Add support for the [Length]attribute.

[satma0745]

Resolves #2756

[martincostello]

Should these be called *WithoutLength?

[satma0745]

I think it should be called *WithLength (as it is right now).
Reason: right above/below these new properties You can see a lot of other properties named in the same manner (and they also don't have attributes), e.g. ArrayWithMinMaxLength property.

I think, this type represents some kind of model schema/structure/skeleton and the following MetadataType type is it's validation rule-set. This reasoning is backed up by:

1. [ModelMetadataType] attribute usage
2. File name (it ends on *ViaMeradataType).

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 25.00000% with 6 lines in your changes are missing coverage. Please review.

Project coverage is 89.87%. Comparing base (a29ba36) to head (9ed91a5).

Files	Patch %	Lines
...ggerGen/SchemaGenerator/OpenApiSchemaExtensions.cs	25.00%	6 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2882      +/-   ##
==========================================
- Coverage   90.06%   89.87%   -0.20%     
==========================================
  Files          72       72              
  Lines        2698     2706       +8     
  Branches      426      428       +2     
==========================================
+ Hits         2430     2432       +2     
- Misses        268      274       +6     

Flag	Coverage Δ
Linux	89.87% <25.00%> (-0.20%)	⬇️
Windows	89.87% <25.00%> (-0.20%)	⬇️
macOS	89.87% <25.00%> (-0.20%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[satma0745]

What else can be covered?

[martincostello]

Codecov seems to think that the majority of your changes aren't actually covered by the tests you added:

[cremor]

How are the builds deciding which test run (so which target framework) is taken for the coverage report? Maybe they are taking the coverage from a non .NET 8 run?

[martincostello]

How are the builds deciding which test run (so which target framework) is taken for the coverage report? Maybe they are taking the coverage from a non .NET 8 run?

Good question - I'm not sure, but they should be being combined together.

[cremor]

If you download the coverage artifact from the build you can see that something is merged. But I'd expect at least 3 coverage files (for the 3 target frameworks used in tests), and it only shows 2:

And according to that report the new method isn't covered at all, which supports my suspected cause:

It's also weird that it shows each method twice in the method list:

[martincostello]

I'm happy that it looks like it's covered/tested.

The yes/no coverage report is likely related to #2836 and the fact that the number of projects and TFMs we have causes lots of file write conflicts.