Update Magick.NET-Q8-AnyCPU to 14.13.1

[Redth]

Note

Are you waiting for the changes in this PR to be merged?
It would be very helpful if you could test the resulting artifacts from this PR and let us know in a comment if this change resolves your issue. Thank you!

Description

Updates Magick.NET-Q8-AnyCPU from 14.12.0 to 14.13.1 in VisualTestUtils.MagickNet.

Testing

• dotnet restore src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
• dotnet build src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -c Release --no-restore

[github-actions]

🚀 Dogfood this PR with:

⚠️ WARNING: Do not do this without first carefully reviewing the code of this PR to satisfy yourself it is safe.

curl -fsSL https://raw.githubusercontent.com/dotnet/maui/main/eng/scripts/get-maui-pr.sh | bash -s -- 35880

Or

• Run remotely in PowerShell:

iex "& { $(irm https://raw.githubusercontent.com/dotnet/maui/main/eng/scripts/get-maui-pr.ps1) } 35880"

[kubaflo]

/review -b feature/enhanced-reviewer -p android

[MauiBot]

AI Review Summary

@Redth — new AI review results are available based on this last commit: 965037b. To request a fresh review after new comments or commits, comment /review rerun.

Review Sessions — click to expand

Gate — Test Before & After Fix

Gate Result: ⚠️ SKIPPED

No tests were detected in this PR.

Recommendation: Add tests to verify the fix using the write-tests-agent.

---

Pre-Flight — Context & Validation

Issue: N/A - No linked issue detected
PR: #35880 - Update Magick.NET-Q8-AnyCPU to 14.13.1
Platforms Affected: Android UI-test host infrastructure (also shared UI test utilities for iOS, MacCatalyst, Windows)
Files Changed: 1 implementation, 0 test

Key Findings

• PR updates src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj from Magick.NET-Q8-AnyCPU 14.12.0 to 14.13.1.
• The package is test infrastructure only; Android impact is host-side visual comparison/native loading, not Android app runtime behavior.
• Gate was already skipped because no tests were detected; no linked issue was found.

Code Review Summary

Verdict: NEEDS_DISCUSSION
Confidence: low
Errors: 0 | Warnings: 0 | Suggestions: 1

Key code review findings:

• 💡 Consider Android host visual-test smoke coverage because restore/build does not prove ImageMagick native-load or screenshot comparison behavior in the Android UI-test host.

Fix Candidates

#	Source	Approach	Test Result	Files Changed	Notes
PR	PR #35880	Directly update Magick.NET-Q8-AnyCPU from 14.12.0 to 14.13.1	⚠️ SKIPPED (Gate: no tests detected)	src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj	Original PR

---

Code Review — Deep Analysis

Code Review — PR #35880

Independent Assessment

What this changes: Updates src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj:9 from Magick.NET-Q8-AnyCPU 14.12.0 to 14.13.1.
Inferred motivation: Security/dependency hygiene. The package is used by visual-test utilities and the newer package version is intended to replace the older dependency.

Reconciliation with PR Narrative

Author claims: Updates Magick.NET-Q8-AnyCPU from 14.12.0 to 14.13.1 in VisualTestUtils.MagickNet, with restore/build validation.
Agreement/disagreement: The PR diff matches the description exactly. No product/runtime MAUI code is modified.

Prior Review Reconciliation

No prior ❌ Error findings found. Public PR comments contain the dogfood bot comment and a /review -b feature/enhanced-reviewer -p android request; public inline review comments are empty.

Blast Radius Assessment

• Runs for all instances: No product instances. It affects all UI test projects that reference VisualTestUtils.MagickNet.
• Startup impact: No app startup impact; possible host-side UI-test startup/native-load impact when ImageMagick is first used.
• Static/shared state: No new static/shared state.

CI Status

• Required-check result: unavailable through gh pr checks --required because GitHub CLI is unauthenticated in this environment.
• Classification: undetermined for this local review; user-provided gate result says gate was skipped because no tests were detected.
• Action taken: capped confidence; did not rerun gate verification.

Findings

No ❌ Error findings.

No ⚠️ Warning findings.

💡 Suggestion — Add Android host visual-test smoke coverage

The package is consumed by Android UI test infrastructure through src/Controls/tests/TestCases.Android.Tests/Controls.TestCases.Android.Tests.csproj and shared UI test code that uses ImageMagick. Restore/build of the utility project proves compile compatibility, but it does not prove native Magick load or screenshot compare/diff behavior in the Android UI-test host.

Failure-Mode Probing

• What happens if this package changes native asset probing on the Android UI-test host? The MAUI app would not crash, but host-side visual comparison could fail when the first MagickImage is created.
• What happens if visual comparison algorithms change subtly? Screenshot comparison thresholds could become more/less sensitive across Android UI tests even though MAUI rendering is unchanged.
• What happens if the large AnyCPU package has restore/feed issues? UI test project restore/build can fail before tests run.

Verdict: NEEDS_DISCUSSION

Confidence: low. The code change is localized and appears sound, but CI/gate evidence is unavailable in this environment and the detected gate was skipped because no tests were present.
Summary: No blocking code issue was found. The main residual risk is Android UI-test host restore/native-load/visual-comparison behavior, which is not covered by the skipped gate.

---

Fix — Analysis & Comparison

Fix Candidates

#	Source	Approach	Test Result	Files Changed	Notes
1	try-fix	Keep Magick.NET-Q8-AnyCPU 14.13.1, but move the version to a local $(MagickNetPackageVersion) MSBuild property.	✅ PASS	1 file	Improves maintainability/auditability, but only marginally better than the direct PR bump.
2	try-fix	Replace Magick.NET-Q8-AnyCPU with narrower Magick.NET-Q8-x64 14.13.1 for host-side Android UI-test use.	❌ FAIL	1 file	Failed with NU1101; package ID is not available from configured feeds.
3	try-fix	Keep Magick.NET-Q8-AnyCPU 14.13.1 and add direct Magick.NET.Core 14.13.1 pin.	✅ PASS	1 file	More explicit transitive dependency pin, but likely redundant because NuGet already resolves the wrapper's core dependency.
PR	PR #35880	Directly update Magick.NET-Q8-AnyCPU from 14.12.0 to 14.13.1.	⚠️ SKIPPED (Gate: no tests detected)	1 file	Original PR; simplest package-servicing fix.

Cross-Pollination

Model	Round	New Ideas?	Details
maui-expert-reviewer	1	Yes	Suggested central/property versioning, host-RID-specific package, validated internal package, dependency-validation path, or suppressing only if the vulnerable path is impossible.
maui-expert-reviewer	2	No	After attempts 1-3, expert reviewer concluded the direct AnyCPU bump is the cleanest appropriate fix; attempt 1 is indirection, attempt 3 is redundant, and attempt 2 is invalid/unavailable and less cross-host safe.

Exhausted: Yes
Selected Fix: PR's fix — It is the simplest viable dependency servicing change. Candidate #1 and #3 passed the available checks but are not demonstrably better: #1 adds indirection without stronger validation, and #3 adds a redundant direct dependency. Candidate #2 failed restore.

Detailed Attempt Narratives

try-fix-1

Approach

Keep Magick.NET-Q8-AnyCPU at 14.13.1, but move the version into a local MSBuild property ($(MagickNetPackageVersion)) and reference that property from the PackageReference.

This is different from PR #35880's direct PackageReference version bump because it makes the dependency version easier to audit/service in the project file while preserving the same resolved package.

Diff

diff --git a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
index 13c824627a..c94e0de741 100644
--- a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
+++ b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
@@ -3,10 +3,11 @@
   <PropertyGroup>
     <TargetFramework>netstandard2.0</TargetFramework>
     <PackageReadmeFile>package-readme.md</PackageReadmeFile>
+    <MagickNetPackageVersion>14.13.1</MagickNetPackageVersion>
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="Magick.NET-Q8-AnyCPU" Version="14.13.1" />
+    <PackageReference Include="Magick.NET-Q8-AnyCPU" Version="$(MagickNetPackageVersion)" />
   </ItemGroup>
 
   <ItemGroup>

Test Results

✅ PASS

Commands:

• dotnet restore src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -v:minimal
• dotnet build src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -c Release --no-restore -v:minimal
• dotnet restore src/Controls/tests/TestCases.Android.Tests/Controls.TestCases.Android.Tests.csproj -v:minimal

Failure Analysis

No failure. This candidate validates restore/build and Android UI-test project restore, but it is only marginally better than the PR's fix: it improves maintainability without improving runtime confidence or adding coverage.

---

try-fix-2

Approach

Replace Magick.NET-Q8-AnyCPU with a narrower host-specific Magick.NET-Q8-x64 package at 14.13.1 to reduce native asset footprint for Android UI tests, which run ImageMagick host-side rather than on-device.

This is different from PR #35880's direct AnyCPU package bump because it attempts to reduce restore/native-load blast radius instead of keeping the broad AnyCPU package.

Diff

diff --git a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
index 13c824627a..cfab8f7824 100644
--- a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
+++ b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
@@ -6,7 +6,7 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="Magick.NET-Q8-AnyCPU" Version="14.13.1" />
+    <PackageReference Include="Magick.NET-Q8-x64" Version="14.13.1" />
   </ItemGroup>
 
   <ItemGroup>

Test Results

❌ FAIL

Commands:

• dotnet restore src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -v:minimal
• dotnet build src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -c Release --no-restore -v:minimal
• dotnet restore src/Controls/tests/TestCases.Android.Tests/Controls.TestCases.Android.Tests.csproj -v:minimal

Failure Analysis

Restore failed with NU1101: Unable to find package Magick.NET-Q8-x64 from the configured feeds. The narrower-package idea is not viable in this repository/feed configuration unless a real supported package ID is identified and ingested.

---

try-fix-3

Approach

Keep Magick.NET-Q8-AnyCPU at 14.13.1, and add an explicit direct Magick.NET.Core 14.13.1 reference so the core dependency is pinned at the same serviced version as the native wrapper package.

This is different from PR #35880's direct wrapper-only bump because it makes the transitive core dependency explicit and reduces ambiguity if other dependency graph changes later influence Magick.NET.Core resolution.

Diff

diff --git a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
index 13c824627a..fb70b6b5ef 100644
--- a/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
+++ b/src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj
@@ -7,6 +7,7 @@
 
   <ItemGroup>
     <PackageReference Include="Magick.NET-Q8-AnyCPU" Version="14.13.1" />
+    <PackageReference Include="Magick.NET.Core" Version="14.13.1" />
   </ItemGroup>
 
   <ItemGroup>

Test Results

✅ PASS

Commands:

• dotnet restore src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -v:minimal
• dotnet build src/TestUtils/src/VisualTestUtils.MagickNet/VisualTestUtils.MagickNet.csproj -c Release --no-restore -v:minimal
• dotnet restore src/Controls/tests/TestCases.Android.Tests/Controls.TestCases.Android.Tests.csproj -v:minimal

Failure Analysis

No failure. This candidate passes the available checks. It is more explicit than the PR's fix but adds a direct dependency that NuGet already resolves transitively at the same version under normal lowest-applicable-version rules, so the practical benefit is limited.

---

---

Report — Final Recommendation

Comparative Report — PR #35880

Candidates compared

Candidate	Approach	Regression result	Assessment
pr	Directly update Magick.NET-Q8-AnyCPU from 14.12.0 to 14.13.1.	⚠️ Gate skipped: no tests detected	Simplest viable servicing fix. No expert-review blocking findings.
pr-plus-reviewer	PR fix plus expert reviewer feedback.	⚠️ Gate skipped: no tests detected	Identical to pr because the expert reviewer found no actionable inline findings to apply.
try-fix-1	Keep Magick.NET-Q8-AnyCPU 14.13.1, but move the version to $(MagickNetPackageVersion).	✅ PASS	Functionally equivalent to the PR fix, with a small maintainability indirection. It does not improve runtime/native-load confidence.
try-fix-2	Replace Magick.NET-Q8-AnyCPU with Magick.NET-Q8-x64 14.13.1.	❌ FAIL	Ranked below passing candidates because restore failed with NU1101; the package ID is not available from configured feeds.
try-fix-3	Keep Magick.NET-Q8-AnyCPU 14.13.1 and add explicit Magick.NET.Core 14.13.1.	✅ PASS	More explicit, but redundant because the wrapper package already resolves the aligned core dependency.

Expert PR evaluation

The expert reviewer found no actionable file:line findings against the PR diff. The reviewer also reported that Magick.NET-Q8-AnyCPU 14.13.1 preserves the relevant framework support and native runtime asset shape and aligns Magick.NET.Core to the same version.

Because there was no actionable feedback, pr-plus-reviewer does not differ from pr.

Ranking

1. pr — Best candidate. It is the smallest correct dependency-servicing fix and avoids redundant project-file indirection or direct transitive pins.
2. pr-plus-reviewer — Equivalent code to pr; ranked just below only because it adds no distinct improvement.
3. try-fix-1 — Passed available checks, but the MSBuild property adds indirection without centralizing the dependency version repository-wide or reducing risk.
4. try-fix-3 — Passed available checks, but the explicit Magick.NET.Core pin is redundant with the wrapper package dependency and increases dependency graph surface.
5. try-fix-2 — Failed restore and therefore must rank below all candidates that passed or remained viable.

Winner

Winner: pr

The raw PR fix is the best candidate because it directly addresses the dependency update with the least project-file churn, no failed regression signal, and no actionable expert-review findings. Passing try-fix candidates are not materially safer or more correct, and the failed try-fix candidate is not viable.

---

Future Action — review latest findings

No alternative fix was selected for this run. Review the session findings and CI results before merging.

[kubaflo]

Closing in favour of #35505