maui

maui copied to clipboard

This is hopefully a straightforward change. Similar to https://github.com/dotnet/maui/pull/22092

See: CA1854

/azp run

Azure Pipelines successfully started running 3 pipeline(s).

Not compiling. Needs more changes in the ApplyPropertiesVisitor class.

Fixed.

@Eilon

/azp run

Azure Pipelines successfully started running 3 pipeline(s).

Looks like I should add the Tizen workload locally. Fixed those builds.

/azp run

Azure Pipelines successfully started running 3 pipeline(s).

Some more errors 😢

D:\a\_work\1\s\src\Core\src\Platform\Tizen\StackNavigationManager.cs(205,8): error CS0128: A local variable or function named 'naviPage' is already defined in this scope [D:\a\_work\1\s\src\Core\src\Core.csproj::TargetFramework=net8.0-tizen]
D:\a\_work\1\s\src\Core\src\Platform\Tizen\StackNavigationManager.cs(211,21): error CS8601: Possible null reference assignment. [D:\a\_work\1\s\src\Core\src\Core.csproj::TargetFramework=net8.0-tizen]
D:\a\_work\1\s\src\Core\src\Platform\Tizen\StackNavigationManager.cs(213,11): error CS8603: Possible null reference return. [D:\a\_work\1\s\src\Core\src\Core.csproj::TargetFramework=net8.0-tizen]

/azp run

Azure Pipelines successfully started running 3 pipeline(s).

Rerun failed tests?

Rerun failed tests?

Yup trying now.

This test seems to be failing consistently:

Expected string length 21 but was 12. Strings differ at index 0.
Expected: "parameterless factory"
But was:  "default ctor"
-----------^

Stack trace
   at Microsoft.Maui.Controls.Xaml.UnitTests.FactoryMethods.Tests.TestArgumentlessFactoryMethod(Boolean useCompiledXaml) in /Users/builder/azdo/_work/3/s/src/Controls/tests/Xaml.UnitTests/FactoryMethods.xaml.cs:line 109
   at InvokeStub_Tests.TestArgumentlessFactoryMethod(Object, Span`1)

1)    at Microsoft.Maui.Controls.Xaml.UnitTests.FactoryMethods.Tests.TestArgumentlessFactoryMethod(Boolean useCompiledXaml) in /Users/builder/azdo/_work/3/s/src/Controls/tests/Xaml.UnitTests/FactoryMethods.xaml.cs:line 109
   at InvokeStub_Tests.TestArgumentlessFactoryMethod(Object, Span`1)

So maybe there's some problem in the change itself?

Given that this PR does change code to do with finding factory methods, and that this particular test has failed multiple times in the PR, I think it warrants an investigation of the code in the PR. Maybe there's some subtle change in terms of how the code now locates matches for constructors somehow?