Skip to content

test(desktop): verify the installed connection and Scope journey - #2581

Merged
hidb4ai merged 1 commit into
oceanbase:masterfrom
jackie-cqz:fix/desktop-connection-acceptance
Nov 11, 2026
Merged

hidb4ai merged 1 commit into
oceanbase:masterfrom
jackie-cqz:fix/desktop-connection-acceptance

Conversation

@jackie-cqz

Copy link
Copy Markdown
Contributor

Which issue or RFC does this PR close?

Refs #2101 and #2103. Local setup handoff is being delivered in #2306. Windows 11 and macOS installed qualification remains tracked in #2108/#2109.

Rationale for this change

The installed workflow activates saved profiles immediately and uses a single matching Scope title. It therefore misses regressions where saving/checking activates a connection, title-based selection targets the wrong Scope, or ordinary operations invoke readiness. The installed Server fixture also extracts its wheel before verifying that it matches the qualification used by the native client.

What changes are included in this PR?

  • Add an installed WebDriver connection journey before the existing Memory scenarios. Save and passively check a profile while disconnected, explicitly activate it, select the second of two same-title Scopes by exact ID, save/search/read Memory and independently check its Scope ownership.
  • Verify Scope switching clears private results and the global connection menu opens the exact saved profile while retaining the selected Scope.
  • Count requests to the isolated Server's fixed /health/ready endpoint. Ordinary UI operations preserve the startup count and the explicit probe adds exactly one request. Save successful observations in connectionJourney in the existing installed UI report.
  • Validate the installed fixture's wheel and extract the same captured bytes before starting the Server. Add a regression that rejects a replaced archive before extraction.
  • Document execution evidence and remaining acceptance boundaries in English and Chinese in desktop/VALIDATION.md.

Are there any user-facing changes?

This change extends acceptance tooling for existing Server connections. The new scenario runs through packaged UI controls and native transport on the disposable Windows runner.

How was this change tested?

Local Windows 10 x64 / Python 3.11.15, based on master cc752d8accb0bc4ffade2d4b84ef6e48206e122d.

  • Original installed fixture reproduction fails the new replacement-wheel test at ZIP extraction. The updated fixture rejects it at qualification validation.
  • uv run --no-sync pytest tests/test_desktop_contract.py tests/test_desktop_fixture_qualification.py -q passes 11 tests.
  • pnpm --dir desktop test passes 211 frontend tests with mocked IPC.
  • Strict Ruff checks/formatting and file hooks on changed paths pass. Linux-target type checks pass with the baseline's Windows symlink placeholder excluded while its actual resource source is checked. The full Windows type hook retains its platform and symlink-checkout diagnostics.
  • A real isolated SQLite Server verifies readiness count 1 → 1 → 2 across identity/discovery, Memory save/search/exact read and explicit readiness. This validates instrumentation through actual HTTP execution.
  • Rebuilt native server_probe and the existing real_server.py harness pass all four SQLite modes after the proxy instrumentation change.
  • The exact Server wheel is built from archived source cc752d8a. All 745 runtime files match after CRLF normalization. Wheel SHA-256 is df51abf71e54bd857962ed171cc8fe37a1f28182e8ccd8fc2851d37595dc4ec4.
  • Merge calculation with feat(desktop): implement local Server setup, recovery and management #2306 head 68b91ce9 merges the Desktop changes cleanly. It encounters an existing ZCode test conflict between that branch and current master.

The new installed WebView2 scenario requires the hosted Windows CI run. Local HTTP and mocked-IPC results leave that scenario NOT_RUN locally. Windows 11/macOS release qualification remains pending in the platform tasks.

AI usage statement

OpenAI Codex assisted with implementation, regression reproduction, real Server validation and documentation.

Copilot AI balanced review requested due to automatic review settings November 10, 2026 11:44

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.

@hidb4ai hidb4ai left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed f6571ba. No blocking findings. The installed connection/Scope journey and pre-extraction wheel validation address the stated acceptance gaps without adding database tables or product interfaces. Hosted Windows installed evidence covers the new journey; clean Windows 11/macOS release qualification remains tracked separately.

@hireview001 hireview001 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@hireview001 hireview001 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@hidb4ai
hidb4ai merged commit eeba979 into oceanbase:master Nov 11, 2026
45 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants