fix(search): backport immense-term children mapping fix to 1.12.10#28572
Merged
Conversation
Backport of the flattened-children reindex fix from main (PR #28509). The recursive column/schema `children` field was mapped `flattened`, which caused Lucene's 32,766-byte per-term limit to fail document indexing for entities with large nested column expressions. Remap `children` from `flattened` to `object/enabled:false` across all affected indexes (table, container, dashboardDataModel, worksheet, topic, apiEndpoint, searchEntity) so the subtree is stored but never indexed. Also: - Replace dead `columns.children.name` / `dataModel.columns.children.name` search field references with `columnNamesFuzzy` in the 4 Java index classes - Remove stale children-field entries from the searchSettings.json seed - Add v11210 migration to scrub the same stale entries from DB-stored SearchSettings on upgrade (idempotent)
github-actions
Bot
added
backend
safe to test
Contributor
|
The Python checkstyle failed. Please run You can install the pre-commit hooks with |
Contributor
🟡 Playwright Results — all passed (31 flaky)✅ 3459 passed · ❌ 0 failed · 🟡 31 flaky · ⏭️ 97 skipped
🟡 31 flaky test(s) (passed on retry)
How to debug locally# Download playwright-test-results-<shard> artifact and unzip
npx playwright show-trace path/to/trace.zip # view trace |
…x files The backport was missing the om_analyzer declaration for columnNamesFuzzy in container, table, and dashboard_data_model across all 4 locales (en/jp/ru/zh). Without it the field falls back to dynamic text mapping with the default analyzer instead of om_analyzer, giving weaker tokenization than main. Adds the same 4-line block that the main PR (#28509) included.
Contributor
|
The Python checkstyle failed. Please run You can install the pre-commit hooks with |
…snapshot SearchSettings.spec.ts 'Restore default search settings' compares against the hardcoded snapshot in searchSettingUtils.ts. The field was removed from searchSettings.json seed but the snapshot wasn't updated, causing the test to fail (expected columns.children.name, received nothing).
Contributor
❌ UI Checkstyle Failed❌ Playwright — ESLint + Prettier + Organise ImportsOne or more Playwright test files have linting or formatting issues. ❌ Core Components — ESLint + PrettierOne or more core-component files have linting or formatting issues. Fix locally (fast — only checks files changed in this branch): make ui-checkstyle-changed |
Contributor
|
The Python checkstyle failed. Please run You can install the pre-commit hooks with |
…backport Backport of SearchIndexNestedColumns.spec.ts from main PR #28509. Verifies that a 25-level-deep column with a >32 KB leaf expression indexes without an immense_term failure and is searchable by its deep column name via columnNamesFuzzy.
Contributor
|
The Python checkstyle failed. Please run You can install the pre-commit hooks with |
ShaileshParmar11
approved these changes
Jun 1, 2026
Code Review ✅ ApprovedBackports the immense-term children mapping fix to 1.12.10 by transitioning field types to object/enabled:false and updating search mappings. No issues found. OptionsDisplay: compact → Showing less information. Comment with these commands to change:
Was this helpful? React with 👍 / 👎 | Gitar |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Backport of main PR #28509 to the 1.12.10 release branch.
childrenfromflattened→object/enabled:falseacross all 7 affected index mappings (4 locales × 7 indexes). Theflattenedtype indexed every nested leaf as a keyword term; a leaf longer than Lucene's 32,766-byte limit (e.g. a large Looker/DAX expression) failed the whole document's reindex.object/enabled:falsestores the subtree in_sourcefor display but never indexes it.columns.children.name/dataModel.columns.children.namesearch field withcolumnNamesFuzzyinTableIndex,ContainerIndex,DashboardDataModelIndex,WorksheetIndex.*.childrenfield entries fromsearchSettings.json.SearchSettingson upgrade. Idempotent —removeIfon an already-clean list is a no-op, so clusters that later upgrade to 1.13.0 run the scrub again in v1130 safely.SearchIndexNestedColumns.spec.ts): Creates a 25-level-deep column with a >32 KB leaf expression, verifies the table indexes without animmense_termfailure, and confirms the deep column name is searchable viacolumnNamesFuzzy.What's NOT included
SearchIndexImmenseTermIT,FlattenedChildrenHighlightSearchIT)Test plan
immense_termfailure on a table/container with deeply nested columnscolumns.children.name/dataModel.columns.children.nameare absent from search settings after upgrade migration runsSearchIndexNestedColumns.spec.ts— 25-level oversized nested column indexes and is searchable by deep column name