[Repo Assist] fix(version): use alpha fallback and add alpha update test channel#571
Conversation
…UI test The <Version> in OpenClaw.Tray.WinUI.csproj was stale at 0.4.7 while the latest release tag is v0.5.0. In CI the version is overridden by GitVersion, but local builds and fallback paths displayed the wrong version. Also update A2UIDashboardScaleTest to assert AppVersionInfo.DisplayVersion instead of a hardcoded literal, so the test stays green across future bumps. Closes #570 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
Codex review: needs maintainer review before merge. Reviewed May 28, 2026, 2:00 PM ET / 18:00 UTC. Summary Reproducibility: yes. from source for the stale local fallback and hardcoded A2UI literal: current main still uses 0.4.7 in the csproj fallback and dashboard fixture. I did not run build/test commands because this review is read-only and those commands create artifacts. Review metrics: 1 noteworthy metric.
Merge readiness Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch. Rank-up moves:
Mantis proof suggestion Risk before merge
Maintainer options:
Next step before merge
Security Review detailsBest possible solution: Land the narrow fallback/display cleanup with the opt-in alpha update channel once required CI is green and maintainers accept the documented alpha ZIP-update testing policy. Do we have a high-confidence way to reproduce the issue? Yes from source for the stale local fallback and hardcoded A2UI literal: current main still uses 0.4.7 in the csproj fallback and dashboard fixture. I did not run build/test commands because this review is read-only and those commands create artifacts. Is this the best way to solve the issue? Yes, assuming the posted maintainer alpha plan is accepted: the PR preserves tag-driven release builds, moves display text through AppVersionInfo, and makes alpha update checks opt-in. A live Windows updater smoke would reduce risk but I did not find a blocking patch defect. AGENTS.md: found and applied where relevant. Codex review notes: model gpt-5.5, reasoning high; reviewed against 8edf4eb07bfe. Label changesLabel changes:
Label justifications:
Evidence reviewedWhat I checked:
Likely related people:
What the crustacean ranks mean
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics. How this review workflow works
|
clawsweeper
Bot
added
rating: 🧂 unranked krab
…ue-570-bump-version-0.5.0-197a3346e667a8c7
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
shanselman
changed the title
clawsweeper
Bot
added
proof: sufficient
|
@clawsweeper re-review\n\nCI is now green after rerunning the transient E2E failure; PR body has been updated with the final check status. |
|
🦞👀 I queued a lightweight read-only assist pass. It will post a separate answer comment and will not edit the durable ClawSweeper review comment or trigger close, merge, repair, label, or branch changes. Request: review\n\nCI is green now; please refresh the durable review against head c338a7b. |
|
ClawSweeper assist: This needs a full correctness re-review, so use Evidence:
Suggested next action: Run Source: #571 (comment) |
|
@clawsweeper review\n\nCI is green now; please refresh the durable review against head c338a7b. |
|
🦞👀 Command router queued. I will update this comment with the next step. |
|
ClawSweeper assist: This request is for a full durable correctness review, not a lightweight assist answer. Use Evidence:
Suggested next action: Run the durable Source: #571 (comment) |
|
@clawsweeper review |
|
🦞🧹 I asked ClawSweeper to review this item again. Re-review progress:
|
|
🤖 This is a note from Repo Assist. This PR currently has a merge conflict ( Given that this PR was revised by a maintainer, I'm leaving the rebase for human decision rather than risk disrupting those changes. Flagging here for awareness — please let me know if you'd like me to attempt a rebase.
|
1 participant
🤖 This PR was created by Repo Assist, then revised by a maintainer to match the current alpha testing plan.
Summary
0.6.0-alpha.1instead of the stale0.4.7or stable0.5.0.Versionfrom GitVersion/tag outputs.AppVersionInfo.DisplayVersion.AppVersionInfodisplay/diagnostics while still stripping SourceLink build metadata.OPENCLAW_UPDATE_CHANNEL=alphaenables pre-release checks and release-list fetching.OPENCLAW_TRAY_DATA_DIRruns.Maintainer decision
Current
mastershould use the alpha local fallback (0.6.0-alpha.1) because the next validation target is alpha release/update testing. This is not a routine release-file bump: release artifacts remain tag-driven through CI, and the csproj value is still only the local fallback for dev/test builds.Behavior proof
Release build metadata from this PR branch:
This means UI/diagnostics can show the alpha label after
AppVersionInfostrips+metadata, while updater-facing file/assembly versions stay numeric.Release-channel proof:
Validation
./build.ps1dotnet test ./tests/OpenClaw.Shared.Tests/OpenClaw.Shared.Tests.csproj --no-restore(2057 total / 2028 passed / 29 skipped)dotnet test ./tests/OpenClaw.Tray.Tests/OpenClaw.Tray.Tests.csproj --no-restore(868 passed)./build.ps1 -Configuration Releasegit diff --check15 successful / 12 skipped)Note: this was a fresh linked worktree, so the first local
--no-restoretest invocations hit the known zero-test no-op. I ran each test project once with restore/build enabled, then reran the required--no-restorecommands and got the counts above. The first CIe2etestsattempt hit a fresh-WSL startup timeout; rerunning the failed job passed.Closes #570