feat(tests,bootstrap): add static tests and winget/sophia markers#6
feat(tests,bootstrap): add static tests and winget/sophia markers#6
Conversation
|
Warning Rate limit exceeded@Timpan4 has exceeded the limit for the number of commits that can be reviewed per hour. Please wait 5 minutes and 25 seconds before requesting another review. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. 📒 Files selected for processing (6)
📝 WalkthroughWalkthroughAdds a GitHub Actions workflow to run Pester tests on Windows, makes bootstrap.ps1 idempotent via Changes
Sequence DiagramsequenceDiagram
participant CI as GitHub Actions
participant Bootstrap as bootstrap.ps1
participant FS as Marker File System
participant WinGet as winget (import)
participant Sophia as Sophia-Preset.ps1
participant Desktop as Desktop Assets
CI->>Bootstrap: Trigger (push / PR) or manual run
Bootstrap->>FS: Check `winget.completed` (and compare apps.json hash)
alt winget marker missing or hash differs
Bootstrap->>WinGet: Compute missing packages, create temp manifest if needed
WinGet->>WinGet: Import packages (winget import)
WinGet->>FS: Write `winget.completed` with apps.json hash
else marker matches
Bootstrap->>Bootstrap: Skip WinGet import
end
Bootstrap->>FS: Check `sophia.completed` (compare preset hash)
alt sophia marker missing or hash differs
Bootstrap->>Sophia: Execute `Sophia-Preset.ps1`
Sophia->>FS: On success write `sophia.completed` with preset hash
else marker matches
Bootstrap->>Bootstrap: Skip Sophia preset
end
Bootstrap->>Desktop: Create `Run Windows Setup.lnk` and `Setup Summary.txt`
Bootstrap->>CI: Exit with summary/status
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d8b1bc78c7
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
bootstrap.ps1 (1)
93-140: Network fallback isn’t really a “fallback” on Win11, andwinget importshould handle spaces in temp paths.
Wait-ForNetwork:Invoke-WebRequestis only attempted whenTest-NetConnectionis missing; on Windows 11 it usually exists, so the HTTPS “fallback” path may never run.winget import $tempAppsJson ...should quote or use an args array (TEMP paths can include spaces).Proposed diff
function Wait-ForNetwork { @@ - if ($canTestNetConnection) { - $httpsSuccess = Test-NetConnection -ComputerName $httpsHost -Port $httpsPort -InformationLevel Quiet - if ($httpsSuccess) { - Write-Log "Network connectivity confirmed via HTTPS ($httpsHost)" -Level SUCCESS - return $true - } - } - elseif ($canInvokeWebRequest) { + if ($canTestNetConnection) { + $httpsSuccess = Test-NetConnection -ComputerName $httpsHost -Port $httpsPort -InformationLevel Quiet + if ($httpsSuccess) { + Write-Log "Network connectivity confirmed via HTTPS ($httpsHost)" -Level SUCCESS + return $true + } + } + + if ($canInvokeWebRequest) { try { Invoke-WebRequest -Uri $httpsUri -Method Head -UseBasicParsing -TimeoutSec 5 | Out-Null Write-Log "Network connectivity confirmed via HTTPS ($httpsHost)" -Level SUCCESS return $true } catch { # continue waiting } } @@ - Write-Log "Running: winget import $tempAppsJson --accept-package-agreements --accept-source-agreements" -Level INFO - - $wingetResult = winget import $tempAppsJson --accept-package-agreements --accept-source-agreements 2>&1 + $wingetArgs = @( + 'import', $tempAppsJson, + '--accept-package-agreements', + '--accept-source-agreements' + ) + Write-Log ("Running: winget {0}" -f ($wingetArgs -join ' ')) -Level INFO + + $wingetResult = winget @wingetArgs 2>&1Also applies to: 233-279
🤖 Fix all issues with AI agents
In @bootstrap.ps1:
- Around line 299-324: The Sophia step currently relies on $LASTEXITCODE which
can be sticky and misreport script success; change the success check to use the
PowerShell automatic variable $? to determine if the invoked script
($SophiaPreset via &) succeeded, and only consult $LASTEXITCODE if $? is $false
and $LASTEXITCODE is not $null to include an explicit exit code in the warning
path. Update the conditional around the execution of & $SophiaPreset to mark
success with $? (then create the $SophiaMarker via New-Item and Add-SummaryItem
with ✓), and on failure use $LASTEXITCODE only when set to decide between
warning vs error summary messages (preserve the existing catch block for
exceptions).
🧹 Nitpick comments (4)
tests/BuildIso.Tests.ps1 (1)
1-19: Tighten the$OEM$\$1\Setupassertion and simplify escaping.
Current test only proves$1\Setupappears somewhere; adding$OEM$reduces false positives, and single-quoted regex avoids backtick escaping noise. Based on learnings, build-iso is expected to populate the OEM setup path.Proposed diff
Describe "build-iso.ps1 static checks" { BeforeAll { $scriptPath = Resolve-Path (Join-Path $PSScriptRoot "..\build-iso.ps1") $scriptContent = Get-Content $scriptPath -Raw } It "uses $OEM$ $1 Setup path" { - $scriptContent | Should -Match "`\$1\\Setup" + $scriptContent | Should -Match '\$OEM\$\s*\\\$1\\Setup' }tests/Autounattend.Tests.ps1 (1)
1-14: Consider assertingFirstLogonCommands(not just the path).
This better matches the requirement thatautounattend.xmltriggers bootstrap on first logon. Based on learnings, this is a key contract.Proposed diff
Describe "autounattend.xml static checks" { BeforeAll { $filePath = Resolve-Path (Join-Path $PSScriptRoot "..\autounattend.xml") $fileContent = Get-Content $filePath -Raw } + It "triggers bootstrap via FirstLogonCommands" { + $fileContent | Should -Match "FirstLogonCommands" + } + It "runs bootstrap from C:\\Setup" { $fileContent | Should -Match "C:\\Setup\\bootstrap\.ps1" }tests/Bootstrap.Tests.ps1 (1)
1-28: LGTM: static checks cover the key bootstrap contracts (markers, desktop artifacts, network checks).
If you want to reduce brittleness later, consider matching$SophiaMarker/$WingetMarkervariable names in addition to raw filenames..github/workflows/pester.yml (1)
1-22: Use -CI flag for reliable test failure detection and pin Pester version (also add minimal permissions).
Right now CI pulls "latest Pester", which can break unexpectedly. In Pester v5, useInvoke-Pester -CIwhich automatically exits with non-zero on test failure—this is simpler and the officially recommended approach for CI. Pin Pester to a stable version (e.g., 5.6.1) to avoid breaking changes. Also add thepermissions: contents: readblock to follow least-privilege principle.Proposed diff
name: Pester on: push: pull_request: +permissions: + contents: read + jobs: pester: runs-on: windows-latest steps: - name: Checkout uses: actions/checkout@v4 - name: Install Pester shell: pwsh run: | - Install-Module Pester -Force -Scope CurrentUser + $pesterVersion = '5.6.1' + Install-Module Pester -RequiredVersion $pesterVersion -Force -Scope CurrentUser + Import-Module Pester -RequiredVersion $pesterVersion - name: Run Pester tests shell: pwsh run: | - Invoke-Pester -Path tests -Output Detailed + Invoke-Pester -Path tests -Output Detailed -CI
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
.github/workflows/pester.ymlbootstrap.ps1tests/Autounattend.Tests.ps1tests/Bootstrap.Tests.ps1tests/BuildIso.Tests.ps1
🧰 Additional context used
📓 Path-based instructions (2)
**/*.ps1
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.ps1: Do not create any PowerShell scripts (.ps1) during the planning phase
All scripts must target Windows 11 (22H2+) only; do not implement or test Windows 10 compatibility
Files:
tests/BuildIso.Tests.ps1tests/Bootstrap.Tests.ps1bootstrap.ps1tests/Autounattend.Tests.ps1
bootstrap.ps1
📄 CodeRabbit inference engine (CLAUDE.md)
bootstrap.ps1: bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
bootstrap.ps1 must produce logs: Desktop summary at C:\Users[User]\Desktop\Setup Summary.txt and detailed log at C:\Setup\install.log
Provide a desktop shortcut to re-run bootstrap.ps1 manually
Files:
bootstrap.ps1
🧠 Learnings (7)
📓 Common learnings
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to build-iso.ps1 : build-iso.ps1 must place autounattend.xml at ISO root and populate the $OEM$\$$\Setup directory with bootstrap.ps1, apps.json, and Sophia.ps1
Applied to files:
tests/BuildIso.Tests.ps1tests/Bootstrap.Tests.ps1bootstrap.ps1tests/Autounattend.Tests.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to **/*.ps1 : All scripts must target Windows 11 (22H2+) only; do not implement or test Windows 10 compatibility
Applied to files:
tests/BuildIso.Tests.ps1tests/Bootstrap.Tests.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
Applied to files:
tests/BuildIso.Tests.ps1tests/Bootstrap.Tests.ps1bootstrap.ps1tests/Autounattend.Tests.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must produce logs: Desktop summary at C:\Users\[User]\Desktop\Setup Summary.txt and detailed log at C:\Setup\install.log
Applied to files:
tests/Bootstrap.Tests.ps1bootstrap.ps1tests/Autounattend.Tests.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : Provide a desktop shortcut to re-run bootstrap.ps1 manually
Applied to files:
tests/Bootstrap.Tests.ps1bootstrap.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to autounattend.xml : autounattend.xml must trigger bootstrap via FirstLogonCommands for automatic execution on first boot
Applied to files:
tests/Autounattend.Tests.ps1
🔇 Additional comments (1)
bootstrap.ps1 (1)
27-28: Good: marker files make bootstrap re-runs idempotent.
Matches the repo requirement thatbootstrap.ps1be safe to run multiple times and produce clear per-step status. Based on learnings / coding guidelines.
Timpan4
force-pushed
the
ci-add-pester-github-actions
branch
2 times, most recently
from
102d6bb to
ce38799
Compare
|
@codex review |
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (6)
.github/workflows/pester.yml (2)
14-17: Pin the Pester module version for reproducible builds.Installing Pester without a version constraint can lead to inconsistent test results if future Pester releases introduce breaking changes.
📌 Proposed fix to pin Pester version
- name: Install Pester shell: pwsh run: | - Install-Module Pester -Force -Scope CurrentUser + Install-Module Pester -Force -Scope CurrentUser -MaximumVersion 5.6.1
10-17: Consider caching the Pester module installation.Adding PowerShell module caching can speed up workflow runs by avoiding repeated downloads.
⚡ Optional enhancement to cache Pester module
Add a caching step before installing Pester:
- name: Cache Pester module uses: actions/cache@v4 with: path: ~\Documents\PowerShell\Modules\Pester key: ${{ runner.os }}-pester-${{ hashFiles('**/pester.yml') }} - name: Install Pester shell: pwsh run: | Install-Module Pester -Force -Scope CurrentUser -MaximumVersion 5.6.1bootstrap.ps1 (4)
240-264: Consider more robust marker file reading.The current implementation reads the marker hash with
-ErrorAction SilentlyContinue, which might hide legitimate errors (e.g., permission issues). While the logic handles null/empty values correctly, distinguishing between "file doesn't exist" and "file read error" would improve diagnostics.🔧 Optional enhancement for marker file reading
$appsHash = (Get-FileHash -Path $AppsJson -Algorithm SHA256).Hash $markerHash = $null if (Test-Path $WingetMarker) { - $markerHash = (Get-Content -Path $WingetMarker -ErrorAction SilentlyContinue | Select-Object -First 1).Trim() + try { + $markerHash = (Get-Content -Path $WingetMarker -First 1 -ErrorAction Stop).Trim() + } + catch { + Write-Log "WARNING: Failed to read marker file: $($_.Exception.Message)" -Level WARNING + $markerHash = $null + } }
279-283: Add error handling for marker file write operation.If the marker file write fails (due to permissions, disk space, etc.), the next run will unnecessarily re-install all packages. While re-installation is safe due to idempotency, logging the failure would help diagnose issues.
🛡️ Proposed fix to add error handling
if ($LASTEXITCODE -eq 0) { Write-Log "WinGet import completed successfully" -Level SUCCESS - Set-Content -Path $WingetMarker -Value $appsHash -Force + try { + Set-Content -Path $WingetMarker -Value $appsHash -Force -ErrorAction Stop + } + catch { + Write-Log "WARNING: Failed to write marker file: $($_.Exception.Message)" -Level WARNING + } Add-SummaryItem -Step "WinGet" -Status "✓" -Message "Installed $($missingPackages.Count) packages ($WingetMarker)"
255-264: Add error handling for marker write when packages already installed.Similar to the installation case, the marker write at line 257 lacks error handling. If this write fails, subsequent runs may produce inconsistent status messages.
🛡️ Proposed fix to add error handling
if ($missingPackages.Count -eq 0) { Write-Log "All WinGet packages already installed" -Level SUCCESS - Set-Content -Path $WingetMarker -Value $appsHash -Force + try { + Set-Content -Path $WingetMarker -Value $appsHash -Force -ErrorAction Stop + } + catch { + Write-Log "WARNING: Failed to write marker file: $($_.Exception.Message)" -Level WARNING + } if ($markerHash -and $markerHash -eq $appsHash) {
311-346: Apply consistent error handling to Sophia marker operations.The Sophia Script marker logic has the same robustness concerns as the WinGet marker:
- Marker file reading (line 315) uses
-ErrorAction SilentlyContinuewhich might hide errors- Marker file writing (line 335) lacks error handling
🛡️ Proposed fix for consistent marker error handling
$presetHash = (Get-FileHash -Path $SophiaPreset -Algorithm SHA256).Hash $markerHash = $null if (Test-Path $SophiaMarker) { - $markerHash = (Get-Content -Path $SophiaMarker -ErrorAction SilentlyContinue | Select-Object -First 1).Trim() + try { + $markerHash = (Get-Content -Path $SophiaMarker -First 1 -ErrorAction Stop).Trim() + } + catch { + Write-Log "WARNING: Failed to read Sophia marker: $($_.Exception.Message)" -Level WARNING + $markerHash = $null + } } if ($markerHash -and $markerHash -eq $presetHash) { Write-Log "Sophia Script already applied; skipping" -Level INFO Add-SummaryItem -Step "Sophia" -Status "✓" -Message "Already applied ($SophiaMarker)" } else { if ($markerHash) { Write-Log "Sophia preset changed since last run; reapplying" -Level INFO } try { Write-Log "Found Sophia preset at $SophiaPreset" -Level INFO Write-Log "Executing Sophia Script..." -Level INFO & $SophiaPreset if ($LASTEXITCODE -eq 0 -or $null -eq $LASTEXITCODE) { Write-Log "Sophia Script execution completed" -Level SUCCESS - Set-Content -Path $SophiaMarker -Value $presetHash -Force + try { + Set-Content -Path $SophiaMarker -Value $presetHash -Force -ErrorAction Stop + } + catch { + Write-Log "WARNING: Failed to write Sophia marker: $($_.Exception.Message)" -Level WARNING + } Add-SummaryItem -Step "Sophia" -Status "✓" -Message "Sophia preset applied ($SophiaMarker)"
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (6)
.github/workflows/pester.ymlTODO.mdbootstrap.ps1tests/Autounattend.Tests.ps1tests/Bootstrap.Tests.ps1tests/BuildIso.Tests.ps1
🚧 Files skipped from review as they are similar to previous changes (3)
- tests/Autounattend.Tests.ps1
- tests/BuildIso.Tests.ps1
- tests/Bootstrap.Tests.ps1
🧰 Additional context used
📓 Path-based instructions (3)
**/*.ps1
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.ps1: Do not create any PowerShell scripts (.ps1) during the planning phase
All scripts must target Windows 11 (22H2+) only; do not implement or test Windows 10 compatibility
Files:
bootstrap.ps1
bootstrap.ps1
📄 CodeRabbit inference engine (CLAUDE.md)
bootstrap.ps1: bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
bootstrap.ps1 must produce logs: Desktop summary at C:\Users[User]\Desktop\Setup Summary.txt and detailed log at C:\Setup\install.log
Provide a desktop shortcut to re-run bootstrap.ps1 manually
Files:
bootstrap.ps1
{README.md,TODO.md,RESEARCH.md,SECURITY.md,CLAUDE.md,docs/**/*.md}
📄 CodeRabbit inference engine (CLAUDE.md)
All documentation must assume Windows 11 (22H2+) only and explicitly exclude Windows 10 support
Files:
TODO.md
🧠 Learnings (6)
📓 Common learnings
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to build-iso.ps1 : build-iso.ps1 must place autounattend.xml at ISO root and populate the $OEM$\$$\Setup directory with bootstrap.ps1, apps.json, and Sophia.ps1
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each step
Applied to files:
bootstrap.ps1TODO.md
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must produce logs: Desktop summary at C:\Users\[User]\Desktop\Setup Summary.txt and detailed log at C:\Setup\install.log
Applied to files:
bootstrap.ps1TODO.md
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to bootstrap.ps1 : Provide a desktop shortcut to re-run bootstrap.ps1 manually
Applied to files:
bootstrap.ps1TODO.md
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to build-iso.ps1 : build-iso.ps1 must place autounattend.xml at ISO root and populate the $OEM$\$$\Setup directory with bootstrap.ps1, apps.json, and Sophia.ps1
Applied to files:
bootstrap.ps1TODO.md
📚 Learning: 2025-10-02T09:50:23.485Z
Learnt from: CR
Repo: Timpan4/declarative-windows PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-10-02T09:50:23.485Z
Learning: Applies to autounattend.xml : autounattend.xml must trigger bootstrap via FirstLogonCommands for automatic execution on first boot
Applied to files:
TODO.md
🔇 Additional comments (4)
bootstrap.ps1 (1)
27-28: LGTM! Marker paths properly defined.The marker file paths are correctly constructed and stored in the C:\Setup directory alongside other setup files, supporting the idempotency requirements.
Based on learnings, this aligns with the requirement that bootstrap.ps1 must be idempotent.
TODO.md (3)
46-47: Documentation accurately reflects implementation.The error handling for Sophia Script execution is properly implemented in bootstrap.ps1 (lines 327-345) with try-catch blocks and appropriate logging.
67-76: Bootstrap Script task completions look accurate.The marked completions for idempotent WinGet/Sophia flows, marker tracking, desktop assets, network checks, and logging align with the implementation in bootstrap.ps1.
Based on learnings, these completions correctly reflect the idempotency requirements.
143-147: Consider updating Pester tests task status.Line 147 shows "Add Pester smoke tests for scripts" as incomplete, but this PR introduces a Pester workflow (
.github/workflows/pester.yml) and test suites referenced in the AI summary. Consider marking this task as complete or updating it to reflect the current implementation status.⛔ Skipped due to learnings
Learnt from: CR Repo: Timpan4/declarative-windows PR: 0 File: CLAUDE.md:0-0 Timestamp: 2025-10-02T09:50:23.485Z Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must be idempotent: skip already-installed apps and already-applied tweaks, and produce clear status for each stepLearnt from: CR Repo: Timpan4/declarative-windows PR: 0 File: CLAUDE.md:0-0 Timestamp: 2025-10-02T09:50:23.485Z Learning: Applies to bootstrap.ps1 : bootstrap.ps1 must produce logs: Desktop summary at C:\Users\[User]\Desktop\Setup Summary.txt and detailed log at C:\Setup\install.log
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: ce3879971e
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Timpan4
force-pushed
the
ci-add-pester-github-actions
branch
2 times, most recently
from
ae2f26e to
a5263b5
Compare
Add static Pester tests for autounattend.xml and bootstrap.ps1 to assert that the bootstrap script is invoked from C:\Setup, execution policy is set, WinGet and Sophia completion markers are tracked, a desktop summary is created, and desktop shortcut is created. The bootstrap tests also check network probing via ping (1.1.1.1), Test-NetConnection and HTTPS fallback with Invoke-WebRequest. Introduce Sophia and WinGet completion marker files in bootstrap.ps1 ($SophiaMarker and $WingetMarker). Update WinGet install logic to skip work when a winget.completed marker exists, and reorganize the branch to first check for network availability, parse apps.json only when needed, and compute missing packages before attempting installation. These changes improve test coverage for static expectations and make the bootstrap flow idempotent and more efficient by avoiding redundant WinGet work when markers indicate completion.
Timpan4
force-pushed
the
ci-add-pester-github-actions
branch
from
a5263b5 to
19ab7a1
Compare
1 participant
Add static Pester tests for autounattend.xml and bootstrap.ps1 asserting:
Introduce Sophia and WinGet completion marker files in bootstrap.ps1 ($SophiaMarker and $WingetMarker)
Update WinGet install logic to:
Improve idempotency and efficiency by avoiding redundant WinGet work when markers indicate completion
Summary by CodeRabbit
New Features
Tests
Chores
✏️ Tip: You can customize this high-level summary in your review settings.