Added another case in the switch block to handle the object returned by Mono which differs from the object returned by CoreCLR

[Venkad000]

Fixes #115295 . Added another case in the switch block to handle the object returned by Mono which differs from the object returned by CoreCLR.

When Create(FieldInfo fieldInfo) is called upon an object of NullabilityInfoContext, the function creates a new object of NullabilityAttributeStateParser by calling CreateParser with a List<CustomAttributes> as argument to it. The NullabilityAttributeStateParser has a member private readonly object? _nullableAttributeArgument which is set to attribute.ConstructorArguments[0].Value in the CreateParser function. CoreCLR returns attribute.ConstructorArguments[0].Value as an object of type ReadOnlyCollection<CustomAttributeTypedArgument> whereas Mono returns it as byte[]. Therefore, the type which is reflected in the switch block is never matched against and hence gets the wrong NullabilityState set to it.

[Venkad000]

@uweigand @giritrivedi @saitama951

[AaronRobinsonMSFT]

@Venkad000 Can you also please add a test for this? Someplace under https://github.com/dotnet/runtime/tree/main/src/libraries/System.Runtime/tests/System.Reflection.Tests would make the most sense.

/cc @ericstj

[Venkad000]

wouldn't it be better to add it here https://github.com/dotnet/runtime/blob/main/src/libraries/System.Runtime/tests/System.Runtime.Tests/System/Reflection/NullabilityInfoContextTests.cs as this specifically deals with tests regarding NullabilityInfoContext?

[AaronRobinsonMSFT]

wouldn't it be better to add it here https://github.com/dotnet/runtime/blob/main/src/libraries/System.Runtime/tests/System.Runtime.Tests/System/Reflection/NullabilityInfoContextTests.cs as this specifically deals with tests regarding NullabilityInfoContext?

Oh yes, much better. Thanks!

[AaronRobinsonMSFT]

LGTM

[Venkad000]

@AaronRobinsonMSFT Can we merge this?

[AaronRobinsonMSFT]

/ba-g Unrelated timeout