How to understand that the library works and all logger calls use the source generator?

[odysseus1973]

Hi!

How to understand that the library works and all logger calls use the source generator?

Thanks.

[stbychkov]

Hey, there are a few ways to check if the source generator is doing its thing:

1. Obviously debugging is one of the options. Step into the Log* method and see where it goes. If it goes through a LoggerMessage class, then the generator's working.
2. Check generated sources. In Rider, you can view generated code right in the IDE. If you see a method named after your class and line number, you're good. If you're not using Rider, you can still find the generated files in the obj folder. If they’re not showing up, try adding this to your .csproj:

<PropertyGroup>
    <EmitCompilerGeneratedFiles>true</EmitCompilerGeneratedFiles>
    <CompilerGeneratedFilesOutputPath>Generated</CompilerGeneratedFilesOutputPath>
</PropertyGroup>

3. Some IDEs (like Rider) used to show interceptors right in the editor if they were based on file path + line + column. Since switching to hash-based ones, that doesn’t show up, but I’m guessing they'll bring it back in a future update (if they haven’t already).

The main rule of thumb is that if the generator can’t apply this approach to your log statement, you’ll usually run into one of two things:

1. Build error. It's quite unlikely, but still can happen in some edge cases the generator doesn’t support yet.
2. Fallback to default. If it can’t generate the interceptor, it just falls back to the regular Logger.Log* call. That’s by design as the generator was built to be implicit and not break anything if it can’t do its job.

Let me know if you hit any issues.