Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 67.59%. Comparing base (819ecb7) to head (a23e583).
⚠️ Report is 305 commits behind head on feature/v3.

@@              Coverage Diff               @@
##           feature/v3    #2794      +/-   ##
==============================================
- Coverage       67.62%   67.59%   -0.03%     
==============================================
  Files             818      818              
  Lines           30745    30763      +18     
  Branches         5179     5180       +1     
==============================================
+ Hits            20789    20792       +3     
- Misses           8564     8583      +19     
+ Partials         1392     1388       -4     

Is my understanding correct that another solution would be just to change schemas for -logs and -traces manually?

Yes, there will not be a single source of truth but we would avoid all complicated changes here.

And where does schema for logs for example actually come from (tbh I don't know)? The way I see it we should just use that (whatever it is). We shouldn't on our side decide that "oh actually this should be the same as in RUM". Logs might know nothing about RUM backend.

Is my understanding correct that another solution would be just to change schemas for -logs and -traces manually?

Yes, there will not be a single source of truth but we would avoid all complicated changes here.

Well, not exactly. We need to have same schema as iOS has. So in case if we just manually adjust *.json files we still gonna have lot of changes in generated code and as well as in mapper logic. So the only difference will be actually in schema file. (And we could avoid adding clone plugin integration). But on the other hand every schema change will force us manually compare changes and align schema, which might lead to potential issues

And where does schema for logs for example actually come from (tbh I don't know)? The way I see it we should just use that (whatever it is). We shouldn't on our side decide that "oh actually this should be the same as in RUM". Logs might know nothing about RUM backend.

Ageed, and we don't actually use RUM schema for logs. We use RUM schema for device and os object. I think it's worth to fix the mess in schema repo and extract some common files which will have all common object, that we could rely on from different features. But such change gonna require to adjust browser and iOS SDKs which is too much for this task at the moment