Return merged Resource on schema conflict

[MrAlias]

Resolve #2341

Why not just log?

By returning a partial result (the merged conflicting resource), the user is able to decide what they want to do. If we just logged this as a warning, the user may unknowingly use a Resource they would rather have dropped or fixed.

Specification compliance

The specification states:

The resulting resource will have the Schema URL calculated as follows:

• If the old resource's Schema URL is empty then the resulting resource's Schema URL will be set to the Schema URL of the updating resource,
• Else if the updating resource's Schema URL is empty then the resulting resource's Schema URL will be set to the Schema URL of the old resource,
• Else if the Schema URLs of the old and updating resources are the same then that will be the Schema URL of the resulting resource,
• Else this is a merging error (this is the case when the Schema URL of the old and updating resources are not empty and are different). The resulting resource is undefined, and its contents are implementation-specific.

Based on this language we are given a fair amount of freedom to handle this conflict in the way we see fit. Looking at other language implementations, Go is the only one that will actually return an error and does not make the merge:

Language	Behavior
Java	logs conflict and returns merged
Javascript	resources do not have schema URL
Python	logs conflict and returns unmerged
Rust	ignore conflict
Ruby	resources do not have schema URL
PHP	logs conflict and returns merged
.NET	resources do not have a schema URL
C++	ignored conflict

Given the specification also states:

The resulting resource MUST have all attributes that are on any of the two input resources. If a key exists on both the old and updating resource, the value of the updating resource MUST be picked (even if the updated value is empty).

Trying to resolve this conflict by actually using the schemas to make a transformation is not compliant. Additionally, given the reluctance to stabilize these transformations, it does not seem prudent to not provide a better user story here.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (f4e1e04) 82.6% compared to head (cd62d36) 82.6%.

❗ Current head cd62d36 differs from pull request most recent head 2f74074. Consider uploading reports for the commit 2f74074 to get more accurate results

Additional details and impacted files

@@          Coverage Diff          @@
##            main   #4876   +/-   ##
=====================================
  Coverage   82.6%   82.6%           
=====================================
  Files        232     232           
  Lines      18862   18868    +6     
=====================================
+ Hits       15595   15601    +6     
  Misses      2977    2977           
  Partials     290     290           

Files	Coverage Δ
sdk/resource/auto.go	91.6% <100.0%> (+0.9%)	⬆️
sdk/resource/resource.go	76.6% <100.0%> (+0.1%)	⬆️

[pellared]

LGTM.

#4876 (comment) can be addressed as a follow-up. I can work on it on Tuesday.

[pellared]

I think that detect should be changed to handle conflicts better.

#4876 (comment)

[jufemaiz]

Big step forward! Thanks for the work here!