leave these: tie the fields allowlist in snapshot-evidence.ts to AndroidSnapshotBackendMetadata with satisfies readonly (keyof AndroidSnapshotBackendMetadata)[], so every retained key names a contract field; only 'disabled' has a test among the new acquisition outcomes in snapshot-helper-session-lifecycle.ts, so a lifecycle test that captures twice could assert 'created' then 'reused', and the start-failure test could assert 'backed-off'; the 'created' and 'start-failed' outcomes already have events (session_ready and session_start_failed), so should 'acquisition' keep only the outcomes no existing phase covers? The scenario-ID check in scenario-start.ts could be a field the scenario declares; the test name in live-automation-scenario.test.ts promises request correlation but asserts only session and time window, and the deviceKey redaction fixture change in snapshot-helper-session.test.ts passes without this PR; and the empty catch {} in runtime.ts leaves no trace when an evidence write fails, so a one-line note in the artifact dir would help.

@thymikee thymikee added the ready-for-human Valid work that needs human implementation, judgment, or maintainer merge label Nov 17, 2026
@thymikee
thymikee merged commit 9717295 into 0.22 Nov 17, 2026
15 checks passed
@thymikee
thymikee deleted the t3/android-helper-transport-failure branch November 17, 2026 17:01
thymikee added a commit that referenced this pull request Nov 19, 2026
* test(android): retain snapshot helper session diagnostics

* test(android): preserve verdict when recording snapshot evidence
thymikee added a commit that referenced this pull request Nov 20, 2026
* test(android): retain snapshot helper session diagnostics

* test(android): preserve verdict when recording snapshot evidence
thymikee added a commit that referenced this pull request Nov 20, 2026
* test(android): retain snapshot helper session diagnostics

* test(android): preserve verdict when recording snapshot evidence
thymikee added a commit that referenced this pull request Nov 21, 2026
* test(android): retain snapshot helper session diagnostics

* test(android): preserve verdict when recording snapshot evidence
thymikee added a commit that referenced this pull request Nov 21, 2026
* test(android): retain snapshot helper session diagnostics

* test(android): preserve verdict when recording snapshot evidence
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready-for-human Valid work that needs human implementation, judgment, or maintainer merge

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant