[repo] Move OpenTelemetry.SemanticConventions project from main repo

[joegoldman2]

Fixes #1657.

Changes

Reintroduce OpenTelemetry.SemanticConventions project that has been removed with open-telemetry/opentelemetry-dotnet#5539.
PR created as draft just to see if I'm going in the right direction and because the generation script is throwing an exception at the moment.

For significant contributions please make sure you have completed the following items:

• Appropriate CHANGELOG.md updated for non-trivial changes
• Design discussion issue #
• Changes in public API reviewed

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 72.55%. Comparing base (71655ce) to head (22f9809).
Report is 216 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1672      +/-   ##
==========================================
- Coverage   73.91%   72.55%   -1.37%     
==========================================
  Files         267      240      -27     
  Lines        9615     9261     -354     
==========================================
- Hits         7107     6719     -388     
- Misses       2508     2542      +34     

Flag	Coverage Δ
unittests-Solution	79.97% <ø> (?)

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

see 227 files with indirect coverage changes

[CodeBlanch]

@joegoldman2

Would you mind adding a CHANGELOG.md file into the project? Let's seed it with something like...

# Changelog

## Unreleased

* Updated to `1.25.0` release of OpenTelemetry Semantic Conventions.
  ([#1672](https://github.com/open-telemetry/opentelemetry-dotnet-contrib/pull/1672))

* Moved from `https://github.com/open-telemetry/opentelemetry-dotnet/` to
  `https://github.com/open-telemetry/opentelemetry-dotnet-contrib/`.

[Kielek]

Consider adding CI/infrqastructure code similar to 9fbfdf2

Also it will be great to have component owner under .github repository for this project.

[lmolkova]

Having all attributes in one place would be problematic in the long term: different stability, huge file, hard to use

I suggest to reuse new approach we've implemented in Java and Python - group attributes by the root namespace following the discussion in open-telemetry/semantic-conventions#551

It's being documented here https://github.com/open-telemetry/semantic-conventions/blob/18325605937cb0a15fde89faabf48c857eab0ca5/supplementary-guidelines/semantic_conventions_code_generation.md.

Here's how Java does it https://github.com/open-telemetry/semantic-conventions-java.

This is not a requirement, and from semconv perspective we want to give language SIGs a freedom to decide how they want to do it.

Happy to help with Jinja if needed.

[lmolkova]

Once .NET allows to set schema URL, having a single one would be problematic - different instrumentations can produce different schema versions.

[lmolkova]

you might want different template attributes (like http.request.header.<key>) to appear differently than regular ones

[CodeBlanch]

@Kielek I pushed changes here to tweak Component.BuildTest.yml so that it skips code coverage upload if there isn't any data available.

Also there was a typo in codecov.yml which is actually breaking all code coverage upload currently 😬 Opened: #1688

[CodeBlanch]

LGTM

[lmolkova]

nit: it should be possible to get rid of print_value macro and use print_member_value filter provided by the otel/semconvgen

Suggested change

	public const {{ type }} {{ member.member_id | to_camelcase(True) }} = {{ print_value(type, member.value) }};
	public const {{ type }} {{ member.member_id | to_camelcase(True) }} = {{ print_member_value(member) }};

[lmolkova]

Thank you!

[joaopgrassi]

Hi all, and @joegoldman2

I'm following up on this as a semconv maintainer and helping migrate to the new tooling (Weaver). I notice that
the script says semconv 1.25, but I see attributes that are waaaaay old, like the ones in AndroidAttributes and I'm a bit confused.

Was this PR really updated with semconv version 1.25? Or was it just migrated as is from the main repo? (with spec version 1.13)

[ben-wilson-mews]

@joegoldman2 I notice that this package isn't published to nuget, any timeline on it being published? The instructions I followed in the readme for the package.

For anyone wondering, the team are working on a polygot generator here. The aim being to republish this soon but with best practices of versioning them to provide a stable interface

[joaopgrassi]

@ben-wilson-mews so you are already working on migrating this to weaver? I also started working on it #1944. How far long are you? I can maybe give what I have and you can continue if you want. Let me know.