feat: add MapDerivedType for existing target type mapping

[TimothyMakkison]

feat: partially add MapDerivedType for existing target type mapping

Description

Partially add MapDerivedType for void mapping methods. Doesn't support object (runtimeTargetTypeMapping???) or generic mapping, I'll do it in a separate PR.

• Need to update docs, exception and xml comments.
• Haven't added tests.
• Not sure if my indetation is correct.
• User defined void mappings should be added in a different PR.

Example generated code

public partial void Map(global::A src, global::B trg)
{
    switch (src, trg)
    {
        case (global::AImpl1 source, global::BImpl1 target):
            target.BaseValue = source.BaseValue;
            target.Value1 = source.Value1;
            break;
        case (global::AImpl2 source, global::BImpl2 target):
            target.BaseValue = source.BaseValue;
            target.Value2 = source.Value2;
            break;
        default:
            throw new System.ArgumentException($"Cannot map {src.GetType()} to B as there is no known derived type mapping", nameof(src));
    }
}

Fixes #694

Checklist

• The existing code style is followed
• The commit message follows our guidelines
• Performed a self-review of my code
• Hard-to-understand areas of my code are commented
• The documentation is updated (as applicable)
• Unit tests are added/updated
• Integration tests are added/updated (as applicable, especially if feature/bug depends on roslyn or framework version in use)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (7628cf3) 91.29% compared to head (da6c1ff) 91.38%.
Report is 1 commits behind head on main.

❗ Current head da6c1ff differs from pull request most recent head ba720dc. Consider uploading reports for the commit ba720dc to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #918      +/-   ##
==========================================
+ Coverage   91.29%   91.38%   +0.08%     
==========================================
  Files         222      223       +1     
  Lines        7165     7239      +74     
  Branches      907      909       +2     
==========================================
+ Hits         6541     6615      +74     
  Misses        412      412              
  Partials      212      212              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[TimothyMakkison]

• Haven't been able to replicate all the tests from DerivedTypeTests as Mapperly should use user defined void mapping methods #626 is required.
• Considering mentioning this as a feature under derived-type-mapping
• Runtime target type mapping should be made into a separate issue.

I intentionally kept bool duplicatedSourceTypesAllowed to make it easier to add generic and runtime mapping (assuming the logic carries over).

[latonz]

Thank you for this PR! 😊 I added my feedback 😊

[TimothyMakkison]

Thanks for the review 👍 I think I've made your changes, really liked the simplification of DerivedTypeMappingBuilder.

Opted to not create SwitchDefaultSection although I can add it if you want.

[latonz]

Thank you for the updates, looks almost ready to merge, few smaller things I noted.
Also an integration test should be added and the documentation should be updated (as you already noted in the PR description ToDo-List 😊 )

[TimothyMakkison]

Can't (easily) do the integration tests rn because of a weird bug - can't find the correct file in the debug artifacts folder.

Which docs section should I update?

[latonz]

Can't (easily) do the integration tests rn because of a weird bug - can't find the correct file in the debug artifacts folder.

Can you elaborate on the bug? Probably run dotnet clean; dotnet build. Note: we recently switched to the simplified .NET 8.0 artifacts output: #917

Which docs section should I update?

I'd add a sentence in derived-type-mapping.md that derived type mappings are also supported for existing target object mappings with a reference to the existing target object mapping page.

[TimothyMakkison]

Can you elaborate on the bug? Probably run dotnet clean; dotnet build. Note: we recently switched to the simplified .NET 8.0 artifacts output: #917

Thanks dotnet clean; dotnet build fixed it for me. I've lost the error but it looks like Verify was trying to call a file for the wrong .NET version.

I'd add a sentence in derived-type-mapping.md that derived type mappings are also supported for existing target object mappings with a reference to the existing target object mapping page.

👍

[TimothyMakkison]

Looks like the benchmark check is failing. Can't run a benchmark right now, but I don't think this PR should have a massive impact on performance.

[latonz]

@TimothyMakkison not sure what we should do about the benchmark test, it's failing quite often recently without any notable performance issues.

[TimothyMakkison]

Strange 🤷 perhaps github pipelines are running slow? For now we could change the threshold percentage

[latonz]

@TimothyMakkison created #969 to increase the threshold.

[TimothyMakkison]

Thanks for the fixing the benchamark, no clue why the integration failed.

Hopefully that's the end of the CI issues 🤞 although code coverage is being weird.

[latonz]

@TimothyMakkison codecov has been super unreliable lately, I think some other open source projects replaced it recently. I created #930 a few days ago to look for alternatives and probably switch to another coverage provider or remove it completely. If you know a good solution, please add a comment there 😊

[github-actions]

🎉 This PR is included in version 3.3.0-next.6 🎉

The release is available on:

• GitHub release
• v3.3.0-next.6

Your semantic-release bot 📦🚀

[github-actions]

🎉 This PR is included in version 3.3.0 🎉

The release is available on:

• GitHub release
• v3.3.0

Your semantic-release bot 📦🚀