[chore] fix check-codeowners; add components for otelcol and confighttp

[axw]

Description

Add components for otelcol and confighttp. The former is for #13950, while the latter is needed to fix .chloggen/13783.yaml.

Link to tracking issue

N/A

Testing

N/A

Documentation

N/A

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.64%. Comparing base (0cdc78b) to head (721983d).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #13951      +/-   ##
==========================================
- Coverage   91.67%   91.64%   -0.04%     
==========================================
  Files         653      653              
  Lines       42554    42554              
==========================================
- Hits        39013    38997      -16     
- Misses       2733     2745      +12     
- Partials      808      812       +4     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dnaeon]

Ah, another tools leftover:

• https://github.com/open-telemetry/opentelemetry-collector/actions/runs/18366967010/job/52321557924?pr=13951

cc: @codeboten , @mx-psi : will submit a PR to get this one fixed asap.

EDIT:

Just noticed, it got fixed by @axw in 48100ee#diff-33e518bda3b8106fe0cb33b8ed49e65aa45b7c9ffd07efcb06e9bc00b575ab17R60

IMO, calling out to githubgen would be better done via a Makefile target.

[axw]

IMO, calling out to githubgen would be better done via a Makefile target.

Agreed. The existing Makefile target is slightly different (it passes -skipgithub), so I think would need to be parameterised?

[dnaeon]

Having a separate Makefile target for this one would be better, IMHO.

Or perhaps the existing generate-codeowners may be adapted to expect additional args (such as -skipgithub, etc.), so it can be re-used?

• opentelemetry-collector/Makefile

  Lines 469 to 471 in 0cdc78b

  	.PHONY: generate-codeowners
  	generate-codeowners:
  	$(GO_TOOL) githubgen --default-codeowner "open-telemetry/collector-approvers" -skipgithub codeowners

[axw]

Yeah I was being a little bit lazy. I'll add an args env var for the target and use that.

[axw]

721983d

[dnaeon]

Wouldn't it make more sense to verify whether we are running in CI or not, and then set opts for codeowner generation respectively?

For example, this would not need to change the existing workflow.

WDYT?

diff --git a/Makefile b/Makefile
index 3ce7d1c98..a0bc35971 100644
--- a/Makefile
+++ b/Makefile
@@ -335,11 +335,13 @@ checkdoc:
 apidiff-build:
        @$(foreach pkg,$(ALL_PKGS),$(call exec-command,./internal/buildscripts/gen-apidiff.sh -p $(pkg)))

-# If we are running in CI, change input directory
+# If we are running in CI, change input directory. Also configure opts for generating codeowners
 ifeq ($(CI), true)
 APICOMPARE_OPTS=$(COMPARE_OPTS)
+GENCODEOWNERS_OPTS=
 else
 APICOMPARE_OPTS=-d "./internal/data/apidiff"
+GENCODEOWNERS_OPTS=-skipgithub
 endif

 # Compare API state snapshots
@@ -468,7 +470,7 @@ generate-gh-issue-templates:

 .PHONY: generate-codeowners
 generate-codeowners:
-       $(GO_TOOL) githubgen --default-codeowner "open-telemetry/collector-approvers" -skipgithub codeowners
+       $(GO_TOOL) githubgen --default-codeowner "open-telemetry/collector-approvers" $(GENCODEOWNERS_OPTS) codeowners

 .PHONY: gengithub
 gengithub: generate-codeowners generate-gh-issue-templates

[axw]

I'm not so sure about that, for two reasons:

• we need to pass in -folder to make it look in the PR branch checkout
• "running in CI" isn't quite enough I think, or at least could be brittle: in this case we can check codeowners against the GitHub API because of the elevated access, so it's not just a matter of checking that we're in CI, but knowing that it's running with a GITHUB_TOKEN with additional access

[axw]

The confighttp changelog entry has been fixed in another PR, and the codeowners one is unrelated to adding an otelcol component. I'll separate that out.

[axw]

#13987