fix(search): resolve entity-specific aliases to canonical index names

[mohityadav766]

Closes #27761.

Problem

When the UI migrated from index=table_search_index to index=table, every search for tables started returning column docs too. ES alias expansion is bidirectional through the parent/child graph in indexMapping.json: column_search_index is created with table as one of its aliases (because tableColumn lists "table" in its parentAliases), so when ES resolves alias table it expands to both table_search_index AND column_search_index.

Same bleed exists for every alias whose entities also act as a parent for some other entity — testCase, testSuite, database, …

Fix

One method. SearchRepository.getIndexOrAliasName(String) now resolves entity-specific aliases to their canonical *_search_index names so we hand ES literal index names and bypass alias expansion entirely.

Input	Output	Notes
"table" (entity alias)	"<cluster>_table_search_index"	The fix. Entity-specific alias resolves to canonical, no children leak.
"dataAsset" / "all" (compound)	"<cluster>_dataAsset"	Passes through. ES expands natively. UI widgets that span the umbrella keep working.
"table_search_index" (canonical)	"<cluster>_table_search_index"	Legacy callers unchanged.
"<cluster>_table_search_index"	unchanged	Idempotent — internal code that hands a resolved value back doesn't double-prefix.
"table," / ","	empty tokens dropped	All-empty input preserved so ES surfaces "unknown index" instead of an empty-target failure.

That's the entire change. No new query params, no schema changes, no method signatures touched. Every existing caller of getIndexOrAliasName picks up the new behavior:

• /v1/search/query, /v1/search/export, /v1/search/preview, /v1/search/nlq/query — resource layer already pre-resolves via this method.
• /v1/search/aggregate (GET + POST), /v1/search/fieldQuery, /v1/search/entityTypeCounts — ES/OS managers already resolve via this method.

Why no flag mechanism

A previous draft of this fix (PR #27762, fix-cluster-aliasing branch) added fetchParentsAliases / fetchChildAliases flags so callers could opt into selective expansion. Audit of every caller showed nobody actually needs that — UI sites with SearchIndex.TABLE / TOPIC / etc. want only that type; UI sites that need mixed entity types use SearchIndex.ALL / SearchIndex.DATA_ASSET (compound aliases, unchanged by this fix); internal Java callers (RBAC, propagation, DataInsightSystemChartRepository) pass entity-specific aliases for entity-specific operations. Adding a flag for a use case nobody has just enlarges the API surface.

Test plan

• Unit tests in SearchRepositoryBehaviorTest pin: entity-alias → canonical resolution; compound-alias passthrough; idempotent already-prefixed token; comma-separated independent resolution; empty-token / all-empty handling; existing indexNameHelpersRespectClusterAlias test (canonical-name input) unchanged.
• Integration smoke against a running stack:

  curl '…/v1/search/query?q=*&index=table&size=20' -H "Authorization: Bearer $TOKEN" \
    | jq '[.hits.hits[]._source.entityType] | unique'
  # Expect: ["table"]   (no "column")
  curl '…/v1/search/query?q=*&index=dataAsset&size=20' -H "Authorization: Bearer $TOKEN" \
    | jq '[.hits.hits[]._source.entityType] | unique'
  # Expect: includes table, topic, dashboard, … (compound alias still works)

• Multi-tenant deployment with clusterAlias configured: ?index=table resolves to <tenant>_table_search_index exactly once (no double-prefix).

🤖 Generated with Claude Code

[gitar-bot]

💡 Edge Case: No test coverage for non-clustered (null/empty clusterAlias) path

All new tests in SearchRepositoryBehaviorTest use clusterAlias = "cluster". The fix changes behavior for non-clustered deployments too: previously getIndexOrAliasName("table") with no cluster alias returned "table" (passthrough), now it returns "table_search_index" (resolved canonical name). This is the core of the fix and arguably the more common deployment mode, but it has zero dedicated test coverage. If IndexMapping.getIndexName(null) or the prefix == null path in resolveSingleAliasToken ever regresses, no test will catch it.

Suggested fix:

@Test
void getIndexOrAliasNameResolvesAliasWithoutClusterAlias() {
  SearchRepository noPrefixRepo =
      newRepository(
          Map.of(Entity.TABLE, TABLE_MAPPING, Entity.DOMAIN, DOMAIN_MAPPING),
          null);
  assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table"));
  assertEquals("dataAsset", noPrefixRepo.getIndexOrAliasName("dataAsset"));
  assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table_search_index"));
}

_{Was this helpful? React with 👍 / 👎 | Reply gitar fix to apply this suggestion}

[gitar-bot]

Code Review 👍 Approved with suggestions 0 resolved / 1 findings

Resolves entity-specific aliases to canonical index names within the search repository. Add test coverage for the non-clustered (null or empty clusterAlias) execution path to ensure robust behavior.

💡 Edge Case: No test coverage for non-clustered (null/empty clusterAlias) path

📄 openmetadata-service/src/test/java/org/openmetadata/service/search/SearchRepositoryBehaviorTest.java:284-298

All new tests in SearchRepositoryBehaviorTest use clusterAlias = "cluster". The fix changes behavior for non-clustered deployments too: previously getIndexOrAliasName("table") with no cluster alias returned "table" (passthrough), now it returns "table_search_index" (resolved canonical name). This is the core of the fix and arguably the more common deployment mode, but it has zero dedicated test coverage. If IndexMapping.getIndexName(null) or the prefix == null path in resolveSingleAliasToken ever regresses, no test will catch it.

Suggested fix

@Test
void getIndexOrAliasNameResolvesAliasWithoutClusterAlias() {
  SearchRepository noPrefixRepo =
      newRepository(
          Map.of(Entity.TABLE, TABLE_MAPPING, Entity.DOMAIN, DOMAIN_MAPPING),
          null);
  assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table"));
  assertEquals("dataAsset", noPrefixRepo.getIndexOrAliasName("dataAsset"));
  assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table_search_index"));
}

🤖 Prompt for agents

Code Review: Resolves entity-specific aliases to canonical index names within the search repository. Add test coverage for the non-clustered (null or empty clusterAlias) execution path to ensure robust behavior.

1. 💡 Edge Case: No test coverage for non-clustered (null/empty clusterAlias) path
   Files: openmetadata-service/src/test/java/org/openmetadata/service/search/SearchRepositoryBehaviorTest.java:284-298

   All new tests in `SearchRepositoryBehaviorTest` use `clusterAlias = "cluster"`. The fix changes behavior for **non-clustered** deployments too: previously `getIndexOrAliasName("table")` with no cluster alias returned `"table"` (passthrough), now it returns `"table_search_index"` (resolved canonical name). This is the core of the fix and arguably the more common deployment mode, but it has zero dedicated test coverage. If `IndexMapping.getIndexName(null)` or the `prefix == null` path in `resolveSingleAliasToken` ever regresses, no test will catch it.

   Suggested fix:
   @Test
   void getIndexOrAliasNameResolvesAliasWithoutClusterAlias() {
     SearchRepository noPrefixRepo =
         newRepository(
             Map.of(Entity.TABLE, TABLE_MAPPING, Entity.DOMAIN, DOMAIN_MAPPING),
             null);
     assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table"));
     assertEquals("dataAsset", noPrefixRepo.getIndexOrAliasName("dataAsset"));
     assertEquals("table_search_index", noPrefixRepo.getIndexOrAliasName("table_search_index"));
   }

Options

Display: compact → Showing less information.

Comment with these commands to change:

Compact

gitar display:verbose

_{Was this helpful? React with 👍 / 👎 | Gitar}

[Copilot]

Pull request overview

This PR fixes search-result “bleed” caused by Elasticsearch/OpenSearch alias expansion by resolving entity-specific index aliases (e.g., table) to their canonical *_search_index names before issuing queries.

Changes:

• Updated SearchRepository.getIndexOrAliasName to resolve entity-specific aliases via entityIndexMap to canonical index names (while leaving compound aliases like all/dataAsset to expand natively).
• Added unit tests in SearchRepositoryBehaviorTest to cover entity-alias resolution, compound passthrough, idempotence, comma-separated handling, and empty-token behavior.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
openmetadata-service/src/main/java/org/openmetadata/service/search/SearchRepository.java	Implements entity-alias → canonical-index resolution and safer token parsing/prefixing logic.
openmetadata-service/src/test/java/org/openmetadata/service/search/SearchRepositoryBehaviorTest.java	Adds regression and behavior tests validating the new index/alias resolution rules.

[sonarqubecloud]

Quality Gate passed for 'open-metadata-ingestion'

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[github-actions]

🟡 Playwright Results — all passed (16 flaky)

✅ 3966 passed · ❌ 0 failed · 🟡 16 flaky · ⏭️ 86 skipped

Shard	Passed	Failed	Flaky	Skipped
🟡 Shard 1	298	0	1	4
🟡 Shard 2	742	0	3	8
🟡 Shard 3	752	0	3	7
🟡 Shard 4	757	0	2	18
🟡 Shard 5	685	0	2	41
🟡 Shard 6	732	0	5	8

🟡 16 flaky test(s) (passed on retry)

• Features/DataAssetRulesDisabled.spec.ts › Database Schema (shard 1, 1 retry)
• Features/ActivityAPI.spec.ts › Activity event is created when description is updated (shard 2, 1 retry)
• Features/ActivityAPI.spec.ts › Activity event shows the actor who made the change (shard 2, 1 retry)
• Features/DataProductRenameConsolidation.spec.ts › Rename then change owner - assets should be preserved (shard 2, 1 retry)
• Features/RTL.spec.ts › Verify Following widget functionality (shard 3, 1 retry)
• Flow/AddRoleAndAssignToUser.spec.ts › Verify assigned role to new user (shard 3, 1 retry)
• Flow/PersonaFlow.spec.ts › Set default persona for team should work properly (shard 3, 1 retry)
• Pages/DataContracts.spec.ts › Create Data Contract and validate for Chart (shard 4, 1 retry)
• Pages/DataContracts.spec.ts › Add and update Security and SLA tabs (shard 4, 1 retry)
• Pages/EntityDataConsumer.spec.ts › Tier Add, Update and Remove (shard 5, 1 retry)
• Pages/EntityDataConsumer.spec.ts › Tier Add, Update and Remove (shard 5, 1 retry)
• Features/AutoPilot.spec.ts › Agents created by AutoPilot should be deleted (shard 6, 1 retry)
• Pages/Lineage/DataAssetLineage.spec.ts › Column lineage for searchIndex -> searchIndex (shard 6, 1 retry)
• Pages/Lineage/LineageFilters.spec.ts › Verify lineage schema filter selection (shard 6, 1 retry)
• Pages/Lineage/LineageRightPanel.spec.ts › Verify custom properties tab IS visible for supported type: searchIndex (shard 6, 1 retry)
• Pages/Users.spec.ts › Check permissions for Data Steward (shard 6, 1 retry)

📦 Download artifacts

How to debug locally

# Download playwright-test-results-<shard> artifact and unzip
npx playwright show-trace path/to/trace.zip    # view trace