SOLR-16396: Align v2 full-config section keys with pluralised forms by jaykay12 · Pull Request #5175 · apache/solr

jaykay12 · 2026-10-16T22:51:20Z

https://issues.apache.org/jira/browse/SOLR-16396

Description

Align v2 full-config section keys with pluralised forms

Solution

Self Detected Bug that got introduced here : SOLR-16396 | Migrate GET config sections query, searchComponent & queryResponseWriter to JAX-RS #5125

Since, this is an ongoing initiative, imo no CL is required here.

Tests

Please describe the tests you've developed or run to confirm this patch implements the feature or solves the problem.

Checklist

Please review the following and check all that apply:

I have reviewed the guidelines for How to Contribute and my code conforms to the standards described there to the best of my ability.
I have created a Jira issue and added the issue ID to my pull request title.
I have given Solr maintainers access to contribute to my PR branch. (optional but recommended, not available for branches on forks living under an organisation)
I have developed this patch against the main branch.
I have run ./gradlew check.
I have added tests for my changes.
I have added documentation for the Reference Guide
I have added a changelog entry for my change

jaykay12 · 2026-10-16T22:52:10Z

/no-changelog

jaykay12 · 2026-10-16T23:02:32Z

Testing Evidences

Scenario	Main	Feature	Veridict
V2 full config			Fixed now ✅
-	-	-	-
V2 filtered config			Was already correct 👍🏽

jaykay12

@epugh please help in triggering workflows for this fix.

epugh · 2026-10-17T14:36:22Z

Agreed no changelog needed

epugh

This makes sense.... however do we need to document in the ref guide the two different out puts? Between v1 and v2?

The V1/V2 full-config output comparison only called out the children-key difference; add the searchComponent/queryResponseWriter pluralization this PR introduces, shown the same way as the existing children-key example (separate V1/V2 JSON output per tab). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

The new V1/V2 full-config diff duplicated the elevator/xslt example and explanation already given a few dozen lines down, where the per-section GET endpoints are documented. Drop the redundant JSON snippets and keep a single forward-pointing sentence instead. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

reqhandlertests() is shared between v1 and v2 harness runs (toggled randomly in @before), and asserted the singular "searchComponent" key against the bare /config endpoint regardless of API version. Since GetConfig#getConfig now pluralizes that key under v2, the test failed whenever the random toggle picked v2 (reproduced with CI's exact seed 177EED04838EDD3C). Pick the expected key based on the harness's URL. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

jaykay12 · 2026-10-17T20:06:07Z

This makes sense.... however do we need to document in the ref guide the two different out puts? Between v1 and v2?

thanks @epugh for doing ref guide updates this time. I will try keeping them in sync from next PRs.

can we retrigger the workflows & verify once.

# Conflicts: # solr/core/src/java/org/apache/solr/handler/admin/api/GetConfig.java

epugh · 2026-10-20T12:43:30Z

Merged `upstream/main` in and resolved the one conflict (in `GetConfig.java` — `getConfig`'s pluralization logic needed to call the now-1-arg `buildConfigMap(expandParams)` signature from #5168, instead of the old 2-arg one). CI re-triggered and approved — running now.

Ran the full relevant local test suite post-merge (48 tests across `GetConfigTest`, `GetConfigSolrCloudTest`, `V2ConfigAPIMappingTest`, `TestSolrConfigHandler`, `TestSolrConfigHandlerCloud`, `PackageToolTest`, `PackageToolPicocliTest`, package-manager tests) — all green, spotless clean.

The full /config dump already pluralized searchComponent/queryResponseWriter for v2 clients, but left requestHandler singular even though the dedicated GET /config/requestHandlers endpoint (and its response key) has used the plural form since SOLR-16396's requestHandler migration. Add requestHandler -> requestHandlers to the same external/internal name map so the full-config remapping picks it up consistently. v1's GET /config is unaffected (explicitly opts out of the renaming via useExternalComponentNames=false). Also fixes two test assertions (TestSolrConfigHandler.testProperty and reqhandlertests) that checked the full "/config" response for a hardcoded singular "requestHandler" key with no v1/v2 branching - these would have started failing intermittently against the V2 harness once this change landed, the same race the existing searchComponentKey branching was already guarding against. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

epugh · 2026-10-20T13:09:05Z

Also pluralized `requestHandler` -> `requestHandlers` in the full v2 `/config` response, for consistency with the dedicated `GET /config/requestHandlers` endpoint (which has used the plural form since the requestHandler migration) and with `searchComponent`/`queryResponseWriter` already being pluralized here. v1's `/config` is unaffected.

Also caught and fixed two test assertions (`TestSolrConfigHandler.testProperty` and `reqhandlertests`) that checked the full `/config` response for a hardcoded singular `"requestHandler"` key without v1/v2 branching — these would have started failing intermittently against the randomized V2 harness runs once this landed, the same class of issue the existing `searchComponentKey` branching already guarded against. Reproduced the failure first (3-4/4 runs failing), then fixed and reran clean (4/4, 6/6 across separate runs).

Updated the ref guide to describe this as a third V1/V2 difference in the full-config output, alongside the existing `children`-key and `searchComponent`/`queryResponseWriter` notes.

CI re-approved and running.

jaykay12

final local testing done with most latest changes, lgtm ✅

/simplify findings applied: - GetConfig.getConfig's boolean useExternalComponentNames parameter encoded "which API version is calling" as an unlabeled boolean, with the only caller (SolrConfigHandler's v1 path) reading as a bare "getConfig(expandParams, false)". Replaced the boolean overload with two distinctly-named methods: getConfig (v2, always pluralizes) and getConfigWithInternalNames (v1, unchanged names) - self-documenting at the one call site that needed it. - EXTERNAL_TO_INTERNAL_COMPONENT_NAME was doing double duty: it drives the GET /config/{component} path-routing translation (only searchComponent/queryResponseWriter ever route through there) and, separately, the full /config dump's top-level key renaming (which also needs requestHandler). Adding requestHandler to the shared map made it look reachable via the path-routing lookup even though requestHandler has its own dedicated endpoint and never goes through getConfigComponent(ByName) - a future path-regex change "completing" the pattern based on this map would have silently lost implicit-handler merging and useParams expansion. Split into two maps scoped to their actual callers. - Factored the duplicated "pick internal vs external key name based on whether this test harness is V1 or V2" ternary (independently written out in both testProperty and reqhandlertests) into one shared fullConfigKey helper in TestSolrConfigHandler. No behavior change - verified via repeated runs of TestSolrConfigHandler (whose @before randomly toggles V1/V2 per run) and the rest of the GetConfig/SolrConfigHandler test suite. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

…ntries The previous comment claimed this map was "a superset of EXTERNAL_TO_INTERNAL_COMPONENT_NAME", but it was actually just a second, independently-written Map.of() literal with the same two entries copy-pasted in. Nothing enforced that relationship - a future edit to one map without the other would silently make the comment false, with no compiler or test catching the drift. Build the full-config map from the component-routing map instead, adding only the one extra entry (requestHandler) that's unique to it. The "superset" claim is now true by construction. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

jaykay12 flagged the comments on EXTERNAL_TO_INTERNAL_COMPONENT_NAME and EXTERNAL_TO_INTERNAL_FULL_CONFIG_NAME as too much code commenting. Cut both down to the one non-obvious fact each carries (requestHandler is intentionally excluded from the routing map; the full-config map is derived, not duplicated), dropping restatement of what the code/identifiers already make clear. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RzzUgiebQ8ZhhgE2E7ZZU5

…5175) Co-authored-by: Eric Pugh <epugh@opensourceconnections.com> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> (cherry picked from commit d5a1701)

jaykay12 added 2 commits October 17, 2026 04:08

minor bug fix

1fdae89

tidy

3379188

github-actions Bot added the tests label Oct 16, 2026

github-actions Bot added the no-changelog label Oct 16, 2026

jaykay12 commented Oct 16, 2026

View reviewed changes

Merge branch 'main' into SOLR-16396-plural-config-all-fix

219e0cf

epugh reviewed Oct 17, 2026

View reviewed changes

github-actions Bot added the documentation Improvements or additions to documentation label Oct 17, 2026

epugh and others added 2 commits October 17, 2026 11:10

jaykay12 requested a review from epugh October 17, 2026 20:06

jaykay12 and others added 2 commits October 18, 2026 12:43

Merge branch 'main' into SOLR-16396-plural-config-all-fix

f53e8b1

Merge remote-tracking branch 'upstream/main' into pr/5175

7be4646

# Conflicts: # solr/core/src/java/org/apache/solr/handler/admin/api/GetConfig.java

Merge branch 'main' into SOLR-16396-plural-config-all-fix

f7ae782

jaykay12 commented Oct 20, 2026

View reviewed changes

Comment thread solr/core/src/java/org/apache/solr/handler/admin/api/GetConfig.java Outdated

epugh and others added 2 commits October 20, 2026 15:12

epugh added this to the 10.x milestone Oct 20, 2026

epugh merged commit d5a1701 into apache:main Oct 20, 2026
6 checks passed

jaykay12 deleted the SOLR-16396-plural-config-all-fix branch October 20, 2026 20:44

Navigation Menu

Repository navigation

SOLR-16396: Align v2 full-config section keys with pluralised forms - #5175

SOLR-16396: Align v2 full-config section keys with pluralised forms#5175
epugh merged 13 commits into
apache:mainfrom
jaykay12:SOLR-16396-plural-config-all-fix

jaykay12 commented Oct 16, 2026 •

edited

Loading

Uh oh!

jaykay12 commented Oct 16, 2026

Uh oh!

jaykay12 commented Oct 16, 2026

Uh oh!

jaykay12 left a comment •

edited

Loading

Uh oh!

epugh commented Oct 17, 2026

Uh oh!

epugh left a comment

Uh oh!

jaykay12 commented Oct 17, 2026

Uh oh!

epugh commented Oct 20, 2026

Uh oh!

epugh commented Oct 20, 2026

Uh oh!

jaykay12 left a comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

jaykay12 commented Oct 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Solution

Tests

Checklist

Uh oh!

jaykay12 commented Oct 16, 2026

Uh oh!

jaykay12 commented Oct 16, 2026

Testing Evidences

Uh oh!

jaykay12 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

epugh commented Oct 17, 2026

Uh oh!

epugh left a comment

Choose a reason for hiding this comment

Uh oh!

jaykay12 commented Oct 17, 2026

Uh oh!

epugh commented Oct 20, 2026

Uh oh!

epugh commented Oct 20, 2026

Uh oh!

jaykay12 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

jaykay12 commented Oct 16, 2026 •

edited

Loading

jaykay12 left a comment •

edited

Loading